From groff-admin  Sun Oct  1 18:10:13 2000
Received: from aua2.aua.auc.dk (aua2.aua.auc.dk [130.225.54.5])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA10091
	for <groff@ffii.org>; Sun, 1 Oct 2000 18:10:13 +0200
Received: from nightowl (nightowl.abk.auc.dk [192.168.80.38])
 by aua.auc.dk (PMDF V5.2-32 #30369)
 with SMTP id <01JUTYWUCC988Y5F2R@aua.auc.dk> for groff@ffii.org; Sun,
 1 Oct 2000 18:13:01 MET-DST
Date: Sun, 01 Oct 2000 18:13:11 +0200
From: "Frank G. Hahn" <efgeho@cs.auc.dk>
To: groff@ffii.org
Message-id: <003b01c02bc2$7e9ad520$2650a8c0@abk.auc.dk>
MIME-version: 1.0
X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
Content-type: multipart/alternative;
	boundary="----=_NextPart_000_0038_01C02BD3.4212DC40"
X-Priority: 3
X-MSMail-priority: Normal
Subject: [Groff] Arrow between ellipses - How to chop (remove overlap) ?
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1909
Lines: 62

This is a multi-part message in MIME format.

------=_NextPart_000_0038_01C02BD3.4212DC40
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi

I have quite some difficulties getting this right.

.PS
E1: ellipse "$DoDo$" at (0,0)
E2: ellipse "$NoNo$" at (5, 5)
arrow from E1 to E2 chop
.PE

Now my problem can you define chop (or synonym) so that the overlap of =
the arrows over the ellipses does not occur ?

/Frank



------=_NextPart_000_0038_01C02BD3.4212DC40
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 5.50.4207.2601" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Hi</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I have quite some difficulties getting =
this=20
right.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>.PS</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>E1: ellipse "$DoDo$" at =
(0,0)</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>E2: ellipse "$NoNo$" at (5, =
5)</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>arrow from E1 to E2 chop</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>.PE</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Now my problem can you define chop (or =
synonym) so=20
that the overlap of the arrows over the ellipses does not occur =
?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>/Frank</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV></BODY></HTML>

------=_NextPart_000_0038_01C02BD3.4212DC40--


From groff-admin  Sun Oct  1 18:49:59 2000
Received: from mail1.svr.pol.co.uk (mail1.svr.pol.co.uk [195.92.193.18])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA10251
	for <groff@ffii.org>; Sun, 1 Oct 2000 18:49:59 +0200
Received: from [195.92.198.123] (helo=mail17.svr.pol.co.uk)
	by mail1.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13fmM3-00028E-00
	for groff@ffii.org; Sun, 01 Oct 2000 17:52:47 +0100
Received: from modem-111.anadune.dialup.pol.co.uk ([62.136.106.111] helo=inputplus.demon.co.uk)
	by mail17.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13fmM2-00080u-00
	for groff@ffii.org; Sun, 01 Oct 2000 17:52:47 +0100
Received: from inputplus (ralph@localhost [127.0.0.1])
	by inputplus.demon.co.uk (8.8.7/8.8.7) with ESMTP id RAA12178;
	Sun, 1 Oct 2000 17:52:43 +0100
Message-Id: <200010011652.RAA12178@inputplus.demon.co.uk>
To: "Frank G. Hahn" <efgeho@cs.auc.dk>
cc: groff@ffii.org
Subject: Re: [Groff] Arrow between ellipses - How to chop (remove overlap) ? 
In-Reply-To: Message from "Frank G. Hahn" <efgeho@cs.auc.dk> 
   of "Sun, 01 Oct 2000 18:13:11 +0200." <003b01c02bc2$7e9ad520$2650a8c0@abk.auc.dk> 
Date: Sun, 01 Oct 2000 17:52:42 +0100
From: Ralph Corderoy <ralph@inputplus.demon.co.uk>
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 548
Lines: 24

Hi,

> I have quite some difficulties getting this right.
> 
> .PS
> E1: ellipse "$DoDo$" at (0,0)
> E2: ellipse "$NoNo$" at (5, 5)
> arrow from E1 to E2 chop
> .PE
> 
> Now my problem can you define chop (or synonym) so that the overlap
> of the arrows over the ellipses does not occur ?

Try

    arrow from E1 to E2 chop 0.3

and see also the definition of chop around page 10 of Kernighan's _PIC
-- A Graphics Language for Typesetting_ mentioned at the end of groff's
pic man page.  <URL:http://cm.bell-labs.com/cm/cs/cstr/116.ps.gz>


Ralph.


From groff-admin  Sun Oct  1 20:51:29 2000
Received: from aua2.aua.auc.dk (aua2.aua.auc.dk [130.225.54.5])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id UAA10911
	for <groff@ffii.org>; Sun, 1 Oct 2000 20:51:29 +0200
Received: from nightowl (nightowl.abk.auc.dk [192.168.80.38])
 by aua.auc.dk (PMDF V5.2-32 #30369)
 with SMTP id <01JUU4JSCH8Y8Y5G8S@aua.auc.dk> for groff@ffii.org; Sun,
 1 Oct 2000 20:54:18 MET-DST
Date: Sun, 01 Oct 2000 20:54:27 +0200
From: "Frank G. Hahn" <efgeho@cs.auc.dk>
Subject: Re: [Groff] Arrow between ellipses - How to chop (remove overlap) ?
To: groff@ffii.org
Message-id: <000f01c02bd9$05f4c9c0$2650a8c0@abk.auc.dk>
MIME-version: 1.0
X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
Content-type: text/plain;	charset="iso-8859-1"
Content-transfer-encoding: 7bit
X-Priority: 3
X-MSMail-priority: Normal
References: <200010011652.RAA12178@inputplus.demon.co.uk>
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1441
Lines: 49

----- Original Message -----
From: "Ralph Corderoy" <ralph@inputplus.demon.co.uk>
To: "Frank G. Hahn" <efgeho@cs.auc.dk>
Cc: <groff@ffii.org>
Sent: Sunday, October 01, 2000 6:52 PM
Subject: Re: [Groff] Arrow between ellipses - How to chop (remove overlap) ?


>
> Hi,
>
> > I have quite some difficulties getting this right.
> >
> > .PS
> > E1: ellipse "$DoDo$" at (0,0)
> > E2: ellipse "$NoNo$" at (5, 5)
> > arrow from E1 to E2 chop
> > .PE
> >
> > Now my problem can you define chop (or synonym) so that the overlap
> > of the arrows over the ellipses does not occur ?
>
> Try
>
>     arrow from E1 to E2 chop 0.3

This would have been nice .. but I need something more generic than that.
I got a program reading a hugin (Bayesian Network) file and generating the
output in PIC-format.
Which results in "many" ellipses scattered about on a page. Some of these
are connected (using arrows).
How do I get pic to chop the arrow correctly in BOTH ends.

I can't chop them using a constant, because I don't really know how the
ellipses will be placed relative to one another.

/Frank

> and see also the definition of chop around page 10 of Kernighan's _PIC
> -- A Graphics Language for Typesetting_ mentioned at the end of groff's
> pic man page.  <URL:http://cm.bell-labs.com/cm/cs/cstr/116.ps.gz>
>
>
> Ralph.
>
> _______________________________________________
> Groff maillist  -  Groff@ffii.org
> http://ffii.org/mailman/listinfo/groff


From groff-admin  Sun Oct  1 23:09:55 2000
Received: from cmailg1.svr.pol.co.uk (cmailg1.svr.pol.co.uk [195.92.195.171])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id XAA11588
	for <groff@ffii.org>; Sun, 1 Oct 2000 23:09:54 +0200
Received: from [195.92.198.123] (helo=mail17.svr.pol.co.uk)
	by cmailg1.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13fqPc-0000QI-00
	for groff@ffii.org; Sun, 01 Oct 2000 22:12:44 +0100
Received: from modem-57.aratan.dialup.pol.co.uk ([62.136.120.185] helo=inputplus.demon.co.uk)
	by mail17.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13fqPR-0003HQ-00
	for groff@ffii.org; Sun, 01 Oct 2000 22:12:34 +0100
Received: from inputplus (ralph@localhost [127.0.0.1])
	by inputplus.demon.co.uk (8.8.7/8.8.7) with ESMTP id WAA23396;
	Sun, 1 Oct 2000 22:12:30 +0100
Message-Id: <200010012112.WAA23396@inputplus.demon.co.uk>
To: "Frank G. Hahn" <efgeho@cs.auc.dk>
cc: groff@ffii.org
Subject: Re: [Groff] Arrow between ellipses - How to chop (remove overlap) ? 
In-Reply-To: Message from "Frank G. Hahn" <efgeho@cs.auc.dk> 
   of "Sun, 01 Oct 2000 20:54:27 +0200." <000f01c02bd9$05f4c9c0$2650a8c0@abk.auc.dk> 
Date: Sun, 01 Oct 2000 22:12:29 +0100
From: Ralph Corderoy <ralph@inputplus.demon.co.uk>
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 960
Lines: 29

Hi,

> > Try
> >
> >     arrow from E1 to E2 chop 0.3
> 
> This would have been nice .. but I need something more generic than
> that.  I got a program reading a hugin (Bayesian Network) file and
> generating the output in PIC-format.  Which results in "many"
> ellipses scattered about on a page. Some of these are connected
> (using arrows).  How do I get pic to chop the arrow correctly in BOTH
> ends.
> 
> I can't chop them using a constant, because I don't really know how
> the ellipses will be placed relative to one another.

If you're leaving the placement of the nodes up to pic then pic knows.

> > and see also the definition of chop around page 10 of Kernighan's
> > _PIC -- A Graphics Language for Typesetting_ mentioned at the end
> > of groff's pic man page.
> > <URL:http://cm.bell-labs.com/cm/cs/cstr/116.ps.gz>

Did you read this fine document?  It explains that pic's trig functions
should be used.  Will that work in your case?


Ralph.


From groff-admin  Mon Oct  2 15:30:47 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA32234
	for <groff@ffii.org>; Mon, 2 Oct 2000 15:30:47 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Mon, 2 Oct 2000 15:33:39 +0200
Received: from rigel.univie.ac.at (dial-143182.HRZ.Uni-Dortmund.DE [129.217.143.182])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e92DXbC15327
	for <groff@ffii.org>; Mon, 2 Oct 2000 15:33:38 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id PAA00506
	for <groff@ffii.org>; Mon, 2 Oct 2000 15:33:25 +0200
Date: Mon, 02 Oct 2000 15:33:24 +0200 (CEST)
Message-Id: <20001002.153324.129787925.wl@gnu.org>
To: groff@ffii.org
From: Werner LEMBERG <wl@gnu.org>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] strange if "..."..." behaviour
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 499
Lines: 23

Consider the following code snippet:


.ds xx \fR|\fP
.ie "|"\*(xx" \
.  tm true
.el \
.  tm false


Would you expect that the result is `true'?  I didn't.

The original troff manual says:

  ... if the strings compare identically (including motions and
  character size and font), _anything_ is accepted as input ...

I would interpret this that the comparison is done by a simple
strcmp(), and not interpreting the strings.  Can someone please check
this with an original AT&T troff?


    Werner

From groff-admin  Mon Oct  2 16:07:49 2000
Received: from auemlsrv.firewall.lucent.com (auemail1.lucent.com [192.11.223.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA00300
	for <groff@ffii.org>; Mon, 2 Oct 2000 16:07:47 +0200
X-Authentication-Warning: genba.ffii.org: Host auemail1.lucent.com [192.11.223.161] claimed to be auemlsrv.firewall.lucent.com
Received: from auemlsrv.firewall.lucent.com (localhost [127.0.0.1])
	by auemlsrv.firewall.lucent.com (Pro-8.9.3/8.9.3) with ESMTP id KAA28076
	for <groff@ffii.org>; Mon, 2 Oct 2000 10:10:41 -0400 (EDT)
Received: from ru0028exch01.spb.lucent.com (h135-171-148-105.lucent.com [135.171.148.105])
	by auemlsrv.firewall.lucent.com (Pro-8.9.3/8.9.3) with ESMTP id KAA27929
	for <groff@ffii.org>; Mon, 2 Oct 2000 10:10:34 -0400 (EDT)
Received: by ru0028exch01.spb.lucent.com with Internet Mail Service (5.5.2650.21)
	id <SKKZX00P>; Mon, 2 Oct 2000 18:10:35 +0400
Message-ID: <402099F49BEED211999700805FC7359F760274@ru0028exch01.spb.lucent.com>
From: "Konovalov, Vadim" <vkonovalov@lucent.com>
To: Werner LEMBERG <wl@gnu.org>, groff@ffii.org
Subject: RE: [Groff] strange if "..."..." behaviour
Date: Mon, 2 Oct 2000 18:10:33 +0400 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="koi8-r"
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 657
Lines: 25

> Consider the following code snippet:
> 
> 
> .ds xx \fR|\fP
> .ie "|"\*(xx" \
> .  tm true
> .el \
> .  tm false
> 
> 
> Would you expect that the result is `true'?  I didn't.
> 
> The original troff manual says:
> 
>   ... if the strings compare identically (including motions and
>   character size and font), _anything_ is accepted as input ...
> 
> I would interpret this that the comparison is done by a simple
> strcmp(), and not interpreting the strings.  Can someone please check
> this with an original AT&T troff?

On SUN, System V Release 4.0 it outputs true for me.

<!ENTITY Vadim REALLIFE "Vadim V.Konovalov, St.Petersburg, Russia">
&Vadim;

From groff-admin  Mon Oct  2 16:34:38 2000
Received: from sdrc.com (heimdall.sdrc.com [146.122.132.195])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA00587
	for <groff@ffii.org>; Mon, 2 Oct 2000 16:34:37 +0200
X-Authentication-Warning: genba.ffii.org: Host heimdall.sdrc.com [146.122.132.195] claimed to be sdrc.com
Received: from tyr.sdrc.com (mailhub-cvg.sdrc.com [146.122.142.31])
	by sdrc.com (8.9.1/8.9.1) with ESMTP id KAA09380;
	Mon, 2 Oct 2000 10:37:24 -0400 (EDT)
Received: from thor.sdrc.com (scjones@thor.sdrc.com [146.122.62.108])
	by tyr.sdrc.com (8.8.6 (PHNE_17190)/8.8.5) with ESMTP id KAA06781;
	Mon, 2 Oct 2000 10:37:22 -0400 (EDT)
Received: (from scjones@localhost)
	by thor.sdrc.com (8.9.0/8.9.0) id KAA00475;
	Mon, 2 Oct 2000 10:37:23 -0400 (EDT)
Message-Id: <200010021437.KAA00475@thor.sdrc.com>
Subject: Re: [Groff] strange if "..."..." behaviour
To: wl@gnu.org (Werner LEMBERG)
Date: Mon, 2 Oct 2000 10:37:22 -0400 (EDT)
Cc: groff@ffii.org
In-Reply-To: <20001002.153324.129787925.wl@gnu.org> from "Werner LEMBERG" at Oct 2, 0 03:33:24 pm
From: larry.jones@sdrc.com (Larry Jones)
X-Mailer: ELM [version 2.4 PL25]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1095
Lines: 32

Werner LEMBERG writes:
> 
> .ds xx \fR|\fP
> .ie "|"\*(xx" \
> .  tm true
> .el \
> .  tm false
> 
> Would you expect that the result is `true'?  I didn't.

Yes.

> The original troff manual says:
> 
>   ... if the strings compare identically (including motions and
>   character size and font), _anything_ is accepted as input ...
> 
> I would interpret this that the comparison is done by a simple
> strcmp(), and not interpreting the strings.  Can someone please check
> this with an original AT&T troff?

No, what it's saying is that the resulting motions, character sizes, and
fonts have to match, not that the individual motion, size and font
requests have to match.  In this case ``|'' and ``\fR|\fP'' both result
in a roman ``|'' character with the same point size and at the same
location on the page, so the strings are equal.  If you added a ``.ft
I'' before the ``.ie'', then it's false because ``|'' produces an italic
``|'' rather than a roman one.  AT&T troff works exactly the same way.

-Larry Jones

Any game without push-ups, hits, burns or noogies is a sissy game. -- Calvin

From groff-admin  Mon Oct  2 18:14:35 2000
Received: from Hesperia (ppp63-brest.libertysurf.fr [213.36.44.63])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA01585
	for <groff@ffii.org>; Mon, 2 Oct 2000 18:14:34 +0200
X-Authentication-Warning: genba.ffii.org: Host ppp63-brest.libertysurf.fr [213.36.44.63] claimed to be Hesperia
Received: from baruchel by Hesperia with local (Exim 2.05 #1 (Debian))
	id 13fn4t-0001Ft-00; Sun, 1 Oct 2000 19:39:07 +0200
Date: Sun, 1 Oct 2000 19:39:07 +0200
From: Thomas Baruchel <baruchel@libertysurf.fr>
To: groff@ffii.org
Message-ID: <20001001193907.A4817@Hesperia>
Reply-To: Thomas Baruchel <baruchel@libertysurf.fr>
Mail-Followup-To: groff@ffii.org
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
Machine: Hesperia sur Debian (Linux)
Subject: [Groff] still exploring the last version: what is gremlin ?
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 448
Lines: 13

Brest, le dimanche 1er octobre

after having used an old version of groff for a while, I just downloaded the
last one. I discovered 'grn'. What exactly is the 'gremlin' format ? What is it
good for ? Do I need extra tool to make gremlin files ? I have 'grn' but not
gremlin(1). Where can I download it ? What is the purpose of this program ?

-- 
   .~.   Thomas Baruchel
   /V\   baruchel@libertysurf.fr
  // \\    Brest
 /(   )\   FRANCE
  ^`~'^

From groff-admin  Mon Oct  2 19:55:31 2000
Received: from srv1.tsoft.net (news.tsoft.net [198.144.192.3])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id TAA02197
	for <groff@ffii.org>; Mon, 2 Oct 2000 19:55:30 +0200
X-Authentication-Warning: genba.ffii.org: Host news.tsoft.net [198.144.192.3] claimed to be srv1.tsoft.net
Received: from bicho.SynchroDS.COM (daniel.ppp.tsoft.com [198.144.199.225])
	by srv1.tsoft.net (8.8.7/8.8.7) with ESMTP id KAA10207;
	Mon, 2 Oct 2000 10:58:23 -0700 (PDT)
Received: (from daniel@localhost)
	by bicho.SynchroDS.COM (8.8.8/8.8.8) id KAA14051;
	Mon, 2 Oct 2000 10:58:05 -0700 (PDT)
Date: Mon, 2 Oct 2000 10:58:05 -0700 (PDT)
From: Daniel Senderowicz <daniel@bicho.SynchroDS.COM>
Message-Id: <200010021758.KAA14051@bicho.SynchroDS.COM>
To: baruchel@libertysurf.fr
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
Cc: groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 969
Lines: 21

>Brest, le dimanche 1er octobre
Looks good and probably sounds good in French!

>after having used an old version of groff for a while, I just downloaded the
>last one. I discovered 'grn'. What exactly is the 'gremlin' format ? What is it
>good for ? Do I need extra tool to make gremlin files ? I have 'grn' but not
>gremlin(1). Where can I download it ? What is the purpose of this program ?

Grn is a graphical interface to *roff, like pic.  The good thing
is that you don't have to type the commands. You can generate
graphically the files with gremlin that is very much like "Fig".
There are a couple of versions of gremlin around, both for X.  If
you're doing your own drawings, it's more flexible than including
postscript files, you can put equations in your drawings, a
significant enhancement.  Can't remember where, but you may find
them with archie or some search engine. If you don't, I'll dig
around to give you more precise directions.

Shana tova!

Dan

From groff-admin  Tue Oct  3 01:50:48 2000
Received: from srv1.tsoft.net (news.tsoft.net [198.144.192.3])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id BAA05456
	for <groff@ffii.org>; Tue, 3 Oct 2000 01:50:47 +0200
X-Authentication-Warning: genba.ffii.org: Host news.tsoft.net [198.144.192.3] claimed to be srv1.tsoft.net
Received: from bicho.SynchroDS.COM (daniel.ppp.tsoft.com [198.144.199.225])
	by srv1.tsoft.net (8.8.7/8.8.7) with ESMTP id QAA20087;
	Mon, 2 Oct 2000 16:53:42 -0700 (PDT)
Received: (from daniel@localhost)
	by bicho.SynchroDS.COM (8.8.8/8.8.8) id QAA15034;
	Mon, 2 Oct 2000 16:53:24 -0700 (PDT)
Date: Mon, 2 Oct 2000 16:53:24 -0700 (PDT)
From: Daniel Senderowicz <daniel@bicho.SynchroDS.COM>
Message-Id: <200010022353.QAA15034@bicho.SynchroDS.COM>
To: baruchel@libertysurf.fr
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
Cc: groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1317
Lines: 31

I found xgremlin-2.0 in
ftp://boss.cae.wisc.edu/hpux/X11/Drawing/xgremlin-2.0

Incidentally, I think that this version might solve an X problem
that Werner observed with his Linux system (which I didn't with my
NetBSD box) with an older version of the program. I just compiled
this one and it also works fine. Good luck.

Dan

>>Brest, le dimanche 1er octobre
>Looks good and probably sounds good in French!
>
>>after having used an old version of groff for a while, I just downloaded the
>>last one. I discovered 'grn'. What exactly is the 'gremlin' format ? What is it
>>good for ? Do I need extra tool to make gremlin files ? I have 'grn' but not
>>gremlin(1). Where can I download it ? What is the purpose of this program ?
>
>Grn is a graphical interface to *roff, like pic.  The good thing
>is that you don't have to type the commands. You can generate
>graphically the files with gremlin that is very much like "Fig".
>There are a couple of versions of gremlin around, both for X.  If
>you're doing your own drawings, it's more flexible than including
>postscript files, you can put equations in your drawings, a
>significant enhancement.  Can't remember where, but you may find
>them with archie or some search engine. If you don't, I'll dig
>around to give you more precise directions.
>
>Shana tova!
>
>Dan

From groff-admin  Tue Oct  3 02:34:44 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id CAA06063
	for <groff@ffii.org>; Tue, 3 Oct 2000 02:34:43 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Tue, 3 Oct 2000 02:37:31 +0200
Received: from rigel.univie.ac.at (dial-143165.HRZ.Uni-Dortmund.DE [129.217.143.165])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e930bTC00734;
	Tue, 3 Oct 2000 02:37:29 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id VAA01167;
	Mon, 2 Oct 2000 21:35:46 +0200
Date: Mon, 02 Oct 2000 21:35:45 +0200 (CEST)
Message-Id: <20001002.213545.105162767.wl@gnu.org>
To: larry.jones@sdrc.com
Cc: groff@ffii.org
Subject: Re: [Groff] strange if "..."..." behaviour
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <200010021437.KAA00475@thor.sdrc.com>
References: <20001002.153324.129787925.wl@gnu.org>
	<200010021437.KAA00475@thor.sdrc.com>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1038
Lines: 25

> > The original troff manual says:
> > 
> >   ... if the strings compare identically (including motions and
> >   character size and font), _anything_ is accepted as input ...
> > 
> > I would interpret this that the comparison is done by a simple
> > strcmp(), and not interpreting the strings.  Can someone please
> > check this with an original AT&T troff?
> 
> No, what it's saying is that the resulting motions, character sizes,
> and fonts have to match, not that the individual motion, size and
> font requests have to match.  In this case ``|'' and ``\fR|\fP''
> both result in a roman ``|'' character with the same point size and
> at the same location on the page, so the strings are equal.  If you
> added a ``.ft I'' before the ``.ie'', then it's false because ``|''
> produces an italic ``|'' rather than a roman one.  AT&T troff works
> exactly the same way.

Thanks for the clarification.  Nevertheless, the wording in the
original troff manual is not ideal.

I'll add a few words to the groff documentation.


    Werner

From groff-admin  Mon Oct  2 20:12:27 2000
Received: from mail1.svr.pol.co.uk (mail1.svr.pol.co.uk [195.92.193.18])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id UAA02440
	for <groff@ffii.org>; Mon, 2 Oct 2000 20:12:26 +0200
Received: from [195.92.198.123] (helo=mail17.svr.pol.co.uk)
	by mail1.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13gA7W-0007ra-00
	for groff@ffii.org; Mon, 02 Oct 2000 19:15:22 +0100
Received: from modem-11.annatar.dialup.pol.co.uk ([62.136.115.139] helo=inputplus.demon.co.uk)
	by mail17.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13gA7U-0008Cb-00
	for groff@ffii.org; Mon, 02 Oct 2000 19:15:21 +0100
Received: from inputplus (ralph@localhost [127.0.0.1])
	by inputplus.demon.co.uk (8.8.7/8.8.7) with ESMTP id TAA15841
	for <groff@ffii.org>; Mon, 2 Oct 2000 19:15:19 +0100
Resent-Message-Id: <200010021815.TAA15841@inputplus.demon.co.uk>
Received: from localhost (ralph@localhost [127.0.0.1])
	by inputplus.demon.co.uk (8.8.7/8.8.7) with ESMTP id TAA14304
	for <ralph@localhost>; Mon, 2 Oct 2000 19:09:17 +0100
Received: from pop3.demon.co.uk
	by localhost with POP3 (fetchmail-4.7.0)
	for ralph@localhost (single-drop); Mon, 02 Oct 2000 18:09:18 +0000 (UTC)
Received: from punt-1.mail.demon.net by mailstore
          for ralph@inputplus.demon.co.uk id 970436913:10:22834:0;
          Sun, 01 Oct 2000 21:48:33 GMT
Received: from aua2.aua.auc.dk ([130.225.54.5]) by punt-1.mail.demon.net
           id aa1021508; 1 Oct 2000 21:48 GMT
Received: from nightowl (nightowl.abk.auc.dk [192.168.80.38])
 by aua.auc.dk (PMDF V5.2-32 #30369)
 with SMTP id <01JUUAMB0O1K8Y5IG4@aua.auc.dk> for ralph@inputplus.demon.co.uk;
 Sun, 1 Oct 2000 23:48:07 MET-DST
Date: Sun, 01 Oct 2000 23:48:16 +0200
From: "Frank G. Hahn" <efgeho@cs.auc.dk>
Subject: Re: [Groff] Arrow between ellipses - How to chop (remove overlap) ?
To: Ralph Corderoy <ralph@inputplus.demon.co.uk>
Message-id: <002b01c02bf1$4e786680$2650a8c0@abk.auc.dk>
MIME-version: 1.0
X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
Content-type: text/plain;	charset="iso-8859-1"
Content-transfer-encoding: 7bit
X-Priority: 3
X-MSMail-priority: Normal
References: <200010012112.WAA23396@inputplus.demon.co.uk>
Resent-To: groff@ffii.org
Resent-Date: Mon, 02 Oct 2000 19:15:19 +0100
Resent-From: Ralph Corderoy <ralph@inputplus.demon.co.uk>
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1708
Lines: 55

----- Original Message -----
From: "Ralph Corderoy" <ralph@inputplus.demon.co.uk>
To: "Frank G. Hahn" <efgeho@cs.auc.dk>
Cc: <groff@ffii.org>
Sent: Sunday, October 01, 2000 11:12 PM
Subject: Re: [Groff] Arrow between ellipses - How to chop (remove overlap) ?


>
> Hi,
>
> > > Try
> > >
> > >     arrow from E1 to E2 chop 0.3
> >
> > This would have been nice .. but I need something more generic than
> > that.  I got a program reading a hugin (Bayesian Network) file and
> > generating the output in PIC-format.  Which results in "many"
> > ellipses scattered about on a page. Some of these are connected
> > (using arrows).  How do I get pic to chop the arrow correctly in BOTH
> > ends.
> >
> > I can't chop them using a constant, because I don't really know how
> > the ellipses will be placed relative to one another.
>
> If you're leaving the placement of the nodes up to pic then pic knows.
>
> > > and see also the definition of chop around page 10 of Kernighan's
> > > _PIC -- A Graphics Language for Typesetting_ mentioned at the end
> > > of groff's pic man page.
> > > <URL:http://cm.bell-labs.com/cm/cs/cstr/116.ps.gz>
>
> Did you read this fine document?  It explains that pic's trig functions
> should be used.  Will that work in your case?
>

I'm not sure. Currently I'm reading up on Ellipse-Line-intersection.

I got:
Ellipse 1:  center(x0, y0), size(width, height)
Ellipse 2:  center(x1, y1), size(width, height);
Line: (x0, y0) through (x1, y1).

If I can decide/calculate the closest intersection-points I can use
arrow from (ix0, iy0) to (ix1, iy1)

but this does not seem to be quite as easy to find the correct formula for
as I thought..

Thanks for the doc-link.

/Frank


> Ralph.

From groff-admin  Wed Oct  4 15:35:39 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA23672
	for <groff@ffii.org>; Wed, 4 Oct 2000 15:35:39 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Wed, 4 Oct 2000 15:38:26 +0200
Received: from rigel.univie.ac.at (dial-143193.HRZ.Uni-Dortmund.DE [129.217.143.193])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e94DbnC12200;
	Wed, 4 Oct 2000 15:37:56 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id PAA03126;
	Wed, 4 Oct 2000 15:30:50 +0200
Date: Wed, 04 Oct 2000 15:30:50 +0200 (CEST)
Message-Id: <20001004.153050.129347863.wl@gnu.org>
To: daniel@bicho.SynchroDS.COM
Cc: baruchel@libertysurf.fr, groff@ffii.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <200010022353.QAA15034@bicho.SynchroDS.COM>
References: <200010022353.QAA15034@bicho.SynchroDS.COM>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 755
Lines: 20

> I found xgremlin-2.0 in
> ftp://boss.cae.wisc.edu/hpux/X11/Drawing/xgremlin-2.0
> 
> Incidentally, I think that this version might solve an X problem
> that Werner observed with his Linux system (which I didn't with my
> NetBSD box) with an older version of the program. I just compiled
> this one and it also works fine. Good luck.

Lucky guy!  This version doesn't compile out of the box.

I still have the same problem -- instead of text, only black boxes are
shown in the text subwindow.  My knowledge of X11 is too limited to
fix that :-( It has probably something to do with the transition from
X11R4 -> X11R6.  Any help appreciated.

I've put an xgremlin2 version to ftp.ffii.org:/pub/groff which compiles
just fine on my Linux box.


    Werner

From groff-admin  Wed Oct  4 18:28:04 2000
Received: from srv1.tsoft.net (news.tsoft.net [198.144.192.3])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA25354
	for <groff@ffii.org>; Wed, 4 Oct 2000 18:28:03 +0200
X-Authentication-Warning: genba.ffii.org: Host news.tsoft.net [198.144.192.3] claimed to be srv1.tsoft.net
Received: from bicho.SynchroDS.COM (daniel.ppp.tsoft.com [198.144.199.225])
	by srv1.tsoft.net (8.8.7/8.8.7) with ESMTP id JAA18487;
	Wed, 4 Oct 2000 09:31:01 -0700 (PDT)
Received: (from daniel@localhost)
	by bicho.SynchroDS.COM (8.8.8/8.8.8) id JAA19823;
	Wed, 4 Oct 2000 09:30:45 -0700 (PDT)
Date: Wed, 4 Oct 2000 09:30:45 -0700 (PDT)
From: Daniel Senderowicz <daniel@bicho.SynchroDS.COM>
Message-Id: <200010041630.JAA19823@bicho.SynchroDS.COM>
To: daniel@bicho.SynchroDS.COM, wl@gnu.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
Cc: baruchel@libertysurf.fr, groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1069
Lines: 30

Sorry, but I'm not an X expert myself, and I don't even have a
linux box to fiddle around (I'm running NetBSD on a DEC5000/240).
Strange that you have that problem, because here I also run
X11R6, and haven't seen any problem. Maybe some good soul with a
linux box can help you.

Regards,

Dan

>> I found xgremlin-2.0 in
>> ftp://boss.cae.wisc.edu/hpux/X11/Drawing/xgremlin-2.0
>> 
>> Incidentally, I think that this version might solve an X problem
>> that Werner observed with his Linux system (which I didn't with my
>> NetBSD box) with an older version of the program. I just compiled
>> this one and it also works fine. Good luck.
>
>Lucky guy!  This version doesn't compile out of the box.
>
>I still have the same problem -- instead of text, only black boxes are
>shown in the text subwindow.  My knowledge of X11 is too limited to
>fix that :-( It has probably something to do with the transition from
>X11R4 -> X11R6.  Any help appreciated.
>
>I've put an xgremlin2 version to ftp.ffii.org:/pub/groff which compiles
>just fine on my Linux box.
>
>
>    Werner

From groff-admin  Wed Oct  4 23:40:00 2000
Received: from polo.po.opole.pl (root@polo.po.opole.pl [193.59.208.10])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id XAA27804
	for <groff@ffii.org>; Wed, 4 Oct 2000 23:39:58 +0200
Received: from ukulele.linux.home (root@dial02.po.opole.pl [193.59.208.16])
	by polo.po.opole.pl (8.8.8/8.8.6/ag-polo.981012-sz+rchk1.22) with ESMTP id XAA29798
	for <groff@ffii.org>; Wed, 4 Oct 2000 23:42:59 +0200 (MET DST)
Received: from guitar.linux.home (root@guitar.linux.home [10.0.0.12])
	by ukulele.linux.home (8.9.3/8.9.3/Debian/GNU) with ESMTP id XAA12283
	for <groff@ffii.org>; Wed, 4 Oct 2000 23:09:52 +0200
Received: from localhost (luke@localhost)
	by guitar.linux.home (8.8.6/8.8.7) with SMTP id WAA00339
	for <groff@ffii.org>; Wed, 4 Oct 2000 22:51:47 +0100
Date: Wed, 4 Oct 2000 22:51:47 +0100 (MET)
From: Lukasz Wiechec <magic@bet.po.opole.pl>
To: groff@ffii.org
Message-ID: <Pine.LNX.3.96.1001004224318.313A-100000@guitar.linux.home>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Groff] Hello to you all !
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 969
Lines: 24

Hello All, 

this is my first posting to this list, so please excuse if my ignorance...
I've searched the archives of this mailing list, but didn't seem to find
anything to help me solve my problem.

I'm trying to make 'groff' generate valid postscript output from
Latin-2-encoded text. I've installed proper Latin-2 Postscript fonts and
they work -- well, almost all of them :-(. The problem is that some
(exacly three) Polish characters are being interpreted as symbols (and
therefore displayed using Symbol special font). This is done only using
'-Tps' (when using '-Tlatin1', tty output is OK).

The question is: how to tell 'groff' that those characters should be
displayed normally ? I would also like to be able to use Symbol font too.
I've tried modifying Symbol font description files -- to no avail.

Greetings from Opole, Poland, 

		------------------------
		MAGIC aka Lukasz Wiechec
		Internet mailto:magic@bet.po.opole.pl
		tel/fax  +(48 77) 4 555 113 


From groff-admin  Thu Oct  5 10:03:45 2000
Received: from atlantis.erlm.siemens.de (atlantis.erlm.siemens.de [212.114.202.14])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id KAA13400
	for <groff@ffii.org>; Thu, 5 Oct 2000 10:03:45 +0200
Received: from mail01.erlm.siemens.de (mail01.erlm.siemens.de [146.254.160.12])
	by atlantis.erlm.siemens.de (8.9.3/8.9.3) with ESMTP id JAA25772
	for <groff@ffii.org>; Thu, 5 Oct 2000 09:55:29 +0200 (MET DST)
Received: from nbgv101a.nbg9.siemens.de (nbgv101a.nbg9.siemens.de [157.163.80.101])
	by mail01.erlm.siemens.de (8.9.3+Sun/8.9.3) with ESMTP id JAA19289
	for <groff@ffii.org>; Thu, 5 Oct 2000 09:55:29 +0200 (MET DST)
Received: by NBGV101A with Internet Mail Service (5.5.2650.21)
	id <4FTNTT6M>; Thu, 5 Oct 2000 09:55:27 +0200
Message-ID: <11DD9734B036D311947400805F0DD607BD8AB1@NBGV101A>
From: Freudenberger Joerg <Joerg.Freudenberger@nbg9.siemens.de>
To: "'groff Mailinglist'" <groff@ffii.org>
Date: Thu, 5 Oct 2000 09:55:26 +0200 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by genba.ffii.org id KAA13401
Subject: [Groff] FW: ISS Security Advisory: GNU Groff utilities read untrusted com
 mands from current working directory
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 8545
Lines: 196

Hallo Groffers

I pass this from Bugtraq for your information

Yours
	Jörg

-----Original Message-----
From: Aleph One [mailto:aleph1@UNDERGROUND.ORG] 
Sent: Thursday, October 05, 2000 12:51 AM
To: BUGTRAQ@SECURITYFOCUS.COM
Subject: ISS Security Advisory: GNU Groff utilities read untrusted
commands from current working directory


-----BEGIN PGP SIGNED MESSAGE-----

Internet Security Systems Security Advisory
October 4, 2000

GNU Groff utilities read untrusted commands from current working
directory

Synopsis:

Internet Security Systems (ISS) has identified vulnerabilities in several
utilities that ship as part of the Groff document formatting system package.

By default, the "troff" program reads its "troffrc" initialization file from
the current working directory.  From a security standpoint, it would be
desirable to restrict the searchable path for this file to the invoker's
home directory and/or a trusted system. Unfortunately, this could present
problems for programs that depend on the current behavior.

The "groff" program, a front-end for troff, has a similar problem.  It looks
for the appropriate device description file (as given by the -T parameter,
or
"ps" by default) using devname/DESC in the current working directory.  The
device description file may contain an optional "postpro" directive, which
defines a command to be run after normal processing. A malicious user could
place a trojan device description file in a world-writable directory
(i.e. /tmp), after which any invocations of groff from that directory are
unsafe.

Impact:

Unsuspecting users, including root, could be coerced into running arbitrary
commands on the system.

The vulnerability is particularly dangerous in Linux distributions that have
the "lesspipe" feature.  By default, a "LESSOPEN" environment variable is
set
which points to a wrapper script for the "less" pager program named
"/usr/bin/lesspipe.sh".  If less is passed a filename with any of the
extensions ".1" through ".9", ".n", or ".man", it automatically calls groff
to handle the file.

Description:

Troff is a document processor that ships with most Unix systems.  Among
other
functions, it formats system manual pages into human-readable form.  The GNU
Groff package includes "troff", the main processing program, and "groff", a
front-end for troff.  Typically, troff is invoked by groff.

Troff supports a set of potentially dangerous macros: "open", "opena",
"pso",
"sy", and "pi", which provide the means to write to files and execute
external commands.  For example, "opena" opens a file for writing in append
mode and "sy" performs a C system() call with the specified argument.

The default in groff is that these dangerous macros are disabled.  This is
accomplished by another macro defined in the file "tmac.safer".  Unless
overridden by the -U (unsafe) flag, the groff program passes troff the
flag "-msafer", which instructs troff to process the tmac.safer macro
before the input file.  However, before troff processes the tmac.safer
macro,
it first looks for a "troffrc" initialization file.  If one is found, it
executes the commands found therein first, bypassing the dangerous macro
protection.  As mentioned above, troff looks for this initialization file in
the current directory, creating a potentially dangerous situation.

Groff (speaking of the actual program now, not the package as a whole) is
a front-end for troff.  It supports a variety of devices. For example, the
PostScript device is named "ps" and allows groff to generate output that
is fit to print on PostScript printers. There is a device for HTML, and one
called "ascii" that's used to pretty-print text on typewriter-like devices.

Each device supported by groff has a corresponding directory of the name
"dev<name>", where <name> is "ps", "ascii", etc.  These directories are
typically installed under some trusted path on the system, i.e., /usr/lib.
The device description file is named "dev<name>/DESC".  Since groff blindly
trusts "DESC" files contained under the current directory hierarchy, an
attacker may be able to fool another user into running any arbitrary
command using the "postpro" directive.

Solar Designer <solar@false.com> points out that the aforementioned files
are not alone in the set that may be accessed from the current directory.
Other hard-coded filenames, such as "troffrc-end", could fall within the `.'
search path as well (troffrc-end is loaded after the -msafer macros,
though).
In fact, the macro files themselves reference other files that could reside
in the current directory.

Recommendations:

Both administrators and users should exercise caution and not run "groff",
"troff", or even the "man" command from untrusted directories.

Internet Security Systems has not received a response from the current GNU
Groff maintainer. In the interest of accelerating the elimination of these
vulnerabilities, this advisory is being disseminated to the open source
community for public discussion.

Internet Security Systems recognizes that reading from the current directory
is traditional groff/troff behavior, and that in many document-creating
scenarios it is actually a useful `feature'. One possibility could be to not
trust the current directory at all by default, perhaps requiring a special
command line option to revert to the old behavior. At any rate, the fix is
not obvious, as per Solar Designer's analysis.

Note that troff's -R option ("Don't load troffrc") does not eliminate the
problem.

Additional Information:

The dangerous Troff macros were discussed on the BUGTRAQ mailing list in
July, 1999 on a thread under the subject heading of "Troff dangerous".  A
searchable archive of the BUGTRAQ list is at: http://www.securityfocus.com.

The Groff package can be found at the following FTP location:

	ftp://ftp.gnu.org/pub/gnu/groff

The Common Vulnerabilities and Exposures (CVE) project has assigned the name
CAN-2000-0803 to this issue.  This is a candidate for inclusion in the CVE
list (http://cve.mitre.org), which standardizes names for security problems.


Credits:

This vulnerability was discovered and researched by Aaron Campbell and
Allen Wilson of the ISS X-Force. Internet Security Systems would like to
acknowledge Solar Designer for his analysis of this problem.

_______

About Internet Security Systems (ISS)
Internet Security Systems (ISS) is a leading global provider of security
management solutions for the Internet. By providing industry-leading
SAFEsuite security software, remote managed security services, and
strategic consulting and education offerings, ISS is a trusted security
provider to its customers, protecting digital assets and ensuring safe
and uninterrupted e-business. ISS' security management solutions protect
more than 5,500 customers worldwide including 21 of the 25 largest U.S.
commercial banks, 10 of the largest telecommunications companies and
over 35 government agencies. Founded in 1994, ISS is headquartered in
Atlanta, GA, with additional offices throughout North America and
international operations in Asia, Australia, Europe, Latin America and
the Middle East. For more information, visit the Internet Security
Systems web site at www.iss.net or call 888-901-7477.

Copyright (c) 2000 by Internet Security Systems, Inc.

Permission is hereby granted for the redistribution of this Alert
electronically. It is not to be edited in any way without express
consent of the X-Force. If you wish to reprint the whole or any part of
this Alert in any other medium excluding electronic medium, please
e-mail xforce@iss.net for permission.

Disclaimer

The information within this paper may change without notice. Use of this
information constitutes acceptance for use in an AS IS condition. There
are NO warranties with regard to this information. In no event shall the
author be liable for any damages whatsoever arising out of or in
connection with the use or spread of this information. Any use of this
information is at the user's own risk.

X-Force PGP Key available at: http://xforce.iss.net/sensitive.php as well
as on MIT's PGP key server and PGP.com's key server.

Please send suggestions, updates, and comments to: X-Force
xforce@iss.net of Internet Security Systems, Inc.


-----BEGIN PGP SIGNATURE-----
Version: 2.6.3a
Charset: noconv

iQCVAwUBOdtMgDRfJiV99eG9AQEqXAP8CH24N8b+q0Die6b44NRGEsHe1byU+OKt
VfTRPZMo/Ag8gv/ENSFuXt6B2TFKLEvM7uZlJuKnmAnKtIQIpaFh6J6TXPWki2V3
UghnlMQaIftSBNI/dKaxjMvzjHngNrR8dgrUdJj/I8lRK6UlRJy5U9jeFBXcVuJ6
HwJU3t9ySIo=
=FxuE
-----END PGP SIGNATURE-----

From groff-admin  Thu Oct  5 18:29:52 2000
Received: from srv1.tsoft.net (news.tsoft.net [198.144.192.3])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA18018
	for <groff@ffii.org>; Thu, 5 Oct 2000 18:29:51 +0200
X-Authentication-Warning: genba.ffii.org: Host news.tsoft.net [198.144.192.3] claimed to be srv1.tsoft.net
Received: from bicho.SynchroDS.COM (daniel.ppp.tsoft.com [198.144.199.225])
	by srv1.tsoft.net (8.8.7/8.8.7) with ESMTP id JAA23498;
	Thu, 5 Oct 2000 09:33:02 -0700 (PDT)
Received: (from daniel@localhost)
	by bicho.SynchroDS.COM (8.8.8/8.8.8) id JAA02457;
	Thu, 5 Oct 2000 09:32:45 -0700 (PDT)
Date: Thu, 5 Oct 2000 09:32:45 -0700 (PDT)
From: Daniel Senderowicz <daniel@bicho.SynchroDS.COM>
Message-Id: <200010051632.JAA02457@bicho.SynchroDS.COM>
To: wl@gnu.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
Cc: baruchel@libertysurf.fr, groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 599
Lines: 19

I just discovered that a change made in main.c for xgremlin-2.0
in line 592:

		readmask = ConnectionNumber(display);/*1 << display->fd;*/ 

slows down the program VERY much, so I went back to the old line,
that is:

		readmask = 1 << display->fd;

Those of you guys with linux system should try both options.

Dan

>Sorry, but I'm not an X expert myself, and I don't even have a
>linux box to fiddle around (I'm running NetBSD on a DEC5000/240).
>Strange that you have that problem, because here I also run
>X11R6, and haven't seen any problem. Maybe some good soul with a
>linux box can help you.

From groff-admin  Fri Oct  6 18:45:22 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA09200
	for <groff@ffii.org>; Fri, 6 Oct 2000 18:45:22 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Fri, 6 Oct 2000 18:48:11 +0200
Received: from rigel.univie.ac.at (dial-143166.HRZ.Uni-Dortmund.DE [129.217.143.166])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e96Gm7C27269;
	Fri, 6 Oct 2000 18:48:07 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id IAA04435;
	Fri, 6 Oct 2000 08:33:57 +0200
Date: Fri, 06 Oct 2000 08:33:57 +0200 (CEST)
Message-Id: <20001006.083357.122620281.wl@gnu.org>
To: daniel@bicho.synchrods.com
Cc: baruchel@libertysurf.fr, groff@ffii.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <200010051632.JAA02457@bicho.SynchroDS.COM>
References: <200010051632.JAA02457@bicho.SynchroDS.COM>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 628
Lines: 20

> I just discovered that a change made in main.c for xgremlin-2.0
> in line 592:
> 
>   readmask = ConnectionNumber(display);/*1 << display->fd;*/ 
> 
> slows down the program VERY much, so I went back to the old line,
> that is:
> 
> 		readmask = 1 << display->fd;
> 
> Those of you guys with linux system should try both options.

How does it slow down?  This wasn't my change, BTW, I've taken it from
the HP-UX port (the link you've given in one of your recent mails),
and according to the various man pages, using ConnectionNumber() seems
the right thing to do to be portable.  Can someone comment this
please?


    Werner

From groff-admin  Fri Oct  6 18:46:03 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA09218
	for <groff@ffii.org>; Fri, 6 Oct 2000 18:46:03 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Fri, 6 Oct 2000 18:49:14 +0200
Received: from rigel.univie.ac.at (dial-143166.HRZ.Uni-Dortmund.DE [129.217.143.166])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e96GnCC27302;
	Fri, 6 Oct 2000 18:49:12 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id JAA03742;
	Thu, 5 Oct 2000 09:17:20 +0200
Date: Thu, 05 Oct 2000 09:17:20 +0200 (CEST)
Message-Id: <20001005.091720.106825793.wl@gnu.org>
To: magic@bet.po.opole.pl
Cc: groff@ffii.org
Subject: Re: [Groff] Hello to you all !
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <Pine.LNX.3.96.1001004224318.313A-100000@guitar.linux.home>
References: <Pine.LNX.3.96.1001004224318.313A-100000@guitar.linux.home>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 297
Lines: 9

> The problem is that some (exacly three) Polish characters are being
> interpreted as symbols (and therefore displayed using Symbol special
> font).  This is done only using '-Tps' (when using '-Tlatin1', tty
> output is OK).

Which characters (or rather, glyphs)?  Example, please.


    Werner

From groff-admin  Fri Oct  6 18:46:03 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA09220
	for <groff@ffii.org>; Fri, 6 Oct 2000 18:46:03 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Fri, 6 Oct 2000 18:49:12 +0200
Received: from rigel.univie.ac.at (dial-143166.HRZ.Uni-Dortmund.DE [129.217.143.166])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e96Gn4C27298;
	Fri, 6 Oct 2000 18:49:05 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id WAA04157;
	Thu, 5 Oct 2000 22:37:31 +0200
Date: Thu, 05 Oct 2000 22:37:31 +0200 (CEST)
Message-Id: <20001005.223731.37301821.wl@gnu.org>
To: Joerg.Freudenberger@nbg9.siemens.de
Cc: groff@ffii.org, xforce@iss.net, solar@false.com
Subject: Re: [Groff] FW: ISS Security Advisory: GNU Groff utilities read
 untrusted com mands from current working directory
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <11DD9734B036D311947400805F0DD607BD8AB1@NBGV101A>
References: <11DD9734B036D311947400805F0DD607BD8AB1@NBGV101A>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by genba.ffii.org id SAA09222
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2578
Lines: 59

[Jörg sent this security alert to the groff list]

> I pass this from Bugtraq for your information

Thanks a lot!  There are really serious problems...

> By default, the "troff" program reads its "troffrc" initialization
> file from the current working directory.  From a security
> standpoint, it would be desirable to restrict the searchable path
> for this file to the invoker's home directory and/or a trusted
> system.  Unfortunately, this could present problems for programs
> that depend on the current behavior.

My suggestion is to restrict the location of troffrc and troffrc-end
to `~' and groff's default tmac directory
(e.g. /usr/local/share/groff/tmac) if the -U flag isn't given.
Additionally, I'll implement tmac.safer internally, i.e., without the
-U flag these requests will produce error messages.  tmac.safer will
then no longer exist.

> The "groff" program, a front-end for troff, has a similar problem.
> It looks for the appropriate device description file (as given by
> the -T parameter, or "ps" by default) using devname/DESC in the
> current working directory.

Ditto.  The current directory is only allowed if the -U option is
given.

> The device description file may contain an optional "postpro"
> directive, which defines a command to be run after normal
> processing.  A malicious user could place a trojan device
> description file in a world-writable directory (i.e. /tmp), after
> which any invocations of groff from that directory are unsafe.

I imagine that the restriction to `trusted' directories (without the
-U flag) closes this security hole.  Please correct me if I'm wrong.

> Internet Security Systems has not received a response from the
> current GNU Groff maintainer.  In the interest of accelerating the
> elimination of these vulnerabilities, this advisory is being
> disseminated to the open source community for public discussion.

At least I haven't received anything from ISS -- to which address have
you sent a mail?  The correct address for bug reports to groff is
`bug-groff@gnu.org'.

> Internet Security Systems recognizes that reading from the current
> directory is traditional groff/troff behavior, and that in many
> document-creating scenarios it is actually a useful `feature'.  One
> possibility could be to not trust the current directory at all by
> default, perhaps requiring a special command line option to revert
> to the old behavior.  At any rate, the fix is not obvious, as per
> Solar Designer's analysis.

I believe that the suggestions given above solve the problem.  Please
comment.


    Werner

From groff-admin  Fri Oct  6 20:07:51 2000
Received: from srv1.tsoft.net (news.tsoft.net [198.144.192.3])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id UAA09887
	for <groff@ffii.org>; Fri, 6 Oct 2000 20:07:50 +0200
X-Authentication-Warning: genba.ffii.org: Host news.tsoft.net [198.144.192.3] claimed to be srv1.tsoft.net
Received: from bicho.SynchroDS.COM (daniel.ppp.tsoft.com [198.144.199.225])
	by srv1.tsoft.net (8.8.7/8.8.7) with ESMTP id LAA00816;
	Fri, 6 Oct 2000 11:11:03 -0700 (PDT)
Received: (from daniel@localhost)
	by bicho.SynchroDS.COM (8.8.8/8.8.8) id LAA05355;
	Fri, 6 Oct 2000 11:10:47 -0700 (PDT)
Date: Fri, 6 Oct 2000 11:10:47 -0700 (PDT)
From: Daniel Senderowicz <daniel@bicho.SynchroDS.COM>
Message-Id: <200010061810.LAA05355@bicho.SynchroDS.COM>
To: wl@gnu.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
Cc: baruchel@libertysurf.fr, groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 933
Lines: 25

For example, if you're drawing a line, when you set one point and
slide the mouse to the following point it will take for ever to
draw the line. I agree that probably ConnectionNumber() is the
proper way to go, however in my case (NetBSD-5000/240) doesn't
work well. Maybe in your system is OK.

Dan

>> I just discovered that a change made in main.c for xgremlin-2.0
>> in line 592:
>> 
>>   readmask = ConnectionNumber(display);/*1 << display->fd;*/ 
>> 
>> slows down the program VERY much, so I went back to the old line,
>> that is:
>> 
>> 		readmask = 1 << display->fd;
>> 
>> Those of you guys with linux system should try both options.
>
>How does it slow down?  This wasn't my change, BTW, I've taken it from
>the HP-UX port (the link you've given in one of your recent mails),
>and according to the various man pages, using ConnectionNumber() seems
>the right thing to do to be portable.  Can someone comment this
>please?

From groff-admin  Fri Oct  6 20:36:07 2000
Received: from sdrc.com (heimdall.sdrc.com [146.122.132.195])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id UAA10017
	for <groff@ffii.org>; Fri, 6 Oct 2000 20:36:06 +0200
X-Authentication-Warning: genba.ffii.org: Host heimdall.sdrc.com [146.122.132.195] claimed to be sdrc.com
Received: from tyr.sdrc.com (mailhub-cvg.sdrc.com [146.122.142.31])
	by sdrc.com (8.9.1/8.9.1) with ESMTP id OAA25115;
	Fri, 6 Oct 2000 14:39:16 -0400 (EDT)
Received: from thor.sdrc.com (scjones@thor.sdrc.com [146.122.62.108])
	by tyr.sdrc.com (8.8.6 (PHNE_17190)/8.8.5) with ESMTP id OAA20457;
	Fri, 6 Oct 2000 14:39:15 -0400 (EDT)
Received: (from scjones@localhost)
	by thor.sdrc.com (8.9.0/8.9.0) id OAA03761;
	Fri, 6 Oct 2000 14:39:14 -0400 (EDT)
Message-Id: <200010061839.OAA03761@thor.sdrc.com>
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
To: wl@gnu.org (Werner LEMBERG)
Date: Fri, 6 Oct 2000 14:39:14 -0400 (EDT)
Cc: daniel@bicho.synchrods.com, baruchel@libertysurf.fr, groff@ffii.org
In-Reply-To: <20001006.083357.122620281.wl@gnu.org> from "Werner LEMBERG" at Oct 6, 0 08:33:57 am
From: larry.jones@sdrc.com (Larry Jones)
X-Mailer: ELM [version 2.4 PL25]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 657
Lines: 16

Werner LEMBERG writes:
> 
> How does it slow down?  This wasn't my change, BTW, I've taken it from
> the HP-UX port (the link you've given in one of your recent mails),
> and according to the various man pages, using ConnectionNumber() seems
> the right thing to do to be portable.  Can someone comment this
> please?

In X11R6 at least (I no longer have X11R5 online), ConnectionNumber is
defined as a macro in X11/Xlib.h that expands to essentially the same
code, so it's hard to imagine it causing any slow down unless someone
neglected to #include X11/Xlib.h.

-Larry Jones

Hello, local Navy recruitment office?  Yes, this is an emergency... -- Calvin

From groff-admin  Fri Oct  6 20:44:25 2000
Received: from sdrc.com (heimdall.sdrc.com [146.122.132.195])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id UAA10059
	for <groff@ffii.org>; Fri, 6 Oct 2000 20:44:24 +0200
X-Authentication-Warning: genba.ffii.org: Host heimdall.sdrc.com [146.122.132.195] claimed to be sdrc.com
Received: from tyr.sdrc.com (mailhub-cvg.sdrc.com [146.122.142.31])
	by sdrc.com (8.9.1/8.9.1) with ESMTP id OAA25802;
	Fri, 6 Oct 2000 14:47:39 -0400 (EDT)
Received: from thor.sdrc.com (scjones@thor.sdrc.com [146.122.62.108])
	by tyr.sdrc.com (8.8.6 (PHNE_17190)/8.8.5) with ESMTP id OAA27263;
	Fri, 6 Oct 2000 14:47:38 -0400 (EDT)
Received: (from scjones@localhost)
	by thor.sdrc.com (8.9.0/8.9.0) id OAA03800;
	Fri, 6 Oct 2000 14:47:38 -0400 (EDT)
Message-Id: <200010061847.OAA03800@thor.sdrc.com>
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
To: wl@gnu.org (Werner LEMBERG)
Date: Fri, 6 Oct 2000 14:47:37 -0400 (EDT)
Cc: daniel@bicho.synchrods.com, baruchel@libertysurf.fr, groff@ffii.org
In-Reply-To: <20001006.083357.122620281.wl@gnu.org> from "Werner LEMBERG" at Oct 6, 0 08:33:57 am
From: larry.jones@sdrc.com (Larry Jones)
X-Mailer: ELM [version 2.4 PL25]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 602
Lines: 23

> I just discovered that a change made in main.c for xgremlin-2.0
> in line 592:
> 
>   readmask = ConnectionNumber(display);/*1 << display->fd;*/ 
> 
> slows down the program VERY much, so I went back to the old line,
> that is:
> 
> 		readmask = 1 << display->fd;

I just realized that the change is wrong -- it should be:

	readmask = 1 << ConnectionNumber(display);

Actually, assuming that readmask is a mask for select, it should
probably be more like:

	FD_ZERO(&readmask);
	FD_SET(ConnectionNumber(display), &readmask);

-Larry Jones

You're just trying to get RID of me, aren't you? -- Calvin

From groff-admin  Fri Oct  6 21:32:51 2000
Received: from srv1.tsoft.net (news.tsoft.net [198.144.192.3])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id VAA10279
	for <groff@ffii.org>; Fri, 6 Oct 2000 21:32:50 +0200
X-Authentication-Warning: genba.ffii.org: Host news.tsoft.net [198.144.192.3] claimed to be srv1.tsoft.net
Received: from bicho.SynchroDS.COM (daniel.ppp.tsoft.com [198.144.199.225])
	by srv1.tsoft.net (8.8.7/8.8.7) with ESMTP id MAA03120;
	Fri, 6 Oct 2000 12:36:02 -0700 (PDT)
Received: (from daniel@localhost)
	by bicho.SynchroDS.COM (8.8.8/8.8.8) id MAA05949;
	Fri, 6 Oct 2000 12:35:46 -0700 (PDT)
Date: Fri, 6 Oct 2000 12:35:46 -0700 (PDT)
From: Daniel Senderowicz <daniel@bicho.SynchroDS.COM>
Message-Id: <200010061935.MAA05949@bicho.SynchroDS.COM>
To: larry.jones@sdrc.com, wl@gnu.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
Cc: baruchel@libertysurf.fr, groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 722
Lines: 26

>> I just discovered that a change made in main.c for xgremlin-2.0
>> in line 592:
>> 
>>   readmask = ConnectionNumber(display);/*1 << display->fd;*/ 
>> 
>> slows down the program VERY much, so I went back to the old line,
>> that is:
>> 
>> 		readmask = 1 << display->fd;
>
>I just realized that the change is wrong -- it should be:
>
>	readmask = 1 << ConnectionNumber(display);

This one works fine.

>Actually, assuming that readmask is a mask for select, it should
>probably be more like:
>
>	FD_ZERO(&readmask);
>	FD_SET(ConnectionNumber(display), &readmask);

This couple of lines give the following error message:

main.c:595: request for member `fds_bits' in something not a structure or union
*** Error code 1

From groff-admin  Fri Oct  6 21:54:59 2000
Received: from sdrc.com (heimdall.sdrc.com [146.122.132.195])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id VAA10392
	for <groff@ffii.org>; Fri, 6 Oct 2000 21:54:58 +0200
X-Authentication-Warning: genba.ffii.org: Host heimdall.sdrc.com [146.122.132.195] claimed to be sdrc.com
Received: from tyr.sdrc.com (mailhub-cvg.sdrc.com [146.122.142.31])
	by sdrc.com (8.9.1/8.9.1) with ESMTP id PAA03343;
	Fri, 6 Oct 2000 15:58:00 -0400 (EDT)
Received: from thor.sdrc.com (scjones@thor.sdrc.com [146.122.62.108])
	by tyr.sdrc.com (8.8.6 (PHNE_17190)/8.8.5) with ESMTP id PAA23149;
	Fri, 6 Oct 2000 15:57:58 -0400 (EDT)
Received: (from scjones@localhost)
	by thor.sdrc.com (8.9.0/8.9.0) id PAA03907;
	Fri, 6 Oct 2000 15:57:58 -0400 (EDT)
Message-Id: <200010061957.PAA03907@thor.sdrc.com>
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
To: daniel@bicho.SynchroDS.COM (Daniel Senderowicz)
Date: Fri, 6 Oct 2000 15:57:58 -0400 (EDT)
Cc: wl@gnu.org, baruchel@libertysurf.fr, groff@ffii.org
In-Reply-To: <200010061935.MAA05949@bicho.SynchroDS.COM> from "Daniel Senderowicz" at Oct 6, 0 12:35:46 pm
From: larry.jones@sdrc.com (Larry Jones)
X-Mailer: ELM [version 2.4 PL25]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 951
Lines: 24

Daniel Senderowicz writes (quoting me):
> 
> >Actually, assuming that readmask is a mask for select, it should
> >probably be more like:
> >
> >	FD_ZERO(&readmask);
> >	FD_SET(ConnectionNumber(display), &readmask);
> 
> This couple of lines give the following error message:
> 
> main.c:595: request for member `fds_bits' in something not a structure or union
> *** Error code 1

Which means that readmask isn't declared correctly -- it's probably
declared as an unsigned int or an unsigned long (or an array of them),
rather than being declared as an fd_set.  The original definition of
select used primitive types, but it quickly became apparent that that
created all kinds of portability problems so the fd_set type and the
FD_* macros were added to encapsulate all of the messy stuff. 
Unfortunately, a lot of programmers never bothered to learn to use them.

-Larry Jones

Things are never quite as scary when you've got a best friend. -- Calvin

From groff-admin  Fri Oct  6 22:57:03 2000
Received: from cmailg1.svr.pol.co.uk (cmailg1.svr.pol.co.uk [195.92.195.171])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id WAA10823
	for <groff@ffii.org>; Fri, 6 Oct 2000 22:57:03 +0200
Received: from [195.92.198.123] (helo=mail17.svr.pol.co.uk)
	by cmailg1.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13hebR-0007h3-00
	for groff@ffii.org; Fri, 06 Oct 2000 22:00:26 +0100
Received: from modem-59.angrod.dialup.pol.co.uk ([62.136.114.59] helo=inputplus.demon.co.uk)
	by mail17.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13hebR-00069d-00
	for groff@ffii.org; Fri, 06 Oct 2000 22:00:25 +0100
Received: from inputplus (ralph@localhost [127.0.0.1])
	by inputplus.demon.co.uk (8.8.7/8.8.7) with ESMTP id VAA08721;
	Fri, 6 Oct 2000 21:23:02 +0100
Message-Id: <200010062023.VAA08721@inputplus.demon.co.uk>
To: Werner LEMBERG <wl@gnu.org>
cc: daniel@bicho.synchrods.com, baruchel@libertysurf.fr, groff@ffii.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ? 
In-Reply-To: Message from Werner LEMBERG <wl@gnu.org> 
   of "Fri, 06 Oct 2000 08:33:57 +0200." <20001006.083357.122620281.wl@gnu.org> 
Date: Fri, 06 Oct 2000 21:23:02 +0100
From: Ralph Corderoy <ralph@inputplus.demon.co.uk>
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1336
Lines: 40

Hi,

> > I just discovered that a change made in main.c for xgremlin-2.0
> > in line 592:
> > 
> >   readmask = ConnectionNumber(display);/*1 << display->fd;*/ 
> > 
> > slows down the program VERY much, so I went back to the old line,
> > that is:
> > 
> >               readmask = 1 << display->fd;
> 
> ...according to the various man pages, using ConnectionNumber() seems
> the right thing to do to be portable.  Can someone comment this
> please?

ConnectionNumber() is the way to get the display's file descriptor.  It
is traditionally just a macro in Xlib.h that gives `display->fd' but
the macro is meant to be used instead.  There's also a function
equivalent, XConnectionNumber().

    % g ConnectionNumber /usr/include/X11/Xlib.h
    #define ConnectionNumber(dpy)   (((_XPrivDisplay)dpy)->fd)
    extern int XConnectionNumber(
    extern Status XInternalConnectionNumbers(

The problem here is ConnectionNumber() and display->fd give a file
descriptor.  readmask presumably needs a mask, hence `1 <<
display->fd'.  If ConnectionNumber() is used then the bitwise shift is
still required to create a mask.

    readmask = 1 << ConnectionNumber(display);

Since display->fd is probably something with a few bits set the
checking of what bits select(2) or something else has left set is
probably taking the extra time.


Ralph.


From groff-admin  Sat Oct  7 09:16:25 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA30652
	for <groff@ffii.org>; Sat, 7 Oct 2000 09:16:25 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sat, 7 Oct 2000 09:19:31 +0200
Received: from rigel.univie.ac.at (dial-143195.HRZ.Uni-Dortmund.DE [129.217.143.195])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e977JSC07320;
	Sat, 7 Oct 2000 09:19:28 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id JAA10150;
	Sat, 7 Oct 2000 09:19:21 +0200
Date: Sat, 07 Oct 2000 09:19:21 +0200 (CEST)
Message-Id: <20001007.091921.41635733.wl@gnu.org>
To: larry.jones@sdrc.com
Cc: daniel@bicho.synchrods.com, baruchel@libertysurf.fr, groff@ffii.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <200010061847.OAA03800@thor.sdrc.com>
References: <20001006.083357.122620281.wl@gnu.org>
	<200010061847.OAA03800@thor.sdrc.com>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 451
Lines: 17

> I just realized that the change is wrong -- it should be:
> 
> 	readmask = 1 << ConnectionNumber(display);
> 
> Actually, assuming that readmask is a mask for select, it should
> probably be more like:
> 
> 	FD_ZERO(&readmask);
> 	FD_SET(ConnectionNumber(display), &readmask);

And this is exactly the code used in the xgremlin version I've
uploaded to ftp.ffii.org:/pub/groff.

Daniel, have you actually tried this particular version?


    Werner

From groff-admin  Sat Oct  7 10:26:26 2000
Received: from srv1.tsoft.net (news.tsoft.net [198.144.192.3])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id KAA31548
	for <groff@ffii.org>; Sat, 7 Oct 2000 10:26:25 +0200
X-Authentication-Warning: genba.ffii.org: Host news.tsoft.net [198.144.192.3] claimed to be srv1.tsoft.net
Received: from bicho.SynchroDS.COM (daniel.ppp.tsoft.com [198.144.199.225])
	by srv1.tsoft.net (8.8.7/8.8.7) with ESMTP id BAA21243;
	Sat, 7 Oct 2000 01:29:45 -0700 (PDT)
Received: (from daniel@localhost)
	by bicho.SynchroDS.COM (8.8.8/8.8.8) id BAA07148;
	Sat, 7 Oct 2000 01:29:29 -0700 (PDT)
Date: Sat, 7 Oct 2000 01:29:29 -0700 (PDT)
From: Daniel Senderowicz <daniel@bicho.SynchroDS.COM>
Message-Id: <200010070829.BAA07148@bicho.SynchroDS.COM>
To: larry.jones@sdrc.com, wl@gnu.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
Cc: baruchel@libertysurf.fr, daniel@bicho.SynchroDS.COM, groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1481
Lines: 45

>From groff-admin@ffii.org Sat Oct  7 01:21:32 2000
>Date: Sat, 07 Oct 2000 09:19:21 +0200 (CEST)
>To: larry.jones@sdrc.com
>Cc: daniel@bicho.synchrods.com, baruchel@libertysurf.fr, groff@ffii.org
>Subject: Re: [Groff] still exploring the last version: what is gremlin ?
>From: Werner LEMBERG <wl@gnu.org>
>In-Reply-To: <200010061847.OAA03800@thor.sdrc.com>
>References: <20001006.083357.122620281.wl@gnu.org>
>	<200010061847.OAA03800@thor.sdrc.com>
>X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
>Mime-Version: 1.0
>Content-Type: Text/Plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>Sender: groff-admin@ffii.org
>Errors-To: groff-admin@ffii.org
>X-BeenThere: groff@ffii.org
>X-Mailman-Version: 2.0beta2.local2
>Precedence: bulk
>List-Id: GNU troff discussion list <groff.ffii.org>
>
>> I just realized that the change is wrong -- it should be:
>> 
>> 	readmask = 1 << ConnectionNumber(display);
>> 
>> Actually, assuming that readmask is a mask for select, it should
>> probably be more like:
>> 
>> 	FD_ZERO(&readmask);
>> 	FD_SET(ConnectionNumber(display), &readmask);
>
>And this is exactly the code used in the xgremlin version I've
>uploaded to ftp.ffii.org:/pub/groff.
>
>Daniel, have you actually tried this particular version?

Yes. I had that slowdown problem with the code in
question. After I made the value of

	readmask = 1 << ConnectionNumber(display);

everything went smoothly. Now I look like Michelangelo but on the
computer.

Dan


From groff-admin  Sat Oct  7 13:42:29 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id NAA32544
	for <groff@ffii.org>; Sat, 7 Oct 2000 13:42:28 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sat, 7 Oct 2000 13:45:55 +0200
Received: from rigel.univie.ac.at (dial-143198.HRZ.Uni-Dortmund.DE [129.217.143.198])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e97BjeC10816
	for <groff@ffii.org>; Sat, 7 Oct 2000 13:45:44 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA10422
	for <groff@ffii.org>; Sat, 7 Oct 2000 10:43:02 +0200
Date: Sat, 07 Oct 2000 10:43:02 +0200 (CEST)
Message-Id: <20001007.104302.104041739.wl@gnu.org>
To: groff@ffii.org
From: Werner LEMBERG <wl@gnu.org>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] new, new, new
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 583
Lines: 37

Dear friends,


I've finally committed various new requests and a new escape:

  .  dei xx yy

     Define indirect.  The sequence

       .ds xx aa
       .ds yy bb
       .dei xx yy

     is equivalent to

       .de aa bb

  .  ecs

     Save current escape character

  .  ecr

     Restore escape character saved with `ecs'.

  .  \B'anything'

     This is an analogon to `\A': If _anything_ is a valid numeric
     expression, return `1', and `0' otherwise.

tmac.trace will now use .dei, .ecs, and .ecr to be independent from
the escape character.

Please test!


    Werner

From groff-admin  Sun Oct  8 11:04:36 2000
Received: from ukulele.linux.home (root@dial01.po.opole.pl [193.59.208.15])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id LAA03881
	for <groff@ffii.org>; Sun, 8 Oct 2000 11:04:30 +0200
X-Authentication-Warning: genba.ffii.org: Host root@dial01.po.opole.pl [193.59.208.15] claimed to be ukulele.linux.home
Received: from guitar.linux.home (root@guitar.linux.home [10.0.0.12])
	by ukulele.linux.home (8.9.3/8.9.3/Debian/GNU) with ESMTP id LAA15944;
	Sat, 7 Oct 2000 11:39:06 +0200
Received: from localhost (luke@localhost)
	by guitar.linux.home (8.8.6/8.8.7) with SMTP id LAA00197;
	Sat, 7 Oct 2000 11:18:44 +0100
Date: Sat, 7 Oct 2000 11:18:44 +0100 (MET)
From: Lukasz Wiechec <magic@bet.po.opole.pl>
To: Werner LEMBERG <wl@gnu.org>
cc: Groff mailinglist <groff@ffii.org>
Subject: Re: [Groff] Hello to you all !
In-Reply-To: <20001005.091720.106825793.wl@gnu.org>
Message-ID: <Pine.LNX.3.96.1001007110612.172A-100000@guitar.linux.home>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1249
Lines: 31

On Thu, 5 Oct 2000, Werner LEMBERG wrote:

> > The problem is that some (exacly three) Polish characters are being
> > interpreted as symbols (and therefore displayed using Symbol special
> > font).  This is done only using '-Tps' (when using '-Tlatin1', tty
> > output is OK).
> 
> Which characters (or rather, glyphs)?  Example, please.

Actually, I've got a problem with four: char177 (which should be aogonek
but displays as plusminus [+-]), char179 (lslash, displaying as
3-superscript or power 3), char188 (zslash, displays as 1/4 or onequarter)
and char166 (Sslash, displaying as vertical bar or '|').

These chars are standard Latin-2 encoded polish characters.

Not all of abovementioned glyphs are displayed using Symbol fonts (I've
checked groff's postscript output, and as far as I can make it out, only
aogonek [or char 177] is displayed as symbol), but anyway, I would like to
make groff to interpret literally what's been Latin-2-encoded.

I don't know whether attaching postscript would be helpful (and lawful;
it's 5K long :-), but if so, let me know.

Big thanks for the response, greetings from Poland,

		------------------------
		MAGIC aka Lukasz Wiechec
		Internet mailto:magic@bet.po.opole.pl
		tel/fax  +(48 77) 4 555 113 


From groff-admin  Sun Oct  8 11:42:11 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id LAA03961
	for <groff@ffii.org>; Sun, 8 Oct 2000 11:42:11 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sun, 8 Oct 2000 11:45:27 +0200
Received: from rigel.univie.ac.at (dial-143179.HRZ.Uni-Dortmund.DE [129.217.143.179])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e989jPC23310;
	Sun, 8 Oct 2000 11:45:25 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id NAA11236;
	Sun, 8 Oct 2000 13:36:09 +0200
Date: Sun, 08 Oct 2000 13:36:08 +0200 (CEST)
Message-Id: <20001008.133608.112617136.wl@gnu.org>
To: magic@bet.po.opole.pl
Cc: groff@ffii.org
Subject: Re: [Groff] Hello to you all !
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <Pine.LNX.3.96.1001007110612.172A-100000@guitar.linux.home>
References: <20001005.091720.106825793.wl@gnu.org>
	<Pine.LNX.3.96.1001007110612.172A-100000@guitar.linux.home>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 855
Lines: 22

> > Which characters (or rather, glyphs)?  Example, please.
> 
> Actually, I've got a problem with four: char177 (which should be
> aogonek but displays as plusminus [+-]), char179 (lslash, displaying
> as 3-superscript or power 3), char188 (zslash, displays as 1/4 or
> onequarter) and char166 (Sslash, displaying as vertical bar or '|').
> 
> These chars are standard Latin-2 encoded polish characters.

In case you are trying to use Latin-2 as an input character set you
will have some difficulties currently since Latin-1 is hard-coded into
groff (I'll try to change this in the near future)...

Or are you using a patched groff with Latin-2 support?

> I don't know whether attaching postscript would be helpful (and
> lawful; it's 5K long :-), but if so, let me know.

Please send me an input and output file (compressed if necessary).


    Werner

From groff-admin  Sun Oct  8 15:44:39 2000
Received: from Hesperia (ppp64-brest.libertysurf.fr [213.36.44.64])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA05330
	for <groff@ffii.org>; Sun, 8 Oct 2000 15:44:37 +0200
X-Authentication-Warning: genba.ffii.org: Host ppp64-brest.libertysurf.fr [213.36.44.64] claimed to be Hesperia
Received: from baruchel by Hesperia with local (Exim 2.05 #1 (Debian))
	id 13iDij-00033f-00; Sun, 8 Oct 2000 12:30:17 +0200
Date: Sun, 8 Oct 2000 12:30:16 +0200
From: Thomas Baruchel <baruchel@libertysurf.fr>
To: groff@ffii.org
Subject: Re: [Groff] Hello to you all ! a digression...
Message-ID: <20001008123016.A11740@Hesperia>
Reply-To: Thomas Baruchel <baruchel@libertysurf.fr>
Mail-Followup-To: groff@ffii.org
References: <20001005.091720.106825793.wl@gnu.org> <Pine.LNX.3.96.1001007110612.172A-100000@guitar.linux.home> <20001008.133608.112617136.wl@gnu.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <20001008.133608.112617136.wl@gnu.org>; from wl@gnu.org on Sun, Oct 08, 2000 at 01:36:08PM +0200
X-Machine: Hesperia sur Debian (Linux)
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1583
Lines: 39

On Sun, Oct 08, 2000 at 01:36:08PM +0200, Werner LEMBERG wrote:
> > > Which characters (or rather, glyphs)?  Example, please.
> > 
> > Actually, I've got a problem with four: char177 (which should be
> > aogonek but displays as plusminus [+-]), char179 (lslash, displaying
> > as 3-superscript or power 3), char188 (zslash, displays as 1/4 or
> > onequarter) and char166 (Sslash, displaying as vertical bar or '|').
> > 
> > These chars are standard Latin-2 encoded polish characters.
> 
> In case you are trying to use Latin-2 as an input character set you
> will have some difficulties currently since Latin-1 is hard-coded into
> groff (I'll try to change this in the near future)...
> 
> Or are you using a patched groff with Latin-2 support?

Brest, le dimanche 8 octobre

No real relation with the current subject... sorry :-(
(still a little ;-)

I just bought a (french) book about (french) ponctuation: history, usage, etc.
Very interesting, but... it is told about characters
> > > (or rather, glyphs)?
that people used centuries ago but which have today disappeared. Does someone
know what is for instance a fly-foot (in french: "pied-de-mouche" where "mouche"
which I have translated by "fly" is the little insect). It was used to mark a
separation between twn paragraphs when copists didn't have enoug place on their
page to write a real paragraph. Is their an URL where I could see images of the
old punctuation characters ?

Sorry to have disturbed the list :-(

-- 
   .~.   Thomas Baruchel
   /V\   baruchel@libertysurf.fr
  // \\    Brest
 /(   )\   FRANCE
  ^`~'^

From groff-admin  Mon Oct  9 04:55:39 2000
Received: from polo.po.opole.pl (root@polo.po.opole.pl [193.59.208.10])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id EAA09900
	for <groff@ffii.org>; Mon, 9 Oct 2000 04:55:38 +0200
Received: from ukulele.linux.home (postman@dial01.po.opole.pl [193.59.208.15])
	by polo.po.opole.pl (8.8.8/8.8.6/ag-polo.981012-sz+rchk1.22) with ESMTP id EAA16788;
	Mon, 9 Oct 2000 04:59:04 +0200 (MET DST)
Received: from guitar.linux.home (root@guitar.linux.home [10.0.0.12])
	by ukulele.linux.home (8.9.3/8.9.3/Debian/GNU) with ESMTP id TAA19354;
	Sun, 8 Oct 2000 19:23:11 +0200
Received: from localhost (luke@localhost)
	by guitar.linux.home (8.8.6/8.8.7) with SMTP id TAA00265;
	Sun, 8 Oct 2000 19:16:59 +0100
Date: Sun, 8 Oct 2000 19:16:59 +0100 (MET)
From: Lukasz Wiechec <magic@bet.po.opole.pl>
To: Werner LEMBERG <wl@gnu.org>
cc: Groff mailinglist <groff@ffii.org>
Subject: Re: [Groff] Latin-2 woes...
In-Reply-To: <20001008.133608.112617136.wl@gnu.org>
Message-ID: <Pine.LNX.3.96.1001008190006.211B-201000@guitar.linux.home>
MIME-Version: 1.0
Content-Type: MULTIPART/MIXED; BOUNDARY="658432-708338397-971029019=:211"
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 9861
Lines: 182

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.
  Send mail to mime@docserver.cac.washington.edu for more info.

--658432-708338397-971029019=:211
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Sun, 8 Oct 2000, Werner LEMBERG wrote:

> > > Which characters (or rather, glyphs)?  Example, please.
> > 
> > Actually, I've got a problem with four: char177 (which should be
> > aogonek but displays as plusminus [+-]), char179 (lslash, displaying
> > as 3-superscript or power 3), char188 (zslash, displays as 1/4 or
> > onequarter) and char166 (Sslash, displaying as vertical bar or '|').
> > 
> > These chars are standard Latin-2 encoded polish characters.
> 
> In case you are trying to use Latin-2 as an input character set you
> will have some difficulties currently since Latin-1 is hard-coded into
> groff (I'll try to change this in the near future)...

So maybe there is a way to use control sequences
 
> Or are you using a patched groff with Latin-2 support?

No, right now I'm using standard 1.16 release... is any available ?  

> Please send me an input and output file (compressed if necessary).

OK, I'm attaching it, but to display it properly you'll need polish
Postscript font (which I took out of 'ogonkify' package...), but if you
look on line 205, where charcodes are displayed you'll see that 7th byte
is 0x7c (decimal 124), and it should be 166 (character for Sacute). In the
same line the font is being changed to Symbol, without being asked to...

Thanks for the response and greetings from Poland,
 
		------------------------
		MAGIC aka Lukasz Wiechec
		Internet mailto:magic@bet.po.opole.pl
		tel/fax  +(48 77) 4 555 113 

--658432-708338397-971029019=:211
Content-Type: APPLICATION/PostScript; name="p1.ps"
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.LNX.3.96.1001008191659.211C@guitar.linux.home>
Content-Description: Postscript output
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--658432-708338397-971029019=:211
Content-Type: TEXT/PLAIN; charset=ISO-8859-2; name="p1.mm"
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.LNX.3.96.1001008191659.211D@guitar.linux.home>
Content-Description: groff source

LnBzICs1DQouY2UNCi5mcCAwIE9UUg0KVGhpcyBpcyBMYXRpbi0yLWVuY29k
ZWQgdGV4dCBjb250YWluaW5nIGFsbCBQb2xpc2ggY2hhcmFjdGVycy4NCg0K
LlwiIEZvbnR5IENvdXJpZXINCi5mdCBSDQqhxsqj0dOmrK9cZjGx5uqz8fO2
vL8NCg==
--658432-708338397-971029019=:211--

From groff-admin  Mon Oct  9 13:00:17 2000
Received: from clyde.ref.collins.co.uk (clyde.ref.collins.co.uk [192.153.153.29])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id NAA31556
	for <groff@ffii.org>; Mon, 9 Oct 2000 13:00:13 +0200
Received: from ref.collins.co.uk (helmsdale [193.119.38.15])
	by clyde.ref.collins.co.uk (8.9.3/8.9.3) with ESMTP id MAA17301;
	Mon, 9 Oct 2000 12:03:10 +0100 (BST)
Received: from ref.collins.co.uk (localhost [127.0.0.1])
	by ref.collins.co.uk (8.9.3+Sun/8.9.1) with ESMTP id MAA01036;
	Mon, 9 Oct 2000 12:02:05 +0100 (BST)
Message-ID: <39E1A5AD.79C61EE3@ref.collins.co.uk>
Date: Mon, 09 Oct 2000 12:02:05 +0100
From: "Stewart C. Russell" <stewart@ref.collins.co.uk>
Organization: HarperCollins Publishers, Bilingual Division
X-Mailer: Mozilla 4.61 [en] (X11; U; SunOS 5.7 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
To: groff@ffii.org
Subject: Re: [Groff] Hello to you all ! a digression...
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1305
Lines: 32

Thomas Baruchel wrote:
> Does someone
> know what is for instance a fly-foot (in french: "pied-de-mouche" where "mouche"
> which I have translated by "fly" is the little insect). It was used to mark a
> separation between two paragraphs

I think we call them pilcrow, and they look like this: ¶ (source:
http://members.aol.com/alembicprs/glocase.htm)

Alternatively, this may be the character (I think) we call the
'viniculum', that little leafy-thing that's in Zapf Dingbats. That's
what I think English* typesetters used. I shall check in the Cambridge
Encyclopedia of the English Language, which has a nifty bit on
historical typesetting.

French typography is interesting. Its spacing is very different from
British/American. Glad to see that someone somewhere has stuck to their
guns.

 Stewart


*: we Scots tended to use French typesetters. They were cheap and quick,
the DTP of the day. Sadly we lost a few of our more colourful letters to
misunderstandings. We had a character that looked a bit like a cursive Z
which was sounded 'ng'; French typesetters just used a Z, and now we
have wacky things like "Menzies" pronounced "Mingus"...

-- 
Stewart C. Russell		Senior Analyst, Dictionary Division
stewart@ref.collins.co.uk	HarperCollins Publishers
use Disclaimer; my $opinion;	Glasgow, Scotland

From groff-admin  Mon Oct  9 21:18:31 2000
Received: from burns.conectiva (brutus.conectiva.com.br [200.250.58.146])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with SMTP id VAA03507
	for <groff@ffii.org>; Mon, 9 Oct 2000 21:18:29 +0200
X-Authentication-Warning: genba.ffii.org: Host brutus.conectiva.com.br [200.250.58.146] claimed to be burns.conectiva
Received: (qmail 17676 invoked from network); 9 Oct 2000 19:21:31 -0000
Received: from soneca.conectiva (HELO conectiva.com.br) (10.0.2.41)
  by burns.conectiva with SMTP; 9 Oct 2000 19:21:31 -0000
Message-ID: <39E21AD7.1023DDE2@conectiva.com.br>
Date: Mon, 09 Oct 2000 16:21:59 -0300
From: Ricardo <ricardo@conectiva.com.br>
Reply-To: ricardo@conectiva.com.br
Organization: Conectiva SA
X-Mailer: Mozilla 4.75 [pt_BR] (X11; U; Linux 2.2.16-13cl i586)
X-Accept-Language: pt-BR
MIME-Version: 1.0
To: Ted.Harding@nessie.mcc.ac.uk
CC: groff@ffii.org
Subject: Re: [Groff] top of page
References: <XFMail.000930100450.Ted.Harding@nessie.mcc.ac.uk>
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 222
Lines: 12

Hi,

Sorry the delay ;)

I will check your file, and if the "rendering" works, i warn you ;)

Hugs,
-- 
------------------------------
Ricardo Soares Guimarães
http://ldp-br.conectiva.com.br
------------------------------

From groff-admin  Mon Oct  9 21:40:21 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id VAA03672
	for <groff@ffii.org>; Mon, 9 Oct 2000 21:40:20 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Mon, 9 Oct 2000 21:43:49 +0200
Received: from rigel.univie.ac.at (dial-143137.HRZ.Uni-Dortmund.DE [129.217.143.137])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e99JhgC08091;
	Mon, 9 Oct 2000 21:43:42 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id OAA12416;
	Mon, 9 Oct 2000 14:02:14 +0200
Date: Mon, 09 Oct 2000 14:02:14 +0200 (CEST)
Message-Id: <20001009.140214.112291815.wl@gnu.org>
To: ricardo@conectiva.com.br
Cc: bernd@mycity.at, groff@ffii.org, Ted.Harding@nessie.mcc.ac.uk
Subject: Re: [Groff] Re: GROFF
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <39D0938E.A783DB6E@conectiva.com.br>
References: <200009252230.AAA00316@schleppo.sun.universe>
	<39D0938E.A783DB6E@conectiva.com.br>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1028
Lines: 26

> Bernd Salbrechter wrote:
> 
> > If this is added, it would make sense to have a way to access with
> > troff request also, to make the macro packages i18n.  I.e., to
> > print "Inhaltsverzeichnis" for "Table of Contence".
> > 
> > Question: can a part of the string namespace be reserved for that.
> > I.e., reserve "\*[nls.catalog.set.msg]" to get value a nls message
> > (sorry I don't know gettext enough for giving an example with that
> > tool).
> 
> Yes. I use some nls macros in my work. in the moment that i call
> groff, i pass my "generic" macro, and my "nls" macro (tmac.pt_BR or
> tmac.es_ES or tmac.fr...).

No.  gettext stuff should be used only for warning and error messages
coming from the troff program and its friends, sent to stderr.  It
makes no sense IMHO to provide a special interface so that macro
packages can access these warning messages -- this is not possible for
English either.

Rather, it has to be solved on the macro level similar to LaTeX.
Ricardo's solution is the right one.


    Werner

From groff-admin  Mon Oct  9 21:40:21 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id VAA03673
	for <groff@ffii.org>; Mon, 9 Oct 2000 21:40:20 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Mon, 9 Oct 2000 21:43:51 +0200
Received: from rigel.univie.ac.at (dial-143137.HRZ.Uni-Dortmund.DE [129.217.143.137])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e99JhnC08095;
	Mon, 9 Oct 2000 21:43:49 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA12345;
	Mon, 9 Oct 2000 10:16:40 +0200
Date: Mon, 09 Oct 2000 10:16:39 +0200 (CEST)
Message-Id: <20001009.101639.48526987.wl@gnu.org>
To: Andrej.Borsenkow@mow.siemens.ru, groff@ffii.org
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <000201c029ee$15eddfc0$21c9ca95@mow.siemens.ru>
References: <20000928.175421.10906604.wl@gnu.org>
	<000201c029ee$15eddfc0$21c9ca95@mow.siemens.ru>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] Re: Current CVS: incorrect re-decrlaration of errno
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 218
Lines: 6

I've removed the `#ifndef errno ...' stuff altogether.  I believe that
modern compilers no longer need this.  If you find an old compiler
which fails please tell me so that I can add appropriate #ifdef's.


    Werner

From groff-admin  Mon Oct  9 21:41:07 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id VAA03702
	for <groff@ffii.org>; Mon, 9 Oct 2000 21:41:06 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Mon, 9 Oct 2000 21:44:11 +0200
Received: from rigel.univie.ac.at (dial-143137.HRZ.Uni-Dortmund.DE [129.217.143.137])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e99JhsC08102;
	Mon, 9 Oct 2000 21:43:54 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA12103;
	Mon, 9 Oct 2000 10:05:50 +0200
Date: Mon, 09 Oct 2000 10:05:50 +0200 (CEST)
Message-Id: <20001009.100550.115917093.wl@gnu.org>
To: daniel@bicho.synchrods.com
Cc: larry.jones@sdrc.com, baruchel@libertysurf.fr, groff@ffii.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <200010070829.BAA07148@bicho.SynchroDS.COM>
References: <200010070829.BAA07148@bicho.SynchroDS.COM>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 539
Lines: 19

> >> Actually, assuming that readmask is a mask for select, it should
> >> probably be more like:
> >> 
> >> 	FD_ZERO(&readmask);
> >> 	FD_SET(ConnectionNumber(display), &readmask);
> 
> After I made the value of
> 
> 	readmask = 1 << ConnectionNumber(display);
> 
> everything went smoothly. Now I look like Michelangelo but on the
> computer.

What shall I do?  I believe that the code in gremlin is OK, and that
your problem is related to your version of BSD -- can you please ask a
BSD guru how to implement it correctly?


    Werner

From groff-admin  Mon Oct  9 22:46:22 2000
Received: from sdrc.com (heimdall.sdrc.com [146.122.132.195])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id WAA04211
	for <groff@ffii.org>; Mon, 9 Oct 2000 22:46:21 +0200
X-Authentication-Warning: genba.ffii.org: Host heimdall.sdrc.com [146.122.132.195] claimed to be sdrc.com
Received: from tyr.sdrc.com (mailhub-cvg.sdrc.com [146.122.142.31])
	by sdrc.com (8.9.1/8.9.1) with ESMTP id QAA13758;
	Mon, 9 Oct 2000 16:49:39 -0400 (EDT)
Received: from thor.sdrc.com (scjones@thor.sdrc.com [146.122.62.108])
	by tyr.sdrc.com (8.8.6 (PHNE_17190)/8.8.5) with ESMTP id QAA12813;
	Mon, 9 Oct 2000 16:49:37 -0400 (EDT)
Received: (from scjones@localhost)
	by thor.sdrc.com (8.9.0/8.9.0) id QAA01208;
	Mon, 9 Oct 2000 16:49:36 -0400 (EDT)
Message-Id: <200010092049.QAA01208@thor.sdrc.com>
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
To: wl@gnu.org (Werner LEMBERG)
Date: Mon, 9 Oct 2000 16:49:36 -0400 (EDT)
Cc: daniel@bicho.synchrods.com, baruchel@libertysurf.fr, groff@ffii.org
In-Reply-To: <20001009.100550.115917093.wl@gnu.org> from "Werner LEMBERG" at Oct 9, 0 10:05:50 am
From: larry.jones@sdrc.com (Larry Jones)
X-Mailer: ELM [version 2.4 PL25]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 492
Lines: 13

Werner LEMBERG writes:
> 
> What shall I do?  I believe that the code in gremlin is OK, and that
> your problem is related to your version of BSD -- can you please ask a
> BSD guru how to implement it correctly?

I also believe that the most recent gremlin code is correct, but that
his problem is that he really hasn't tried that version.  Once again,
it's ftp://ftp.ffii.org/pub/groff/xgremlin2-20001003.tar.bz2.

-Larry Jones

I just can't identify with that kind of work ethic. -- Calvin

From groff-admin  Tue Oct 10 09:08:22 2000
Received: from clyde.ref.collins.co.uk (clyde.ref.collins.co.uk [192.153.153.29])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA21501
	for <groff@ffii.org>; Tue, 10 Oct 2000 09:08:20 +0200
Received: from ref.collins.co.uk (helmsdale [193.119.38.15])
	by clyde.ref.collins.co.uk (8.9.3/8.9.3) with ESMTP id IAA19149
	for <groff@ffii.org>; Tue, 10 Oct 2000 08:11:21 +0100 (BST)
Received: from ref.collins.co.uk (localhost [127.0.0.1])
	by ref.collins.co.uk (8.9.3+Sun/8.9.1) with ESMTP id IAA03011
	for <groff@ffii.org>; Tue, 10 Oct 2000 08:10:52 +0100 (BST)
Message-ID: <39E2C0FC.6BCEAB46@ref.collins.co.uk>
Date: Tue, 10 Oct 2000 08:10:52 +0100
From: "Stewart C. Russell" <stewart@ref.collins.co.uk>
Organization: HarperCollins Publishers, Bilingual Division
X-Mailer: Mozilla 4.61 [en] (X11; U; SunOS 5.7 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
To: groff mail list <groff@ffii.org>
Subject: Re: [Groff] Hello to you all ! a digression...
References: <39E1A5AD.79C61EE3@ref.collins.co.uk>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1407
Lines: 46

"Stewart C. Russell" wrote:
> 
> Alternatively, this may be the character (I think) we call the
> 'viniculum'

As usual, I was talking through my hat. It's called the hedera (ivy
leaf) and used for brief pauses in the text, not paragraph marking. Zapf
Dingbats glyph a107, if you're interested. Looks like this (if you're
using a monospaced mail reader, as one should):

 
     #####   #####       ### 
    ####### ### ###      ####
   #     ### #   #       ####
  #        ###   ##      ####
 ###         ### ##      ### 
  #           ## ##       ## 
 ##            ###       ### 
 ##              ####   ##   
 ###             # # ###     
          ###   #            
       #######     # #       
       ########   ######     
      ########## #######     
      ######### #########    
     #####################   
     #####################   
     #####################   
     ####################    
     #####################   
      ###################    
     ###################     
      #################      
       #############         
       ############          
       #########       #     
         #######      ###    
         #######     ###     
           #####     ###     
           # ###   # ###     
              #######        

(piping a very low res pbm file through "pnmnoraw | tail +3 |\
 tr '01' ' #'" is your friend. Much better than pbmtoascii)

 Stewart

From groff-admin  Tue Oct 10 09:14:28 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA21587
	for <groff@ffii.org>; Tue, 10 Oct 2000 09:14:28 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Tue, 10 Oct 2000 09:17:29 +0200
Received: from rigel.univie.ac.at (dial-143140.HRZ.Uni-Dortmund.DE [129.217.143.140])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9A7HOC18187;
	Tue, 10 Oct 2000 09:17:24 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id XAA14018;
	Mon, 9 Oct 2000 23:26:02 +0200
Date: Mon, 09 Oct 2000 23:26:01 +0200 (CEST)
Message-Id: <20001009.232601.95508706.wl@gnu.org>
To: magic@bet.po.opole.pl
Cc: groff@ffii.org
Subject: Re: [Groff] Latin-2 woes...
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <Pine.LNX.3.96.1001008190006.211B-201000@guitar.linux.home>
References: <20001008.133608.112617136.wl@gnu.org>
	<Pine.LNX.3.96.1001008190006.211B-201000@guitar.linux.home>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-2
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by genba.ffii.org id JAA21588
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 936
Lines: 29

> > > Actually, I've got a problem with four: char177 (which should be
> > > aogonek but displays as plusminus [+-]), char179 (lslash,
> > > displaying as 3-superscript or power 3), char188 (zslash,
> > > displays as 1/4 or onequarter) and char166 (Sslash, displaying
> > > as vertical bar or '|').

This particular problem can be solved easily by saying

  groff <your macro package(s)> -mpsnew ...

Nevertheless, the general solution is to write a proper input encoding
file to map Latin-2 to glyph names; something like

  .char £ \[/L]
  .char ³ \[/l]
  ...

(and inventing glyph names if necessary).  In combination with
properly added Polish PS fonts this should work as expected.

> > Or are you using a patched groff with Latin-2 support?
> 
> No, right now I'm using standard 1.16 release... is any available ?  

Yes, but I no longer have the reference; maybe Staszek
<grciszew@cyf-kr.edu.pl> can give a pointer.


    Werner

From groff-admin  Tue Oct 10 15:08:42 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA24697
	for <groff@ffii.org>; Tue, 10 Oct 2000 15:08:41 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Tue, 10 Oct 2000 15:11:56 +0200
Received: from rigel.univie.ac.at (dial-143131.HRZ.Uni-Dortmund.DE [129.217.143.131])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9ADBsC05419
	for <groff@ffii.org>; Tue, 10 Oct 2000 15:11:54 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id NAA14503
	for <groff@ffii.org>; Tue, 10 Oct 2000 13:53:19 +0200
Date: Tue, 10 Oct 2000 13:53:18 +0200 (CEST)
Message-Id: <20001010.135318.78489442.wl@gnu.org>
To: groff@ffii.org
From: Werner LEMBERG <wl@gnu.org>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] new string comparison?
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 428
Lines: 18

Dear friends,


would it make sense to add a new condition to the `if' request (and
its siblings) to compare strings literally?  Something like

  .ds xxx aaa
  .ds yyy aaa
  .
  .if s "\*[xxx]"\*[yyy]" \
  .  tm true

The strings won't be processed, just expanded (in copy-mode).  For
most macro packages, `processing' of strings isn't necessary;
additionally, comparing strings literally would be (much?) faster.


    Werner

From groff-admin  Tue Oct 10 16:48:47 2000
Received: from cmailg1.svr.pol.co.uk (cmailg1.svr.pol.co.uk [195.92.195.171])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA25249
	for <groff@ffii.org>; Tue, 10 Oct 2000 16:48:47 +0200
Received: from [195.92.198.123] (helo=mail17.svr.pol.co.uk)
	by cmailg1.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13j0lZ-0004ye-00
	for groff@ffii.org; Tue, 10 Oct 2000 15:52:29 +0100
Received: from modem-35.aradan.dialup.pol.co.uk ([62.136.117.163] helo=inputplus.demon.co.uk)
	by mail17.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13j0lY-00001Q-00
	for groff@ffii.org; Tue, 10 Oct 2000 15:52:28 +0100
Received: from inputplus (ralph@localhost [127.0.0.1])
	by inputplus.demon.co.uk (8.8.7/8.8.7) with ESMTP id PAA14664;
	Tue, 10 Oct 2000 15:41:10 +0100
Message-Id: <200010101441.PAA14664@inputplus.demon.co.uk>
To: Werner LEMBERG <wl@gnu.org>
cc: groff@ffii.org
Subject: Re: [Groff] new string comparison? 
In-Reply-To: Message from Werner LEMBERG <wl@gnu.org> 
   of "Tue, 10 Oct 2000 13:53:18 +0200." <20001010.135318.78489442.wl@gnu.org> 
Date: Tue, 10 Oct 2000 15:41:10 +0100
From: Ralph Corderoy <ralph@inputplus.demon.co.uk>
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 881
Lines: 28

Hi,

> would it make sense to add a new condition to the `if' request (and
> its siblings) to compare strings literally?  Something like
> 
>   .ds xxx aaa
>   .ds yyy aaa
>   .
>   .if s "\*[xxx]"\*[yyy]" \
>   .  tm true
> 
> The strings won't be processed, just expanded (in copy-mode).  For
> most macro packages, `processing' of strings isn't necessary;
> additionally, comparing strings literally would be (much?) faster.

My 2p's worth;  I'd be wary of adding yet another thing, especially one
that is so similar to something that is already there.  I've never
found the current comparision wanting and unless the speed increase
helps an awful lot with slow documents I don't think it's worth the
bloat.  Not bloat so much in the code sense, but in people having to
know yet another thing.

How often has it been asked for on the mailing list?  I'm a new
arrival.


Ralph.


From groff-admin  Tue Oct 10 17:30:34 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id RAA25561
	for <groff@ffii.org>; Tue, 10 Oct 2000 17:30:34 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Tue, 10 Oct 2000 17:33:52 +0200
Received: from rigel.univie.ac.at (dial-143183.HRZ.Uni-Dortmund.DE [129.217.143.183])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9AFXoC09562;
	Tue, 10 Oct 2000 17:33:50 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id RAA14890;
	Tue, 10 Oct 2000 17:21:43 +0200
Date: Tue, 10 Oct 2000 17:21:43 +0200 (CEST)
Message-Id: <20001010.172143.66060456.wl@gnu.org>
To: ralph@inputplus.demon.co.uk
Cc: groff@ffii.org
Subject: Re: [Groff] new string comparison? 
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <200010101441.PAA14664@inputplus.demon.co.uk>
References: <wl@gnu.org>
	<200010101441.PAA14664@inputplus.demon.co.uk>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 976
Lines: 28

> My 2p's worth; I'd be wary of adding yet another thing, especially
> one that is so similar to something that is already there.  I've
> never found the current comparision wanting and unless the speed
> increase helps an awful lot with slow documents I don't think it's
> worth the bloat.  Not bloat so much in the code sense, but in people
> having to know yet another thing.

Well, as told you in an earlier mail I'm working on a new version of
mdoc.  To improve readability, I've replaced numerical class values
with strings which means that I don't have tests like

  .if (\n[...] == 1)

but

  .if "\*[...]"macro"

This happens very often in the code; a quick test has shown that
numerical comparison is more than twice as fast as string comparison.
I haven't investigated yet (i.e., not written the code) whether my
proposal really gives a useful speed improvement.

> How often has it been asked for on the mailing list?

Not in the last three years :-)


    Werner

From groff-admin  Thu Oct 12 05:51:17 2000
Received: from polo.po.opole.pl (root@polo.po.opole.pl [193.59.208.10])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id FAA20012
	for <groff@ffii.org>; Thu, 12 Oct 2000 05:51:16 +0200
Received: from ukulele.linux.home (dial01.po.opole.pl [193.59.208.15])
	by polo.po.opole.pl (8.8.8/8.8.6/ag-polo.981012-sz+rchk1.22) with ESMTP id FAA24744;
	Thu, 12 Oct 2000 05:54:49 +0200 (MET DST)
Received: from guitar.linux.home (root@guitar.linux.home [10.0.0.12])
	by ukulele.linux.home (8.9.3/8.9.3/Debian/GNU) with ESMTP id WAA03652;
	Mon, 31 Jul 2000 22:32:45 +0200
Received: from localhost (luke@localhost)
	by guitar.linux.home (8.8.6/8.8.7) with SMTP id XAA00439;
	Wed, 11 Oct 2000 23:21:48 +0100
Date: Wed, 11 Oct 2000 23:21:48 +0100 (MET)
From: Lukasz Wiechec <magic@bet.po.opole.pl>
To: Werner LEMBERG <wl@gnu.org>
cc: Groff mailinglist <groff@ffii.org>
Subject: Re: [Groff] Latin-2 woes...
In-Reply-To: <20001009.232601.95508706.wl@gnu.org>
Message-ID: <Pine.LNX.3.96.1001011231811.347A-100000@guitar.linux.home>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=ISO-8859-2
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by genba.ffii.org id FAA20013
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 957
Lines: 30

On Mon, 9 Oct 2000, Werner LEMBERG wrote:

> > > > Actually, I've got a problem with four: char177 (which should be
> > > > aogonek but displays as plusminus [+-]), char179 (lslash,
> > > > displaying as 3-superscript or power 3), char188 (zslash,
> > > > displays as 1/4 or onequarter) and char166 (Sslash, displaying
> > > > as vertical bar or '|').
> 
> This particular problem can be solved easily by saying
> 
>   groff <your macro package(s)> -mpsnew ...

Wow ! It did a trick. Thanks !!! Still, the best solutions are usual right
before your nose... but to err is human :-)

> Nevertheless, the general solution is to write a proper input encoding
> file to map Latin-2 to glyph names; something like
> 
>   .char £ \[/L]
>   .char ³ \[/l]
>   ...

I don't quite follow. '.char' isn't a groff macro, is it ? (4 letters ?)


		------------------------
		MAGIC aka Lukasz Wiechec
		Internet mailto:magic@bet.po.opole.pl
		tel/fax  +(48 77) 4 555 113 


From groff-admin  Thu Oct 12 10:29:39 2000
Received: from cmailg1.svr.pol.co.uk (cmailg1.svr.pol.co.uk [195.92.195.171])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id KAA02746
	for <groff@ffii.org>; Thu, 12 Oct 2000 10:29:39 +0200
Received: from [195.92.198.123] (helo=mail17.svr.pol.co.uk)
	by cmailg1.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13jdo1-0000ll-00
	for groff@ffii.org; Thu, 12 Oct 2000 09:33:37 +0100
Received: from modem-6.alqualonde.dialup.pol.co.uk ([62.136.102.134] helo=inputplus.demon.co.uk)
	by mail17.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13jdo0-0006er-00
	for groff@ffii.org; Thu, 12 Oct 2000 09:33:37 +0100
Received: from inputplus (ralph@localhost [127.0.0.1])
	by inputplus.demon.co.uk (8.8.7/8.8.7) with ESMTP id JAA02392;
	Thu, 12 Oct 2000 09:33:30 +0100
Message-Id: <200010120833.JAA02392@inputplus.demon.co.uk>
To: Lukasz Wiechec <magic@bet.po.opole.pl>
cc: Werner LEMBERG <wl@gnu.org>, Groff mailinglist <groff@ffii.org>
Subject: Re: [Groff] Latin-2 woes... 
In-Reply-To: Message from Lukasz Wiechec <magic@bet.po.opole.pl> 
   of "Wed, 11 Oct 2000 23:21:48 BST." <Pine.LNX.3.96.1001011231811.347A-100000@guitar.linux.home> 
Date: Thu, 12 Oct 2000 09:33:30 +0100
From: Ralph Corderoy <ralph@inputplus.demon.co.uk>
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1725
Lines: 40

Hi,

> > Nevertheless, the general solution is to write a proper input
> > encoding file to map Latin-2 to glyph names; something like
> > 
> >   .char £ \[/L]
> >   .char ³ \[/l]
> >   ...
> 
> I don't quite follow. '.char' isn't a groff macro, is it ? (4 letters ?)

Sure is.  Look at groff's troff(1) man page.

    .char c string
        Define  character c to be string.  Every time char- acter c
        needs to be printed, string  will  be  pro- cessed  in  a
        temporary environment and the result will be wrapped up into a
        single object.   Compati- bility mode will be turned off and
        the escape char- acter will be set to \ while string is  being
        pro- cessed.  Any emboldening, constant spacing or track
        kerning will be applied to this object rather  than to
        individual  characters  in string.  A character defined by this
        request can be  used  just  like  a normal character provided
        by the output device.  In particular other characters can be
        translated to it with  the  tr  request;  it  can be made the
        leader character by the lc request; repeated patterns  can be
        drawn  with  the  character using the \l and \L escape
        sequences; words  containing  the  character can  be
        hyphenated correctly, if the hcode request is used to give the
        character a  hyphenation  code.  There  is  a special
        anti-recursion feature: use of character within the character's
        definition will be handled  like  normal  characters  not
        defined with char.  A character definition can be  removed
        with the rchar request.

The two letter restriction has gone for many things in groff.


Ralph.


From groff-admin  Thu Oct 12 11:07:37 2000
Received: from probity.mcc.ac.uk (probity.mcc.ac.uk [130.88.200.94])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id LAA02973
	for <groff@ffii.org>; Thu, 12 Oct 2000 11:07:36 +0200
Received: from tedistan.annex.mcc.ac.uk
	([194.66.22.200] helo=tedistan.annex.umist.ac.uk ident=root)
	by probity.mcc.ac.uk with esmtp (Exim 2.05 #4)
	id 13jeOj-0009kn-00; Thu, 12 Oct 2000 10:11:33 +0100
Received: (from efh@localhost)
	by eels.fort.knox.uk (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) id KAA05178;
	Thu, 12 Oct 2000 10:10:00 +0100
Message-ID: <XFMail.001012101000.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: XFMail 1.3 [p0] on Linux
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
In-Reply-To: <Pine.LNX.3.96.1001011231811.347A-100000@guitar.linux.home>
Date: Thu, 12 Oct 2000 10:10:00 +0100 (BST)
Reply-To: Ted.Harding@nessie.mcc.ac.uk
From: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
To: Lukasz Wiechec <magic@bet.po.opole.pl>
Subject: Re: [Groff] Latin-2 woes...
Cc: Groff mailinglist <groff@ffii.org>, Werner LEMBERG <wl@gnu.org>
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 4994
Lines: 138

On 11-Oct-00 Lukasz Wiechec wrote:
>> Nevertheless, the general solution is to write a proper input encoding
>> file to map Latin-2 to glyph names; something like
>> 
>>   .char £ \[/L]
>>   .char ³ \[/l]
>>   ...
> 
> I don't quite follow. '.char' isn't a groff macro, is it ? (4 letters
> ?)

1. '.char' is indeed a groff macro (groff names can be any length)
   but it will not work in "compatibility" mode since traditional
   troff names can be at most two letters long.

   .char \[name] string

   defines a character whose name is "name" and which is constructed
   by formatting "string". You then use it by putting '\[name]' where
   you want it in the input text.

   If a character with name "name" already exists, then it is replaced
   by the definition. Any character can be treated in this way
   (for instance, you can redefine the ordinary English character "a",
   as I do for Cyrillic, for instance:

   .de Cyrillic
   .ft AntCy
   .ftr Cy AntCy
   .char \(yu \N'192'
   .char a \N'193'
   .............


2. The groff command '.rchar \[name]' removes the definition. This
   gives rise to one of two situations.

   a) The character-name "name" can be found in the font file for
      the current font or a currently-searchable Special font. In
      this case the character with that name will be used as though
      the ".char" definition had never been given.

   b) The character-name "name" will disappear for ever. In particular,
      if you give one definition for '.char \[name] ... ' and then
      another definition '.char \[name] ... ' followed later by
      '.rchar \[name]' then the first definition is gone too.
      Character definitions do not "stack", and if you want the first
      one again then you must redefine it.

Therefore, for instance, if (using my .Cyrillic which defines Cyrillic
characters as above, and my ./Cyrillic which undoes all the definitions)
I write

   .Cyrillic
   Vladimir Putin
   ./Cyrillic
   Vladimir Putin

the first "Vladimir Putin" will print in Cyrillic, and the second will
print in English, since all these characters can still be found in the
standard fonts. On the other hand,

   .Cyrillic
   \[Ch]e\[ch]ni\[ya]
   ./Cyrillic
   \[Ch]e\[ch]ni\[ya]

will first print the Cyrillic version of the name which is written
"Chechniya" in English, and then print "eni" in English, since \[Ch],
\[ch] and \[ya] have been removed and are not in the standard font files,
while e, n and i can still be found.

Werner's suggestion anounts to the following.

In the standard PostScript fonts, there are characters with PostScript
names "Lslash" and "lslash": the glyphs for these form part of standard
PostScript and so do not need separate definition.

If you look in one of the devps font files (say .../groff/font/devps/TR)
you will find

   ....
   /l      278,683 2       0234    lslash
   ....
   /L      611,662 2       0237    Lslash
   ....

so these characters have groff names \[/l] and \[/L] already defined
in the font files: therefore ".char anything \[/L]" means that whenever
a character named "anything" occurs in the input stream, the string
"\[/L]" is used instead. Therefore this will print as intended.

In the Latin-1 encoding (which is what groff recognises), the "£" sign
occurs in the same position as "Lslash" in Latin-2. Therefore, when
groff sees the input-byte corresponding to "£" in Latin-1, it consults
the ".char" definition and replaces it with "\[/L]. This does not involve
any builtin "knowledge" of Latin-2 encoding: it is the user who has
supplied this in the ".char" definition:

  .char £ \[/L]

Unfortunately for Poles and others, the ready-made glyphs in the
standard PostScript fonts do not cover all the needed possibilities,
and you will then have to define your characters with strings which
cause them to be directly constructed. For this, fortunately the
standard PostScript fonts contain glyphs for a variety of accents.


For instance, the ms macros allow a character to be used as an
"accent-under" for the preceding character, in particular "ogonek"
(PostScript name) for which the groff name is "\[ho]" ("hook"):

   ho    333,0,165       0       0230    ogonek

So you can define the string \*[ogonek] which performs the
correct placement of this accent under the preceding character by

  .acc*under-def ogonek \[ho]

  .char \[o,] o\*[ogonek]

and then use \[o,] whenever you need "o-ogonek".

I had been meaning to pick up on this topic earlier, giving a
complete repertoire of Latin-2 translations, but I have been too
taken up with other things recently to finish it. However, I will
do it later.

Meanwhile, I hope the above at least points the way.

Best wishes,
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 284 7749
Date: 12-Oct-00                                       Time: 10:10:00
------------------------------ XFMail ------------------------------

From groff-admin  Thu Oct 12 11:07:54 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id LAA02988
	for <groff@ffii.org>; Thu, 12 Oct 2000 11:07:54 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Thu, 12 Oct 2000 11:11:01 +0200
Received: from rigel.univie.ac.at (dial-143175.HRZ.Uni-Dortmund.DE [129.217.143.175])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9C9AxC07343;
	Thu, 12 Oct 2000 11:10:59 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA00735;
	Thu, 12 Oct 2000 10:25:59 +0200
Date: Thu, 12 Oct 2000 10:25:58 +0200 (CEST)
Message-Id: <20001012.102558.63133857.wl@gnu.org>
To: magic@bet.po.opole.pl
Cc: groff@ffii.org
Subject: Re: [Groff] Latin-2 woes...
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <Pine.LNX.3.96.1001011231811.347A-100000@guitar.linux.home>
References: <20001009.232601.95508706.wl@gnu.org>
	<Pine.LNX.3.96.1001011231811.347A-100000@guitar.linux.home>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=iso-8859-2
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by genba.ffii.org id LAA02989
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 373
Lines: 15

> > Nevertheless, the general solution is to write a proper input
> > encoding file to map Latin-2 to glyph names; something like
> > 
> >   .char £ \[/L]
> >   .char ³ \[/l]
> >   ...
> 
> I don't quite follow. '.char' isn't a groff macro, is it ? (4
> letters ?)

No, it's a new request specific to troff.  Please read the man page of
GNU troff for details.


    Werner

From groff-admin  Fri Oct 13 15:03:09 2000
Received: from Hesperia (ppp219-global8.libertysurf.fr [213.36.127.219])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA28165
	for <groff@ffii.org>; Fri, 13 Oct 2000 15:03:06 +0200
X-Authentication-Warning: genba.ffii.org: Host ppp219-global8.libertysurf.fr [213.36.127.219] claimed to be Hesperia
Received: from baruchel by Hesperia with local (Exim 2.05 #1 (Debian))
	id 13k49v-0000LN-00; Fri, 13 Oct 2000 14:41:59 +0200
Date: Fri, 13 Oct 2000 14:41:59 +0200
From: Thomas Baruchel <baruchel@libertysurf.fr>
To: groff@ffii.org
Message-ID: <20001013144159.A1301@Hesperia>
Reply-To: Thomas Baruchel <baruchel@libertysurf.fr>
Mail-Followup-To: groff@ffii.org
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
X-Machine: Hesperia sur Debian (Linux)
Subject: [Groff] What were Scribe and Mint ?
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 600
Lines: 18

Brest, le vendredi 13 octobre

Hi, I am a LaTeX2groff man. Since My wife has a thesis to write, I anyway told
her to use LaTeX :-( I really don't think she could have liked groff :-(
I found a book on LaTeX that I read several years ago, and in the introduction
was a list of "ancesters" (troff of course, but also scribe and mint)

What were Scribe and Mint ? What was their output formats ? Is is still
possible to download them and compile them (on linux) ? Are they free ?

just to see ;-)

-- 
   .~.   Thomas Baruchel
   /V\   baruchel@libertysurf.fr
  // \\    Brest
 /(   )\   FRANCE
  ^`~'^

From groff-admin  Sat Oct 14 01:05:27 2000
Received: from mystic.false.com (root@false.com [198.65.171.171])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id BAA00587
	for <groff@ffii.org>; Sat, 14 Oct 2000 01:05:26 +0200
X-Authentication-Warning: genba.ffii.org: Host root@false.com [198.65.171.171] claimed to be mystic.false.com
Received: from false.com (root@false.com [198.65.171.171]) by mystic.false.com (8.9.3/8.9.3) with ESMTP id SAA28810; Fri, 13 Oct 2000 18:09:22 -0500
Received: (from solar@localhost) by false.com (8.8.5/8.8.5) id DAA03585; Sat, 14 Oct 2000 03:06:43 +0400
From: Solar Designer <solar@false.com>
Message-Id: <200010132306.DAA03585@false.com>
Subject: Re: [Groff] FW: ISS Security Advisory: GNU Groff utilities read untrusted commands from current working directory
To: wl@gnu.org
Date: Sat, 14 Oct 2000 03:06:42 +0400 (MSD)
Cc: groff@ffii.org
X-Mailer: ELM [version 2.4ME+ PL31 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1742
Lines: 38

I am forwarding my comments on an earlier version of the ISS advisory,
just so that you get full context.  I was assuming that I'd have to
produce a fix for the groff package I include in a distribution, so
that is why I didn't want to change the code too heavily.

----- Forwarded message from Solar Designer -----

> GNU Groff utilities read untrusted commands from current working
> directory

It really is non-obvious how this should be fixed.  The trivial fix
would be to remove the dots from fontpath and tmacpath, but I expect
this to break things.  I've considered restricting this change to the
safer mode, but (1) there're cases where the safer mode is useful and
the input file is in the current directory (well, they could use the
full path) and (2) safer_flag isn't properly initialized by the time
search_path constructors are called (don't want to be changing too
much of the code in a security patch).  So this doesn't seem like a
good idea.

> By default, the "troff" program reads its "troffrc" initialization file from
> the current working directory.  From a security standpoint, it would be 
> desirable to restrict the searchable path for this file to the invoker's 
> home directory and/or a trusted system path.

Unfortunately, it's not just the "troffrc", "troffrc-end" and other
hard-coded filenames that may be accessed from the current directory.
The macro files themselves reference other files:

.\" Load hyphenation patterns from `hyphen.us' (in the tmac directory).
.do hpf hyphen.us

There appears to be no way to specify that a file should be loaded
from "the tmac directory".  Hard-coding the full path in too many
files (some of which may come from other packages) is inconvenient.

Signed,
Solar Designer

From groff-admin  Sat Oct 14 01:21:55 2000
Received: from mystic.false.com (root@false.com [198.65.171.171])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id BAA00631
	for <groff@ffii.org>; Sat, 14 Oct 2000 01:21:54 +0200
X-Authentication-Warning: genba.ffii.org: Host root@false.com [198.65.171.171] claimed to be mystic.false.com
Received: from false.com (root@false.com [198.65.171.171]) by mystic.false.com (8.9.3/8.9.3) with ESMTP id SAA28958; Fri, 13 Oct 2000 18:25:59 -0500
Received: (from solar@localhost) by false.com (8.8.5/8.8.5) id DAA03620; Sat, 14 Oct 2000 03:21:13 +0400
From: Solar Designer <solar@false.com>
Message-Id: <200010132321.DAA03620@false.com>
Subject: Re: [Groff] FW: ISS Security Advisory: GNU Groff utilities read untrusted com mands from current working directory
In-Reply-To: <20001005.223731.37301821.wl@gnu.org> from Werner LEMBERG at "Oct 5, 0 10:37:31 pm"
To: wl@gnu.org (Werner LEMBERG)
Date: Sat, 14 Oct 2000 03:21:13 +0400 (MSD)
Cc: groff@ffii.org
X-Mailer: ELM [version 2.4ME+ PL31 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1127
Lines: 27

> > By default, the "troff" program reads its "troffrc" initialization
> > file from the current working directory.  From a security
> > standpoint, it would be desirable to restrict the searchable path
> > for this file to the invoker's home directory and/or a trusted
> > system.  Unfortunately, this could present problems for programs
> > that depend on the current behavior.
> 
> My suggestion is to restrict the location of troffrc and troffrc-end
> to `~' and groff's default tmac directory
> (e.g. /usr/local/share/groff/tmac) if the -U flag isn't given.

What about the files specified from the command line (these should be
opened relative to the current directory, or existing lesspipe.sh
would break)?  What about files referenced from troffrc and possibly
from other files?

I think it is safe to allow relative paths on the command line, but
everything referenced from troffrc should be restricted.

> Additionally, I'll implement tmac.safer internally, i.e., without the
> -U flag these requests will produce error messages.  tmac.safer will
> then no longer exist.

This is a good idea.

Signed,
Solar Designer

From groff-admin  Sat Oct 14 16:49:32 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA19588
	for <groff@ffii.org>; Sat, 14 Oct 2000 16:49:32 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sat, 14 Oct 2000 16:53:41 +0200
Received: from rigel.univie.ac.at (dial-143131.HRZ.Uni-Dortmund.DE [129.217.143.131])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9EErdC21399;
	Sat, 14 Oct 2000 16:53:39 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA04774;
	Sat, 14 Oct 2000 10:43:57 +0200
Date: Sat, 14 Oct 2000 10:43:56 +0200 (CEST)
Message-Id: <20001014.104356.63130402.wl@gnu.org>
To: solar@false.com
Cc: groff@ffii.org
Subject: Re: [Groff] FW: ISS Security Advisory: GNU Groff utilities read
 untrusted com mands from current working directory
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <200010132321.DAA03620@false.com>
References: <20001005.223731.37301821.wl@gnu.org>
	<200010132321.DAA03620@false.com>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 245
Lines: 10

> > Additionally, I'll implement tmac.safer internally, i.e., without
> > the -U flag these requests will produce error messages.
> > tmac.safer will then no longer exist.
> 
> This is a good idea.

I've implemented this right now.


    Werner

From groff-admin  Sat Oct 14 16:49:32 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA19590
	for <groff@ffii.org>; Sat, 14 Oct 2000 16:49:32 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sat, 14 Oct 2000 16:53:43 +0200
Received: from rigel.univie.ac.at (dial-143131.HRZ.Uni-Dortmund.DE [129.217.143.131])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9EErfC21402;
	Sat, 14 Oct 2000 16:53:41 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA04785;
	Sat, 14 Oct 2000 10:57:42 +0200
Date: Sat, 14 Oct 2000 10:57:42 +0200 (CEST)
Message-Id: <20001014.105742.104027168.wl@gnu.org>
To: solar@false.com
Cc: groff@ffii.org
Subject: Re: [Groff] FW: ISS Security Advisory: GNU Groff utilities read
 untrusted com mands from current working directory
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <200010132321.DAA03620@false.com>
References: <20001005.223731.37301821.wl@gnu.org>
	<200010132321.DAA03620@false.com>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1124
Lines: 27

> > > By default, the "troff" program reads its "troffrc"
> > > initialization file from the current working directory.  From a
> > > security standpoint, it would be desirable to restrict the
> > > searchable path for this file to the invoker's home directory
> > > and/or a trusted system.  Unfortunately, this could present
> > > problems for programs that depend on the current behavior.
> > 
> > My suggestion is to restrict the location of troffrc and
> > troffrc-end to `~' and groff's default tmac directory
> > (e.g. /usr/local/share/groff/tmac) if the -U flag isn't given.
> 
> What about the files specified from the command line (these should
> be opened relative to the current directory, or existing lesspipe.sh
> would break)?

I think such files won't cause any harm since all unsafe requests are
now disabled in a secure manner.

> What about files referenced from troffrc and possibly from other
> files?

An easy solution would be to disable the .so request for troffrc and
troffrc-end; only .mso would work.  Hyphenation patterns already use
the macro path, so this wouldn't cause problems.


    Werner

From groff-admin  Sat Oct 14 16:49:33 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA19594
	for <groff@ffii.org>; Sat, 14 Oct 2000 16:49:33 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sat, 14 Oct 2000 16:53:39 +0200
Received: from rigel.univie.ac.at (dial-143131.HRZ.Uni-Dortmund.DE [129.217.143.131])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9EEraC21396;
	Sat, 14 Oct 2000 16:53:37 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id MAA04860;
	Sat, 14 Oct 2000 12:58:45 +0200
Date: Sat, 14 Oct 2000 12:58:45 +0200 (CEST)
Message-Id: <20001014.125845.11595642.wl@gnu.org>
To: solar@false.com
Cc: groff@ffii.org
Subject: Re: [Groff] FW: ISS Security Advisory: GNU Groff utilities read
 untrusted commands from current working directory
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <200010132306.DAA03585@false.com>
References: <200010132306.DAA03585@false.com>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1309
Lines: 33

> > GNU Groff utilities read untrusted commands from current working
> > directory
> 
> It really is non-obvious how this should be fixed.  The trivial fix
> would be to remove the dots from fontpath and tmacpath, but I expect
> this to break things.  I've considered restricting this change to
> the safer mode, but (1) there're cases where the safer mode is
> useful and the input file is in the current directory (well, they
> could use the full path) and (2) safer_flag isn't properly
> initialized by the time search_path constructors are called (don't
> want to be changing too much of the code in a security patch).  So
> this doesn't seem like a good idea.

But this is the way to go IMHO.  As mentioned in a previous mail,
restricting the font and tmac patch to `~' and the default path in
safer mode can easily be implemented.  Additionally, I will add a
warning message if GROFF_{TMAC,FONT}_PATH is set that these two
environment variables are ignored in safer mode.

> The macro files themselves reference other files:
> 
> .\" Load hyphenation patterns from `hyphen.us' (in the tmac directory).
> .do hpf hyphen.us

.hpf uses the same search patch as the .mso request.

> There appears to be no way to specify that a file should be loaded
> from "the tmac directory".

There is: .mso


   Werner

From groff-admin  Mon Oct 16 08:31:53 2000
Received: from mailgw1.be.to (mailgw1.be.to [210.235.212.5])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id IAA09127
	for <groff@ffii.org>; Mon, 16 Oct 2000 08:31:51 +0200
Received: from mail1.be.to (mail1.be.to [210.235.212.2])
	by mailgw1.be.to (8.11.0+3.3W/8.11.0/BETO3.11-20000907025324) with ESMTP id e9G6Zxd16182
	for <groff@ffii.org>; Mon, 16 Oct 2000 15:35:59 +0900
	(envelope-from okazaki@be.to)
Received: from REISHI (mitaka1-76.kcom.ne.jp [210.174.84.76])
	by mail1.be.to (8.9.3+3.2W/BETO.2.0-2000090621000000) with SMTP id PAA19027
	for <groff@ffii.org>; Mon, 16 Oct 2000 15:35:58 +0900
Received: (qmail 852 invoked from network); 16 Oct 2000 06:33:03 -0000
Received: from localhost (HELO acidrain.localnet) (127.0.0.1)
  by localhost with SMTP; 16 Oct 2000 06:33:03 -0000
Date: Mon, 16 Oct 2000 15:33:01 +0900
Message-ID: <86r95hthwi.wl@dolphin.be.to>
From: OKAZAKI Tetsurou <okazaki@be.to>
To: groff@ffii.org
References: <14826.25637.472475.26482Z@surfchem0.riken.go.jp>
	<14826.26984.796923.29321F@surfchem0.riken.go.jp>
User-Agent: Wanderlust/2.3.0 (Roam) REMI/1.14.3 (Matsudai) Chao/1.14.1
 (=?ISO-8859-4?Q?Rokujiz=F2?=) APEL/10.2 Emacs/20.7 (i386--freebsd) MULE/4.0
 (HANANOEN)
Organization: Unknown
MIME-Version: 1.0 (generated by REMI 1.14.3 - "Matsudai")
Content-Type: multipart/mixed;
 boundary="Multipart_Mon_Oct_16_15:33:01_2000-1"
Subject: [Groff] Forward: groff: radical re-implementation
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 4592
Lines: 116

--Multipart_Mon_Oct_16_15:33:01_2000-1
Content-Type: text/plain; charset=US-ASCII

Hi,

The message below is from debian-i18n.

Tetsurou


--Multipart_Mon_Oct_16_15:33:01_2000-1
Content-Type: message/rfc822

Resent-Date: Mon, 16 Oct 2000 11:31:22 +0900
Date: Mon, 16 Oct 2000 11:35:20 +0900
Message-ID: <14826.26984.796923.29321F@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org
Cc: cjh@kr.FreeBSD.org
Subject: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Mon, 16 Oct 2000 11:12:53 +0900"
	<14826.25637.472475.26482Z@surfchem0.riken.go.jp>
References: <14826.25637.472475.26482Z@surfchem0.riken.go.jp>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Resent-Message-ID: <_y1KSB.A.x3G.2hm65@murphy>
Resent-From: debian-i18n@lists.debian.org
X-Mailing-List: <debian-i18n@lists.debian.org> archive/latest/654
X-Loop: debian-i18n@lists.debian.org
Precedence: list
Resent-Sender: debian-i18n-request@lists.debian.org

Hi,

This thread is about the 'radical re-implementation' of groff.

We first thought that we have to fix Japanese patch so that
it checks EUC encoding only in Japanese mode (-Tnippon).  However,
we immediately found that the original design of groff is
very confusing.  I think the developer of Japanese patch could
not avoid inheriting the confusingness.

Why 'ascii' and 'latin1' are treated as 'device type'?  The device
type should be 'tty' or so.  Because of this confusing design,
we have no way to treat, for example, Japanese X11 output or
Korean PostScript.  You know, THIS IS NOT DUE TO LACK OF 
IMPLEMENTATION BUT DUE TO CONFUSED DESIGN.  Should we type
'groff -Tlatin1 -Tx75' for X11 output with latin1 encoding?
Entirely No!

The ideal implementation will be using 'wchar_t' for reading.
Abolish device types of 'ascii', 'ascii8', 'latin1', 'nippon',
and 'utf8' and introduce a new device type such as 'tty'.
Ukai has surveyed roughly the source code of groff and posted
a brief but long list of needed works (in debian-devel@debian.or.jp 
mailing list in Japanese).

  http://www.debian.or.jp/Lists-Archives/debian-devel/200010/msg00072.html

Fortunately, fgetwc(), putwchar(), wprintf(), swprintf(), and so on 
are available in new Glibc 2.2.  mbstowcs() and so on are also
available since older Glibc.  These functions are locale-sensible
and can handle any encodings.  Note that they can also treat UTF-8
under UTF-8 locale, though the current Debian locales package does
not include any UTF-8 locales.  We should not give UTF-8 special
treatment.  Discussion is in progress about this new design of
groff at debian-devel@debian.or.jp mailing list (in Japanese)
and personal communication.

   --- * --- * ---

However, I found a big problem.  I can set my LANG as ja_JP.eucJP,
ja_JP.SJIS, or ja_JP.UTF-8.  An American person can set LANG as
en_US.ISO8859-1 and en_US.UTF-8.  Note that only changing LANG
variable must change the behavior of the OS.  However, the source
of manpages are written in a certain encoding for each language.
For example, when a user sets LANG as en_US.ISO8859-1 and the
manpages are written in UTF-8, Groff will read manpages as 
ISO8859-1 and the manpages will not displayed properly.

The solution may be like this:

- Groff to have a new command option, like '--input-encoding'
    to specify the encoding of the input file.
- Man-db to specify the encoding when invoking Groff.
- To have a Policy on encoding of manpages for each language.
  That will be one of them:
   (a) All manpages to be written in UTF-8.
   (b) To decide encoding for each language.  This may be
       same as the present state.
  This Policy will be implemented to Man-db package.

I can imagine another solution:

- Groff assumes the input as the encoding of current locale.
- Man-db to invoke Groff through iconv(1).

Problem of the latter idea is that the current version of locales 
package does not have any UTF-8 locale.  How UTF-8 --> wchar_t 
conversion can be achieved without UTF-8 locale?  WE MUST NOT 
ASSUME THAT INTERNAL EXPRESSION OF WCHAR_T IS UCS-4, though this
is true for Glibc.

---
Tomohiro KUBOTA <kubota@debian.org>


--  
To UNSUBSCRIBE, email to debian-i18n-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

--Multipart_Mon_Oct_16_15:33:01_2000-1--

From groff-admin  Mon Oct 16 15:25:41 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA11842
	for <groff@ffii.org>; Mon, 16 Oct 2000 15:25:41 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Mon, 16 Oct 2000 15:30:07 +0200
Received: from rigel.univie.ac.at (dial-143159.HRZ.Uni-Dortmund.DE [129.217.143.159])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9GDU6w16878
	for <groff@ffii.org>; Mon, 16 Oct 2000 15:30:06 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id LAA07713
	for <groff@ffii.org>; Mon, 16 Oct 2000 11:59:08 +0200
Date: Mon, 16 Oct 2000 11:59:07 +0200 (CEST)
Message-Id: <20001016.115907.13775213.wl@gnu.org>
To: groff@ffii.org
From: Werner LEMBERG <wl@gnu.org>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] fontpath during installation
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 321
Lines: 15

Dear friends,


currently, the font path during configuration will be set to

  fontpath=.:$(fontdir):/usr/lib/font

The `.' will be handled specially in the near future to solve security
problems.

But what about /usr/lib/font?  Is there any reason to retain this
absolute path?  I would like to remove it.


    Werner

From groff-admin  Mon Oct 16 21:17:10 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id VAA14490
	for <groff@ffii.org>; Mon, 16 Oct 2000 21:17:10 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Mon, 16 Oct 2000 21:21:11 +0200
Received: from rigel.univie.ac.at (dial-143189.HRZ.Uni-Dortmund.DE [129.217.143.189])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9GJKkw29550;
	Mon, 16 Oct 2000 21:20:46 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id QAA08329;
	Mon, 16 Oct 2000 16:41:36 +0200
Date: Mon, 16 Oct 2000 16:41:35 +0200 (CEST)
Message-Id: <20001016.164135.88348722.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14826.26984.796923.29321F@surfchem0.riken.go.jp>
References: <14826.25637.472475.26482Z@surfchem0.riken.go.jp>
	<14826.26984.796923.29321F@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] Re: groff: radical re-implementation
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 4739
Lines: 113

[I'm CC'ing this mail to the groff@ mailing list.  May I ask to move
the discussion about improvments/changings of groff to this list?]


From: Tomohiro KUBOTA <tkubota@riken.go.jp>
Subject: groff: radical re-implementation
Date: Mon, 16 Oct 2000 11:35:20 +0900

> This thread is about the 'radical re-implementation' of groff.
> 
> We first thought that we have to fix Japanese patch so that it
> checks EUC encoding only in Japanese mode (-Tnippon).  However, we
> immediately found that the original design of groff is very
> confusing.  I think the developer of Japanese patch could not avoid
> inheriting the confusingness.
> 
> Why 'ascii' and 'latin1' are treated as 'device type'?  The device
> type should be 'tty' or so.  Because of this confusing design, we
> have no way to treat, for example, Japanese X11 output or Korean
> PostScript.  You know, THIS IS NOT DUE TO LACK OF IMPLEMENTATION BUT
> DUE TO CONFUSED DESIGN.  Should we type 'groff -Tlatin1 -Tx75' for
> X11 output with latin1 encoding?  Entirely No!

As you may know, this confusion has historical origins.  I'm not
willing to add new `devices' like `latin-2' or even `nippon' due to
this currently.

I plan to separate input encodings, output encodings, and character
sets from devices.  Then, we will have real devices like tty, ps, or
dvi.  Input characters will be converted to glyph names by troff, and
these glyph names will be mapped to output encodings (for ttys)
resp. fonts (for everything else) according to the device and font
data.

> The ideal implementation will be using 'wchar_t' for reading.

But this will fail for some compilers...

> Abolish device types of 'ascii', 'ascii8', 'latin1', 'nippon', and
> 'utf8' and introduce a new device type such as 'tty'.

This will come.  Of course, patches are welcome :-) If you or your
partners have serious interest in improving groff I can give you write
access to the CVS.

> Ukai has surveyed roughly the source code of groff and posted
> a brief but long list of needed works (in debian-devel@debian.or.jp 
> mailing list in Japanese).
> 
>   http://www.debian.or.jp/Lists-Archives/debian-devel/200010/msg00072.html
> 
> Fortunately, fgetwc(), putwchar(), wprintf(), swprintf(), and so on
> are available in new Glibc 2.2.  mbstowcs() and so on are also
> available since older Glibc.  These functions are locale-sensible
> and can handle any encodings.  Note that they can also treat UTF-8
> under UTF-8 locale, though the current Debian locales package does
> not include any UTF-8 locales.  We should not give UTF-8 special
> treatment.  Discussion is in progress about this new design of groff
> at debian-devel@debian.or.jp mailing list (in Japanese) and personal
> communication.

Please bear in mind that groff shall work on non-GNU systems also!  My
idea is to only accept UTF8, ascii, latin1, and ebcdic as input
encodings (the latter three for historical reasons only).

Maybe on systems with a recent glibc, iconv() and friends can be used
to do more, but generally I prefer an iconv-preprocessor so that groff
itself has not to deal with encoding conversions.

> 
>    --- * --- * ---
> 
> However, I found a big problem.  I can set my LANG as ja_JP.eucJP,
> ja_JP.SJIS, or ja_JP.UTF-8.  An American person can set LANG as
> en_US.ISO8859-1 and en_US.UTF-8.  Note that only changing LANG
> variable must change the behavior of the OS.  However, the source
> of manpages are written in a certain encoding for each language.
> For example, when a user sets LANG as en_US.ISO8859-1 and the
> manpages are written in UTF-8, Groff will read manpages as 
> ISO8859-1 and the manpages will not displayed properly.
> 
> The solution may be like this:
> 
> - Groff to have a new command option, like '--input-encoding'
>     to specify the encoding of the input file.

Yes, an iconv preprocessor for GNU systems to convert input files to
Unicode.

> - Man-db to specify the encoding when invoking Groff.
> - To have a Policy on encoding of manpages for each language.
>   That will be one of them:
>    (a) All manpages to be written in UTF-8.
>    (b) To decide encoding for each language.  This may be
>        same as the present state.
>   This Policy will be implemented to Man-db package.
> 
> I can imagine another solution:
> 
> - Groff assumes the input as the encoding of current locale.

This is probably not correctly set everywhere.

> - Man-db to invoke Groff through iconv(1).
> 
> Problem of the latter idea is that the current version of locales 
> package does not have any UTF-8 locale.  How UTF-8 --> wchar_t 
> conversion can be achieved without UTF-8 locale?  WE MUST NOT 
> ASSUME THAT INTERNAL EXPRESSION OF WCHAR_T IS UCS-4, though this
> is true for Glibc.


    Werner

From groff-admin  Mon Oct 16 21:43:18 2000
Received: from smtp5.dti.ne.jp (smtp5.dti.ne.jp [202.216.228.40])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id VAA14632
	for <groff@ffii.org>; Mon, 16 Oct 2000 21:43:16 +0200
Received: from localhost (PPP81.sagami-ap3.dti.ne.jp [210.159.176.227]) by smtp5.dti.ne.jp (8.9.3/3.7W) with ESMTP id EAA26278; Tue, 17 Oct 2000 04:47:11 +0900 (JST)
Message-Id: <200010161947.EAA26278@smtp5.dti.ne.jp>
To: wl@gnu.org
Cc: tkubota@riken.go.jp, debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org,
        groff@ffii.org
From: GOTO Masanori <gotom@debian.or.jp>
In-Reply-To: Your message of "Mon, 16 Oct 2000 16:41:35 +0200 (CEST)"
References: <20001016.164135.88348722.wl@gnu.org>
X-Mailer: Mew version 1.70 on Emacs 19.34.1 / Mule 2.3
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Date: Tue, 17 Oct 2000 04:48:28 +0900
X-Dispatcher: impost version 0.99i (Apr. 6, 1997)
Subject: [Groff] Re: groff: radical re-implementation
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 3313
Lines: 78

Hi, All.

From: Werner LEMBERG <wl@gnu.org>
Date: Mon, 16 Oct 2000 16:41:35 +0200 (CEST)
> From: Tomohiro KUBOTA <tkubota@riken.go.jp>
> Subject: groff: radical re-implementation
> Date: Mon, 16 Oct 2000 11:35:20 +0900
> > Why 'ascii' and 'latin1' are treated as 'device type'?  The device
> > type should be 'tty' or so.  Because of this confusing design, we
> > have no way to treat, for example, Japanese X11 output or Korean
> > PostScript.  You know, THIS IS NOT DUE TO LACK OF IMPLEMENTATION BUT
> > DUE TO CONFUSED DESIGN.  Should we type 'groff -Tlatin1 -Tx75' for
> > X11 output with latin1 encoding?  Entirely No!
> 
> As you may know, this confusion has historical origins.  I'm not
> willing to add new `devices' like `latin-2' or even `nippon' due to
> this currently.
> 
> I plan to separate input encodings, output encodings, and character
> sets from devices.  Then, we will have real devices like tty, ps, or
> dvi.  Input characters will be converted to glyph names by troff, and
> these glyph names will be mapped to output encodings (for ttys)
> resp. fonts (for everything else) according to the device and font
> data.

It's nice news!

> > The ideal implementation will be using 'wchar_t' for reading.
> 
> But this will fail for some compilers...

Hmm, ISO C99 now becomes standard, but ...

> > Ukai has surveyed roughly the source code of groff and posted
> > a brief but long list of needed works (in debian-devel@debian.or.jp 
> > mailing list in Japanese).
> > 
> >   http://www.debian.or.jp/Lists-Archives/debian-devel/200010/msg00072.html
> > 
> > Fortunately, fgetwc(), putwchar(), wprintf(), swprintf(), and so on
> > are available in new Glibc 2.2.  mbstowcs() and so on are also
> > available since older Glibc.  These functions are locale-sensible
> > and can handle any encodings.  Note that they can also treat UTF-8
> > under UTF-8 locale, though the current Debian locales package does
> > not include any UTF-8 locales.  We should not give UTF-8 special
> > treatment.  Discussion is in progress about this new design of groff
> > at debian-devel@debian.or.jp mailing list (in Japanese) and personal
> > communication.
> 
> Please bear in mind that groff shall work on non-GNU systems also!  My
> idea is to only accept UTF8, ascii, latin1, and ebcdic as input
> encodings (the latter three for historical reasons only).
> 
> Maybe on systems with a recent glibc, iconv() and friends can be used
> to do more, but generally I prefer an iconv-preprocessor so that groff
> itself has not to deal with encoding conversions.

There is portable iconv implementation developed by Bruno Haible,
which is derivered from glibc iconv().
In the case of using iconv(), such a portable library may be helpful.
Another solution is to use GNU recode as iconv-preprocessor.

> > - Man-db to invoke Groff through iconv(1).
> > 
> > Problem of the latter idea is that the current version of locales 
> > package does not have any UTF-8 locale.  How UTF-8 --> wchar_t 
> > conversion can be achieved without UTF-8 locale?  WE MUST NOT 
> > ASSUME THAT INTERNAL EXPRESSION OF WCHAR_T IS UCS-4, though this
> > is true for Glibc.

No. UTF-8 locale is easy to include into locales package.
And what is the problem? Using iconv() is not needed any UTF-8
locales...?

Regards,
-- GOTO Masanori



From groff-admin  Tue Oct 17 03:18:38 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id DAA17674
	for <groff@ffii.org>; Tue, 17 Oct 2000 03:18:37 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 9CA5F1C122; Tue, 17 Oct 2000 10:27:16 +0900 (JST)
Date: Tue, 17 Oct 2000 10:27:12 +0900
Message-ID: <14827.43760.126423.72159A@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Mon, 16 Oct 2000 16:41:35 +0200 (CEST)"
	<20001016.164135.88348722.wl@gnu.org>
References: <14826.25637.472475.26482Z@surfchem0.riken.go.jp>
	<14826.26984.796923.29321F@surfchem0.riken.go.jp>
	<20001016.164135.88348722.wl@gnu.org>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Subject: [Groff] Re: groff: radical re-implementation
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 4061
Lines: 103

Hi,

At Mon, 16 Oct 2000 16:41:35 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> [I'm CC'ing this mail to the groff@ mailing list.  May I ask to move
> the discussion about improvments/changings of groff to this list?]

Ok, I joined groff@ffii.org mailing list, though I send this message
also for debian-i18n list to inform that I agreed to move.


>> The ideal implementation will be using 'wchar_t' for reading.
> But this will fail for some compilers...

Now wchar_t is supported by many systems.  It is a mandatory for
internationalization.

The merit of wchar_t is that:  write once and work for every
encodings, uncluding UTF-8.  Otherwise, you have to write
similar source codes many times for Latin-1, EBCDIC, UTF-8,
and so on so on.  Especially, I will insist that Groff should
support EUC-* multibyte encodings for CJK languages.  This is
what the current Groff cannot handle entirely.  (CJK people
also uses ISO-2022-* encodings.)

The other merit of wchar_t is user-friendliness.  Once a user
set LANG variable, every softwares work under the specified 
encoding.  If not, you have to specify encodings for every software.
We don't want to have ~/.groffrc, ~/.greprc, ~/.bashrc, ~/.xtermrc,
and so on so on to specify 'encoding=ISO8859-1' or 'encoding=UTF-8'.



>> Abolish device types of 'ascii', 'ascii8', 'latin1', 'nippon', and
>> 'utf8' and introduce a new device type such as 'tty'.

I suppose you don't know about 'ascii8' device.  This is a local
patch for Debian's Groff that is 8-bit clean (like latin1) but
doesn't assume that 8-bit part is latin1 encoding.  For example,
'-' is used for hyphenation and '\(co' is converted into '(C)'.
This is for 8-bit encodings other than latin1, i.e., ISO8859-2,3,..,
and KOI8-R.  (Not for CJK multibyte languages).



> Please bear in mind that groff shall work on non-GNU systems also!  My
> idea is to only accept UTF8, ascii, latin1, and ebcdic as input
> encodings (the latter three for historical reasons only).

I wrote about Glibc because the message is to Debian mailing list.
Of course I think of portability.  wchar_t is portable.  I recommend
to implement wchar_t as a new architecture and ascii, latin-1, and
ebcdic as historical encodings.  (We may add 'UTF8' as a historical
one.)

I think what is 'historical' is systems which don't support wchar_t.



> Maybe on systems with a recent glibc, iconv() and friends can be used
> to do more, but generally I prefer an iconv-preprocessor so that groff
> itself has not to deal with encoding conversions.

I think this works well.  However, who invokes iconv-preprocessor?
A user or wrapper-software?  What determines the command option for
iconv?



>> - Groff assumes the input as the encoding of current locale.
> This is probably not correctly set everywhere.

How a system can be configured by a user, in ways other than locale?
A user who want to specify his/her language and encoding will set 
LANG variable.  Or, having many ~/.foobarrc for every softwares or 
specifying --encoding=foobar everytime (s)he invokes a software?  
I think setting LANG is a reasonable way.


One compromise is that:
 - to use UCS-4 for internal processing, not wchar_t.
 - a small part of input and output to be encoding-sensible.
 - command options for encodings of input and output to be added.
 - a compile-time option I18N to be introduced.
 - when I18N is off, default input is latin-1 and default output
   is also latin-1.
 - when I18N is on, default input and default output are sensible
   to LC_CTYPE locale.
 - Of course these default encodings can be overrided by command
   options.
 - Groff can be compiled with I18N off for systems without 
   internationalization functions such as setlocale().
 - iconv(3) to be used for converting between input/output encodings
   and internal UCS-4 encoding, if available (I18N=true).
 - if I18N is false, conversion process to be hard-coded for
   Latin-1, EBCDIC, and UTF-8.

Do you think this can be achieved?

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Tue Oct 17 06:19:38 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id GAA31533
	for <groff@ffii.org>; Tue, 17 Oct 2000 06:19:36 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 638381C122; Tue, 17 Oct 2000 13:28:16 +0900 (JST)
Date: Tue, 17 Oct 2000 13:28:09 +0900
Message-ID: <14827.54616.909851.72159A@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Tue, 17 Oct 2000 10:27:12 +0900"
	<14827.43760.126423.72159A@surfchem0.riken.go.jp>
References: <14826.25637.472475.26482Z@surfchem0.riken.go.jp>
	<14826.26984.796923.29321F@surfchem0.riken.go.jp>
	<20001016.164135.88348722.wl@gnu.org>
	<14827.43760.126423.72159A@surfchem0.riken.go.jp>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1905
Lines: 51

Additional comments on my 'compromise' idea.

> One compromise is that:
>  - to use UCS-4 for internal processing, not wchar_t.
>  - a small part of input and output to be encoding-sensible.

A small part of the source code of Groff related to I/O
has to be encoding-sensible.  This part can handle Latin-1,
EBCDIC, and UTF-8.  Additionally, if Groff is compiled within
internationalized OS (i.e. setlocale(), iconv(), nl_langinfo(),
and so on are available), the part also has locale-sensible
file I/O.

The input routine outputs the contents with UCS-4 encoding.
The output routine inputs the contents with UCS-4 encoding.

Other almost part of the source code is written to handle UCS-4.

For example: typedef long ucs4_t; and substitute char with ucs4_t.

>  - command options for encodings of input and output to be added.

For example, --input-encoding and --output-encoding.

>  - a compile-time option I18N to be introduced.

Autoconf may be used for this purpose.

>  - when I18N is off, default input is latin-1 and default output
>    is also latin-1.
>  - when I18N is on, default input and default output are sensible
>    to LC_CTYPE locale.
>  - Of course these default encodings can be overrided by command
>    options.

Thus, valid encoding names for --input-encoding and --output-encoding
command options are: 
  - latin1, ascii, utf8, (ascii8)  [compiled without I18N]
  - latin1, ascii, utf8, (ascii8), and encodings supported by OS
    [compiled with I18N]

>  - Groff can be compiled with I18N off for systems without 
>    internationalization functions such as setlocale().
>  - iconv(3) to be used for converting between input/output encodings
>    and internal UCS-4 encoding, if available (I18N=true).
>  - if I18N is false, conversion process to be hard-coded for
>    Latin-1, EBCDIC, and UTF-8.

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Tue Oct 17 09:27:09 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA00637
	for <groff@ffii.org>; Tue, 17 Oct 2000 09:27:09 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Tue, 17 Oct 2000 09:30:54 +0200
Received: from rigel.univie.ac.at (dial-143144.HRZ.Uni-Dortmund.DE [129.217.143.144])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9H7Uew13942;
	Tue, 17 Oct 2000 09:30:40 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id JAA09051;
	Tue, 17 Oct 2000 09:29:47 +0200
Date: Tue, 17 Oct 2000 09:29:47 +0200 (CEST)
Message-Id: <20001017.092947.10251051.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14827.54616.909851.72159A@surfchem0.riken.go.jp>
References: <20001016.164135.88348722.wl@gnu.org>
	<14827.43760.126423.72159A@surfchem0.riken.go.jp>
	<14827.54616.909851.72159A@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1437
Lines: 34

> A small part of the source code of Groff related to I/O has to be
> encoding-sensible.  This part can handle Latin-1, EBCDIC, and UTF-8.
> Additionally, if Groff is compiled within internationalized OS
> (i.e. setlocale(), iconv(), nl_langinfo(), and so on are available),
> the part also has locale-sensible file I/O.

As mentioned in another mail I would rather like to have this moved to
the iconv preprocessor also.  Latin-1, EBCDIC, and UTF-8 should be
hard-coded if glibc isn't available.

> Other almost part of the source code is written to handle UCS-4.

Maybe we are just arguing about terms, but UCS-4 makes only sense for
character codes, but groff handles glyphs...  So UCS-4 shall *not* be
used internally in troff.

> For example: typedef long ucs4_t; and substitute char with ucs4_t.

Yes, a `long' type is the right one; this gives 31bits for glyphs and
negative values for the `special' characters in troff.

> Thus, valid encoding names for --input-encoding and --output-encoding
> command options are: 
>   - latin1, ascii, utf8, (ascii8)  [compiled without I18N]
>   - latin1, ascii, utf8, (ascii8), and encodings supported by OS
>     [compiled with I18N]

You are thinking only of the tty device, aren't you?  UTF-8 used as an
output encoding for a PS printer doesn't make any sense...  If
possible please refer to the LaTeX model of input and output encodings
if you are talking about these terms.


    Werner

From groff-admin  Tue Oct 17 09:27:10 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA00640
	for <groff@ffii.org>; Tue, 17 Oct 2000 09:27:09 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Tue, 17 Oct 2000 09:30:54 +0200
Received: from rigel.univie.ac.at (dial-143144.HRZ.Uni-Dortmund.DE [129.217.143.144])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9H7Uiw13960;
	Tue, 17 Oct 2000 09:30:45 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id JAA09047;
	Tue, 17 Oct 2000 09:20:37 +0200
Date: Tue, 17 Oct 2000 09:20:37 +0200 (CEST)
Message-Id: <20001017.092037.48403856.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14827.43760.126423.72159A@surfchem0.riken.go.jp>
References: <14826.26984.796923.29321F@surfchem0.riken.go.jp>
	<20001016.164135.88348722.wl@gnu.org>
	<14827.43760.126423.72159A@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 6430
Lines: 166

> The merit of wchar_t is that: write once and work for every
> encodings, uncluding UTF-8.  Otherwise, you have to write similar
> source codes many times for Latin-1, EBCDIC, UTF-8, and so on so on.
> Especially, I will insist that Groff should support EUC-* multibyte
> encodings for CJK languages.  This is what the current Groff cannot
> handle entirely.  (CJK people also uses ISO-2022-* encodings.)

Well, I insist that GNU troff doesn't support multibyte enodings at
all :-) troff itself should work on a glyph basis only.  It has to
work with *glyph names*, be it CJK entities or whatever.  Currently,
the conversion from input encoding to glyph entities and the further
processing of glyphs is not clearly separated.  From a modular point
of view it makes sense if troff itself is restricted to a single input
encoding (UTF-8) which is basically only meant as a wrapper to glyph
names (cf. \U'xxxx' to enter Unicode encoded characters).  Everything
else should be moved to a preprocessor.

> The other merit of wchar_t is user-friendliness.  Once a user set
> LANG variable, every softwares work under the specified encoding.
> If not, you have to specify encodings for every software.  We don't
> want to have ~/.groffrc, ~/.greprc, ~/.bashrc, ~/.xtermrc, and so on
> so on to specify 'encoding=ISO8859-1' or 'encoding=UTF-8'.

I must admit that I've never worked with wchar_t and friends.  I see
the benefits, though.  And with glibc, we have a nice ready-to-use
library which enormously simplifies the implementation.

> >> Abolish device types of 'ascii', 'ascii8', 'latin1', 'nippon', and
> >> 'utf8' and introduce a new device type such as 'tty'.
> 
> I suppose you don't know about 'ascii8' device.

I know it :-)  I've studied the Japanese patch in detail.
 
> Of course I think of portability.  wchar_t is portable.  I recommend
> to implement wchar_t as a new architecture and ascii, latin-1, and
> ebcdic as historical encodings.  (We may add 'UTF8' as a historical
> one.)

Yes.

> > Maybe on systems with a recent glibc, iconv() and friends can be
> > used to do more, but generally I prefer an iconv-preprocessor so
> > that groff itself has not to deal with encoding conversions.
> 
> I think this works well.  However, who invokes iconv-preprocessor?

groff (the program).

> A user or wrapper-software?

groff *is* the wrapper of troff.

> What determines the command option for iconv?

I suggest various possibilities.

  . By default, locales should be used.  A --locale option might be a
    good idea also since this has implications on text processing;
    example: Portuguese doesn't use fi and fl ligatures.  The problem
    is that a locale influences both input encoding and text
    processing which might not always be the right choice.

  . Another approach to influence text processing (not input encoding)
    is the OpenType way: --script, --language, --feature.

  . Command lines shall be able to override input encoding
    (--input-encoding).

  . Finally, we need to divide the -T option into a --device and
    --output-encoding.

> > > - Groff assumes the input as the encoding of current locale.
> > This is probably not correctly set everywhere.
> 
> How a system can be configured by a user, in ways other than locale?
> A user who want to specify his/her language and encoding will set
> LANG variable.  Or, having many ~/.foobarrc for every softwares or
> specifying --encoding=foobar everytime (s)he invokes a software?  I
> think setting LANG is a reasonable way.

Believe me, most professional UNIX users in Germany don't have LANG
set correctly (including me).  For example, I don't like to see German
error messages since I'm used to the English ones.  In fact, I never
got used to the German way of handling computers.  The German keyboard
is awkward for programmers...

Note that I consider groff not only as a tool to format man pages but
as a text processing tool also which must be able to work completely
independent from the locale.

> One compromise is that:
>  - to use UCS-4 for internal processing, not wchar_t.

The internal processing within groff should use glyph IDs resp. glyph
names.

>  - a small part of input and output to be encoding-sensible.

???  Please explain.

>  - command options for encodings of input and output to be added.

Yes.

>  - a compile-time option I18N to be introduced.

Yes.  The `iconv' preprocessor would then do some trivial, hard-coded
conversion.

>  - when I18N is off, default input is latin-1 and default output
>    is also latin-1.

Sounds reasonable.

>  - when I18N is on, default input and default output are sensible
>    to LC_CTYPE locale.

Default output?  Really?  Please explain.

>  - iconv(3) to be used for converting between input/output encodings
>    and internal UCS-4 encoding, if available (I18N=true).
>  - if I18N is false, conversion process to be hard-coded for
>    Latin-1, EBCDIC, and UTF-8.

Yes.

> Do you think this can be achieved?

Yes it can, but not tomorrow...

My roadmap is roughly as follows.  It won't introduce radical changes
immediately; probably this results in even more work.  But given my
time constraints I believe this route is better.

Volunteers are highly welcome.  And please comment whether you like my
ideas at all.

I'll start with that after I've finished my reimplementation of the
mdoc package.  Don't expect this happen too quickly :-(

  . Separate input encodings from font encodings.  Remove all
    references to charXXX from the font definition files.  Provide
    input encoding files similar to LaTeX.

  . Make troff work internally with 32bit glyph entities.  I believe
    this is the hardest job of all subtasks due to the complexity of
    changes.

  . Write the input encoding preprocessor.  If you like, you can
    immediately start with it since it is completely independent from
    troff itself.

  . Update all groff utilities.

My ideas follow Omega, one of TeX's successors.  I don't intend to
make groff that powerful (for example, I don't have plans to support
various writing directions), but the conceptual separation will be
similar: The iconv preprocessor will be equal to Omega's input OTPs.
Options like --language or --feature will influence troff's internal
engine.  If necessary (for tty devices only, I think), iconv-like
routines are used to map glyph names back to characters so that groff
can act as a filter.


    Werner

From groff-admin  Tue Oct 17 11:52:02 2000
Received: from clyde.ref.collins.co.uk (clyde.ref.collins.co.uk [192.153.153.29])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id LAA01470
	for <groff@ffii.org>; Tue, 17 Oct 2000 11:52:00 +0200
Received: from ref.collins.co.uk (helmsdale [193.119.38.15])
	by clyde.ref.collins.co.uk (8.9.3/8.9.3) with ESMTP id KAA15564
	for <groff@ffii.org>; Tue, 17 Oct 2000 10:55:26 +0100 (BST)
Received: from ref.collins.co.uk (localhost [127.0.0.1])
	by ref.collins.co.uk (8.9.3+Sun/8.9.1) with ESMTP id KAA10648
	for <groff@ffii.org>; Tue, 17 Oct 2000 10:54:54 +0100 (BST)
Message-ID: <39EC21EE.92337A7@ref.collins.co.uk>
Date: Tue, 17 Oct 2000 10:54:54 +0100
From: "Stewart C. Russell" <stewart@ref.collins.co.uk>
Organization: HarperCollins Publishers, Bilingual Division
X-Mailer: Mozilla 4.61 [en] (X11; U; SunOS 5.7 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
To: groff mail list <groff@ffii.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] pdfmarks
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 280
Lines: 10

Has anyone here done work with pdfmarks and groff? I'd like to put some
watermarks, cropping instructions and keywords in a distilled groff
file, and was wondering if:

- anyone had any macros they'd like to share, and/or

- if there were any problems encountered.

tvm,
 Stewart

From groff-admin  Tue Oct 17 11:52:55 2000
Received: from clyde.ref.collins.co.uk (clyde.ref.collins.co.uk [192.153.153.29])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id LAA01491
	for <groff@ffii.org>; Tue, 17 Oct 2000 11:52:52 +0200
Received: from ref.collins.co.uk (helmsdale [193.119.38.15])
	by clyde.ref.collins.co.uk (8.9.3/8.9.3) with ESMTP id KAA15572
	for <groff@ffii.org>; Tue, 17 Oct 2000 10:57:17 +0100 (BST)
Received: from ref.collins.co.uk (localhost [127.0.0.1])
	by ref.collins.co.uk (8.9.3+Sun/8.9.1) with ESMTP id KAA10652
	for <groff@ffii.org>; Tue, 17 Oct 2000 10:56:45 +0100 (BST)
Message-ID: <39EC225D.35C3EE3E@ref.collins.co.uk>
Date: Tue, 17 Oct 2000 10:56:45 +0100
From: "Stewart C. Russell" <stewart@ref.collins.co.uk>
Organization: HarperCollins Publishers, Bilingual Division
X-Mailer: Mozilla 4.61 [en] (X11; U; SunOS 5.7 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
To: groff mail list <groff@ffii.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] type42 fonts
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 189
Lines: 7

I'm sure groff will have no problem with Type42 fonts (TrueType fonts in
a PS font wrapper) -- will it?

I've been burned by seemingly-savvy applications failing on these
things.

 Stewart

From groff-admin  Tue Oct 17 15:10:35 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA03093
	for <groff@ffii.org>; Tue, 17 Oct 2000 15:10:32 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 755B31C122; Tue, 17 Oct 2000 22:19:15 +0900 (JST)
Date: Tue, 17 Oct 2000 22:19:09 +0900
Message-ID: <14828.20941.944311.81928B@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Tue, 17 Oct 2000 09:20:37 +0200 (CEST)"
	<20001017.092037.48403856.wl@gnu.org>
References: <14826.26984.796923.29321F@surfchem0.riken.go.jp>
	<20001016.164135.88348722.wl@gnu.org>
	<14827.43760.126423.72159A@surfchem0.riken.go.jp>
	<20001017.092037.48403856.wl@gnu.org>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 8455
Lines: 212

Hi,

At Tue, 17 Oct 2000 09:20:37 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> Well, I insist that GNU troff doesn't support multibyte enodings at
> all :-) troff itself should work on a glyph basis only.  It has to
> work with *glyph names*, be it CJK entities or whatever.  Currently,
> the conversion from input encoding to glyph entities and the further
> processing of glyphs is not clearly separated.  From a modular point
> of view it makes sense if troff itself is restricted to a single input
> encoding (UTF-8) which is basically only meant as a wrapper to glyph
> names (cf. \U'xxxx' to enter Unicode encoded characters).  Everything
> else should be moved to a preprocessor.

This paragraph says two things:
 - GNU troff will support UTF-8 only.  Thus, multibyte encodings
   will be not supported.  [Though UTF-8 is multibyte :-p ]
 - Groff handles glyph, not character.
I don't understand relationship between these two.  UTF-8 is a code
for character, not glyph.  ISO8859-1 and EUC-JP are also codes for
character.  No difference among UTF-8, ISO8859-1, and EUC-JP.

However, I won't stick to wchar_t or ucs-4 for internal code,
though I have no idea about your '31bit glyph code'.  (Maybe I
have to study Omega...)



> > The other merit of wchar_t is user-friendliness.  Once a user set
> > LANG variable, every softwares work under the specified encoding.
> > If not, you have to specify encodings for every software.  We don't
> > want to have ~/.groffrc, ~/.greprc, ~/.bashrc, ~/.xtermrc, and so on
> > so on to specify 'encoding=ISO8859-1' or 'encoding=UTF-8'.
> 
> I must admit that I've never worked with wchar_t and friends.  I see
> the benefits, though.  And with glibc, we have a nice ready-to-use
> library which enormously simplifies the implementation.

Ok, the merit can be achieved by your idea of preprocessor.
Please note that glibc is not a only system which supporte locale.
I suppose that almost commercial UNIX systems support locale.
I suppose BSD variants (Free-, Net-, and Open-) also, though I 
don't know them.

You know, Sun Microsystems recently offered an internationalization
technology to open source community.  (http://www.sun.com/solaris/global/)
This means Solaris supports locale.


> > I suppose you don't know about 'ascii8' device.
> 
> I know it :-)  I've studied the Japanese patch in detail.

Wonderful!  However, though it is true that I wrote the
'ascii8' patch, this patch has no relation to the Japanese patch.

My patch consists of two parts:
 - `cp -r devlatin1 devascii8` and added 'char***' for >0x80.
 - hard-coded soft-hyphenation character (which is valid only 
   for Latin-1) is changed (this seems to be fixed by Groff 1.16).



> > > Maybe on systems with a recent glibc, iconv() and friends can be
> > > used to do more, but generally I prefer an iconv-preprocessor so
> > > that groff itself has not to deal with encoding conversions.
> > I think this works well.  However, who invokes iconv-preprocessor?
> groff (the program).
> > A user or wrapper-software?
> groff *is* the wrapper of troff.

I understand.  This works, since it is not likely that a user invokes
troff directly.


> > What determines the command option for iconv?
> 
> I suggest various possibilities.

At first I say that I agree with your idea in almost part.


>   . By default, locales should be used.  A --locale option might be a
>     good idea also since this has implications on text processing;
>     example: Portuguese doesn't use fi and fl ligatures.  The problem
>     is that a locale influences both input encoding and text
>     processing which might not always be the right choice.

The name '--locale' is confusing since it has no relation to
locale, i.e., a term which refer to a certain standard technology.

Though you may already know, please note that
 - Japanese and Chinese text contains few whitespace characters.
   (Japanese and Chinese words are not separated by whitespace).
   Therefore, different line-breaking algorithm should be used.
   (Hyphen character is not used when a word is broken into lines.)
   (Modern Korean language contains whitespace characters between
   words --- though not words, strictly speaking.)
 - Hyphenation algorithm differs from language to language.
 - Almost CJK characters (ideographics, hiragana, katakana, hangul,
   and so on) have double width on tty.  Since you won't use wchar_t,
   you cannot use wcwidth() to get the width for characters.
   The source code for Xterm multiwidth/combining character extention
   patch may help you, though it is based on unicode.
   http://www.ecs.soton.ac.uk/~rwb197/xterm/
 - Latin-1 people may use 0xa9 for '\(co'.  However, this character
   cannot be read in other encodings.  The current Groff convert
   '\(co' to 0xa9 in latin1 device and to '(C)' in ascii device.
   How it works for future Groff?  Use u+00a9?  The postprocessor
   (see below) cannot convert u+00a9 to '(C)' because the width is
   different and typesetting is broken.  It is very difficult to
   design to avoid this problem...

>   . Another approach to influence text processing (not input encoding)
>     is the OpenType way: --script, --language, --feature.
Though I don't know them, I hope these can handle Asian languages.

>   . Command lines shall be able to override input encoding
>     (--input-encoding).
Yes.

>   . Finally, we need to divide the -T option into a --device and
>     --output-encoding.

What is the default encoding for tty?  I suggest this should be
locale-sensible.  (Or, this can be UTF-8 and Groff can invoke a
postprocessor.)


> Believe me, most professional UNIX users in Germany don't have LANG
> set correctly (including me).  For example, I don't like to see German
> error messages since I'm used to the English ones.  In fact, I never
> got used to the German way of handling computers.  The German keyboard
> is awkward for programmers...

I think all of you will need LANG variable when UTF-8 becomes popular.
This is because UTF-8 is multibyte.  There are already many softwares
which are internationalized using locale mechanism.  These softwares
support UTF-8 if you set LANG variable.

If you like English message, you may set 'LANG=en_US.UTF-8'.


> > One compromise is that:
> >  - to use UCS-4 for internal processing, not wchar_t.
> 
> The internal processing within groff should use glyph IDs resp. glyph
> names.

As I wrote above, I won't stick to the internal code, though I 
have no idea what is glyph code.


> >  - a small part of input and output to be encoding-sensible.
> ???  Please explain.

Sorry, I meant:
  A small part of the source code of Groff related to I/O
  has to be encoding-sensible.
as I wrote in the previous mail.  You already wrote a comment
on this point.  Since I agree with your idea of preprocessor,
discussion on this idea has finished.


> >  - a compile-time option I18N to be introduced.
> 
> Yes.  The `iconv' preprocessor would then do some trivial, hard-coded
> conversion.

You mean, the preprocessor is iconv(1) ?
The preprocessor, provisional name 'gpreconv', will be designed as:
 - includes hard-coded converter for latin1, ebcdic, and utf8.
 - uses iconv(3) if possible (compiled within internationalized OS).
 - parses --input-encoding option.
 - default input is latin1 if compiled within non-internationalized OS.
 - default input is locale-sensible if compiled within internationalized OS.
I think it is easy to implement 'gpreconv'.


> >  - when I18N is off, default input is latin-1 and default output
> >    is also latin-1.
> Sounds reasonable.

Thus I designed the above 'gpreconv'.  Oh, I have to design 'gpostconv'
also.


> >  - when I18N is on, default input and default output are sensible
> >    to LC_CTYPE locale.
> Default output?  Really?  Please explain.

Default output encoding for tty device.  This can be achieved by
'gpostconv' postprocessor.


> I'll start with that after I've finished my reimplementation of the
> mdoc package.  Don't expect this happen too quickly :-(

I see.  All of us are volunteers and have no deadline.


>   . Write the input encoding preprocessor.  If you like, you can
>     immediately start with it since it is completely independent from
>     troff itself.

Ok, I think I can write that in a few weeks.  It is easy, though
the source code of Groff itself is difficult for me :-)

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Tue Oct 17 17:01:54 2000
Received: from probity.mcc.ac.uk (probity.mcc.ac.uk [130.88.200.94])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id RAA04013
	for <groff@ffii.org>; Tue, 17 Oct 2000 17:01:53 +0200
Received: from tedistan.annex.mcc.ac.uk
	([194.66.22.200] helo=tedistan.annex.umist.ac.uk ident=root)
	by probity.mcc.ac.uk with esmtp (Exim 2.05 #4)
	id 13lYJt-000I3N-00; Tue, 17 Oct 2000 16:06:25 +0100
Received: (from efh@localhost)
	by eels.fort.knox.uk (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) id PAA27944;
	Tue, 17 Oct 2000 15:01:11 +0100
Message-ID: <XFMail.001017150111.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: XFMail 1.3 [p0] on Linux
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
In-Reply-To: <39EC21EE.92337A7@ref.collins.co.uk>
Date: Tue, 17 Oct 2000 15:01:11 +0100 (BST)
Reply-To: Ted.Harding@nessie.mcc.ac.uk
From: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
To: "Stewart C. Russell" <stewart@ref.collins.co.uk>
Subject: RE: [Groff] pdfmarks
Cc: groff mail list <groff@ffii.org>
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1871
Lines: 55

On 17-Oct-00 Stewart C. Russell wrote:
> Has anyone here done work with pdfmarks and groff? I'd like to put some
> watermarks, cropping instructions and keywords in a distilled groff
> file, and was wondering if:
> 
> - anyone had any macros they'd like to share, and/or
> 
> - if there were any problems encountered.

The basic mechanism for including pdfmarks is very simple.

First. doctor your "prologue" file in .../devps to include, near the
beginning, the code

/pdfmark where {pop} {userdict /pdfmark /cleartomark load put} ifelse

The following is where I've put it myself:

/setpacking where {
        pop
        currentpacking
        true setpacking
} if
/pdfmark where {pop} {userdict /pdfmark /cleartomark load put} ifelse
/grops 120 dict dup begin 

Next, whenever you want to plant a pdfmark, simply use the
\X'ps: exec ....' code to do it:

\X'ps: exec [ +++++++++++++ pdfmark'

where the "+++++++++++" denotes the content of the pdfmark.

Now the fun starts. You need to convert groff's PostScript output into
PDF using a converter which will do a proper job with pdfmarks. I believe
Adobe's Distiller will always do a proper job; at any rate, on those
occasions when I have asked Distiller-enabled friends to do it for
me, it has never failed.

Other than that, 'ps2pdf' in the latest ghostscript now does a very
reasonable job with most variants of pdfmark, I understand.

So also, with some gaps, does the latest version of 'pstill'.

In both cases, however, I have had pdfmarks which were not converted
properly. Your mileage may vary.

Good luck,
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 284 7749
Date: 17-Oct-00                                       Time: 15:01:11
------------------------------ XFMail ------------------------------

From groff-admin  Tue Oct 17 17:44:14 2000
Received: from lavender.ukai.org (ns.ukai.debian.gr.jp [210.157.158.39])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id RAA04254
	for <groff@ffii.org>; Tue, 17 Oct 2000 17:44:13 +0200
X-Authentication-Warning: genba.ffii.org: Host ns.ukai.debian.gr.jp [210.157.158.39] claimed to be lavender.ukai.org
Received: from lichee.ukai.org.ukai.org (unknown [211.123.26.7])
	by lavender.ukai.org (Postfix) with ESMTP
	id 0363952A02; Wed, 18 Oct 2000 00:48:28 +0900 (JST)
Date: Wed, 18 Oct 2000 00:46:26 +0900
Message-ID: <87lmvnv5bh.wl@lichee.ukai.org>
From: Fumitoshi UKAI <ukai@debian.or.jp>
To: groff@ffii.org
Cc: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org
Subject: Re: [Groff] Re: groff: radical re-implementation
In-Reply-To: In your message of "Tue, 17 Oct 2000 22:19:09 +0900"
	<14828.20941.944311.81928B@surfchem0.riken.go.jp>
References: <14826.26984.796923.29321F@surfchem0.riken.go.jp>
	<20001016.164135.88348722.wl@gnu.org>
	<14827.43760.126423.72159A@surfchem0.riken.go.jp>
	<20001017.092037.48403856.wl@gnu.org>
	<14828.20941.944311.81928B@surfchem0.riken.go.jp>
User-Agent: Wanderlust/2.2.15 (More Than Words) EMIKO/1.13.9 (Euglena tripteris) FLIM/1.13.2 (Kasanui) APEL/10.2 Emacs/20.7 (i386-debian-linux-gnu) MULE/4.0 (HANANOEN)
Organization: Debian JP Project
MIME-Version: 1.0 (generated by EMIKO 1.13.9 - "Euglena tripteris")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 3898
Lines: 88

At Tue, 17 Oct 2000 22:19:09 +0900,
Tomohiro KUBOTA <tkubota@riken.go.jp> wrote:

> Though you may already know, please note that
>  - Japanese and Chinese text contains few whitespace characters.
>    (Japanese and Chinese words are not separated by whitespace).
>    Therefore, different line-breaking algorithm should be used.
>    (Hyphen character is not used when a word is broken into lines.)
>    (Modern Korean language contains whitespace characters between
>    words --- though not words, strictly speaking.)

As regards line breaking algorithm, I think we need some more cflags,
at least for Japanese.  That is,
 
   - lines must not be broken before the character
   - lines must not be broken after the character

These seems to be implemented as PRE_KINSOKU and POST_KINSOKU in jgroff,
but it's done by hardcoded.  I think this should be done by tmac.<lang>, 
so I think it's good idea to have some mechanisms to load language 
specific tmac files.

BTW, what do you think about code name for multibyte character/wide
character or glyph code what you said?  In jgroff, it seems it used
wchar<EUCcode>.

>  - Hyphenation algorithm differs from language to language.

This is already implemented by .hla <language>, isn't it?

>  - Almost CJK characters (ideographics, hiragana, katakana, hangul,
>    and so on) have double width on tty.  Since you won't use wchar_t,
>    you cannot use wcwidth() to get the width for characters.
>    The source code for Xterm multiwidth/combining character extention
>    patch may help you, though it is based on unicode.
>    http://www.ecs.soton.ac.uk/~rwb197/xterm/

I think we could use font description information for it. 
jgroff provides "fixedkanji" directive in font description. But, 
the code of font description loader depends on EUC<->KuTen mapping, 
and it's not good idea for i18n.   I think it would be better to provide
"wcharset" directive which support code range.  However, code range 
couldn't be used with EUC encoding or something like that,  and not used
for Unicode, because we couldn't expect character codes for some language
are in succession.

Anyway, jgroff provides new font "M" and "G", which are "Mincho" and
"Gothic" respectively, for wide characters.  What is the right way
to add i18n support in groff about font description?

>  - Latin-1 people may use 0xa9 for '\(co'.  However, this character
>    cannot be read in other encodings.  The current Groff convert
>    '\(co' to 0xa9 in latin1 device and to '(C)' in ascii device.
>    How it works for future Groff?  Use u+00a9?  The postprocessor
>    (see below) cannot convert u+00a9 to '(C)' because the width is
>    different and typesetting is broken.  It is very difficult to
>    design to avoid this problem...

Is it impossible to be implemented by character translation mechanism
in groff?  I think it's the right way to select which translation is
used by output device, that is, -T<device> option.  It might be
check $TERM environment for tty devices, because terminal emulator
for Japanese wouldn't show u+00a9 correctly, so it requires some translation.

> >   . Another approach to influence text processing (not input encoding)
> >     is the OpenType way: --script, --language, --feature.
> Though I don't know them, I hope these can handle Asian languages.
> 
> >   . Command lines shall be able to override input encoding
> >     (--input-encoding).
> Yes.

How about creating new request (.encoding "encoding-name") in roff 
source?

> > > One compromise is that:
> > >  - to use UCS-4 for internal processing, not wchar_t.
> > 
> > The internal processing within groff should use glyph IDs resp. glyph
> > names.

Could you explain (or point us which source code in groff) how to map
glyph IDs to output code, please?
For tty output device, UTF-8 and postprocessor may be possible, but
how about X, ps, or dvi?

Regards,
Fumitoshi UKAI

From groff-admin  Tue Oct 17 18:40:25 2000
Received: from mailout1.hananet.net (mailout1.hananet.net [210.220.163.34])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA04624
	for <groff@ffii.org>; Tue, 17 Oct 2000 18:40:17 +0200
Received: from gnomaniac.myhome ([211.108.30.127]) by
          mailout1.hananet.net (Netscape Messaging Server 4.15) with ESMTP
          id G2L2IK02.V05; Wed, 18 Oct 2000 01:44:45 +0900 
Received: (from cjh@localhost)
	by gnomaniac.myhome (8.11.1/8.11.0) id e9HGhdD01877;
	Wed, 18 Oct 2000 01:43:39 +0900 (KST)
	(envelope-from cjh@kr.FreeBSD.org)
X-Authentication-Warning: gnomaniac.myhome: cjh set sender to cjh@kr.FreeBSD.org using -f
To: Tomohiro KUBOTA <tkubota@riken.go.jp>
Cc: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
References: <14826.26984.796923.29321F@surfchem0.riken.go.jp>
	<20001016.164135.88348722.wl@gnu.org>
	<14827.43760.126423.72159A@surfchem0.riken.go.jp>
	<20001017.092037.48403856.wl@gnu.org>
	<14828.20941.944311.81928B@surfchem0.riken.go.jp>
From: CHOI Junho <cjh@kr.FreeBSD.org>
Date: 18 Oct 2000 01:43:38 +0900
In-Reply-To: Tomohiro KUBOTA's message of "Tue, 17 Oct 2000 22:19:09 +0900"
Message-ID: <863dhve7ut.fsf@gnomaniac.myhome>
User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.7
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1546
Lines: 28

>>>>> "TK" == Tomohiro KUBOTA <tkubota@riken.go.jp> writes:

    >  The other merit of wchar_t is user-friendliness.  Once a user set
    >  LANG variable, every softwares work under the specified encoding.
    >  If not, you have to specify encodings for every software.  We don't
    >  want to have ~/.groffrc, ~/.greprc, ~/.bashrc, ~/.xtermrc, and so on
    >  so on to specify 'encoding=ISO8859-1' or 'encoding=UTF-8'.
    >  
    >  I must admit that I've never worked with wchar_t and friends.  I see
    >  the benefits, though.  And with glibc, we have a nice ready-to-use
    >  library which enormously simplifies the implementation.

    TK> Ok, the merit can be achieved by your idea of preprocessor.
    TK> Please note that glibc is not a only system which supporte locale.
    TK> I suppose that almost commercial UNIX systems support locale.
    TK> I suppose BSD variants (Free-, Net-, and Open-) also, though I 
    TK> don't know them.

FreeBSD locale system don't have UTF-8 encoding. NetBSD and OpenBSD is
same or worse than FreeBSD. So in case of this, iconv() should come from
external libraries(FreeBSD ports has it, converters/iconv). There are
plans to have iconv() internally near future, though.

-- 
 +++ Any opinions in this posting are my own and not those of my employers +++
 CHOI Junho                             <cjh@FreeBSD.ORG> <cjh@kr.FreeBSD.ORG>
 KFUG <http://www.kr.FreeBSD.org>         Web Data Bank <http://www.wdb.co.kr>
 FreeBSD, GNU/Linux Developer                   http://www.kr.FreeBSD.org/~cjh

From groff-admin  Tue Oct 17 18:49:43 2000
Received: from mailout1.hananet.net (mailout1.hananet.net [210.220.163.34])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA04700
	for <groff@ffii.org>; Tue, 17 Oct 2000 18:49:41 +0200
Received: from gnomaniac.myhome ([211.108.30.127]) by
          mailout1.hananet.net (Netscape Messaging Server 4.15) with ESMTP
          id G2L2YD00.S03; Wed, 18 Oct 2000 01:54:13 +0900 
Received: (from cjh@localhost)
	by gnomaniac.myhome (8.11.1/8.11.0) id e9HGrCj01993;
	Wed, 18 Oct 2000 01:53:12 +0900 (KST)
	(envelope-from cjh@kr.FreeBSD.org)
X-Authentication-Warning: gnomaniac.myhome: cjh set sender to cjh@kr.FreeBSD.org using -f
To: Fumitoshi UKAI <ukai@debian.or.jp>
Cc: groff@ffii.org, debian-i18n@lists.debian.org
Subject: Re: [Groff] Re: groff: radical re-implementation
References: <14826.26984.796923.29321F@surfchem0.riken.go.jp>
	<20001016.164135.88348722.wl@gnu.org>
	<14827.43760.126423.72159A@surfchem0.riken.go.jp>
	<20001017.092037.48403856.wl@gnu.org>
	<14828.20941.944311.81928B@surfchem0.riken.go.jp>
	<87lmvnv5bh.wl@lichee.ukai.org>
From: CHOI Junho <cjh@kr.FreeBSD.org>
Date: 18 Oct 2000 01:53:10 +0900
In-Reply-To: Fumitoshi UKAI's message of "Wed, 18 Oct 2000 00:46:26 +0900"
Message-ID: <86u2abcsuh.fsf@gnomaniac.myhome>
User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.7
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1381
Lines: 30

>>>>> "FU" == Fumitoshi UKAI <ukai@debian.or.jp> writes:

    FU> At Tue, 17 Oct 2000 22:19:09 +0900,
    FU> Tomohiro KUBOTA <tkubota@riken.go.jp> wrote:

    >  Though you may already know, please note that
    >   - Japanese and Chinese text contains few whitespace characters.
    >     (Japanese and Chinese words are not separated by whitespace).
    >     Therefore, different line-breaking algorithm should be used.
    >     (Hyphen character is not used when a word is broken into lines.)
    >     (Modern Korean language contains whitespace characters between
    >     words --- though not words, strictly speaking.)

    FU> As regards line breaking algorithm, I think we need some more cflags,
    FU> at least for Japanese.  That is,
 
[...]

    FU> For tty output device, UTF-8 and postprocessor may be possible, but
    FU> how about X, ps, or dvi?

It sounds to me that we should make the beast again :< How about
making a good roff -> html converter and call mozilla to do proper
conversion(X11, ps, text, etc...)? :) Just kidding. sorry.

-- 
 +++ Any opinions in this posting are my own and not those of my employers +++
 CHOI Junho                             <cjh@FreeBSD.ORG> <cjh@kr.FreeBSD.ORG>
 KFUG <http://www.kr.FreeBSD.org>         Web Data Bank <http://www.wdb.co.kr>
 FreeBSD, GNU/Linux Developer                   http://www.kr.FreeBSD.org/~cjh

From groff-admin  Tue Oct 17 19:16:23 2000
Received: from cmailg7.svr.pol.co.uk (cmailg7.svr.pol.co.uk [195.92.195.177])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id TAA04866
	for <groff@ffii.org>; Tue, 17 Oct 2000 19:16:22 +0200
Received: from [195.92.67.23] (helo=mail18.svr.pol.co.uk)
	by cmailg7.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13laQ4-00069H-00
	for groff@ffii.org; Tue, 17 Oct 2000 18:20:56 +0100
Received: from modem-65.porcupine-puffer.dialup.pol.co.uk ([62.137.50.193] helo=merlin.freeserve.net)
	by mail18.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13laQ2-0001vA-00
	for groff@ffii.org; Tue, 17 Oct 2000 18:20:54 +0100
Received: by merlin.freeserve.net (Linux Smail3.2.0.101 #1)
	id m13lafz-000ix9C; Tue, 17 Oct 2000 18:37:23 +0100 (BST)
Message-Id: <m13lafz-000ix9C@merlin.freeserve.net>
Date: Tue, 17 Oct 2000 18:37:23 +0100 (BST)
From: Gaius Mulley <gaius@glam.ac.uk>
To: groff@ffii.org
Subject: [Groff] patches for html2 and specific help requested
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 5892
Lines: 199

Hi Werner,

Here are some patches for html2. These patches allow groff -Thtml2
to create minimal png files. They also provide a fix to html w.r.t
eqn handling.

I wonder whether I could ask for some help in the following areas:

tmac/tmac.html-tags:

could someone check that the tag generation mechanism is
non intrusive.

src/preproc/html2/pre-html.cc: in function createImage
I'm having difficulty producing transparent png files.
If anyone has done this before - maybe you can drop me
an email.

in src/src/devices/grohtml/html.cc the function
assign_min_max_for_arc works out the smallest box which
will encompass an arc. I wonder whether a groff friend
who is also a mathematician can find a better solution
than my 16 tests :-) this code is also needed in troff
src/roff/troff/node.cc:determine_line_limits. But
I've not copied it yet..

many thanks Gaius


--- groff-cvs/ChangeLog	Sun Oct 15 00:57:32 2000
+++ groff-html/ChangeLog	Tue Oct 17 17:17:59 2000
@@ -1,3 +1,9 @@
+2000-10-17  Gaius Mulley  <gaius@glam.ac.uk>
+
+	* src/roff/troff/node.cc fixed calculation of opminx
+	and fixed non intrusive eol marker.
+	Added method determine_line_limits.
+	
 2000-10-14  Werner LEMBERG  <wl@gnu.org>
 
 	Replace tmac.safer with a real secure solution.
--- groff-cvs/src/roff/troff/node.cc	Fri Aug 25 08:44:08 2000
+++ groff-html/src/roff/troff/node.cc	Tue Oct 17 17:18:51 2000
@@ -752,6 +752,7 @@
   void really_on();
   void really_off();
   void draw(char, hvpair *, int, font_size);
+  void determine_line_limits (char code, hvpair *point, int npoints);
   int get_hpos() { return hpos; }
   int get_vpos() { return vpos; }
 };
@@ -896,6 +897,9 @@
     put(tbuf_kern);
     put(' ');
   }
+  
+  check_output_limits(output_hpos, output_vpos);
+
   for (int i = 0; i < tbuf_len; i++)
     put(tbuf[i]);
   put('\n');
@@ -1061,6 +1065,42 @@
   current_tfont = tf;
 }
 
+/*
+ *  determine_line_limits - works out the smallest box which will contain
+ *                          the entity, code, built from the point array.
+ */
+
+void troff_output_file::determine_line_limits (char code, hvpair *point, int npoints)
+{
+  int i, x, y;
+
+  switch (code) {
+
+  case 'c':
+  case 'C':
+    /* only the h field is used when defining a circle */
+    check_output_limits(output_hpos, output_vpos-point[0].h.to_units()/2);
+    check_output_limits(output_hpos+point[0].h.to_units(), output_vpos+point[0].h.to_units()/2);
+    break;
+  case 'E':
+  case 'e':
+    check_output_limits(output_hpos, output_vpos-point[1].v.to_units()/2);
+    check_output_limits(output_hpos+point[0].h.to_units(), output_vpos+point[1].v.to_units()/2);
+    break;
+  default:
+    /*
+     *  remember this doesn't work for arc..
+     */
+    x=output_hpos;
+    y=output_vpos;
+    for (i=0; i<npoints; i++) {
+      x += point[i].h.to_units();
+      y += point[i].v.to_units();
+      check_output_limits(x, y);
+    }
+  }
+}
+
 void troff_output_file::draw(char code, hvpair *point, int npoints,
 			     font_size fsize)
 {
@@ -1088,6 +1128,9 @@
       put(' ');
       put(point[i].v.to_units());
     }
+
+  determine_line_limits(code, point, npoints);
+
   for (i = 0; i < npoints; i++)
     output_hpos += point[i].h.to_units();
   hpos = output_hpos;
@@ -1324,14 +1367,6 @@
 {
   if (printing && output_on)
     really_print_line(x, y, n, before, after, width);
-
-  if (before.to_units() < after.to_units()) {
-    check_output_limits(x.to_units()                 , y.to_units()+before.to_units());
-    check_output_limits(x.to_units()+width.to_units(), y.to_units()+after.to_units()+n->size());
-  } else {
-    check_output_limits(x.to_units()                 , y.to_units()+after.to_units());
-    check_output_limits(x.to_units()+width.to_units(), y.to_units()+before.to_units()+n->size());
-  }
   delete_node_list(n);
 }
 
--- groff-cvs/src/roff/troff/env.cc	Fri Aug 25 08:44:08 2000
+++ groff-html/src/roff/troff/env.cc	Tue Oct 17 15:59:41 2000
@@ -1946,8 +1946,9 @@
       if (!illegal_input_char((unsigned char)*p))
 	m->append(*p);
     output_pending_lines();
-    output_line(new special_node(*m), 0);
-  }  
+    output(new special_node(*m), !fill, 0, 0, 0);
+    output_pending_lines();
+  }
 }
 
 void environment::do_break()
--- groff-cvs/src/preproc/eqn/main.cc	Fri Mar 10 00:35:32 2000
+++ groff-html/src/preproc/eqn/main.cc	Mon Oct 16 09:43:34 2000
@@ -110,8 +110,8 @@
       }
       restore_compatibility();
       printf(".lf %d\n", current_lineno);
-      graphic_end();
       put_string(linebuf, stdout);
+      graphic_end();
     }
     else if (start_delim != '\0' && linebuf.search(start_delim) >= 0
 	     && inline_equation(fp, linebuf, str))
--- groff-cvs/src/preproc/html2/pre-html.cc	Mon Oct  9 20:41:59 2000
+++ groff-html/src/preproc/html2/pre-html.cc	Tue Oct 17 17:50:10 2000
@@ -39,6 +39,10 @@
 #include <unistd.h>
 #endif
 
+#ifndef errno
+extern int errno;
+#endif
+
 #ifdef _POSIX_VERSION
 #include <sys/wait.h>
 #define PID_T pid_t
@@ -267,7 +271,7 @@
   char_block *t=head;
   int r;
 
-  writeString(".nr html2enable 1\n");
+  writeString(".nr html2enable 0\n");
   writeString(".nr htmlflip 0\n");
   if (t != 0) {
     do {
@@ -383,14 +387,17 @@
 	    image_device,
 	    image_res,
 	    psFileName,
-	    i->X1*image_res/POSTSCRIPTRES,
+	    i->X1*image_res/POSTSCRIPTRES-IMAGE_BOARDER_PIXELS,
 	    i->Y1*image_res/POSTSCRIPTRES-IMAGE_BOARDER_PIXELS,
-	    (i->X2-i->X1)*image_res/POSTSCRIPTRES+IMAGE_BOARDER_PIXELS,
-	    (i->Y2-i->Y1)*image_res/POSTSCRIPTRES+IMAGE_BOARDER_PIXELS,
+	    (i->X2-i->X1)*image_res/POSTSCRIPTRES+2*IMAGE_BOARDER_PIXELS,
+	    (i->Y2-i->Y1)*image_res/POSTSCRIPTRES+2*IMAGE_BOARDER_PIXELS,
 	    TRANSPARENT,
 	    i->imageName);
-    fprintf(stderr, buffer);
+    // fprintf(stderr, buffer);
     system(buffer);
+  } else {
+    fprintf(stderr, "ignoring image as x1 coord is -1\n");
+    fflush(stderr);
   }
 }
 

From groff-admin  Wed Oct 18 00:43:44 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id AAA06686
	for <groff@ffii.org>; Wed, 18 Oct 2000 00:43:43 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Wed, 18 Oct 2000 00:47:57 +0200
Received: from rigel.univie.ac.at (dial-143174.HRZ.Uni-Dortmund.DE [129.217.143.174])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9HMlnw14636;
	Wed, 18 Oct 2000 00:47:49 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id AAA09782;
	Wed, 18 Oct 2000 00:46:47 +0200
Date: Wed, 18 Oct 2000 00:46:46 +0200 (CEST)
Message-Id: <20001018.004646.101498877.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14828.20941.944311.81928B@surfchem0.riken.go.jp>
References: <14827.43760.126423.72159A@surfchem0.riken.go.jp>
	<20001017.092037.48403856.wl@gnu.org>
	<14828.20941.944311.81928B@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 7469
Lines: 170

> > Well, I insist that GNU troff doesn't support multibyte enodings
> > at all :-) troff itself should work on a glyph basis only.  It has
> > to work with *glyph names*, be it CJK entities or whatever.
> > Currently, the conversion from input encoding to glyph entities
> > and the further processing of glyphs is not clearly separated.
> > From a modular point of view it makes sense if troff itself is
> > restricted to a single input encoding (UTF-8) which is basically
> > only meant as a wrapper to glyph names (cf. \U'xxxx' to enter
> > Unicode encoded characters).  Everything else should be moved to a
> > preprocessor.
> 
> This paragraph says two things:
>  - GNU troff will support UTF-8 only.  Thus, multibyte encodings
>    will be not supported.  [Though UTF-8 is multibyte :-p ]

This was a typo, sorry.  I've meant that I don't want to support
multiple multibyte encodings.

>  - Groff handles glyph, not character.
> I don't understand relationship between these two.  UTF-8 is a code
> for character, not glyph.  ISO8859-1 and EUC-JP are also codes for
> character.  No difference among UTF-8, ISO8859-1, and EUC-JP.

Well, this is *very* important.  The most famous example is that the
character `f', followed by character `i' will be translated into a
single glyph `fi' (which has incidentally a Unicode number for
historical reasons).  A lot of other ligatures don't have a character
code.  Or consider a font which has 10 or more versions of the `&'
character (such a font really exists).  Do you see the difference?  A
font can have multiple glyphs for a single character.  For other
scripts like Arabic it is necessary to do a lot of contextual analysis
to get the right glyphs.  Indic scripts like Tamil have about 50 input
character codes which map up to 3000 glyphs!

Consider the CJK part of Unicode.  A lot of Chinese, Korean, Japanese,
and Vietnamese glyphs have been unified, but you have to select a
proper locale to get the right glyph -- many Japanese people have been
misled because a lot of glyphs in the Unicode book which have a JIS
character code don't look `right' for Japanese.

For me, groff is primarily a text processing tool, and such a program
works with glyphs to be printed on paper.  A `character' is an
abstract concept, basically.  Your point of view, I think, is
completely different: You treat groff as a filter which just
inserts/removes some spaces, newline characters etc.

> However, I won't stick to wchar_t or ucs-4 for internal code, though
> I have no idea about your '31bit glyph code'.  (Maybe I have to
> study Omega...)

A `glyph code' is just an arbitrary registration number for a glyph
specified in the font definition file.  It is invariable from the
input encoding.  Adobe has `official' glyph lists like `Adobe
standard' or `Adobe Japan1'.  CID encoded PostScript fonts use CMaps
to map the input encoding to these glyph IDs.

> The name '--locale' is confusing since it has no relation to locale,
> i.e., a term which refer to a certain standard technology.

I welcome any suggestions for better names...

>  - Japanese and Chinese text contains few whitespace characters.
>    (Japanese and Chinese words are not separated by whitespace).
>    Therefore, different line-breaking algorithm should be used.
>    (Hyphen character is not used when a word is broken into lines.)
>    (Modern Korean language contains whitespace characters between
>    words --- though not words, strictly speaking.)

Not really a different line breaking algorithm but more glyph
properties (to be set with `.cflags'): disallowing breaks after or
before a glyph for implementing kinsoku shori; for implementing
shibuaki properly we probably need to extend the .cflags syntax to set
glyph properties for whole glyph classes.

For the non-CJK experts: `kinsoku shori' means that some CJK glyphs
must not start a line (for example, an ideographic comma or closing
bracket) resp. end a line (opening brackets).  `shibuaki' means
`quarter space'; this is the space between CJK characters and Latin
characters -- there are Japanese standards which defines all these
things in great detail.

>  - Hyphenation algorithm differs from language to language.

What exactly do you mean?  The only real difficult language which
could be easily supported with groff is Thai (and similar languages).
You need at least a dictionary to find word breaks.  All other
languages can easily be managed with the current algorithm, I believe.

>  - Almost CJK characters (ideographics, hiragana, katakana, hangul,
>    and so on) have double width on tty.  Since you won't use wchar_t,
>    you cannot use wcwidth() to get the width for characters.

This is not a problem.  Just give the proper glyph width in the tty
font definition files.

>  - Latin-1 people may use 0xa9 for '\(co'.  However, this character
>    cannot be read in other encodings.  The current Groff convert
>    '\(co' to 0xa9 in latin1 device and to '(C)' in ascii device.
>    How it works for future Groff?  Use u+00a9?  The postprocessor
>    (see below) cannot convert u+00a9 to '(C)' because the width is
>    different and typesetting is broken.  It is very difficult to
>    design to avoid this problem...

For tty devices, the route is as follows.  Let's assume that the input
encoding is Latin-1.  Then the input character code `0xa9' will be
converted to Unicode character `U+00a9' (by the preprocessor).  A
hard-coded table maps this character code to a glyph with the name
`co'.  Now troff looks up the metric info in the font definition file.
If the target device is an ASCII-capable terminal, the width is three
characters (the glyph `co' is defined with the .char request to be
equal to `(C)'); if it is a Unicode-capable terminal, the width is one
character.  After formatting, a hard-coded table maps the glyphs back
to Unicode.

Note that the last step may fail for glyphs which have no
corresponding Unicode value.

> >   . Finally, we need to divide the -T option into a --device and
> >     --output-encoding.
> 
> What is the default encoding for tty?  I suggest this should be
> locale-sensible.  (Or, this can be UTF-8 and Groff can invoke a
> postprocessor.)

I favor UTF-8 + postprocessor.  Terminal capabilities should be
selected with macro packages; for example, an ASCII terminal would get
the options

  -m ascii --device=tty --output-encoding=ascii

the tmac.ascii file would be very similar to tmac.tty + tmac.tty-char.

A latin-2 terminal would be

  -m latin2 --device=tty --output-encoding=latin2

A Unicode terminal emulating an ASCII terminal would be

  -m ascii --device=tty --output-encoding=utf8

etc.

Using a postprocessor we need only a single font definition file for
tty devices.

> > Yes.  The `iconv' preprocessor would then do some trivial, hard-coded
> > conversion.
> 
> You mean, the preprocessor is iconv(1) ?

Basically yes, with some adaptations to groff.

> The preprocessor, provisional name 'gpreconv', will be designed as:
> - includes hard-coded converter for latin1, ebcdic, and utf8.
> - uses iconv(3) if possible (compiled within internationalized OS).
> - parses --input-encoding option.
> - default input is latin1 if compiled within non-internationalized
>   OS.
> - default input is locale-sensible if compiled within
>   internationalized OS.

Exactly.

> Thus I designed the above 'gpreconv'.  Oh, I have to design
> 'gpostconv' also.

It should be very similar to the preprocessor.


    Werner

From groff-admin  Wed Oct 18 05:29:46 2000
Received: from srv1.tsoft.net (news.tsoft.net [198.144.192.3])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id FAA08175
	for <groff@ffii.org>; Wed, 18 Oct 2000 05:29:45 +0200
X-Authentication-Warning: genba.ffii.org: Host news.tsoft.net [198.144.192.3] claimed to be srv1.tsoft.net
Received: from bicho.SynchroDS.COM (daniel.ppp.tsoft.com [198.144.199.225])
	by srv1.tsoft.net (8.8.7/8.8.7) with ESMTP id UAA26059;
	Tue, 17 Oct 2000 20:34:16 -0700 (PDT)
Received: (from daniel@localhost)
	by bicho.SynchroDS.COM (8.8.8/8.8.8) id UAA12168;
	Tue, 17 Oct 2000 20:34:37 -0700 (PDT)
Date: Tue, 17 Oct 2000 20:34:37 -0700 (PDT)
From: Daniel Senderowicz <daniel@bicho.SynchroDS.COM>
Message-Id: <200010180334.UAA12168@bicho.SynchroDS.COM>
To: larry.jones@sdrc.com, wl@gnu.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
Cc: baruchel@libertysurf.fr, groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 691
Lines: 18

Werner,

>> What shall I do?  I believe that the code in gremlin is OK, and that
>> your problem is related to your version of BSD -- can you please ask a
>> BSD guru how to implement it correctly?

I'm not an X erudite, nor I have a linux box to fiddle around (I
run NetBSD on a DECstation) but is it possible that the program
cannot find the required fonts?

>I also believe that the most recent gremlin code is correct, but that
>his problem is that he really hasn't tried that version.  Once again,
>it's ftp://ftp.ffii.org/pub/groff/xgremlin2-20001003.tar.bz2.

I just downloaded that version to compare, can someone tell me what
the extension '.bz2' is? Gunzip doesn't unpack it.

Dan

From groff-admin  Wed Oct 18 05:56:32 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id FAA08234
	for <groff@ffii.org>; Wed, 18 Oct 2000 05:56:31 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 9E8971C122; Wed, 18 Oct 2000 13:05:20 +0900 (JST)
Date: Wed, 18 Oct 2000 13:05:15 +0900
Message-ID: <14829.8570.630819.72159A@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Wed, 18 Oct 2000 00:46:46 +0200 (CEST)"
	<20001018.004646.101498877.wl@gnu.org>
References: <14827.43760.126423.72159A@surfchem0.riken.go.jp>
	<20001017.092037.48403856.wl@gnu.org>
	<14828.20941.944311.81928B@surfchem0.riken.go.jp>
	<20001018.004646.101498877.wl@gnu.org>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 3933
Lines: 90

Hi,

At Wed, 18 Oct 2000 00:46:46 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

>>  - GNU troff will support UTF-8 only.  Thus, multibyte encodings
>>    will be not supported.  [Though UTF-8 is multibyte :-p ]
> This was a typo, sorry.  I've meant that I don't want to support
> multiple multibyte encodings.
>>  - Groff handles glyph, not character.
>> I don't understand relationship between these two.  UTF-8 is a code
>> for character, not glyph.  ISO8859-1 and EUC-JP are also codes for
>> character.  No difference among UTF-8, ISO8859-1, and EUC-JP.

Ah, I meant that 'these two' is the relationship between that Groff
supports UTF-8 only and that Groff processes glyphs.  Sorry for my
poor English.

However, thank you for explaining glyph.  I also understand you 
understand problems on Japanese character codes well. 

I also understand the basic design of groff, though I had to read
the source code of groff myself...  I wonder why people who knows
better than I don't join this list and discuss about
internationalization.

Note that CJK ideographs also has distinction between character and
glyph.  The most famous example is two variants of a 'tall or high'
character.  Japanese people regard these two as the same in daily
use but Japanese people regard these two as different if they are
used in person's names or so on.  I don't know how Chinese and
Korean people treat them.  It may be different.  However, IMHO,
we should neglect this problem now since there are so far no standard
to treat these variants properly.  Though it is important, it is not
in our scope.


> A `glyph code' is just an arbitrary registration number for a glyph
> specified in the font definition file.

Then the 'font definition file' will be irrationally large.  I think
at least CJK ideographics and Korean precompiled Hanguls have to be
treated in different way.  (Ukai has already pointed this problem.
jgroff uses 'wchar<EUCcode>' for glyph names of Japanese characters.)


> For tty devices, the route is as follows.  Let's assume that the input
> encoding is Latin-1.  Then the input character code `0xa9' will be
> converted to Unicode character `U+00a9' (by the preprocessor). 
> A hard-coded table maps this character code to a glyph with the name
> `co'.  Now troff looks up the metric info in the font definition file.

Yes.

> If the target device is an ASCII-capable terminal, the width is three
> characters (the glyph `co' is defined with the .char request to be
> equal to `(C)'); if it is a Unicode-capable terminal, the width is one
> character.  After formatting, a hard-coded table maps the glyphs back
> to Unicode.

How troff knows that the tty device is ASCII-capable or Unicode-capable?
 --- Ok, I understand it by reading the next line:

>  -m ascii --device=tty --output-encoding=ascii

'-m ascii' tells that.  '--output-encoding' will be passed through
for postprocessor.

A problem.  When compiled within internationalized OS, the names
for encodings (for iconv(3) and so on) is implementation-dependent
(You know, there are many implementation-dependent items in 
standard C/C++ language).  A solution is: we can have a hard-coded
translation table between implementation-dependent encoding names
and macro names for -m.  The table must be changed by OS (by
'./configure' script or so).  A minimal table will be translate
every implementation-dependent encoding names into 'ascii' macro,
since almost encodings in the world are superset of ASCII.  A full
table for a OS will cover the list generated by 'iconv --list'.

# Though I think some standardization of names for encoding is needed,
# it is not our topic now.

Since the '-m' option is generated by groff and passed to troff,
groff has to have '#ifdef I18N' code.  (or, the code can be
integrated to the preprocessor if we design the preprocessor to invoke
troff.)

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Wed Oct 18 08:47:16 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id IAA22955
	for <groff@ffii.org>; Wed, 18 Oct 2000 08:47:16 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Wed, 18 Oct 2000 08:51:41 +0200
Received: from rigel.univie.ac.at (dial-143142.HRZ.Uni-Dortmund.DE [129.217.143.142])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9I6pZw21044;
	Wed, 18 Oct 2000 08:51:36 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id CAA09991;
	Wed, 18 Oct 2000 02:18:24 +0200
Date: Wed, 18 Oct 2000 02:18:23 +0200 (CEST)
Message-Id: <20001018.021823.130623821.wl@gnu.org>
To: ukai@debian.or.jp
Cc: groff@ffii.org, debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <87lmvnv5bh.wl@lichee.ukai.org>
References: <20001017.092037.48403856.wl@gnu.org>
	<14828.20941.944311.81928B@surfchem0.riken.go.jp>
	<87lmvnv5bh.wl@lichee.ukai.org>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2523
Lines: 73

> As regards line breaking algorithm, I think we need some more cflags,
> at least for Japanese.  That is,
>  
>    - lines must not be broken before the character
>    - lines must not be broken after the character
> 
> These seems to be implemented as PRE_KINSOKU and POST_KINSOKU in
> jgroff, but it's done by hardcoded.  I think this should be done by
> tmac.<lang>, so I think it's good idea to have some mechanisms to
> load language specific tmac files.

Which mechanism do you suggest?

> BTW, what do you think about code name for multibyte character/wide
> character or glyph code what you said?  In jgroff, it seems it used
> wchar<EUCcode>.

I suggest that we follow the Adobe Glyph List (AGL):

  http://partners.adobe.com/asn/developer/typeforum/unicodegn.html

This means that CJK glyph names would be uni<Unicode>, e.g. `uni635F'.

> jgroff provides "fixedkanji" directive in font description.  But,
> the code of font description loader depends on EUC<->KuTen mapping,
> and it's not good idea for i18n.  I think it would be better to
> provide "wcharset" directive which support code range.  However,
> code range couldn't be used with EUC encoding or something like
> that, and not used for Unicode, because we couldn't expect character
> codes for some language are in succession.

It doesn't matter.  A range directive like `wcharset' (an ugly name,
BTW) just tells troff that glyphs in a given the range have identical
metrics.

A better name is probably `glyphclass':

  glyphclass <sample character> <range begin> <range end>


Example:

  glyphclass uni4F34 uni4E00 uni9FFF

The sample glyph needs a real metrics entry.

> Anyway, jgroff provides new font "M" and "G", which are "Mincho" and
> "Gothic" respectively, for wide characters.  What is the right way
> to add i18n support in groff about font description?

Basically, there is nothing to do.  The only addition needed is a way
to make the font description files smaller, and this is just the
proposed `wcharset' (or `glyphclass') command.

> > >   . Command lines shall be able to override input encoding
> > >     (--input-encoding).
> > Yes.
> 
> How about creating new request (.encoding "encoding-name") in roff 
> source?

Good idea.

> Could you explain (or point us which source code in groff) how to map
> glyph IDs to output code, please?

As explained in a previous mail, a hard-coded mapping table from
glyph-names to Unicode output encoding is needed for tty devices.

For all other devices, nothing will change.


    Werner

From groff-admin  Wed Oct 18 16:06:56 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA25799
	for <groff@ffii.org>; Wed, 18 Oct 2000 16:06:55 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Wed, 18 Oct 2000 16:11:34 +0200
Received: from rigel.univie.ac.at (dial-143177.HRZ.Uni-Dortmund.DE [129.217.143.177])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9IEBVw03412
	for <groff@ffii.org>; Wed, 18 Oct 2000 16:11:32 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id MAA10628
	for <groff@ffii.org>; Wed, 18 Oct 2000 12:01:28 +0200
Date: Wed, 18 Oct 2000 12:01:27 +0200 (CEST)
Message-Id: <20001018.120127.43003101.wl@gnu.org>
To: groff@ffii.org
From: Werner LEMBERG <wl@gnu.org>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] .di behaviour
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 602
Lines: 63

Dear friends,


look at the following input:

xxx
.
.di test
yyy
.br
.di
.
zzz
\*[test]

This yields:

zzz xxx yyy

Another example:

xxx
.
.di test
yyy
.br
zzz
.di
.
aaa
\*[test]

This yields:

zzz aaa xxx yyy


Where is it documented that (a) at the beginning of a diversion,
partial input lines are included in the diversion and (b) that a
diversion must be finished with a line break (which will be dropped)?

With other words, the right way to write the first example is

xxx
.
.di test
.br
yyy
.br
.di
.
zzz
\*[test]

to get

xxx zzz yyy


Is this behaviour identical to UNIX troff?


    Werner

From groff-admin  Wed Oct 18 22:52:21 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id WAA28119
	for <groff@ffii.org>; Wed, 18 Oct 2000 22:52:21 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Wed, 18 Oct 2000 22:56:27 +0200
Received: from rigel.univie.ac.at (dial-143172.HRZ.Uni-Dortmund.DE [129.217.143.172])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9IKuOw13132;
	Wed, 18 Oct 2000 22:56:24 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id QAA00587;
	Wed, 18 Oct 2000 16:56:58 +0200
Date: Wed, 18 Oct 2000 16:56:58 +0200 (CEST)
Message-Id: <20001018.165658.41632154.wl@gnu.org>
To: daniel@bicho.synchrods.com
Cc: larry.jones@sdrc.com, baruchel@libertysurf.fr, groff@ffii.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <200010180334.UAA12168@bicho.SynchroDS.COM>
References: <200010180334.UAA12168@bicho.SynchroDS.COM>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 697
Lines: 19

> >> What shall I do?  I believe that the code in gremlin is OK, and
> >> that your problem is related to your version of BSD -- can you
> >> please ask a BSD guru how to implement it correctly?
> 
> I'm not an X erudite, nor I have a linux box to fiddle around (I run
> NetBSD on a DECstation) but is it possible that the program cannot
> find the required fonts?

Maybe, but I don't think so.  It would be nice if someone on Linux
could try whether gremlin runs as expected...

> I just downloaded that version to compare, can someone tell me what
> the extension '.bz2' is? Gunzip doesn't unpack it.

You need the bzip2/bunzip2 program -- it compresses better than zip
resp. gzip.


    Werner

From groff-admin  Wed Oct 18 22:52:22 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id WAA28122
	for <groff@ffii.org>; Wed, 18 Oct 2000 22:52:21 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Wed, 18 Oct 2000 22:56:38 +0200
Received: from rigel.univie.ac.at (dial-143172.HRZ.Uni-Dortmund.DE [129.217.143.172])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9IKuRw13135;
	Wed, 18 Oct 2000 22:56:27 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id QAA00584;
	Wed, 18 Oct 2000 16:54:53 +0200
Date: Wed, 18 Oct 2000 16:54:53 +0200 (CEST)
Message-Id: <20001018.165453.74757492.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14829.8570.630819.72159A@surfchem0.riken.go.jp>
References: <14828.20941.944311.81928B@surfchem0.riken.go.jp>
	<20001018.004646.101498877.wl@gnu.org>
	<14829.8570.630819.72159A@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2777
Lines: 64

> However, thank you for explaining glyph.  I also understand you 
> understand problems on Japanese character codes well. 

Well, I'm the author of the CJK package for LaTeX, I've written a
ttf2pk converter, and I'm a member of the FreeType core team :-)

> Note that CJK ideographs also has distinction between character and
> glyph.  The most famous example is two variants of a 'tall or high'
> character.  Japanese people regard these two as the same in daily
> use but Japanese people regard these two as different if they are
> used in person's names or so on.

I know these problems too well -- AFAIK, in JIS X 0208 these two
variants are unified.  Do you know details about the new JIS X 0213
standard?

> I don't know how Chinese and Korean people treat them.  It may be
> different.  However, IMHO, we should neglect this problem now since
> there are so far no standard to treat these variants properly.
> Though it is important, it is not in our scope.

If you are working on a terminal you need a character set which
distinguishes the two forms.

> > A `glyph code' is just an arbitrary registration number for a glyph
> > specified in the font definition file.
> 
> Then the 'font definition file' will be irrationally large.  I think
> at least CJK ideographics and Korean precompiled Hanguls have to be
> treated in different way.  (Ukai has already pointed this problem.
> jgroff uses 'wchar<EUCcode>' for glyph names of Japanese
> characters.)

Right.  I think I've answered this problem in my last mail (regarding
a `glyphclass' directive in font description files).

> A problem.  When compiled within internationalized OS, the names for
> encodings (for iconv(3) and so on) is implementation-dependent (You
> know, there are many implementation-dependent items in standard
> C/C++ language).  A solution is: we can have a hard-coded
> translation table between implementation-dependent encoding names
> and macro names for -m.  The table must be changed by OS (by
> './configure' script or so).  A minimal table will be translate
> every implementation-dependent encoding names into 'ascii' macro,
> since almost encodings in the world are superset of ASCII.  A full
> table for a OS will cover the list generated by 'iconv --list'.

I don't think so.  For example, we could restrict to MIME character
set tags which are standardized.

> Since the '-m' option is generated by groff and passed to troff,
> groff has to have '#ifdef I18N' code.  (or, the code can be
> integrated to the preprocessor if we design the preprocessor to
> invoke troff.)

Indeed, the default behaviour should be that the preprocessor adds a

  .mso tmac.<charset>

line or something similar to the document, but there must be a
possibility to override it manually.


    Werner

From groff-admin  Thu Oct 19 03:51:01 2000
Received: from srv1.tsoft.net (news.tsoft.net [198.144.192.3])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id DAA30136
	for <groff@ffii.org>; Thu, 19 Oct 2000 03:51:00 +0200
X-Authentication-Warning: genba.ffii.org: Host news.tsoft.net [198.144.192.3] claimed to be srv1.tsoft.net
Received: from bicho.SynchroDS.COM (daniel.ppp.tsoft.com [198.144.199.225])
	by srv1.tsoft.net (8.8.7/8.8.7) with ESMTP id SAA02316;
	Wed, 18 Oct 2000 18:55:37 -0700 (PDT)
Received: (from daniel@localhost)
	by bicho.SynchroDS.COM (8.8.8/8.8.8) id SAA15485;
	Wed, 18 Oct 2000 18:55:59 -0700 (PDT)
Date: Wed, 18 Oct 2000 18:55:59 -0700 (PDT)
From: Daniel Senderowicz <daniel@bicho.SynchroDS.COM>
Message-Id: <200010190155.SAA15485@bicho.SynchroDS.COM>
To: daniel@bicho.SynchroDS.COM, wl@gnu.org
Subject: Re: [Groff] still exploring the last version: what is gremlin ?
Cc: baruchel@libertysurf.fr, groff@ffii.org, larry.jones@sdrc.com
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 360
Lines: 11

>> I just downloaded that version to compare, can someone tell me what
>> the extension '.bz2' is? Gunzip doesn't unpack it.
>
>You need the bzip2/bunzip2 program -- it compresses better than zip
>resp. gzip.

Just compiled. This gremlin looks good too. The only thing that
the fonts are smaller than in the other version. I don't see any
problem so far.

Dan

From groff-admin  Thu Oct 19 08:24:50 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id IAA12734
	for <groff@ffii.org>; Thu, 19 Oct 2000 08:24:49 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 760621C122; Thu, 19 Oct 2000 15:33:45 +0900 (JST)
Date: Thu, 19 Oct 2000 15:33:38 +0900
Message-ID: <14830.38338.558492.22222V@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Wed, 18 Oct 2000 16:54:53 +0200 (CEST)"
	<20001018.165453.74757492.wl@gnu.org>
References: <14828.20941.944311.81928B@surfchem0.riken.go.jp>
	<20001018.004646.101498877.wl@gnu.org>
	<14829.8570.630819.72159A@surfchem0.riken.go.jp>
	<20001018.165453.74757492.wl@gnu.org>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 3327
Lines: 83

Hi,

At Wed, 18 Oct 2000 16:54:53 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> > However, thank you for explaining glyph.  I also understand you 
> > understand problems on Japanese character codes well. 
> Well, I'm the author of the CJK package for LaTeX, I've written a
> ttf2pk converter, and I'm a member of the FreeType core team :-)

Great!  An another great thing is that you explain basic things to me
so kindly though you are so busy on such many heavy projects...


> I know these problems too well -- AFAIK, in JIS X 0208 these two
> variants are unified.  Do you know details about the new JIS X 0213
> standard?

I heard that JIS X 0208 unites two 'high or tall' variants because
of a policy that JIS X 0208 is a code for characters, not for glyphs.
However, it can change with years whether a set of glyphs are considered
as variants of one character or different characters.

JIS X 0213 has many characters which are also included in JIS X 0212.
It is very confusing.  I guess JIS people think JIS X 0212 is obsolete.
I think it needs time for JIS X 0213 to be popular.  I even don't know
whether JIS X 0213 will become popular or not (like JIS X 0212).
A few characters in JIS X 0213 are not included in the present Unicode.


> > Then the 'font definition file' will be irrationally large.  I think
> > at least CJK ideographics and Korean precompiled Hanguls have to be
> > treated in different way.  (Ukai has already pointed this problem.
> > jgroff uses 'wchar<EUCcode>' for glyph names of Japanese
> > characters.)
> 
> Right.  I think I've answered this problem in my last mail (regarding
> a `glyphclass' directive in font description files).

Then all of these glyphs have to have the same width.  Fortunately,
CJK ideograms and Korean Hanguls have fixed-width glyphs.


> I don't think so.  For example, we could restrict to MIME character
> set tags which are standardized.

I think a table is needed anyway.  This is because Groff should accept
implementation-dependent locale names so that Groff works well together
with other softwares on the OS.  I think we can design Groff so that 
it accepts both implementation-dependent locale names and MIME character
set tags, since it is not likely that a name in implementation-dependent
locale names calls a different encoding in MIME character set tags, nor
vice versa.


> Indeed, the default behaviour should be that the preprocessor adds a
> 
>   .mso tmac.<charset>
> 
> line or something similar to the document, but there must be a
> possibility to override it manually.

Good idea.  Thus '#ifdef I18N' part can be restricted in pre/post-
processors.

Overriding?  well, the current Groff has '-a' option.  I think
this can be used for this purpose.  (Anyway, we can provide
substitution only for non-letter symbols like soft-hyphen, '(C)',
circles, squares, and so on.  I think this is sufficient.)

We have to think about uniting my idea on design of preprocessor
and Ukai's idea of '.encoding "encoding-name"' in roff source.

 - it is the preprocessor that handles the ".encoding" .
 - priority is that
   * --input-encodings wins.
   * .encoding is next.
   * then falling into the default (locale-sensible for i18n OS
     and latin-1 for non-i18n OS).

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Thu Oct 19 14:32:21 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id OAA16324
	for <groff@ffii.org>; Thu, 19 Oct 2000 14:32:20 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Thu, 19 Oct 2000 14:36:57 +0200
Received: from rigel.univie.ac.at (dial-143138.HRZ.Uni-Dortmund.DE [129.217.143.138])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9JCaqw28696;
	Thu, 19 Oct 2000 14:36:53 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA02539;
	Thu, 19 Oct 2000 10:40:35 +0200
Date: Thu, 19 Oct 2000 10:40:35 +0200 (CEST)
Message-Id: <20001019.104035.106257837.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14830.38338.558492.22222V@surfchem0.riken.go.jp>
References: <14829.8570.630819.72159A@surfchem0.riken.go.jp>
	<20001018.165453.74757492.wl@gnu.org>
	<14830.38338.558492.22222V@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2803
Lines: 88

> JIS X 0213 has many characters which are also included in JIS X 0212.
> It is very confusing.  I guess JIS people think JIS X 0212 is
> obsolete.

Basically, only Emacs supports JIS X 0212...

> A few characters in JIS X 0213 are not included in the present
> Unicode.

AFAIK, this will be fixed (or have already been fixed?) in the next
Unicode release where more than 10000 CJK characters are added (in the
surrogate area).

> > > Then the 'font definition file' will be irrationally large.
> >
> > Right.  I think I've answered this problem in my last mail (regarding
> > a `glyphclass' directive in font description files).
> 
> Then all of these glyphs have to have the same width.

Why?  It is intended that `glyphclass' can occur multiple times.  Say,
one glyphclass command for full-width glyphs, and another one for
half-width glyphs.

> > Indeed, the default behaviour should be that the preprocessor adds
> > a
> > 
> >   .mso tmac.<charset>
> > 
> > line or something similar to the document, but there must be a
> > possibility to override it manually.
> 
> Good idea.  Thus '#ifdef I18N' part can be restricted in pre/post-
> processors.

Exactly.

> Overriding?  Well, the current Groff has '-a' option.  I think this
> can be used for this purpose.  (Anyway, we can provide substitution
> only for non-letter symbols like soft-hyphen, '(C)', circles,
> squares, and so on.  I think this is sufficient.)

The `-a' option is almost useless today IMHO.  It will show a tty
approximation of the typeset output:

  groff -a -man -Tdvi troff.man | less

It is *not* the right way to quickly select an ASCII device.  To
override the used macros for the output character set we need a new
option.

Using `-a' is comparable to dvi2tty or similar converters.

> We have to think about uniting my idea on design of preprocessor
> and Ukai's idea of '.encoding "encoding-name"' in roff source.
> 
>  - it is the preprocessor that handles the ".encoding" .
>  - priority is that
>    * --input-encodings wins.
>    * .encoding is next.
>    * then falling into the default (locale-sensible for i18n OS
>      and latin-1 for non-i18n OS).

Exactly.  Compare this to the Emacs model of `local variables'.

Note that such an encoding request has to determine the encoding *and*
character set of a document (similar to Emacs).

I suggest that we don't use `.encoding' but

  -*- charset-encoding: xxx -*-

in the first comment block (almost similar to Emacs).  troff shouldn't
notice encoding issues at all and just accept UTF-8.

If really necessary, we can add two additional commands to select
encoding and character separatedly:

  -*- charset: ...; encoding: ... -*-

Examples:

  .\" -*- charset: JIS-X-0208; encoding: EUC -*-

  .\" -*- charset: JIS-X-0208; encoding: ISO-2022 -*-


    Werner

From groff-admin  Thu Oct 19 16:42:37 2000
Received: from mailout1.hananet.net (mailout1.hananet.net [210.220.163.34])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA17179
	for <groff@ffii.org>; Thu, 19 Oct 2000 16:42:35 +0200
Received: from gnomaniac.myhome ([211.178.105.167]) by
          mailout1.hananet.net (Netscape Messaging Server 4.15) with ESMTP
          id G2OMEK01.P1Y; Thu, 19 Oct 2000 23:47:10 +0900 
Received: (from cjh@localhost)
	by gnomaniac.myhome (8.11.1/8.11.0) id e9JEk3l02074;
	Thu, 19 Oct 2000 23:46:03 +0900 (KST)
	(envelope-from cjh@kr.FreeBSD.org)
X-Authentication-Warning: gnomaniac.myhome: cjh set sender to cjh@kr.FreeBSD.org using -f
To: Tomohiro KUBOTA <tkubota@riken.go.jp>
Cc: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
References: <14828.20941.944311.81928B@surfchem0.riken.go.jp>
	<20001018.004646.101498877.wl@gnu.org>
	<14829.8570.630819.72159A@surfchem0.riken.go.jp>
	<20001018.165453.74757492.wl@gnu.org>
	<14830.38338.558492.22222V@surfchem0.riken.go.jp>
From: CHOI Junho <cjh@kr.FreeBSD.org>
Date: 19 Oct 2000 23:46:03 +0900
In-Reply-To: Tomohiro KUBOTA's message of "Thu, 19 Oct 2000 15:33:38 +0900"
Message-ID: <86d7gwzy6s.fsf@gnomaniac.myhome>
User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.7
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1066
Lines: 20

>>>>> "TK" == Tomohiro KUBOTA <tkubota@riken.go.jp> writes:

    >  Right.  I think I've answered this problem in my last mail (regarding
    >  a `glyphclass' directive in font description files).

    TK> Then all of these glyphs have to have the same width.  Fortunately,
    TK> CJK ideograms and Korean Hanguls have fixed-width glyphs.

Hanguls are usually treated as fixed-width, but it is not true. Many
Korean wordprocessor supports proportional Hangul fonts,
and even Korean Windows (9X/ME/2000) have proportional Hangul TrueType
fonts in default(e.g. Gulim. Gulim has one more variation, Gulim-che,
it is fixed-width for using with english fixed-width fonts). In Unix
world, I didn't see any support for proportional Hangul fonts yet.

-- 
 +++ Any opinions in this posting are my own and not those of my employers +++
 CHOI Junho                             <cjh@FreeBSD.ORG> <cjh@kr.FreeBSD.ORG>
 KFUG <http://www.kr.FreeBSD.org>         Web Data Bank <http://www.wdb.co.kr>
 FreeBSD, GNU/Linux Developer                   http://www.kr.FreeBSD.org/~cjh

From groff-admin  Thu Oct 19 18:28:47 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA17932
	for <groff@ffii.org>; Thu, 19 Oct 2000 18:28:46 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id B553F1C122; Fri, 20 Oct 2000 01:37:47 +0900 (JST)
Date: Fri, 20 Oct 2000 01:37:43 +0900
Message-ID: <14831.9047.710253.72159A@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Thu, 19 Oct 2000 10:40:35 +0200 (CEST)"
	<20001019.104035.106257837.wl@gnu.org>
References: <14829.8570.630819.72159A@surfchem0.riken.go.jp>
	<20001018.165453.74757492.wl@gnu.org>
	<14830.38338.558492.22222V@surfchem0.riken.go.jp>
	<20001019.104035.106257837.wl@gnu.org>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1401
Lines: 39

Hi,

At Thu, 19 Oct 2000 10:40:35 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> Note that such an encoding request has to determine the encoding *and*
> character set of a document (similar to Emacs).
(snip)
> Examples:
>   .\" -*- charset: JIS-X-0208; encoding: EUC -*-
>   .\" -*- charset: JIS-X-0208; encoding: ISO-2022 -*-

No.  only specifying 'encoding' is sufficient.  This is because
'encoding' includes information on which charset to be used.

For example, there are no encodings whose name is 'EUC'.
'EUC' is a generic name for EUC-based encodings (EUC-JP, EUC-KR,
EUC-CN, and EUC-TW).  'EUC' also means a method to build a
encoding which consists of at most four ISO2022-compliant charsets.

Yes, ISO-2022 is a name of encoding.  It consists of many charsets
such as ISO-8859-*, ISO-646-*, JIS-X-0208, KS-X-1001, GB-2312, and
so on so on.  There are many subsets of ISO-2022, such as ISO-2022-JP,
ISO-2022-KR, and so on.  EUC encodings are also subsets of ISO-2022.

Thus, when I specify encoding is ISO-2022-JP, it automatically
says that charsets are US-ASCII, JIS X 0201 (LeftHalf), 
JIS X 0208-1978, and JIS X 0208-1983.  When I specify encoding
is EUC-KR, it automatically says that charsets are US-ASCII and
KS X 1001.


> troff shouldn't notice encoding issues at all and just accept UTF-8.

Yes.

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Thu Oct 19 22:11:58 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id WAA19573
	for <groff@ffii.org>; Thu, 19 Oct 2000 22:11:58 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Thu, 19 Oct 2000 22:16:31 +0200
Received: from rigel.univie.ac.at (dial-143131.HRZ.Uni-Dortmund.DE [129.217.143.131])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9JKGQw14820;
	Thu, 19 Oct 2000 22:16:27 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id WAA04694;
	Thu, 19 Oct 2000 22:15:02 +0200
Date: Thu, 19 Oct 2000 22:15:02 +0200 (CEST)
Message-Id: <20001019.221502.98557204.wl@gnu.org>
To: cjh@kr.FreeBSD.org
Cc: tkubota@riken.go.jp, debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <86d7gwzy6s.fsf@gnomaniac.myhome>
References: <20001018.165453.74757492.wl@gnu.org>
	<14830.38338.558492.22222V@surfchem0.riken.go.jp>
	<86d7gwzy6s.fsf@gnomaniac.myhome>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 651
Lines: 16

> Hanguls are usually treated as fixed-width, but it is not true. Many
> Korean wordprocessor supports proportional Hangul fonts, and even
> Korean Windows (9X/ME/2000) have proportional Hangul TrueType fonts
> in default (e.g. Gulim.  Gulim has one more variation, Gulim-che; it
> is fixed-width for using with english fixed-width fonts).  In Unix
> world, I didn't see any support for proportional Hangul fonts yet.

The same exists for Japanese and Chinese, especially for vertical
writing.

Such fonts must be treated as any other groff font since glyph metric
classes can't be used -- this means very large font declaration
files...


    Werner

From groff-admin  Thu Oct 19 22:11:58 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id WAA19576
	for <groff@ffii.org>; Thu, 19 Oct 2000 22:11:58 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Thu, 19 Oct 2000 22:16:33 +0200
Received: from rigel.univie.ac.at (dial-143131.HRZ.Uni-Dortmund.DE [129.217.143.131])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9JKGUw14824;
	Thu, 19 Oct 2000 22:16:30 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id WAA04691;
	Thu, 19 Oct 2000 22:12:07 +0200
Date: Thu, 19 Oct 2000 22:12:07 +0200 (CEST)
Message-Id: <20001019.221207.70228535.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14831.9047.710253.72159A@surfchem0.riken.go.jp>
References: <14830.38338.558492.22222V@surfchem0.riken.go.jp>
	<20001019.104035.106257837.wl@gnu.org>
	<14831.9047.710253.72159A@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1039
Lines: 25

> > Note that such an encoding request has to determine the encoding *and*
> > character set of a document (similar to Emacs).
> (snip)
> > Examples:
> >   .\" -*- charset: JIS-X-0208; encoding: EUC -*-
> >   .\" -*- charset: JIS-X-0208; encoding: ISO-2022 -*-
> 
> No.  only specifying 'encoding' is sufficient.  This is because
> 'encoding' includes information on which charset to be used.

This is not true.  Encoding does *not* imply the character set.
You are talking about charset/encoding tags.

> Thus, when I specify encoding is ISO-2022-JP, it automatically says
> that charsets are US-ASCII, JIS X 0201 (LeftHalf), JIS X 0208-1978,
> and JIS X 0208-1983.  When I specify encoding is EUC-KR, it
> automatically says that charsets are US-ASCII and KS X 1001.

Again, EUC-KR and friends are charset/encoding tags.  ISO 2022 is not
limited to this, as you certainly know (and as Emacs proves).  You
can, for example, have a character set in the GL or in the GR area.
You can have 7bit ISO 2022 and 8bit ISO 2022, etc.


    Werner

From groff-admin  Fri Oct 20 03:50:51 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id DAA21792
	for <groff@ffii.org>; Fri, 20 Oct 2000 03:50:49 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id E97AF1C122; Fri, 20 Oct 2000 10:59:52 +0900 (JST)
Date: Fri, 20 Oct 2000 10:59:47 +0900
Message-ID: <14831.42770.6994.72159A@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: tkubota@riken.go.jp, debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Thu, 19 Oct 2000 22:15:02 +0200 (CEST)"
	<20001019.221502.98557204.wl@gnu.org>
References: <20001018.165453.74757492.wl@gnu.org>
	<14830.38338.558492.22222V@surfchem0.riken.go.jp>
	<86d7gwzy6s.fsf@gnomaniac.myhome>
	<20001019.221502.98557204.wl@gnu.org>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1930
Lines: 47

Hi,

At Thu, 19 Oct 2000 22:15:02 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> The same exists for Japanese and Chinese, especially for vertical
> writing.

I think *ideograms* have fixed width everywhere.  Of course CJK languages
have their own non-letter symbols which sometimes don't have fixed width.
(In Japanese, non-letter symbols have the fixed width unless two 
non-letter symbols continue.  This applies for both horizonal and
vertical writing.)

Hiraganas and Katakanas are...
I like fixed width, though 'MS P mincho' and 'MS P gothic' fonts
(included in Japanese version of MS Windows) have non-fixed
width. (I don't want to call it 'proportional' because they don't
have good proportion at all for my sense of beauty.)

Note that different font have to be used for horizonal and vertical
writing.  However, you said you won't support vertical, didn't you?


# for people don't know about CJK (Chinese, Japanese, and Korean)
# languages:
# CJK people use ideograms which is called Hanji, Kanji, or Hanja.
# Though they have the same origin, they developed in different way
# in these countries.  In Unicode these ideograms are unified
# and this unification sometimes causes a trouble which Werner pointed
# before.
# Japanese people also use two sets of syllablic letters called Hiragana
# and Katakana.  Korean people also use a set of syllablic letters called
# Hangul.  Chinese people use ideograms only.  Of course CJK people
# use non-letter symbols like parentheses and so on.

Thus we can avoid the worst case to have glyph metrics for all
ideograms.

Hanguls, Hiraganas, Katakanas, ideograms, and almost CJK characters
have 'doublewidth' on tty, except for 'halfwidth' characters.  
'Fullwidth forms' characters also have 'doublewidth' on tty.  Thus 
it is easy to implement a font file for tty using glyph metric classes.

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Fri Oct 20 05:40:37 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id FAA22315
	for <groff@ffii.org>; Fri, 20 Oct 2000 05:40:36 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id D46661C122; Fri, 20 Oct 2000 12:49:41 +0900 (JST)
Date: Fri, 20 Oct 2000 12:49:38 +0900
Message-ID: <14831.49362.330858.29321F@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Thu, 19 Oct 2000 22:12:07 +0200 (CEST)"
	<20001019.221207.70228535.wl@gnu.org>
References: <14830.38338.558492.22222V@surfchem0.riken.go.jp>
	<20001019.104035.106257837.wl@gnu.org>
	<14831.9047.710253.72159A@surfchem0.riken.go.jp>
	<20001019.221207.70228535.wl@gnu.org>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2896
Lines: 63

Hi,

At Thu, 19 Oct 2000 22:12:07 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> This is not true.  Encoding does *not* imply the character set.
> You are talking about charset/encoding tags.

Hmm, I cannot understand your idea...

In Emacs, charsets such as ISO8859-1, JISX0208.1990, and BIG5 are
defined.  Using these charsets, encodings such as euc-japan,
iso-2022-jp, and iso-2022-7bit are defined.  A user can use
these encodings by, for example, M-x set-buffer-file-coding-system,
M-x set-terminal-encoding-system, and so on.  (Names for 'coding-system'
can be followed by '-unix', '-dos', or '-mac' which specify
line-breaking code.)  I can also specify the encoding of a file
using '-*-coding: euc-jp;-*-' in the first line of the file.

You said that encoding names can be specified by MIME charset tag
names.  I write mails in 'charset=us-ascii' or in 'charset=iso-2022-jp'
and web pages in 'charset=us-ascii', 'charset=iso-2022-jp',
'charset=euc-jp', 'charset=utf-8', or so on.  I never specify
encoding and charset separately.  Nor I don't write 'charset=euc'.

ISO-2022 is a encoding which includes many charsets.  Using ISO-2022,
I can write a multilingual text including US-ASCII, ISO 646-*, ISO 8859-*, 
JIS X * (Japanese), CNS 11643 (traditional Chinese), GB 2312 (simplified
Chinese), TIS620 (Thai), and so on.  GL, GR, G0, G1, G2, and G3 can
be used for these charsets with clearly defined escape sequences and
other control codes.  Since the escape sequences and control codes are
clearly defined, we don't need 'charset=' information to read ISO-2022
text.  The preprocesser can work without it (though I won't implement
ISO-2022 converter.  I will implement only four converters --- from
Latin-1, EBCDIC, UTF-8 (no-conversion), and locale encoding (iconv(3))
to UTF-8).

# Note that conversion from Unicode variants to ISO-2022 (not ISO-2022-JP,
# ISO-2022-CN, and so on) contains a problem and almost impossible.
# However, now we are discussing on reading the roff source, not writing.

Indicating JIS-X-0208 and EUC is insufficient to specify an encoding.
Also, telling JIS-X-0208 and ISO-2022 lacks information.  In the former
case, EUC can handle four character sets for GL, GR, SS1, and SS2.
EUC-JP is: ASCII for GL and JIS-X-0208 for GR.  ISO-2022-JP is more
complex.

Practical view; according to your idea, a user can specify 
'charset: KOI8-R; encoding: EUC', which cannot be specified with my 
idea.  However, I don't think this can be a reason your idea is
superior.  Rather, IMHO, such a usage is harmful.

I intend to mean
 - character set: CCS (Coded Character Set) in RFC 2130
 - encoding: CES (Character Encoding Scheme) in RFC 2130

I don't understand on what context you say 'EUC' is an encoding.

And, I think this is the most important, what is the merit of your idea?

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Fri Oct 20 09:40:24 2000
Received: from probity.mcc.ac.uk (probity.mcc.ac.uk [130.88.200.94])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA05252
	for <groff@ffii.org>; Fri, 20 Oct 2000 09:40:24 +0200
Received: from tedistan.annex.mcc.ac.uk
	([194.66.22.200] helo=tedistan.annex.umist.ac.uk ident=root)
	by probity.mcc.ac.uk with esmtp (Exim 2.05 #4)
	id 13mWra-0001ir-00; Fri, 20 Oct 2000 08:45:15 +0100
Received: (from efh@localhost)
	by eels.fort.knox.uk (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) id AAA11831;
	Fri, 20 Oct 2000 00:37:06 +0100
Message-ID: <XFMail.001020003706.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: XFMail 1.3 [p0] on Linux
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
In-Reply-To: <20001018.120127.43003101.wl@gnu.org>
Date: Fri, 20 Oct 2000 00:37:06 +0100 (BST)
Reply-To: Ted.Harding@nessie.mcc.ac.uk
From: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
To: Werner LEMBERG <wl@gnu.org>
Subject: RE: [Groff] .di behaviour
Cc: groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 3370
Lines: 121

Hi Werner and all,

This is subtle; but as far as I can tell Werner's examples are consistent
with standard UNIX troff and the behaviour is indeed at least implicitly
documented. The key is in the description of ".di" (Troff User's Manual
section 7):

  .di xx
    "Divert _output_ to macro xx ..."

together with Section 4:

  4.1 Filling and adjusting. Normally, words are collected from input
  text lines and assembled into an output text line until some word
  does not fit. ...

In other words, output does not occur until a line is filled (unless,
of course, a ".br" is encountered, in which case "the filling of the
line currently being collected is stopped and the line is output without
adjustment"). Therefore that part of an input line preceding a diversion
which forms the beginning of an incomplete output line will be included
in the diversion, since output does not begin until the line is filled
or ".br" is encountered. Similarly at the end of a diversion: if the
end of the body of the diversion will be the beginning of an incomplete
output line, then its output has not commenced and it will not be
diverted, since it is the _output_ which is diverted, not the _input_.

A similar description can be found in "Text Processing and Typesetting
with UNIX" by D. Barron and M. Rees (Addison-Wesley 1987), page 180:

  "Note also that no break is performed: this means that any partial
  line awaiting output will become part of the diversion. ... The
  diversion is terminated using [.di] without an argument ... as at
  the start of the diversion, no break is performed. There is therefore
  the possibility of omitting any partially formed output line from the
  diversion body. To avoid this problem, a suitable request which involves
  a break is recommended ... "

The implication is that if you want the entire contents of the diversion
body to be kept together and output at the point where the diversion macro
is called, then it is wise to begin the diversion with a ".br" and also to
end it with a ".br".

I hope this helps (I confess to finding it confusing on occasion ... ).

Ted.
=======================================================================
On 18-Oct-00 Werner LEMBERG wrote:
> 
> Dear friends,
> 
> 
> look at the following input:
> 
> xxx
> .
> .di test
> yyy
> .br
> .di
> .
> zzz
> \*[test]
> 
> This yields:
> 
> zzz xxx yyy
> 
> Another example:
> 
> xxx
> .
> .di test
> yyy
> .br
> zzz
> .di
> .
> aaa
> \*[test]
> 
> This yields:
> 
> zzz aaa xxx yyy
> 
> 
> Where is it documented that (a) at the beginning of a diversion,
> partial input lines are included in the diversion and (b) that a
> diversion must be finished with a line break (which will be dropped)?
> 
> With other words, the right way to write the first example is
> 
> xxx
> .
> .di test
> .br
> yyy
> .br
> .di
> .
> zzz
> \*[test]
> 
> to get
> 
> xxx zzz yyy
> 
> 
> Is this behaviour identical to UNIX troff?
> 
> 
>     Werner
> _______________________________________________
> Groff maillist  -  Groff@ffii.org
> http://ffii.org/mailman/listinfo/groff

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 284 7749
Date: 20-Oct-00                                       Time: 00:37:06
------------------------------ XFMail ------------------------------

From groff-admin  Fri Oct 20 16:03:11 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA08207
	for <groff@ffii.org>; Fri, 20 Oct 2000 16:03:10 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Fri, 20 Oct 2000 16:07:58 +0200
Received: from rigel.univie.ac.at (dial-143153.HRZ.Uni-Dortmund.DE [129.217.143.153])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9KE7uw22997;
	Fri, 20 Oct 2000 16:07:56 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id OAA00390;
	Fri, 20 Oct 2000 14:00:50 +0200
Date: Fri, 20 Oct 2000 14:00:50 +0200 (CEST)
Message-Id: <20001020.140050.74750218.wl@gnu.org>
To: Ted.Harding@nessie.mcc.ac.uk
Cc: groff@ffii.org
Subject: Re: [Groff] .di behaviour
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <XFMail.001020003706.Ted.Harding@nessie.mcc.ac.uk>
References: <20001018.120127.43003101.wl@gnu.org>
	<XFMail.001020003706.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 284
Lines: 9

> This is subtle; but as far as I can tell Werner's examples are
> consistent with standard UNIX troff and the behaviour is indeed at
> least implicitly documented.

So it is not documented :-)  Thanks for the clarification.  I'll put
your words into the texinfo manual.


    Werner

From groff-admin  Fri Oct 20 16:03:12 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA08212
	for <groff@ffii.org>; Fri, 20 Oct 2000 16:03:12 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Fri, 20 Oct 2000 16:07:41 +0200
Received: from rigel.univie.ac.at (dial-143153.HRZ.Uni-Dortmund.DE [129.217.143.153])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9KE7Yw22978;
	Fri, 20 Oct 2000 16:07:35 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id OAA00429;
	Fri, 20 Oct 2000 14:45:51 +0200
Date: Fri, 20 Oct 2000 14:45:51 +0200 (CEST)
Message-Id: <20001020.144551.30188717.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14831.49362.330858.29321F@surfchem0.riken.go.jp>
References: <14831.9047.710253.72159A@surfchem0.riken.go.jp>
	<20001019.221207.70228535.wl@gnu.org>
	<14831.49362.330858.29321F@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 755
Lines: 22

> > This is not true.  Encoding does *not* imply the character set.
> > You are talking about charset/encoding tags.
> 
> Hmm, I cannot understand your idea...
> 
> I intend to mean
>  - character set: CCS (Coded Character Set) in RFC 2130
>  - encoding: CES (Character Encoding Scheme) in RFC 2130

First of all:  We both mean the same, and we agree how to handle the
problem in groff.  I'm only arguing about technical terms.

Another try.

Consider a PostScript font with its encoding vector.  You have a
single glyph set which can map to multiple encodings.  My intention is
to use the terms `set' and `encoding' in a consistent way -- I want to
avoid that we have to use other words if we are talking about glyphs
instead of characters.


    Werner

From groff-admin  Fri Oct 20 16:03:12 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA08209
	for <groff@ffii.org>; Fri, 20 Oct 2000 16:03:11 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Fri, 20 Oct 2000 16:08:01 +0200
Received: from rigel.univie.ac.at (dial-143153.HRZ.Uni-Dortmund.DE [129.217.143.153])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9KE7ww23001;
	Fri, 20 Oct 2000 16:07:58 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id OAA00397;
	Fri, 20 Oct 2000 14:14:45 +0200
Date: Fri, 20 Oct 2000 14:14:44 +0200 (CEST)
Message-Id: <20001020.141444.71083696.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14831.42770.6994.72159A@surfchem0.riken.go.jp>
References: <86d7gwzy6s.fsf@gnomaniac.myhome>
	<20001019.221502.98557204.wl@gnu.org>
	<14831.42770.6994.72159A@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=iso-2022-jp
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 315
Lines: 14

> > The same exists for Japanese and Chinese, especially for vertical
> > writing.
> 
> I think *ideograms* have fixed width everywhere.

Well, maybe.  But sometimes there is kerning.  Please consult Ken
Lunde's `CJKV Information Processing' for details.  Example:

   $B!;(B
   $B0l(B
   $B!;(B


    Werner

From groff-admin  Fri Oct 20 18:48:42 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA09194
	for <groff@ffii.org>; Fri, 20 Oct 2000 18:48:41 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 6B7541C122; Sat, 21 Oct 2000 01:57:48 +0900 (JST)
Date: Sat, 21 Oct 2000 01:57:43 +0900
Message-ID: <14832.31111.196218.91839F@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Fri, 20 Oct 2000 14:45:51 +0200 (CEST)"
	<20001020.144551.30188717.wl@gnu.org>
References: <14831.9047.710253.72159A@surfchem0.riken.go.jp>
	<20001019.221207.70228535.wl@gnu.org>
	<14831.49362.330858.29321F@surfchem0.riken.go.jp>
	<20001020.144551.30188717.wl@gnu.org>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2075
Lines: 60

Hi,

At Fri, 20 Oct 2000 14:45:51 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> First of all:  We both mean the same, and we agree how to handle the
> problem in groff.  I'm only arguing about technical terms.
> 
> Another try.
> 
> Consider a PostScript font with its encoding vector.  You have a
> single glyph set which can map to multiple encodings.  My intention is
> to use the terms `set' and `encoding' in a consistent way -- I want to
> avoid that we have to use other words if we are talking about glyphs
> instead of characters.

I understand I am confused.  I have to confirm a few points:

1. Your 'charset' and 'encoding' are for troff or for preprocessor?
   I thought both of them are for preprocessor.  The preprocessor
   figures out the way to convert the input to UTF-8 from the information.
2. Which will the pre/postprocessors handle, characters or glyphs?
   Or, is it meaningless to distinguish the object for pre/post-
   processors is character or glyph? (since they handle concrete
   encodings such as Latin-1 and UTF-8.  If the implementation 
   is not affected, it will be meaningless to think about whether
   the Latin-1, UTF-8, and so on are codes for character or glyph.)
3. Your 'charset' is for glyph and 'encoding' is for character?
   I thought both of them are for character, since I thought both 
   of them are for preprocessor.
4. I though we were discussing on (tags in roff souce for) preprocessor.
   Is that right?



Is this chart right (for tty)?


   roff source in any encoding like '\(co'     (character)
          |
          |  preprocessor
          V
   UTF-8 stream like u+00a9                    (character)
          |
          |  troff
          V
   glyph expression like 'co'                  (glyph)
          |
          |  troff (continuing)
          V
   UTF-8 stream like u+00a9 or '(C)'           (character)
          |
          |  postprocessor
          V
   formatted text in any encoding              (character)


---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Fri Oct 20 19:21:05 2000
Received: from tkb.mpl.com (IDENT:root@tkb.mpl.com [198.77.4.83])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id TAA09450
	for <groff@ffii.org>; Fri, 20 Oct 2000 19:21:04 +0200
Received: from tkb.mpl.com (IDENT:tkb@tkb.mpl.com [198.77.4.83])
	by tkb.mpl.com (8.9.3/8.9.3) with ESMTP id NAA23704;
	Fri, 20 Oct 2000 13:23:55 -0400
From: "T. Kurt Bond" <tkb@tkb.mpl.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <14832.32681.975096.519027@tkb.mpl.com>
Date: Fri, 20 Oct 2000 13:23:53 -0400 (EDT)
To: Werner LEMBERG <wl@gnu.org>
Cc: tkubota@riken.go.jp, debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org,
        groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
In-Reply-To: <20001019.104035.106257837.wl@gnu.org>
References: <14829.8570.630819.72159A@surfchem0.riken.go.jp>
	<20001018.165453.74757492.wl@gnu.org>
	<14830.38338.558492.22222V@surfchem0.riken.go.jp>
	<20001019.104035.106257837.wl@gnu.org>
X-Mailer: VM 6.75 under Emacs 20.4.1
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 957
Lines: 24

Werner LEMBERG writes:
> The `-a' option is almost useless today IMHO.  It will show a tty
> approximation of the typeset output:
> 
>   groff -a -man -Tdvi troff.man | less
> 
> It is *not* the right way to quickly select an ASCII device.  To
> override the used macros for the output character set we need a new
> option.
> 
> Using `-a' is comparable to dvi2tty or similar converters.

Exactly.  Sometimes `-a' is explained as producing "ascii" output, but
I think it's better understood as producing "approximate" output.  It
is essentially useful only for debugging and for temporary runs where
the side effects (.write output, .tm output) are more important than
having actual properly formatted pages for output.  In those cases
it's very useful, but in normal cases it's of no use at all.

Would it be useful to add to the texinfo documentation a not
explaining that `-a' should only be used for these situations?
-- 
T. Kurt Bond, tkb@tkb.mpl.com


From groff-admin  Fri Oct 20 21:31:10 2000
Received: from probity.mcc.ac.uk (probity.mcc.ac.uk [130.88.200.94])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id VAA10231
	for <groff@ffii.org>; Fri, 20 Oct 2000 21:31:10 +0200
Received: from tedistan.annex.mcc.ac.uk
	([194.66.22.200] helo=tedistan.annex.umist.ac.uk ident=root)
	by probity.mcc.ac.uk with esmtp (Exim 2.05 #4)
	id 13mhxO-000DSU-00; Fri, 20 Oct 2000 20:35:59 +0100
Received: (from efh@localhost)
	by eels.fort.knox.uk (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) id UAA03637;
	Fri, 20 Oct 2000 20:32:17 +0100
Message-ID: <XFMail.001020203217.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: XFMail 1.3 [p0] on Linux
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
In-Reply-To: <20001017.092037.48403856.wl@gnu.org>
Date: Fri, 20 Oct 2000 20:32:17 +0100 (BST)
Reply-To: Ted.Harding@nessie.mcc.ac.uk
From: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
To: Werner LEMBERG <wl@gnu.org>
Subject: Re: [Groff] Re: groff: radical re-implementation
Cc: groff@ffii.org, debian-i18n@lists.debian.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 7557
Lines: 145

On 17-Oct-00 Werner LEMBERG wrote:
> Well, I insist that GNU troff doesn't support multi-byte encodings at
> all :-) troff itself should work on a glyph basis only.  It has to
> work with *glyph names*, be it CJK entities or whatever.  Currently,
> the conversion from input encoding to glyph entities and the further
> processing of glyphs is not clearly separated.  From a modular point
> of view it makes sense if troff itself is restricted to a single input
> encoding (UTF-8) which is basically only meant as a wrapper to glyph
> names (cf. \U'xxxx' to enter Unicode encoded characters).  Everything
> else should be moved to a preprocessor.

Hi Folks,

I have now managed to read this correspondence (16-20 Oct) and think
about it for a bit. Sorry to have been obliged to leave it on one
side until this week ended.

On the whole I go with the view that Werner has expressed, here and
in other mails. Also, I think that some people are not clear about
the distinction between "character" and "glyph" (not sure that I am,
in all cases, come to that ... ).

I would like to present an even more conservative view than Werner
has stated. And, by the way, in the following "troff" means the
main formatting program "gtroff". "groff" denotes the whole package.

A.1. At present troff accepts 8-bit input, i.e. recognises 256 distinct
entities in the input stream (with a small number of exceptions which
are "illegal").

It does not really matter that these are interpreted, by default, as
iso-latin-1. They could correspond to anything on your screen when you
are typing, and you can set up translation macros in troff to make them
correspond to anything else (using either the ".char" request or the
traditional ".tr" or the new ".trnt" requests).

A.2. The direct correspondence between input bytes and characters is
defined in the font files for the device. In addition, groups of bytes
(such as, represented in ASCII, "\[Do]") can be made to correspond to
specific characters named in the font files.

A.3. What gets printed or displayed is a "glyph" which is defined by the
current font definition for the device. (Even in English, a character
such as "A" could be printed as a Times-Roman "A" glyph, a Helvetica
BoldItalic "A" glyph, \ ZapfChancery-MediumItalic glyph, ... ).
Troff uses the glyph-metric information in the font file to compute
its formatting.

A.4. Troff is not, and was never intended to be, WYSIWIG. Its concept
is that you prepare an input stream (using whatever interface pleases
you, and if this shows you say kanji characters then that's fine,
so long as you don't expect troff to "see" them as kanji) which,
when interpreted by troff, produces printed/displayed output which
bear the marks that you want. I don't see anything wrong (except
possibly in ease of use) in creating an ASCII input stream in
order to generate Japanese output. Preparation of an output
stream to drive a device capable of rendering the output is
the job of the post-processor (and, provided you have installed
appropriate font definition files, I cannot think of anything
that would be beyond the PostScript device "devps").

A: It follows that troff is already language-independent, for all
languages whose typographic conventions can be achieved by the primitive
mechanisms already present in troff. For such languages, there is no
need to change troff at all. For some other languages, there are
minor extra requirements which would require small extensions to
troff which would not interact with existing mechanisms.

Major exceptions to language-independence, at present, include all
the "left-to-right" languages (Hebrew, Arabic, ... ). I have been
studying Dan Berry's implementation of "ffortid" ["ditroff" backwards]
which is a post-processor that allows right-to-left text to be
correctly printed. I believe that a port to groff is quite feasible.

Dan Berry has also done "triroff" [tri-directional troff] for traditional
UNIX troff which can in addition do the top-to-bottom printing for
Chinese etc. To my untutored eye, the results look OK. This could also be
ported to groff.

Extra complications can arise in some languages, such as special
hyphenation rules (as has been mentioned); presence or absence of
particular ligatures [and I think that troff's hard-wired set
of ligatures should be replaced by a user-definable set] (e.g. in
Turkish you never use "fi" ligature since this suppresses the
distinction between dotless-i and i-with-dot); some characters may
not end, or may not begin, a line; some characters have different glyphs
at the beginning, the middle, or the end of words; and so on.
The above are cases where minor extensions of troff are required,
but they do not interact with other features of troff and require
no radical re-implementation.

Some of the complications with specific languages (such as the extra
space separating punctuation marks in French) can be set up on
a language-specific basis by suitable macros, and require no change at
all in troff itself.

B: Troff should be able to cope with multi-lingual documents, where
several different languages occur in the same document. I do NOT
believe that the right way to do this is to extend troff's capacity
to recognise thousands of different input encodings covering all the
languages which it might be called upon to typeset (e.g. by Unicode or
the like).

Troff's multi-character naming convention means that anything you could
possibly need can be defined, and given a name in the troff input
"character set" whenever you really need it, so long as you have the
device resources to render the appropriate glyph. If you want to use a
multi-byte encoding in your input-preparation software, you can
pre-process this with a suitable filter to generate the troff
input-sequences you need (I have done this with WordPerfect
multinational characters, for instance, which are two-byte entities).

C: Error messages and similar communications with the user (which
have nothing directly to do with troff's real job) are irrelevant to
the question of revising groff. If people would like these to appear in
their own language then I'm sure it can be arranged in a way which
would require no change whatever in the fundamental workings of troff.

CONCLUSION: Troff certainly needs some extensions to cope with the
typesetting demands of some languages (of which the major ones that
I can think of have been mentioned above). I also believe that there
are some features of troff which need to be changed in any case, but
these has nothing to do with language or "locale".

Apart from this, I believe that troff has all the primitive functionality
needed to cope with different languages and that any user can define their
own resources for specific languages (including multi-lingual documents).

There is certainly a strong argument for people who are expert both
in troff and in specific languages to prepare _definitive_ language-
specific resources, rather than have different users all doing different
and more-or-less adequate jobs on their own; but that is another issue
and still does not involve any radical re-design of groff.

Therefore, I suggest, troff can basically be left alone; it does not
need radical re-implementation.

Best wishes to all,
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 284 7749
Date: 20-Oct-00                                       Time: 20:32:16
------------------------------ XFMail ------------------------------

From groff-admin  Sat Oct 21 15:31:35 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA29876
	for <groff@ffii.org>; Sat, 21 Oct 2000 15:31:34 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 633861C122; Sat, 21 Oct 2000 22:40:39 +0900 (JST)
Date: Sat, 21 Oct 2000 22:40:32 +0900
Message-ID: <14833.40144.420924.72159A@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: groff@ffii.org, debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Fri, 20 Oct 2000 20:32:17 +0100 (BST)"
	<XFMail.001020203217.Ted.Harding@nessie.mcc.ac.uk>
References: <20001017.092037.48403856.wl@gnu.org>
	<XFMail.001020203217.Ted.Harding@nessie.mcc.ac.uk>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 4077
Lines: 88

Hi,

At Fri, 20 Oct 2000 20:32:17 +0100 (BST),
(Ted Harding) <Ted.Harding@nessie.mcc.ac.uk> wrote:

> It does not really matter that these are interpreted, by default, as
> iso-latin-1. They could correspond to anything on your screen when you
> are typing, and you can set up translation macros in troff to make them
> correspond to anything else (using either the ".char" request or the
> traditional ".tr" or the new ".trnt" requests).

Since I am not familiar with the internals of troff, I don't know
this works well for multibyte encodings such as EUC-*, UTF-8, and so
on.  However, there are encodings which have shift state with escape
sequence, like ISO-2022-JP, ISO-2022-CN, ISO-2022-INT-1, and so on.
Such encodings cannot treated by ".char" .  Do you have any positive
reason not to support such encodings, even when such encodings can
easily supported using standard "locale" technology?

We have to prepare many font files for every encodings in the world.
It is inefficient.  And more, what is the mechanism to choose proper
font file for desired encoding?

I think the answer is "locale" technology.  Groff can check LC_CTYPE
category.  Thus, all what a user has to do is to set LANG, LC_CTYPE,
or LC_ALL environmental for various softwares to work on needed
encoding.

However, I am interested in how Groff 1.16 works for UTF-8 input.
I could not find any code for UTF-8 input, though I found a code for
UTF-8 output in src/devices/grotty/tty.cc .  Am I missing something?
(Of course /font/devutf8/* has no implementation of UTF-8 encoding,
though it seems to have a table for glyph names -> UCS-2.)


> B: Troff should be able to cope with multi-lingual documents, where
> several different languages occur in the same document. I do NOT
> believe that the right way to do this is to extend troff's capacity
> to recognise thousands of different input encodings covering all the
> languages which it might be called upon to typeset (e.g. by Unicode or
> the like).

This is a confusion of language and encoding.  I think UTF-8 support
can and should be achieved via locale technology.  By using locale
technology, a software can be written in encoding-independent way.
The software can support any encodings including UTF-8.  Why
do we have to hard-code UTF-8, though we can use locale technology
to support any encodings including UTF-8 ?

I agree that we should not extend troff to recognize thousands of
encodings if we have to hard-code every encodings.  But it is not true.
Again, do you have any positive reason not to support encodings other
than UTF-8 ?

We have many systems using encodings such as EUC-* and ISO-2022-*.
Some systems will migrate to UTF-8 soon.  Some will migrate after
a certain time.  Users using UTF-8 and EUC-* may share a system.
Some will never migrate into UTF-8.  Groff should support all of them.
I think you will also have trouble if you cannot use 8-bit encodings.


> C: Error messages and similar communications with the user (which
> have nothing directly to do with troff's real job) are irrelevant to
> the question of revising groff. If people would like these to appear in
> their own language then I'm sure it can be arranged in a way which
> would require no change whatever in the fundamental workings of troff.

This is a different topic.  You are right.  We can use gettext to
handle translation without radical re-implementation of troff nor
Groff system.  However, IMO, this topic is much less important.

You are reminded of this topic because I use a word "locale", aren't you?
Sure, locale is also used to change language for messages
(LC_MESSAGES category).  However, what I am discussing is LC_CTYPE 
category.

Supporting "locale" technology (especially LC_CTYPE category) is
important so that Groff can work well with other softwares.


P.S. to CHOI Junho:
I think you'd better subscribe groff@ffii.org or you may lose messages.
I'd like you not to leave this discussion because you are also a
multibyte-language speaker.

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Sat Oct 21 15:40:33 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA29940
	for <groff@ffii.org>; Sat, 21 Oct 2000 15:40:32 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 29BD11C122; Sat, 21 Oct 2000 22:49:48 +0900 (JST)
Date: Sat, 21 Oct 2000 22:49:45 +0900
Message-ID: <14833.40697.666263.26482Z@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Fri, 20 Oct 2000 14:14:44 +0200 (CEST)"
	<20001020.141444.71083696.wl@gnu.org>
References: <86d7gwzy6s.fsf@gnomaniac.myhome>
	<20001019.221502.98557204.wl@gnu.org>
	<14831.42770.6994.72159A@surfchem0.riken.go.jp>
	<20001020.141444.71083696.wl@gnu.org>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=ISO-2022-JP
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1117
Lines: 35

Hi,

At Fri, 20 Oct 2000 14:14:44 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> > I think *ideograms* have fixed width everywhere.
> 
> Well, maybe.  But sometimes there is kerning.  Please consult Ken
> Lunde's `CJKV Information Processing' for details.  Example:
> 
>    $B!;(B
>    $B0l(B
>    $B!;(B

Wow!  This is the first time I received a Japanese mail from
non-Japanese speaker.

Everyone, please notice the header of 
> Content-Type: Text/Plain; charset=iso-2022-jp
> Content-Transfer-Encoding: 7bit
from Werner's mail.  (This mail will have the same header
since I cited the Japanese characters.)

BTW, you may be right, since I am not an expert on typesetting.
However, I prefer fixed-width ideograms even for such cases.
I have never seen any printed matters with non-fixed-width
ideograms even for such cases.

I have seen the book at a few bookstores (not translated in Japanese
yet).  Though I am interested in it, the book is too thick for me...
(Even if the book were in Japanese, it would be heavy for me.)

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Sat Oct 21 16:35:44 2000
Received: from probity.mcc.ac.uk (probity.mcc.ac.uk [130.88.200.94])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA30175
	for <groff@ffii.org>; Sat, 21 Oct 2000 16:35:43 +0200
Received: from tedistan.annex.mcc.ac.uk
	([194.66.22.200] helo=tedistan.annex.umist.ac.uk ident=root)
	by probity.mcc.ac.uk with esmtp (Exim 2.05 #4)
	id 13mzp9-0000Th-00; Sat, 21 Oct 2000 15:40:39 +0100
Received: (from efh@localhost)
	by eels.fort.knox.uk (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) id PAA26585;
	Sat, 21 Oct 2000 15:39:24 +0100
Message-ID: <XFMail.001021153924.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: XFMail 1.3 [p0] on Linux
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
In-Reply-To: <14833.40144.420924.72159A@surfchem0.riken.go.jp>
Date: Sat, 21 Oct 2000 15:39:24 +0100 (BST)
Reply-To: Ted.Harding@nessie.mcc.ac.uk
From: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
To: Tomohiro KUBOTA <tkubota@riken.go.jp>
Subject: Re: [Groff] Re: groff: radical re-implementation
Cc: cjh@kr.FreeBSD.org, debian-i18n@lists.debian.org, groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2598
Lines: 54

On 21-Oct-00 Tomohiro KUBOTA wrote:
> Hi,
> 
> At Fri, 20 Oct 2000 20:32:17 +0100 (BST),
> (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk> wrote:
> 
>> B: Troff should be able to cope with multi-lingual documents, where
>> several different languages occur in the same document. I do NOT
>> believe that the right way to do this is to extend troff's capacity
>> to recognise thousands of different input encodings covering all the
>> languages which it might be called upon to typeset (e.g. by Unicode or
>> the like).
> 
> This is a confusion of language and encoding.  I think UTF-8 support
> can and should be achieved via locale technology.  By using locale
> technology, a software can be written in encoding-independent way.
> The software can support any encodings including UTF-8.  Why
> do we have to hard-code UTF-8, though we can use locale technology
> to support any encodings including UTF-8 ?

To bring the argument I am trying to present face-to-face with the
point you seem to be trying to make:

Someone writing a document about Middle Eastern and related literatures
may wish to use the Arabic, Persian, Hebrew, Turkish (all of which have
different scripts), and also various Central Asian languages (such as
Turkmen) which are often written in Cyrillic or in a variant of Cyrillic.

Can you explain how this could be handled "via locale technology"
in a single document?

Even in its present state, groff can handle such material in a
straightforward way (though with extra work from the user in order to
make the right-to-left languages work correctly -- the basic method would
be to get it formatted correctly with these languages input and printed as
left-to-right, and then edit the input file, using the formatted output
for reference, to reverse the input sequences on a line-by-line basis).

[One of the dangers which I fear if groff were re-structured on a
"locale" basis, or similar mechanism. is that its flexibility, indeed in
principle its universality, would be compromised and limited by the
constraints of that mechanism. It is perhaps not recognised widely
eough that groff, in its present state, is capable of being greatly
extended -- by means of user-defined macros, preprocessors, and
post-processors -- without fundamental change to troff.]

With best wishes,
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 284 7749
Date: 21-Oct-00                                       Time: 15:39:24
------------------------------ XFMail ------------------------------

From groff-admin  Sat Oct 21 18:13:12 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA30838
	for <groff@ffii.org>; Sat, 21 Oct 2000 18:13:11 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id C34121C122; Sun, 22 Oct 2000 01:22:29 +0900 (JST)
Date: Sun, 22 Oct 2000 01:22:24 +0900
Message-ID: <14833.49856.849748.72159A@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: cjh@kr.FreeBSD.org, debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Sat, 21 Oct 2000 15:39:24 +0100 (BST)"
	<XFMail.001021153924.Ted.Harding@nessie.mcc.ac.uk>
References: <14833.40144.420924.72159A@surfchem0.riken.go.jp>
	<XFMail.001021153924.Ted.Harding@nessie.mcc.ac.uk>
User-Agent: Wanderlust/1.0.3 (Notorious) SEMI/1.12.1 ([JR] Nonoichi)
 FLIM/1.12.7 (=?ISO-8859-4?Q?Y=FEzaki?=) Emacs/20.7 (i386-debian-linux-gnu)
 MULE/4.1 (AOI)
MIME-Version: 1.0 (generated by SEMI 1.12.1 - "[JR] Nonoichi")
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1752
Lines: 41

Hi,

At Sat, 21 Oct 2000 15:39:24 +0100 (BST),
(Ted Harding) <Ted.Harding@nessie.mcc.ac.uk> wrote:

> Someone writing a document about Middle Eastern and related literatures
> may wish to use the Arabic, Persian, Hebrew, Turkish (all of which have
> different scripts), and also various Central Asian languages (such as
> Turkmen) which are often written in Cyrillic or in a variant of Cyrillic.
> 
> Can you explain how this could be handled "via locale technology"
> in a single document?

In UTF-8 or ISO-2022 locale, if the OS supports it.
For example, "en_US.UTF-8" .   "en_US" part can be anything which
the OS supports.

Please consult:
  "docs.sun.com: Unicode Support in the Solaris Operiting Environment"
  http://docs.sun.com/ab2/coll.651.1/SOLUNICOSUPPT/@Ab2TocView?Ab2Lang=C&Ab2Enc=iso-8859-1
and you will understand how UTF-8 can be used in "locale" technology.

Ok, the preprocessor which I am writing will support both "locale" mode
and conventional mode (for compatibility).  The conventional mode will
support Latin-1, EBCDIC, and UTF-8.  Thus, you can use UTF-8 even on
OSes which don't support UTF-8 locale.


> [One of the dangers which I fear if groff were re-structured on a
> "locale" basis, or similar mechanism. is that its flexibility, indeed in
> principle its universality, would be compromised and limited by the
> constraints of that mechanism. It is perhaps not recognised widely
> eough that groff, in its present state, is capable of being greatly
> extended -- by means of user-defined macros, preprocessors, and
> post-processors -- without fundamental change to troff.]

Can you write a macro which enable locale-sensible file/tty I/O?

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Sat Oct 21 22:27:27 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id WAA31920
	for <groff@ffii.org>; Sat, 21 Oct 2000 22:27:27 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sat, 21 Oct 2000 22:32:07 +0200
Received: from rigel.univie.ac.at (dial-143140.HRZ.Uni-Dortmund.DE [129.217.143.140])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9LKW3w20944;
	Sat, 21 Oct 2000 22:32:03 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id TAA01332;
	Sat, 21 Oct 2000 19:46:36 +0200
Date: Sat, 21 Oct 2000 19:46:35 +0200 (CEST)
Message-Id: <20001021.194635.08318027.wl@gnu.org>
To: Ted.Harding@nessie.mcc.ac.uk
Cc: groff@ffii.org, debian-i18n@lists.debian.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <XFMail.001020203217.Ted.Harding@nessie.mcc.ac.uk>
References: <20001017.092037.48403856.wl@gnu.org>
	<XFMail.001020203217.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 4138
Lines: 97

> A.1. At present troff accepts 8-bit input, i.e. recognises 256
> distinct entities in the input stream (with a small number of
> exceptions which are "illegal").

We need at least 20 bit (for Unicode BMP + surrogates) and the special
characters.  A 32bit wide number is thus the right choice IMHO.

> It does not really matter that these are interpreted, by default, as
> iso-latin-1.

I plan to remove the hard-coded `charXXX' values, moving them to macro
files.

> A.2. The direct correspondence between input bytes and characters is
> defined in the font files for the device.

But this isn't the right place.  Input character stuff should not be
there at all.

> I don't see anything wrong (except possibly in ease of use) in
> creating an ASCII input stream in order to generate Japanese output.

Not everything can resp. should be handled on the glyph level, for
example hyphenation.

> Preparation of an output stream to drive a device capable of
> rendering the output is the job of the post-processor (and, provided
> you have installed appropriate font definition files, I cannot think
> of anything that would be beyond the PostScript device "devps").

As mentioned in another mail we have to extend the metric directives
to cope with the many CJK characters without make troff too slow.

> A: It follows that troff is already language-independent, for all
> languages whose typographic conventions can be achieved by the
> primitive mechanisms already present in troff. For such languages,
> there is no need to change troff at all. For some other languages,
> there are minor extra requirements which would require small
> extensions to troff which would not interact with existing
> mechanisms.

Correct.  The changes we are discussing only affects the input
character level and not the troff engine itself (except additional
typesetting features for CJK and possibly other languages).

> I think that troff's hard-wired set of ligatures should be replaced
> by a user-definable set.

Definitely.

> Some characters have different glyphs at the beginning, the middle,
> or the end of words; and so on.

Usually, such changes involve contextual analysis which I won't
implement.  In case a preprocessor is doing this, it has to directly
send glyph entities to troff.  So this isn't a problem.

> B: Troff should be able to cope with multi-lingual documents, where
> several different languages occur in the same document. I do NOT
> believe that the right way to do this is to extend troff's capacity
> to recognise thousands of different input encodings covering all the
> languages which it might be called upon to typeset (e.g. by Unicode
> or the like).

This is done by a preprocessor and not visible to troff itself.  troff
will see Unicode only.

> Troff's multi-character naming convention means that anything you
> could possibly need can be defined, and given a name in the troff
> input "character set" whenever you really need it, so long as you
> have the device resources to render the appropriate glyph.

There are only 256 `multi-characters' named `charXXX'.  Everything
else are glyph entities (even if they behave like a character in most
cases).  The reality is that groff doesn't really make a difference
between a character and a glyph, and it has high priority to me to
implement this distinction.  I'll probably start with renaming a lot
of troff internals.

> If you want to use a multi-byte encoding in your input-preparation
> software, you can pre-process this with a suitable filter to
> generate the troff input-sequences you need (I have done this with
> WordPerfect multinational characters, for instance, which are
> two-byte entities).

This filter will be the yet-to-come preprocessor.

> CONCLUSION: Troff certainly needs some extensions to cope with the
> typesetting demands of some languages (of which the major ones that
> I can think of have been mentioned above). I also believe that there
> are some features of troff which need to be changed in any case, but
> these has nothing to do with language or "locale".

Locales support only affects pre- and postprocessors.


    Werner

From groff-admin  Sat Oct 21 22:27:27 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id WAA31918
	for <groff@ffii.org>; Sat, 21 Oct 2000 22:27:27 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sat, 21 Oct 2000 22:32:05 +0200
Received: from rigel.univie.ac.at (dial-143140.HRZ.Uni-Dortmund.DE [129.217.143.140])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9LKVww20941;
	Sat, 21 Oct 2000 22:31:59 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id VAA01392;
	Sat, 21 Oct 2000 21:46:31 +0200
Date: Sat, 21 Oct 2000 21:46:31 +0200 (CEST)
Message-Id: <20001021.214631.45175800.wl@gnu.org>
To: tkb@tkb.mpl.com
Cc: tkubota@riken.go.jp, debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org,
        groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14832.32681.975096.519027@tkb.mpl.com>
References: <14830.38338.558492.22222V@surfchem0.riken.go.jp>
	<20001019.104035.106257837.wl@gnu.org>
	<14832.32681.975096.519027@tkb.mpl.com>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 174
Lines: 7

> Would it be useful to add to the texinfo documentation a note
> explaining that `-a' should only be used for these situations?

I've added some words, thanks.


    Werner

From groff-admin  Sat Oct 21 22:27:50 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id WAA31945
	for <groff@ffii.org>; Sat, 21 Oct 2000 22:27:49 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sat, 21 Oct 2000 22:32:30 +0200
Received: from rigel.univie.ac.at (dial-143140.HRZ.Uni-Dortmund.DE [129.217.143.140])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9LKWQw20956;
	Sat, 21 Oct 2000 22:32:26 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA01111;
	Sat, 21 Oct 2000 10:46:52 +0200
Date: Sat, 21 Oct 2000 10:46:51 +0200 (CEST)
Message-Id: <20001021.104651.85678878.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14832.31111.196218.91839F@surfchem0.riken.go.jp>
References: <14831.49362.330858.29321F@surfchem0.riken.go.jp>
	<20001020.144551.30188717.wl@gnu.org>
	<14832.31111.196218.91839F@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2228
Lines: 70

> 1. Your 'charset' and 'encoding' are for troff or for preprocessor?

In general.  I want to define terms completely independent on any
particular program.  We have

  character set
  character encoding
  glyph set
  glyph encoding

>    I thought both of them are for preprocessor.  The preprocessor
>    figures out the way to convert the input to UTF-8 from the
>    information.

A groff preprocessor will work as you have described.  Under the
assumption that you are talking about input characters, the term
`encoding' indeed implies the character set(s).  After some thinking I
have to correct myself: It is better to say that `EUC' is an `encoding
scheme' which describes which character ranges and how many bytes are
used.  Sorry for the confusion.

> 2. Which will the pre/postprocessors handle, characters or glyphs?

The preprocessor converts from characters to characters (i.e. to
Unicode), grotty + postprocessor convert glyph names back to Unicode
characters (using a hard-coded table), then from characters to
characters.  I don't know yet whether it makes sense to unify the
latter two programs.

> 3. Your 'charset' is for glyph and 'encoding' is for character?
>    I thought both of them are for character, since I thought both 
>    of them are for preprocessor.

My point was to make the distinction clear between `set' and
`encoding'.  Maybe it is only of academic interest, but it (hopefully)
helps to clear up the used terms.

> 4. I though we were discussing on (tags in roff souce for)
>    preprocessor.  Is that right?

Yes.

>    roff source in any encoding like '\(co'     (character)
>           |
>           |  preprocessor
>           V
>    UTF-8 stream like u+00a9                    (character)
>           |
>           |  troff
>           V
>    glyph expression like 'co'                  (glyph)
>           |
>           |  troff (continuing)
>           V

Here is missing a step:

     typeset output                              (glyph)
            |
            |  grotty
            V

>    UTF-8 stream like u+00a9 or '(C)'           (character)
>           |
>           |  postprocessor
>           V
>    formatted text in any encoding              (character)


    Werner

From groff-admin  Sun Oct 22 00:43:32 2000
Received: from probity.mcc.ac.uk (probity.mcc.ac.uk [130.88.200.94])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id AAA32650
	for <groff@ffii.org>; Sun, 22 Oct 2000 00:43:32 +0200
Received: from tedistan.annex.mcc.ac.uk
	([194.66.22.200] helo=tedistan.annex.umist.ac.uk ident=root)
	by probity.mcc.ac.uk with esmtp (Exim 2.05 #4)
	id 13n7RI-0006Sj-00; Sat, 21 Oct 2000 23:48:32 +0100
Received: (from efh@localhost)
	by eels.fort.knox.uk (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) id XAA04845;
	Sat, 21 Oct 2000 23:47:03 +0100
Message-ID: <XFMail.001021234703.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: XFMail 1.3 [p0] on Linux
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
In-Reply-To: <20001021.194635.08318027.wl@gnu.org>
Date: Sat, 21 Oct 2000 23:47:03 +0100 (BST)
Reply-To: Ted.Harding@nessie.mcc.ac.uk
From: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
To: Werner LEMBERG <wl@gnu.org>
Subject: Re: [Groff] Re: groff: radical re-implementation
Cc: debian-i18n@lists.debian.org, groff@ffii.org
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 3810
Lines: 82

Hi Werner (and all)

Thanks for this clarifying explanation. I have a couple of comments,
one explanatory, the other which, I think, may point to the core
of the question.

On 21-Oct-00 Werner LEMBERG wrote:
>> Troff's multi-character naming convention means that anything you
>> could possibly need can be defined, and given a name in the troff
>> input "character set" whenever you really need it, so long as you
>> have the device resources to render the appropriate glyph.
> 
> There are only 256 `multi-characters' named `charXXX'.  Everything
> else are glyph entities (even if they behave like a character in most
> cases).  The reality is that groff doesn't really make a difference
> between a character and a glyph, and it has high priority to me to
> implement this distinction.  I'll probably start with renaming a lot
> of troff internals.

1. Perhaps I should clarify: by "multi-character naming convention"
I mean the fact that you can decide to use the sequence of ASCII
characters, for instance, "\[O-ogonek]" as the name of a "character".

In passing: I see no _logical_ distinction between using a string
of ASCII characters to name a "character", and using a string of
bytes which implements a UTF-8 encoding.

2. Perhaps it is a good point of view to see troff (gtroff) as an
engine which handles _glyphs_, not characters, in a given context of
typographic style and layout. The current glyph is defined by the current
point size, the current font, and the name of the "character" which is to
be rendered, and troff necessarily takes account of the metric information
associated with this glyph.

The fact that ASCII characters and the iso-latin-1 characters
corresponding to byte-values > 128 are (by default) the troff names of
"characters" in a group of European languages -- together with certain
other marks and symbols -- is logically (in my view) an irrelevant
coincidence which happens to be very convenient for people using these
languages; but it is not at all necessary. Nothing at all stops you from
defining

  .char a \*a

as the name of Greek "alpha", and so on, if you want to simply the typing
of input in a passage of Greek using an ASCII interface.

Logically, therefore, troff could be "neutral" about what the byte "a"
stands for. From that point of view, a troff which makes no assumptions
of this kind, amd which consults external tables about the meaning of
its input and about the characteristics of what output that input
implies, purely for the purpose of correct formatting, is perhaps the
pure ideal. And from that point of view, therefore, unifying the input
conventions on the basis of a comprehensive encoding (such as UTF-8
or Unicode is intended to become) would be a great step towards
attaining this neutrality.

However, I wish to think more about this issue.

Meanwhile, interested parties who have not yet studied it may find
the "UTF-8 and Unicode FAQ for Unix/Linux" by Markus Kuhn well worth
reading:

  http://www.cl.cam.ac.uk/~mgk25/unicode.html

By the way, your comment that hyphenation, for instance, is not a "glyph
question" is, I think, not wholly correct. Certainly, hyphenation _rules_
are not a glyph question: as well as being language-dependent, there may
also be "house rules" about it; these come under "typographic style" as
above. But the size of a hyphen and associated spacing are glyph issues,
and these may interact with where a hyphenation occurs or whether it
occurs at all, according to the rules.

An interesting debate!

Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 284 7749
Date: 21-Oct-00                                       Time: 23:47:03
------------------------------ XFMail ------------------------------

From groff-admin  Sun Oct 22 09:05:02 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA30416
	for <groff@ffii.org>; Sun, 22 Oct 2000 09:05:01 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sun, 22 Oct 2000 09:10:01 +0200
Received: from rigel.univie.ac.at (dial-143155.HRZ.Uni-Dortmund.DE [129.217.143.155])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9M79vw28751;
	Sun, 22 Oct 2000 09:09:58 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id WAA01645;
	Sat, 21 Oct 2000 22:43:09 +0200
Date: Sat, 21 Oct 2000 22:43:09 +0200 (CEST)
Message-Id: <20001021.224309.35655881.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: groff@ffii.org, debian-i18n@lists.debian.org, cjh@kr.FreeBSD.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14833.40144.420924.72159A@surfchem0.riken.go.jp>
References: <20001017.092037.48403856.wl@gnu.org>
	<XFMail.001020203217.Ted.Harding@nessie.mcc.ac.uk>
	<14833.40144.420924.72159A@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 490
Lines: 11

> However, I am interested in how Groff 1.16 works for UTF-8 input.
> I could not find any code for UTF-8 input, though I found a code for
> UTF-8 output in src/devices/grotty/tty.cc .  Am I missing something?
> (Of course /font/devutf8/* has no implementation of UTF-8 encoding,
> though it seems to have a table for glyph names -> UCS-2.)

No UTF-8 input support.  This is an urgent need for one of the next
major releases.  grotty can output UTF-8 if activated with -Tutf8.


    Werner

From groff-admin  Sun Oct 22 09:05:02 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA30419
	for <groff@ffii.org>; Sun, 22 Oct 2000 09:05:02 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sun, 22 Oct 2000 09:09:58 +0200
Received: from rigel.univie.ac.at (dial-143155.HRZ.Uni-Dortmund.DE [129.217.143.155])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9M79sw28748;
	Sun, 22 Oct 2000 09:09:55 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id WAA01648;
	Sat, 21 Oct 2000 22:45:37 +0200
Date: Sat, 21 Oct 2000 22:45:37 +0200 (CEST)
Message-Id: <20001021.224537.38717761.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <14833.40697.666263.26482Z@surfchem0.riken.go.jp>
References: <14831.42770.6994.72159A@surfchem0.riken.go.jp>
	<20001020.141444.71083696.wl@gnu.org>
	<14833.40697.666263.26482Z@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=iso-2022-jp
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 711
Lines: 22

> > Well, maybe.  But sometimes there is kerning.  Please consult Ken
> > Lunde's `CJKV Information Processing' for details.  Example:
> > 
> >    $B!;(B
> >    $B0l(B
> >    $B!;(B
> 
> Wow!  This is the first time I received a Japanese mail from
> non-Japanese speaker.

I *like* the Mew mailing program for Emacs :-)

> I have seen the book at a few bookstores (not translated in Japanese
> yet).  Though I am interested in it, the book is too thick for me...
> (Even if the book were in Japanese, it would be heavy for me.)

I strongly recommend to buy it, especially if you are interested in
character sets and encodings.  Of the 1200+ pages (or so), about 900
pages are mapping tables.


    Werner

From groff-admin  Sun Oct 22 14:08:59 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id OAA31587
	for <groff@ffii.org>; Sun, 22 Oct 2000 14:08:58 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Sun, 22 Oct 2000 14:13:58 +0200
Received: from rigel.univie.ac.at (dial-143189.HRZ.Uni-Dortmund.DE [129.217.143.189])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9MCDtw02924;
	Sun, 22 Oct 2000 14:13:55 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id OAA02023;
	Sun, 22 Oct 2000 14:11:58 +0200
Date: Sun, 22 Oct 2000 14:11:58 +0200 (CEST)
Message-Id: <20001022.141158.06884328.wl@gnu.org>
To: Ted.Harding@nessie.mcc.ac.uk
Cc: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <XFMail.001021234703.Ted.Harding@nessie.mcc.ac.uk>
References: <20001021.194635.08318027.wl@gnu.org>
	<XFMail.001021234703.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2241
Lines: 48

> 2. Perhaps it is a good point of view to see troff (gtroff) as an
> engine which handles _glyphs_, not characters, in a given context of
> typographic style and layout. The current glyph is defined by the
> current point size, the current font, and the name of the
> "character" which is to be rendered, and troff necessarily takes
> account of the metric information associated with this glyph.

Exactly.  But the current terminology in gtroff is more than
ambiguous, and I believe that we need a clear separation between
characters and glyphs.

> Logically, therefore, troff could be "neutral" about what the byte
> "a" stands for. From that point of view, a troff which makes no
> assumptions of this kind, amd which consults external tables about
> the meaning of its input and about the characteristics of what
> output that input implies, purely for the purpose of correct
> formatting, is perhaps the pure ideal. And from that point of view,
> therefore, unifying the input conventions on the basis of a
> comprehensive encoding (such as UTF-8 or Unicode is intended to
> become) would be a great step towards attaining this neutrality.

I fully agree.  A single input character set (as universal as
possible) is the right thing, and everything else shall be managed by
preprocessors (and a postprocessor for tty).

> Meanwhile, interested parties who have not yet studied it may find
> the "UTF-8 and Unicode FAQ for Unix/Linux" by Markus Kuhn well worth
> reading:
> 
>   http://www.cl.cam.ac.uk/~mgk25/unicode.html

Yes, Markus is doing an excellent job.

> By the way, your comment that hyphenation, for instance, is not a
> "glyph question" is, I think, not wholly correct. Certainly,
> hyphenation _rules_ are not a glyph question: as well as being
> language-dependent, there may also be "house rules" about it; these
> come under "typographic style" as above. But the size of a hyphen
> and associated spacing are glyph issues, and these may interact with
> where a hyphenation occurs or whether it occurs at all, according to
> the rules.

I mean the algorithm of finding possible breakpoints which must be
based on input characters.  The final decision where a word will be
broken is of course a glyph issue.


    Werner

From groff-admin  Mon Oct 23 04:06:17 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id EAA04298
	for <groff@ffii.org>; Mon, 23 Oct 2000 04:06:15 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 53D3D1C203; Mon, 23 Oct 2000 11:15:25 +0900 (JST)
Date: Mon, 23 Oct 2000 11:15:22 +0900
Message-ID: <87itqkwbet.wl@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Sat, 21 Oct 2000 10:46:51 +0200 (CEST)"
	<20001021.104651.85678878.wl@gnu.org>
References: <14831.49362.330858.29321F@surfchem0.riken.go.jp>
	<20001020.144551.30188717.wl@gnu.org>
	<14832.31111.196218.91839F@surfchem0.riken.go.jp>
	<20001021.104651.85678878.wl@gnu.org>
User-Agent: Wanderlust/1.1.1 (Purple Rain) EMY/1.13.8 (Tastes differ) FLIM/1.13.2 (Kasanui) APEL/10.2 Emacs/20.7 (i386-debian-linux-gnu) MULE/4.1 (AOI)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2606
Lines: 80

Hi,

At Sat, 21 Oct 2000 10:46:51 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> In general.  I want to define terms completely independent on any
> particular program.  We have
> 
>   character set
>   character encoding
>   glyph set
>   glyph encoding

I understand.  Since we are discussing on the preprocessor, let's 
concentrate on character, not glyph.  I think you now will agree to
specify the 'character set/encoding' by a single word such as
'EUC-JP' instead of a pair of 'JIS-X-0208' and 'EUC'.

BTW, I am implementing the preprocessor.  Now it has features of:
 - input from standard input (stdin)
 - output to standard output (stdout)
 - I18N directive to support locale-sensible mode
 - hard-coded converter from Latin1, EBCDIC, and UTF-8 to UTF-8
 - locale-sensible converter from any encodings supported by OS to UTF-8
   (note: UTF-8 has to be supported by iconv(3) )
 - encoding for input is determined by command option or default
 - default is 'latin1' when compiled without I18N or locale-sensible when
   compiled with I18N
However I have to implement
 - encoding has to be determined also by '-*- ... -*-' directive in
   the roff source
 - (I18N mode) encoding has to be able to be specified by MIME-style
   and Emacs-style names.
 - efficiency of memory and CPU usage is not considered yet.
 - input from files besides stdin

I will send the source soon.


> >    roff source in any encoding like '\(co'     (character)
> >           |
> >           |  preprocessor
> >           V
> >    UTF-8 stream like u+00a9                    (character)
> >           |
> >           |  troff
> >           V
> >    glyph expression like 'co'                  (glyph)
> >           |
> >           |  troff (continuing)
> >           V
> 
> Here is missing a step:
> 
>      typeset output                              (glyph)
>             |
>             |  grotty
>             V
> 
> >    UTF-8 stream like u+00a9 or '(C)'           (character)
> >           |
> >           |  postprocessor
> >           V
> >    formatted text in any encoding              (character)


I understand well.  Thank you for your explanation.
BTW, besides TTY output, HTML will need postprocess from glyph to 
character like 'grotty' in tty mode, since HTML is a text file.
I think the encoding for HTML can be always UTF-8.  We can add a
line between <HEAD> and </HEAD>

<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=utf-8">

(I found a code in grohtml.cc to write this line without charset
directive.)

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Mon Oct 23 08:17:03 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id IAA19216
	for <groff@ffii.org>; Mon, 23 Oct 2000 08:17:03 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Mon, 23 Oct 2000 08:22:14 +0200
Received: from rigel.univie.ac.at (dial-143129.HRZ.Uni-Dortmund.DE [129.217.143.129])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9N6MBw17968
	for <groff@ffii.org>; Mon, 23 Oct 2000 08:22:12 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id IAA10686
	for <groff@ffii.org>; Mon, 23 Oct 2000 08:20:26 +0200
Date: Mon, 23 Oct 2000 08:20:25 +0200 (CEST)
Message-Id: <20001023.082025.10290877.wl@gnu.org>
To: groff@ffii.org
From: Werner LEMBERG <wl@gnu.org>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] new groff directory structure
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 255
Lines: 13

Dear friends,


to follow the GNU standards, groff will now by default install its
data files into

  /usr/local/share/groff/<version>

(e.g. /usr/local/share/groff/1.16.1).  Please update your packages, if
necessary, to reflect this change.


    Werner

From groff-admin  Mon Oct 23 09:41:53 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA20327
	for <groff@ffii.org>; Mon, 23 Oct 2000 09:41:52 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Mon, 23 Oct 2000 09:46:52 +0200
Received: from rigel.univie.ac.at (dial-143153.HRZ.Uni-Dortmund.DE [129.217.143.153])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9N7knw24680;
	Mon, 23 Oct 2000 09:46:49 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id JAA10975;
	Mon, 23 Oct 2000 09:42:18 +0200
Date: Mon, 23 Oct 2000 09:42:18 +0200 (CEST)
Message-Id: <20001023.094218.59476979.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <87itqkwbet.wl@surfchem0.riken.go.jp>
References: <14832.31111.196218.91839F@surfchem0.riken.go.jp>
	<20001021.104651.85678878.wl@gnu.org>
	<87itqkwbet.wl@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 856
Lines: 25

> I think you now will agree to specify the 'character set/encoding'
> by a single word such as 'EUC-JP' instead of a pair of 'JIS-X-0208'
> and 'EUC'.

Yes :-)

> BTW, I am implementing the preprocessor.  Now it has features of:

>  - hard-coded converter from Latin1, EBCDIC, and UTF-8 to UTF-8
>  - locale-sensible converter from any encodings supported by OS to UTF-8
>    (note: UTF-8 has to be supported by iconv(3) )

May I suggest that you temporarily implement a hack so that you can
use it with the Japanese patch of groff?  I don't know how long it
will take until the necessary changes for gtroff have been
implemented.

> BTW, besides TTY output, HTML will need postprocess from glyph to 
> character like 'grotty' in tty mode, since HTML is a text file.

Yes and no.  HTML output also supports entities with the &...;
directive.


    Werner

From groff-admin  Tue Oct 24 06:57:04 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id GAA13793
	for <groff@ffii.org>; Tue, 24 Oct 2000 06:57:03 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 6B3521C203; Tue, 24 Oct 2000 14:06:26 +0900 (JST)
Date: Tue, 24 Oct 2000 14:06:18 +0900
Message-ID: <87lmvekeut.wl@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Mon, 23 Oct 2000 09:42:18 +0200 (CEST)"
	<20001023.094218.59476979.wl@gnu.org>
References: <14832.31111.196218.91839F@surfchem0.riken.go.jp>
	<20001021.104651.85678878.wl@gnu.org>
	<87itqkwbet.wl@surfchem0.riken.go.jp>
	<20001023.094218.59476979.wl@gnu.org>
User-Agent: Wanderlust/1.1.1 (Purple Rain) EMY/1.13.8 (Tastes differ) FLIM/1.13.2 (Kasanui) APEL/10.2 Emacs/20.7 (i386-debian-linux-gnu) MULE/4.1 (AOI)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1961
Lines: 51

Hi,

At Mon, 23 Oct 2000 09:42:18 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> >  - hard-coded converter from Latin1, EBCDIC, and UTF-8 to UTF-8
> >  - locale-sensible converter from any encodings supported by OS to UTF-8
> >    (note: UTF-8 has to be supported by iconv(3) )
> 
> May I suggest that you temporarily implement a hack so that you can
> use it with the Japanese patch of groff?  I don't know how long it
> will take until the necessary changes for gtroff have been
> implemented.

What do you think about the merit of preprocessor with the current
Groff which doesn't recognize UTF-8 input?

I think the preprocessor can contribute Groff to be locale-sensible.
However, groff wrapper or troff will need some mechanism to receive
a report on locale from the preprocessor.

The algorithm will be: check locale and use
 - -Tlatin1 for Latin-1 languages
 - -Tnippon for Japanese
 - -Tascii8 for other languages
if groff wrapper is invoked with -Ttty.  (IMO, we should not override
user's specification of -Tlatin1, -Tascii, -Tnippon, and so on).

BTW, do you plan to release Groff with Japanese patch, with my
preprocessor, as a makeshift until Groff with UTF-8 input will be
available?  (I thought so since you seem to be interested in my
preprocessor working with Japanese-patched Groff. :-)


> > BTW, besides TTY output, HTML will need postprocess from glyph to 
> > character like 'grotty' in tty mode, since HTML is a text file.
> 
> Yes and no.  HTML output also supports entities with the &...;
> directive.

Either (UTF-8 or &...;) will be OK.
Eigher have their own merits and demerits.
HTML output will be a ASCII text with &...; .  ASCII is the most
portable character set/encoding in the world.  However, reading
HTML source with &...; will be hard if the most part of the text
consists from non-ASCII characters, such as Japanese, Russian, 
and Greek.

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Tue Oct 24 14:45:26 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id OAA18574
	for <groff@ffii.org>; Tue, 24 Oct 2000 14:45:25 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Tue, 24 Oct 2000 14:50:30 +0200
Received: from rigel.univie.ac.at (dial-143193.HRZ.Uni-Dortmund.DE [129.217.143.193])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9OCoQG17205;
	Tue, 24 Oct 2000 14:50:27 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA13013;
	Tue, 24 Oct 2000 10:32:14 +0200
Date: Tue, 24 Oct 2000 10:32:13 +0200 (CEST)
Message-Id: <20001024.103213.112617592.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <87lmvekeut.wl@surfchem0.riken.go.jp>
References: <87itqkwbet.wl@surfchem0.riken.go.jp>
	<20001023.094218.59476979.wl@gnu.org>
	<87lmvekeut.wl@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1132
Lines: 28

> The algorithm will be: check locale and use
>  - -Tlatin1 for Latin-1 languages
>  - -Tnippon for Japanese
>  - -Tascii8 for other languages
> if groff wrapper is invoked with -Ttty.  (IMO, we should not override
> user's specification of -Tlatin1, -Tascii, -Tnippon, and so on).

What mechanism do you suggest for communication between the
preprocessor and troff?

> BTW, do you plan to release Groff with Japanese patch, with my
> preprocessor, as a makeshift until Groff with UTF-8 input will be
> available?  (I thought so since you seem to be interested in my
> preprocessor working with Japanese-patched Groff. :-)

Do you think this is useful?  It introduces a lot of stuff which has
to be removed afterwards.  Wouldn't it be better to stay with a
separate groff-jp distribution?

> HTML output will be a ASCII text with &...; .  ASCII is the most
> portable character set/encoding in the world.  However, reading HTML
> source with &...; will be hard if the most part of the text consists
> from non-ASCII characters, such as Japanese, Russian, and Greek.

I'll leave this to Gaius who is working on HTML :-)


    Werner

From groff-admin  Tue Oct 24 20:53:51 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id UAA21585
	for <groff@ffii.org>; Tue, 24 Oct 2000 20:53:51 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Tue, 24 Oct 2000 20:59:02 +0200
Received: from rigel.univie.ac.at (dial-143154.HRZ.Uni-Dortmund.DE [129.217.143.154])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9OIx1G01938
	for <groff@ffii.org>; Tue, 24 Oct 2000 20:59:01 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id UAA17660
	for <groff@ffii.org>; Tue, 24 Oct 2000 20:58:56 +0200
Date: Tue, 24 Oct 2000 20:58:55 +0200 (CEST)
Message-Id: <20001024.205855.48522222.wl@gnu.org>
To: groff@ffii.org
From: Werner LEMBERG <wl@gnu.org>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] local tmac directory
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 153
Lines: 8

Dear friends,


similar to Emacs, I've added a local tmac directory called `site-tmac'
which will be scanned before the standard tmac path.


    Werner

From groff-admin  Wed Oct 25 02:02:58 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id CAA23772
	for <groff@ffii.org>; Wed, 25 Oct 2000 02:02:57 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 9A9211C203; Wed, 25 Oct 2000 09:12:39 +0900 (JST)
Date: Wed, 25 Oct 2000 09:12:34 +0900
Message-ID: <87n1ft93t9.wl@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Tue, 24 Oct 2000 10:32:13 +0200 (CEST)"
	<20001024.103213.112617592.wl@gnu.org>
References: <87itqkwbet.wl@surfchem0.riken.go.jp>
	<20001023.094218.59476979.wl@gnu.org>
	<87lmvekeut.wl@surfchem0.riken.go.jp>
	<20001024.103213.112617592.wl@gnu.org>
User-Agent: Wanderlust/1.1.1 (Purple Rain) EMY/1.13.8 (Tastes differ) FLIM/1.13.2 (Kasanui) APEL/10.2 Emacs/20.7 (i386-debian-linux-gnu) MULE/4.1 (AOI)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1708
Lines: 44

Hi,

At Tue, 24 Oct 2000 10:32:13 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

>> BTW, do you plan to release Groff with Japanese patch, with my
>> preprocessor, as a makeshift until Groff with UTF-8 input will be
>> available?  (I thought so since you seem to be interested in my
>> preprocessor working with Japanese-patched Groff. :-)
>
> Do you think this is useful?  It introduces a lot of stuff which has
> to be removed afterwards.  Wouldn't it be better to stay with a
> separate groff-jp distribution?

Ok, I fully agree with you.  Many users of Groff may be confused if
Groff with Japanese patch becomes an 'official' version.


>> The algorithm will be: check locale and use
>>  - -Tlatin1 for Latin-1 languages
>>  - -Tnippon for Japanese
>>  - -Tascii8 for other languages
>> if groff wrapper is invoked with -Ttty.  (IMO, we should not override
>> user's specification of -Tlatin1, -Tascii, -Tnippon, and so on).
>
> What mechanism do you suggest for communication between the
> preprocessor and troff?

Well, I thought again and conclude that the current version of Groff
cannot cooperate well with the preprocessor.  If we want
locale-sensibility before the re-implementation of troff, I suggest
that groff wrapper is suitable for such an implementation.  I think
this is not so big evil.

However, I think the best 'makeshift' will be that troff will support
UTF-8 I/O without drastic change of internal of troff.  I don't know
this can be acheved with minor modification of troff or not, since
I am not familiar with the internal of troff.

Then the pre/postprocessors will be able to cooperate with troff.

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota

From groff-admin  Wed Oct 25 15:28:07 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA10709
	for <groff@ffii.org>; Wed, 25 Oct 2000 15:28:07 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Wed, 25 Oct 2000 15:33:27 +0200
Received: from rigel.univie.ac.at (dial-143136.HRZ.Uni-Dortmund.DE [129.217.143.136])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9PDXLG15826;
	Wed, 25 Oct 2000 15:33:22 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA18444;
	Wed, 25 Oct 2000 10:09:34 +0200
Date: Wed, 25 Oct 2000 10:09:34 +0200 (CEST)
Message-Id: <20001025.100934.105425300.wl@gnu.org>
To: tkubota@riken.go.jp
Cc: debian-i18n@lists.debian.org, groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <87n1ft93t9.wl@surfchem0.riken.go.jp>
References: <87lmvekeut.wl@surfchem0.riken.go.jp>
	<20001024.103213.112617592.wl@gnu.org>
	<87n1ft93t9.wl@surfchem0.riken.go.jp>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1271
Lines: 32

> > What mechanism do you suggest for communication between the
> > preprocessor and troff?
> 
> Well, I thought again and conclude that the current version of Groff
> cannot cooperate well with the preprocessor.  If we want
> locale-sensibility before the re-implementation of troff, I suggest
> that groff wrapper is suitable for such an implementation.  I think
> this is not so big evil.

Hmm.  What about the following temporary solution:

  Run the preprocessor twice; the first time it is called directly by
  the groff program, and it returns an error code which groff can use
  to set the -T switch.  The proprocessor shouldn't do any processing
  except checking the locale setting and testing the first line(s) of
  the input file for an `encoding' directive'.

  The second time just do the normal pipeline.

Such an algorithm can be implemented with just a few lines of code, I
think.

> However, I think the best 'makeshift' will be that troff will
> support UTF-8 I/O without drastic change of internal of troff.  I
> don't know this can be acheved with minor modification of troff or
> not, since I am not familiar with the internal of troff.

There are a lot of things to change, test, and check until 32bit input
characters are accepted...


    Werner

From groff-admin  Wed Oct 25 23:08:56 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id XAA15840
	for <groff@ffii.org>; Wed, 25 Oct 2000 23:08:56 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Wed, 25 Oct 2000 23:14:12 +0200
Received: from rigel.univie.ac.at (dial-143172.HRZ.Uni-Dortmund.DE [129.217.143.172])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9PLEAG07275
	for <groff@ffii.org>; Wed, 25 Oct 2000 23:14:10 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id XAA20686
	for <groff@ffii.org>; Wed, 25 Oct 2000 23:13:43 +0200
Date: Wed, 25 Oct 2000 23:13:43 +0200 (CEST)
Message-Id: <20001025.231343.99195970.wl@gnu.org>
To: groff@ffii.org
From: Werner LEMBERG <wl@gnu.org>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] FOO.tmac vs tmac.FOO
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 500
Lines: 15

Dear friends,


I've reverted the current behaviour of troff: Now it first checks
whether FOO.tmac exists, then it tries tmac.FOO.  The main reason for
doing this is that I can get rid of the annoying renaming orgy for
some macro files which have identical names for AT&T troff (which is
installed in parallel).

I don't think this change (and the fixes that will come in the process
of renaming all macro files to the XXX.scheme) causes any harm since
it is fully backwards compatible.


    Werner

From groff-admin  Thu Oct 26 04:53:21 2000
Received: from surfchem0.riken.go.jp (surfchem0.riken.go.jp [134.160.25.161])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id EAA18715
	for <groff@ffii.org>; Thu, 26 Oct 2000 04:53:20 +0200
Received: from surfchem0.riken.go.jp (localhost [127.0.0.1])
	by surfchem0.riken.go.jp (Postfix) with ESMTP
	id 7895E1C203; Thu, 26 Oct 2000 12:03:08 +0900 (JST)
Date: Thu, 26 Oct 2000 12:03:03 +0900
Message-ID: <87u2a05moo.wl@surfchem0.riken.go.jp>
From: Tomohiro KUBOTA <tkubota@riken.go.jp>
To: debian-i18n@lists.debian.org, groff@ffii.org
Sent-via: debian-i18n@lists.debian.org,groff@ffii.org
Subject: Re: [Groff] Re: groff: radical re-implementation
Mail-Reply-To: tkubota@riken.go.jp
In-Reply-To: In your message of "Wed, 25 Oct 2000 10:09:34 +0200 (CEST)"
	<20001025.100934.105425300.wl@gnu.org>
References: <87lmvekeut.wl@surfchem0.riken.go.jp>
	<20001024.103213.112617592.wl@gnu.org>
	<87n1ft93t9.wl@surfchem0.riken.go.jp>
	<20001025.100934.105425300.wl@gnu.org>
User-Agent: Wanderlust/1.1.1 (Purple Rain) EMY/1.13.8 (Tastes differ) FLIM/1.13.2 (Kasanui) APEL/10.2 Emacs/20.7 (i386-debian-linux-gnu) MULE/4.1 (AOI)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2932
Lines: 80

Hi,

At Wed, 25 Oct 2000 10:09:34 +0200 (CEST),
Werner LEMBERG <wl@gnu.org> wrote:

> Hmm.  What about the following temporary solution:
> 
>   Run the preprocessor twice; the first time it is called directly by
>   the groff program, and it returns an error code which groff can use
>   to set the -T switch.  The proprocessor shouldn't do any processing
>   except checking the locale setting and testing the first line(s) of
>   the input file for an `encoding' directive'.
> 
>   The second time just do the normal pipeline.

I think such a locale-sensibility is achieved with small code,
even without any OS's locale-related functions such as setlocale(3)
and so on.  Just like:

---------------------------
char *locale;
locale = getenv("LC_ALL");
if (locale == NULL) locale=getenv("LC_CTYPE");
if (locale == NULL) locale=getenv("LANG");

struct lang_and_device {
  char *language;
  char *device;
} lang_table[] = {
  {"da", "latin1"},  /* Danish */
  {"de", "latin1"},  /* German */
  {"en", "latin1"},  /* English */
  {"es", "latin1"},  /* Spanish */
  {"fi", "latin1"},  /* Finnish */
  {"fr", "latin1"},  /* French */
  {"ga", "latin1"},  /* Irish */
  {"is", "latin1"},  /* Icelandic */
  {"it", "latin1"},  /* Italian */
  {"ja", "nippon"},  /* Japanese */
  {"nl", "latin1"},  /* Dutch */
  {"no", "latin1"},  /* Norwegian */
  {"pt", "latin1"},  /* Portuguese */
  {"sv", "latin1"},  /* Swedish */
  {NULL, NULL}
};

struct lang_and_device *lang;
char *device = "ascii8";
for (lang=lang_table; lang->language != NULL; lang++) {
  if (!strncmp(locale, lang->language, 2)) device = lang->device;
}
---------------------------

This code is for groff wrapper.  When groff wrapper is invoked with
command option of '-Ttty', this algorithm should be used.

Note:
 - 'utf8' device is not covered with this algorithm since the
   current troff doesn't support utf-8 input.  Since the current
   Groff cannot handle encodings for input output separately,
   it is appropriate that the both encodings for input and output
   are assumed to be the same.  However, the current implementation
   of 'utf8' device is ISO8859-1 input and UTF-8 output.
 - 'tty' is mapped to 'latin1', 'ascii8', and 'nippon' devices.
   It is obvious that this algorithm is useless for the official
   groff which does not support 'ascii8' nor 'nippon'.  Thus,
   if the official groff are to be locale-sensible, we should
   at least support 'ascii8' device.
 - 'ascii' device cannot be used for this purpose because even
   English-speaking people sometimes use ISO-8859-1 charsets.
   (Yes, people all over the world often read English manpages.
   Thus, I think it is evil to use ISO-8859-1 character in
   English manpages.  However, we are talking about user's
   locale environment, not about the language of the roff source.)

How do you think about this code?

---
Tomohiro KUBOTA <kubota@debian.org>
http://surfchem0.riken.go.jp/~kubota/

From groff-admin  Thu Oct 26 16:09:44 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA05720
	for <groff@ffii.org>; Thu, 26 Oct 2000 16:09:44 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Thu, 26 Oct 2000 16:15:10 +0200
Received: from rigel.univie.ac.at (dial-143195.HRZ.Uni-Dortmund.DE [129.217.143.195])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9QEF7G24076;
	Thu, 26 Oct 2000 16:15:09 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id JAA24534;
	Thu, 26 Oct 2000 09:35:12 +0200
Date: Thu, 26 Oct 2000 09:35:11 +0200 (CEST)
Message-Id: <20001026.093511.116354739.wl@gnu.org>
To: groff@ffii.org, fab@prosa.it
From: Werner LEMBERG <wl@gnu.org>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Subject: [Groff] an improved version of man.local
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 723
Lines: 39

Below an improved version of man.local to display man pages without
intermediate headers and footers.

Older versions of this file sometimes exited before printing the
footer line.  The used trick (increasing the page length a bit, then
issuing .fl) has been taken from the mdoc package.


   Enjoy!

     Werner


PS: I get a lot of mail delivery errors for <fab@debye.prosa.it>
    coming from the debian.org machines.  In case Fabrizio's address
    is incorrect: Who knows the new one?


======================================================================

.am TH
.if n \{\
.  wh 0
.  wh -1i
.  wh -.5i
.  an-header
.\}
..
.
.de an-end
.  nr % 1
.  pl +3v
.  fl
.  sp 3
.  an-p-footer
.  pl \\n[nl]u
..
.
.em an-end

From groff-admin  Thu Oct 26 16:30:26 2000
Received: from maxwell.phys.lafayette.edu (maxwell.phys.lafayette.edu [139.147.200.41])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id QAA06311
	for <groff@ffii.org>; Thu, 26 Oct 2000 16:30:25 +0200
Received: from localhost (doughera@localhost)
	by maxwell.phys.lafayette.edu (8.9.3+Sun/8.9.3) with ESMTP id KAA13442
	for <groff@ffii.org>; Thu, 26 Oct 2000 10:36:56 -0400 (EDT)
X-Authentication-Warning: maxwell.phys.lafayette.edu: doughera owned process doing -bs
Date: Thu, 26 Oct 2000 10:36:55 -0400 (EDT)
From: Andy Dougherty <doughera@lafayette.edu>
X-Sender: doughera@maxwell.phys.lafayette.edu
To: groff@ffii.org
Subject: Re: [Groff] new groff directory structure
In-Reply-To: <20001024.205855.48522222.wl@gnu.org>
Message-ID: <Pine.SOL.4.10.10010260950500.13372-100000@maxwell.phys.lafayette.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2602
Lines: 61

> to follow the GNU standards, groff will now by default install its
> data files into

>    /usr/local/share/groff/<version>

>  e.g. /usr/local/share/groff/1.16.1).

This would break nearly every document I have written on my system. I use
the -mm macros extensively.  These macros can be controlled by a number of
parameters that must be set *before* the -mm macros are loaded.  Hence, my
typical prologue is something like:

    .nr N 2                 .\" Page header replaces footer on Page 1.
    .ie \n(.g .so /usr/local/share/groff/tmac/tmac.m
    .el .so /usr/lib/tmac/tmac.m

This style is recommended in section 2.5 of the AT&T -mm documentation
(without the groff reference, of course!).

If a version number gets added, all my old documents won't print without
hand-editing. Worse, the hand-editing will have to be changed every time
the version number gets changed.

I'm curious why this might be considered useful.  It's not at all clear to
me what is being accomplished.  (I helped design and implement perl's
versioned library structure, so I do have considerable experience with the
sorts of issues involved.)

I just checked on http://www.gnu.org/prep/standards.html and didn't see
any specific reference to the issue of libraries and version numbers.
Perhaps there is some other "standard" I should consult for the rationale?

As I see it at first glance, the difference between groff and other
packages that use versioned libraries (such as perl and emacs) is that the
package automatically handles the version numbers; users aren't expected
to load in the files directly.  At least for the -mm macros, directly
importing them is a useful documented idiom.

I see a few possible solutions.  Doubtless there are other better ones
too:

1.  Don't use version-specific directories.  (Current behavior).
    Unless you are really trying to support multiple simultaneous 
    installations of different versions of groff, the current behavior
    seems fine.
2.  Add in a "smarter" .so command that "knows" about the versioned
    directory structure (and probably about the site directories too).
3.  For each of the standard tmac.* files, add in a stub 
    tmac.* file in /usr/local/src/groff/ that sources the one from the
    versioned directory.
4.  Define a string register of some sort to hold the current version
    so a .so command can be written that is robust across version
    changes.

These are just a few suggestions.  Doubtless others may have better ones.

-- 
    Andy Dougherty		doughera@lafayette.edu
    Dept. of Physics
    Lafayette College, Easton PA 18042


From groff-admin  Fri Oct 27 07:49:35 2000
Received: from nx5.HRZ.Uni-Dortmund.DE (nx5.HRZ.Uni-Dortmund.DE [129.217.131.21])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id HAA27321
	for <groff@ffii.org>; Fri, 27 Oct 2000 07:49:34 +0200
Received: from sx2.hrz.uni-dortmund.de by nx5.HRZ.Uni-Dortmund.DE
          via smtp-local with ESMTP; Fri, 27 Oct 2000 07:55:00 +0200
Received: from rigel.univie.ac.at (dial-143174.HRZ.Uni-Dortmund.DE [129.217.143.174])
	by sx2.hrz.uni-dortmund.de (8.10.2+Sun/8.9.3) with ESMTP id e9R5smG11746;
	Fri, 27 Oct 2000 07:54:53 +0200 (MET DST)
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id AAA26674;
	Fri, 27 Oct 2000 00:17:46 +0200
Date: Fri, 27 Oct 2000 00:17:45 +0200 (CEST)
Message-Id: <20001027.001745.131922597.wl@gnu.org>
To: doughera@lafayette.edu
Cc: groff@ffii.org
Subject: Re: [Groff] new groff directory structure
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <Pine.SOL.4.10.10010260950500.13372-100000@maxwell.phys.lafayette.edu>
References: <20001024.205855.48522222.wl@gnu.org>
	<Pine.SOL.4.10.10010260950500.13372-100000@maxwell.phys.lafayette.edu>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2078
Lines: 57

> > to follow the GNU standards, groff will now by default install its
> > data files into
> 
> >    /usr/local/share/groff/<version>
> 
> >  e.g. /usr/local/share/groff/1.16.1).
> 
> This would break nearly every document I have written on my system.
>
> ...
> 
> If a version number gets added, all my old documents won't print
> without hand-editing. Worse, the hand-editing will have to be
> changed every time the version number gets changed.

You must use the .mso request!  The example would then look like this:

  .nr N 2                 .\" Page header replaces footer on Page 1.
  .ie \n(.g .mso tmac.m
  .el .so /usr/lib/tmac/tmac.m

> I'm curious why this might be considered useful.  It's not at all
> clear to me what is being accomplished.  (I helped design and
> implement perl's versioned library structure, so I do have
> considerable experience with the sorts of issues involved.)

It makes sense.  Consider how emacs is working.  I was in Tsukuba
(Japan) this summer, working together with some core developers of
Emacs and XEmacs -- these guys are running up to four different Emacs
and XEmacs versions simultaneously!

> I just checked on http://www.gnu.org/prep/standards.html and didn't
> see any specific reference to the issue of libraries and version
> numbers.  Perhaps there is some other "standard" I should consult
> for the rationale?

All GNU package maintainers received a mail from RMS to do that,
making installation of different versions possible.  It was an easy
fix, and together with the new site-tmac directory you get a clean
solution, I think.

> 2.  Add in a "smarter" .so command that "knows" about the versioned
>     directory structure (and probably about the site directories
>     too). 

.mso

> 4.  Define a string register of some sort to hold the current version
>     so a .so command can be written that is robust across version
>     changes.

Major version number is \n(.x, minor version number is \n(.y, and
revision number is \n(.Y; for version directories, the revision number
is omitted if it is zero.


    Werner

From groff-admin  Fri Oct 27 15:32:02 2000
Received: from maxwell.phys.lafayette.edu (maxwell.phys.lafayette.edu [139.147.200.41])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id PAA30858
	for <groff@ffii.org>; Fri, 27 Oct 2000 15:32:00 +0200
Received: from localhost (doughera@localhost)
	by maxwell.phys.lafayette.edu (8.9.3+Sun/8.9.3) with ESMTP id JAA17746;
	Fri, 27 Oct 2000 09:38:19 -0400 (EDT)
X-Authentication-Warning: maxwell.phys.lafayette.edu: doughera owned process doing -bs
Date: Fri, 27 Oct 2000 09:38:18 -0400 (EDT)
From: Andy Dougherty <doughera@lafayette.edu>
X-Sender: doughera@maxwell.phys.lafayette.edu
To: Werner LEMBERG <wl@gnu.org>
cc: groff@ffii.org
Subject: Re: [Groff] new groff directory structure
In-Reply-To: <20001027.001745.131922597.wl@gnu.org>
Message-ID: <Pine.SOL.4.10.10010270916040.17647-100000@maxwell.phys.lafayette.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2523
Lines: 51

On Fri, 27 Oct 2000, Werner LEMBERG wrote:

> You must use the .mso request!  The example would then look like this:
> 
>   .nr N 2                 .\" Page header replaces footer on Page 1.
>   .ie \n(.g .mso tmac.m
>   .el .so /usr/lib/tmac/tmac.m

Ahh.  Thanks.  I can only plead age and ignorance and failure to read the
manual :-).  (I've been doing that particular idiom for at least 4
different troff's for at least 15 years.  It never occured to me to check
if groff had an enhanced .so request.)  Sorry about that.  I'll still have
to go back and hand-edit documents, but you can reasonably argue that's my
fault, not groff's.  And at least I'll only have to do it once.

> > I'm curious why this might be considered useful.  It's not at all
> > clear to me what is being accomplished.  (I helped design and
> > implement perl's versioned library structure, so I do have
> > considerable experience with the sorts of issues involved.)
> 
> It makes sense.  Consider how emacs is working.  I was in Tsukuba
> (Japan) this summer, working together with some core developers of
> Emacs and XEmacs -- these guys are running up to four different Emacs
> and XEmacs versions simultaneously!

I realized after sending my message that I had failed to consider
cross-system sorts of sharing.  That is, if the files in /usr/share/groff
are truly shared across systems (e.g. Solaris and AIX both sharing the
/usr/share filesystem (or even two Linux/386 systems), then it could well
be that you'd want to have 1.17.1 installed for say the Solaris systems
while you might want to keep 1.16.1 around for the AIX systems (e.g. if
there's some sort of porting or stability issue with AIX that can't be
instantly solved).  The versioned directories do handle that nicely.  

If the goal is to eventually have separate versions of groff all on the
same system (e.g. 1.16.1 and 1.17.1) running simultaneously and sharing a
directory tree, then there are still the problems that the associated
binaries (e.g. troff, pic, eqn, refer, etc.) don't have version numbers
and hence must be segregated somehow else.  It was *this* sort of issue
that puzzled me.  I don't use emacs, but my understanding is that at least
emacs and Xemacs have different binary names so both can be installed
simultaneously in /usr/local/bin.

Thanks for taking the time to explain things.  I appreciate it.
And thanks for the groff development work.

-- 
    Andy Dougherty		doughera@lafayette.edu
    Dept. of Physics
    Lafayette College, Easton PA 18042


From groff-admin  Fri Oct 27 19:58:37 2000
Received: from Hesperia (ppp56-global1.libertysurf.fr [213.36.120.56])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id TAA00314
	for <groff@ffii.org>; Fri, 27 Oct 2000 19:58:36 +0200
X-Authentication-Warning: genba.ffii.org: Host ppp56-global1.libertysurf.fr [213.36.120.56] claimed to be Hesperia
Received: from baruchel by Hesperia with local (Exim 2.05 #1 (Debian))
	id 13pDpt-0003Wx-00; Fri, 27 Oct 2000 20:02:37 +0200
Date: Fri, 27 Oct 2000 20:02:36 +0200
From: Thomas Baruchel <baruchel@libertysurf.fr>
To: groff@ffii.org
Message-ID: <20001027200236.A13566@Hesperia>
Reply-To: Thomas Baruchel <baruchel@libertysurf.fr>
Mail-Followup-To: groff@ffii.org
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
X-Machine: Hesperia sur Debian (Linux)
Subject: [Groff] How to compile a unicode file ?
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 296
Lines: 13

Brest, le vendredi 27 octobre

Hi,

I just discovered the SAM editor, which saves files in the unicode charset.
I wonder if I can compile them with groff without translate them to latin1...

-- 
   .~.   Thomas Baruchel
   /V\   baruchel@libertysurf.fr
  // \\    Brest
 /(   )\   FRANCE
  ^`~'^

From groff-admin  Sat Oct 28 22:13:52 2000
Received: from unet.univie.ac.at (unet.univie.ac.at [131.130.230.7])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id WAA20110
	for <groff@ffii.org>; Sat, 28 Oct 2000 22:13:51 +0200
Received: from rigel.univie.ac.at (root@unet2-69.univie.ac.at [131.130.232.69])
	by unet.univie.ac.at (8.9.3/8.9.3) with ESMTP id WAA203706;
	Sat, 28 Oct 2000 22:19:33 +0200
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id OAA01151;
	Sat, 28 Oct 2000 14:43:49 +0200
Date: Sat, 28 Oct 2000 14:43:49 +0200 (CEST)
Message-Id: <20001028.144349.78703857.wl@gnu.org>
To: doughera@lafayette.edu
Cc: groff@ffii.org
Subject: Re: [Groff] new groff directory structure
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <Pine.SOL.4.10.10010270916040.17647-100000@maxwell.phys.lafayette.edu>
References: <20001027.001745.131922597.wl@gnu.org>
	<Pine.SOL.4.10.10010270916040.17647-100000@maxwell.phys.lafayette.edu>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1224
Lines: 28

> If the goal is to eventually have separate versions of groff all on
> the same system (e.g. 1.16.1 and 1.17.1) running simultaneously and
> sharing a directory tree, then there are still the problems that the
> associated binaries (e.g. troff, pic, eqn, refer, etc.) don't have
> version numbers and hence must be segregated somehow else.  It was
> *this* sort of issue that puzzled me.  I don't use emacs, but my
> understanding is that at least emacs and Xemacs have different
> binary names so both can be installed simultaneously in
> /usr/local/bin.

Indeed.  I'm thinking about this -- the solution is quite easy; just
name the binaries, say, `groff-1.16.1' and add a hard link to
`groff'.  Emacs even goes a step further: It installs binaries which
are called by Emacs only into something like

  /usr/local/libexec/emacs/20.6/i686-pc-linux-gnu

Since many people use the groff programs in a pipe, this option (as
nice it is) can't be used.

Maybe groff (the program) shall check how it is called.  If arg[0] is,
say, groff-1.16.1, it takes the `-1.16.1' postfix and appends it to
all programs it is calling, i.e., pic-1.16.1, grotty-1.16.1, etc.

Has somebody a more elegant solution to this problem?


    Werner

From groff-admin  Sun Oct 29 08:15:43 2000
Received: from unet.univie.ac.at (unet.univie.ac.at [131.130.230.7])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id IAA16932
	for <groff@ffii.org>; Sun, 29 Oct 2000 08:15:42 +0100
Received: from rigel.univie.ac.at (root@unet4-41.univie.ac.at [131.130.233.41])
	by unet.univie.ac.at (8.9.3/8.9.3) with ESMTP id IAA225364;
	Sun, 29 Oct 2000 08:21:31 +0100
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id WAA01376;
	Sat, 28 Oct 2000 22:35:18 +0200
Date: Sat, 28 Oct 2000 22:35:18 +0200 (CEST)
Message-Id: <20001028.223518.39155860.wl@gnu.org>
To: baruchel@libertysurf.fr
Cc: groff@ffii.org
Subject: Re: [Groff] How to compile a unicode file ?
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <20001027200236.A13566@Hesperia>
References: <20001027200236.A13566@Hesperia>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 265
Lines: 10

> I just discovered the SAM editor, which saves files in the unicode
> charset.  I wonder if I can compile them with groff without
> translate them to latin1...

This is indeed a funny editor...

And sorry, Unicode input is not yet available in groff.


    Werner

From groff-admin  Sun Oct 29 11:38:58 2000
Received: from fencepost.gnu.org (we-refuse-to-spy-on-our-users@fencepost.gnu.org [199.232.76.164])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id LAA18048
	for <groff@ffii.org>; Sun, 29 Oct 2000 11:38:57 +0100
From: beowulf@quik.com
Received: from vector.intergate.ca ([207.34.179.20])
	by fencepost.gnu.org with smtp (Exim 3.16 #1 (Debian))
	id 13ppxH-0007yJ-00
	for <groff@gnu.org>; Sun, 29 Oct 2000 05:44:47 -0500
Received: (qmail 20814 invoked by uid 1007); 29 Oct 2000 10:45:52 -0000
Received: from beowulf@quik.com by vector.intergate.ca with qmail-scanner-0.90 (uvscan: v4.0.50/v4054. . Clean. Processed in 0.394918 secs); 29/10/2000 02:45:52
Received: from gateway-208.181.229.173.intergate.ca (HELO quik.com) (208.181.229.173)
  by vector.intergate.ca with SMTP; 29 Oct 2000 10:45:52 -0000
Reply-To: beowulf@quik.com
To: groff@gnu.org
Message-Id: <E13ppxH-0007yJ-00@fencepost.gnu.org>
Date: Sun, 29 Oct 2000 05:44:47 -0500
Subject: [Groff] Attn: Corporate Traing Coordinator - Windows 2000 Courseware Content
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1876
Lines: 30

As you restructure your IT training program and consider how you might
imagine the best way of providing content for your courses, you'll
find yourself appreciating course content that is already prepared
and ready for immediate use in your curriculum.  

We design and develop books and curriculum that support Systems Engineering technologies and certifications including:
· A+
· Network+
· i-Net+
· MCSE
· MCDBA 
(Full course descriptions are available)

Some of our books have been top 10 best sellers through Amazon.com.
Included with this letter are descriptions of the books that we sell (see Study Guide Descriptions).

I thought you might also be interested to hear about our adaptive and non-adaptive test engine slated to be available 3rd quarter, 2000.  The engine is designed for practice and study.  Test questions and answers are easily added, edited or deleted.  

We can help you with your high-tech training needs including books, books on CD, Instructor led classes and distance learning products.  We also create custom high-tech curriculum for companies. 

We are a one-stop solution for high-tech educational content, training and resources. We offer distance learning and instructor lead products for both individuals and groups and our education solutions meet typical corporate and learning center needs.  We are a Microsoft Solution Provider (MSP), Microsoft Certified Technical Education Center (CTEC) and a Computer Technology Industry Association Certified Training Center (CompTIA).  We partner with a major University, an international leader in adult-instructor-lead and distance learning, to offer graduate and undergraduate credit for many of our courses. 

Please feel free to contact me in the mean time.  I look forward speaking with you further.

Casey Lea, Creative Director
or Domhnall Adams, CS
DCGNA, CS and Associates
780-998-4066



From groff-admin  Sun Oct 29 17:59:20 2000
Received: from probity.mcc.ac.uk (probity.mcc.ac.uk [130.88.200.94])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id RAA20850
	for <groff@ffii.org>; Sun, 29 Oct 2000 17:59:19 +0100
Received: from tedistan.annex.mcc.ac.uk
	([194.66.22.200] helo=tedistan.annex.umist.ac.uk ident=root)
	by probity.mcc.ac.uk with esmtp (Exim 2.05 #4)
	id 13pvtP-000FKn-00
	for groff@ffii.org; Sun, 29 Oct 2000 17:05:12 +0000
Received: (from efh@localhost)
	by eels.fort.knox.uk (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) id OAA07086;
	Sun, 29 Oct 2000 14:48:10 GMT
Message-ID: <XFMail.001029144810.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: XFMail 1.3 [p0] on Linux
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
In-Reply-To: <20001028.144349.78703857.wl@gnu.org>
Date: Sun, 29 Oct 2000 14:48:10 -0000 (GMT)
Reply-To: Ted.Harding@nessie.mcc.ac.uk
From: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
To: groff@ffii.org
Subject: Re: [Groff] new groff directory structure
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1786
Lines: 40

On 28-Oct-00 Werner LEMBERG wrote:
> Indeed.  I'm thinking about this -- the solution is quite easy; just
> name the binaries, say, `groff-1.16.1' and add a hard link to
> `groff'.  Emacs even goes a step further: It installs binaries which
> are called by Emacs only into something like
> 
>   /usr/local/libexec/emacs/20.6/i686-pc-linux-gnu
> 
> Since many people use the groff programs in a pipe, this option (as
> nice it is) can't be used.

Not to mention all the people who don't use emacs ...

> Maybe groff (the program) shall check how it is called.  If arg[0] is,
> say, groff-1.16.1, it takes the `-1.16.1' postfix and appends it to
> all programs it is calling, i.e., pic-1.16.1, grotty-1.16.1, etc.
> 
> Has somebody a more elegant solution to this problem?

There are good arguments for having a version-dependent directory tree.

I do NOT like the idea of groff checking how it is called. For one thing,
it complicates groff for unnecessary reasons. Another is that one may
wish to call different components of the groff package (eqn, tbl, pic,
refer, troff, ... ) individually (as I often do). This suggested solution
would have to be applied to each component, if they were respectively
version-dependent also.

Probably the smoothest way to arrange things is to have subtrees with
version-dependent names (both in /usr/bin and /usr/...../groff etc),
together with a rich set of symbolic links in the usual places.
However, this restructuring is prety radical too ...

Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 284 7749
Date: 29-Oct-00                                       Time: 14:48:10
------------------------------ XFMail ------------------------------

From groff-admin  Mon Oct 30 09:30:45 2000
Received: from ramses.erlm.siemens.de (ramses.erlm.siemens.de [212.114.202.13])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA05265
	for <groff@ffii.org>; Mon, 30 Oct 2000 09:30:45 +0100
Received: from mail01.erlm.siemens.de (mail01.erlm.siemens.de [146.254.160.12])
	by ramses.erlm.siemens.de (8.9.3/8.9.3) with ESMTP id JAA17270;
	Mon, 30 Oct 2000 09:36:11 +0100 (MET)
Received: from nbgv101a.nbg9.siemens.de (nbgv101a.nbg9.siemens.de [157.163.80.101])
	by mail01.erlm.siemens.de (8.9.3+Sun/8.9.3) with ESMTP id JAA10184;
	Mon, 30 Oct 2000 09:35:49 +0100 (MET)
Received: by NBGV101A with Internet Mail Service (5.5.2650.21)
	id <4FTN4WCX>; Mon, 30 Oct 2000 09:35:52 +0100
Message-ID: <11DD9734B036D311947400805F0DD607BD8AE1@NBGV101A>
From: Freudenberger Joerg <Joerg.Freudenberger@nbg9.siemens.de>
To: "'Werner LEMBERG'" <wl@gnu.org>
Cc: groff@ffii.org
Subject: RE: [Groff] new groff directory structure
Date: Mon, 30 Oct 2000 09:35:48 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by genba.ffii.org id JAA05266
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 405
Lines: 15

Hallo together

> Has somebody a more elegant solution to this problem?

Have a look to the mechanism of gcc. 
I know it is definitely complex.
But there are many compiler users juggling with -V <Version> 
(this is that what we need) and the other architecture options.

And there is no solution for pic egn tbl and friends ...
I think of an environment variable may be named GROFF_VERSION.

Yours
	Jörg


From groff-admin  Mon Oct 30 11:59:24 2000
Received: from unet.univie.ac.at (unet.univie.ac.at [131.130.230.7])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id LAA08008
	for <groff@ffii.org>; Mon, 30 Oct 2000 11:59:23 +0100
Received: from rigel.univie.ac.at (root@uvo-22.univie.ac.at [131.130.230.22])
	by unet.univie.ac.at (8.9.3/8.9.3) with ESMTP id MAA65248;
	Mon, 30 Oct 2000 12:05:11 +0100
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id KAA02735;
	Mon, 30 Oct 2000 10:11:36 +0100
Date: Mon, 30 Oct 2000 10:11:36 +0100 (CET)
Message-Id: <20001030.101136.124290560.wl@gnu.org>
To: Ted.Harding@nessie.mcc.ac.uk
Cc: groff@ffii.org, eliz@is.elta.co.il
Subject: Re: [Groff] new groff directory structure
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <XFMail.001029144810.Ted.Harding@nessie.mcc.ac.uk>
References: <20001028.144349.78703857.wl@gnu.org>
	<XFMail.001029144810.Ted.Harding@nessie.mcc.ac.uk>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 2094
Lines: 51

[about versioned binaries similar to Emacs]

> > Maybe groff (the program) shall check how it is called.  If arg[0] is,
> > say, groff-1.16.1, it takes the `-1.16.1' postfix and appends it to
> > all programs it is calling, i.e., pic-1.16.1, grotty-1.16.1, etc.
> 
> I do NOT like the idea of groff checking how it is called. For one
> thing, it complicates groff for unnecessary reasons. Another is that
> one may wish to call different components of the groff package (eqn,
> tbl, pic, refer, troff, ... ) individually (as I often do). This
> suggested solution would have to be applied to each component, if
> they were respectively version-dependent also.

My idea was to let groff (the program) manage the versions.  In most
cases this would do a good job.  And no, only groff would have this
logic since none of the components calls other programs.

> Probably the smoothest way to arrange things is to have subtrees
> with version-dependent names (both in /usr/bin and /usr/...../groff
> etc), together with a rich set of symbolic links in the usual
> places.  However, this restructuring is pretty radical too ...

I can't see how this will solve the problem to avoid

  grn-1.16.1 xxx | pic-1.16.1 | ...

if the current version is, say, 1.17 -- or do you intend to add a
directory in front of $PATH pointing to your proposed `bin'
subdirectory?

Having subtrees in /usr/bin is, hmm, very strange.

Another solution (which is IMHO even better than my proposed changes
to groff): Install the pic binary as, say, pic-1.16.1 in the libexec
tree.  `pic' in /usr/local/bin then becomes a script which contains
something like

  : ${GROFF_VERSION-1.16.1}
  /usr/local/libexec/groff/$GROFF_VERSION/pic

If GROFF_VERSION isn't set, pic version 1.16.1 is called.  Otherwise,
the specified version will be used.

For MSDOS (and Windows?) it might be necessary to replace the scripts
with small binaries since batch files don't support pipes, AFAIK.

What about other packages with a couple of versioned binaries
(contrary to Emacs and gcc which basically only have a single binary)?


    Werner

From groff-admin  Mon Oct 30 12:14:51 2000
Received: from clyde.ref.collins.co.uk (clyde.ref.collins.co.uk [192.153.153.29])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id MAA08193
	for <groff@ffii.org>; Mon, 30 Oct 2000 12:14:49 +0100
Received: from ref.collins.co.uk (helmsdale [193.119.38.15])
	by clyde.ref.collins.co.uk (8.9.3/8.9.3) with ESMTP id LAA19308
	for <groff@ffii.org>; Mon, 30 Oct 2000 11:20:45 GMT
Received: from ref.collins.co.uk (localhost [127.0.0.1])
	by ref.collins.co.uk (8.9.3+Sun/8.9.1) with ESMTP id LAA00930
	for <groff@ffii.org>; Mon, 30 Oct 2000 11:20:11 GMT
Message-ID: <39FD596B.23461942@ref.collins.co.uk>
Date: Mon, 30 Oct 2000 11:20:11 +0000
From: "Stewart C. Russell" <stewart@ref.collins.co.uk>
Organization: HarperCollins Publishers, Bilingual Division
X-Mailer: Mozilla 4.61 [en] (X11; U; SunOS 5.7 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
To: groff@ffii.org
Subject: Re: [Groff] new groff directory structure
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 393
Lines: 10

> /usr/local/libexec/groff/

can I make a plea for the incredibly lazy amongst us?

/usr/local/libexec clashes with /usr/local/lib for those of us who rely
on file completion and typeahead to get things done. How difficult would
it have been to choose a unique directory prefix?

Maybe I've spent too long working on system consoles with broken
displays. GUIs? CRTs? Strictly for beginners...

From groff-admin  Mon Oct 30 12:52:22 2000
Received: from atlantis.erlm.siemens.de (atlantis.erlm.siemens.de [212.114.202.14])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id MAA08451
	for <groff@ffii.org>; Mon, 30 Oct 2000 12:52:22 +0100
Received: from tcenter.erlm.siemens.de (tcenter.erlm.siemens.de [146.254.160.14])
	by atlantis.erlm.siemens.de (8.9.3/8.9.3) with ESMTP id MAA20895;
	Mon, 30 Oct 2000 12:41:03 +0100 (MET)
Received: from mail01.erlm.siemens.de (mail01.erlm.siemens.de [146.254.160.12])
	by tcenter.erlm.siemens.de (8.9.0/8.9.0) with ESMTP id MAA17903;
	Mon, 30 Oct 2000 12:41:03 +0100 (MET)
Received: from nbgv101a.nbg9.siemens.de (nbgv101a.nbg9.siemens.de [157.163.80.101])
	by mail01.erlm.siemens.de (8.9.3+Sun/8.9.3) with ESMTP id MAA26727;
	Mon, 30 Oct 2000 12:40:52 +0100 (MET)
Received: by NBGV101A with Internet Mail Service (5.5.2650.21)
	id <4FTN4W30>; Mon, 30 Oct 2000 12:40:56 +0100
Message-ID: <11DD9734B036D311947400805F0DD607BD8AE4@NBGV101A>
From: Freudenberger Joerg <Joerg.Freudenberger@nbg9.siemens.de>
To: "'Werner LEMBERG'" <wl@gnu.org>, Ted.Harding@nessie.mcc.ac.uk
Cc: groff@ffii.org, eliz@is.elta.co.il
Subject: RE: [Groff] new groff directory structure
Date: Mon, 30 Oct 2000 12:40:52 +0100
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by genba.ffii.org id MAA08452
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 461
Lines: 15

> What about other packages with a couple of versioned binaries
> (contrary to Emacs and gcc which basically only have a single binary)?

gcc has a sortiment of wrapper binaries ( cc, c++, f77, gcc, g++, g77 ... )
which are usualy in /bin. (configure with PREFIX=/ !!)
The real work is done by a bundle of binaries in 
/lib/gcc-lib/<arch>/<version>.

The jungest wrapper is able to handle the access to the older versions.

Only meant as comment.

Yours
	Jörg


From groff-admin  Mon Oct 30 18:38:35 2000
Received: from mailgw1.netvision.net.il (mailgw1.netvision.net.il [194.90.1.14])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA12382
	for <groff@ffii.org>; Mon, 30 Oct 2000 18:38:33 +0100
Received: from is.elta.co.il (is.elta.co.il [199.203.121.2])
	by mailgw1.netvision.net.il (8.9.3/8.9.3) with ESMTP id TAA28526;
	Mon, 30 Oct 2000 19:44:29 +0200 (IST)
Received: from is (is [199.203.121.2])
	by is.elta.co.il (8.9.3/8.8.8) with SMTP id TAA15960;
	Mon, 30 Oct 2000 19:43:20 +0200 (IST)
Date: Mon, 30 Oct 2000 19:43:19 +0200 (IST)
From: Eli Zaretskii <eliz@is.elta.co.il>
X-Sender: eliz@is
To: Werner LEMBERG <wl@gnu.org>
cc: Ted.Harding@nessie.mcc.ac.uk, groff@ffii.org
Subject: Re: [Groff] new groff directory structure
In-Reply-To: <20001030.101136.124290560.wl@gnu.org>
Message-ID: <Pine.SUN.3.91.1001030192753.15587H-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1768
Lines: 37

On Mon, 30 Oct 2000, Werner LEMBERG wrote:

> > > Maybe groff (the program) shall check how it is called.  If arg[0] is,
> > > say, groff-1.16.1, it takes the `-1.16.1' postfix and appends it to
> > > all programs it is calling, i.e., pic-1.16.1, grotty-1.16.1, etc.
> > 
> > I do NOT like the idea of groff checking how it is called. For one
> > thing, it complicates groff for unnecessary reasons.

Actually, modifying the program's behavior depending on argv[0] is 
against the current GNU coding standards.  I think standards.texi in its 
latest versions mentions this.

> For MSDOS (and Windows?) it might be necessary to replace the scripts
> with small binaries since batch files don't support pipes, AFAIK.

This depends on the implementation of the Groff pipeline and of the
library functions that are called when the pipeline is run by the groff 
driver.  For example, the DJGPP port of Groff will have no 
redirection-related problems with batch files, because it calls the shell 
to run each batch in sequence, and redirects the standard handles of the 
shell.

There is one problem with batch files, though: that they don't take a 
variable number of arguments too well.  So my vote would be against 
scripts and batch files.

> What about other packages with a couple of versioned binaries

GCC uses version-specific directories to find its subprograms (cc1, 
cc1plus, cpp, etc.).  So you have, say, gcc-2.7.2, gcc-2.8.1, and 
gcc-2.95.2 as 3 executables (one of them is typically hard-linked to 
gcc), and each one knows where to look for its subprograms.

However, there's a difference between GCC and Groff: GCC users never 
invoke cc1 or cc1plus from the command line.  So they don't need to be on 
your $PATH.  With Groff, this won't work, I think.

From groff-admin  Mon Oct 30 18:39:27 2000
Received: from mailgw1.netvision.net.il (mailgw1.netvision.net.il [194.90.1.14])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA12396
	for <groff@ffii.org>; Mon, 30 Oct 2000 18:39:25 +0100
Received: from is.elta.co.il (is.elta.co.il [199.203.121.2])
	by mailgw1.netvision.net.il (8.9.3/8.9.3) with ESMTP id TAA28712;
	Mon, 30 Oct 2000 19:45:23 +0200 (IST)
Received: from is (is [199.203.121.2])
	by is.elta.co.il (8.9.3/8.8.8) with SMTP id TAA15969;
	Mon, 30 Oct 2000 19:44:13 +0200 (IST)
Date: Mon, 30 Oct 2000 19:44:13 +0200 (IST)
From: Eli Zaretskii <eliz@is.elta.co.il>
X-Sender: eliz@is
To: Freudenberger Joerg <Joerg.Freudenberger@nbg9.siemens.de>
cc: "'Werner LEMBERG'" <wl@gnu.org>, Ted.Harding@nessie.mcc.ac.uk,
        groff@ffii.org
Subject: RE: [Groff] new groff directory structure
In-Reply-To: <11DD9734B036D311947400805F0DD607BD8AE4@NBGV101A>
Message-ID: <Pine.SUN.3.91.1001030194326.15587I@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 199
Lines: 5

On Mon, 30 Oct 2000, Freudenberger Joerg wrote:

> gcc has a sortiment of wrapper binaries ( cc, c++, f77, gcc, g++, g77 ... )

Actually, gcc, g++, g77 and others are much more than simple wrappers.

From groff-admin  Tue Oct 31 00:12:14 2000
Received: from unet.univie.ac.at (unet.univie.ac.at [131.130.230.7])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id AAA17387
	for <groff@ffii.org>; Tue, 31 Oct 2000 00:12:13 +0100
Received: from rigel.univie.ac.at (root@unet2-118.univie.ac.at [131.130.232.118])
	by unet.univie.ac.at (8.9.3/8.9.3) with ESMTP id AAA04458;
	Tue, 31 Oct 2000 00:18:12 +0100
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id OAA03018;
	Mon, 30 Oct 2000 14:38:16 +0100
Date: Mon, 30 Oct 2000 14:38:16 +0100 (CET)
Message-Id: <20001030.143816.08315792.wl@gnu.org>
To: stewart@ref.collins.co.uk
Cc: groff@ffii.org
Subject: Re: [Groff] new groff directory structure
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <39FD596B.23461942@ref.collins.co.uk>
References: <39FD596B.23461942@ref.collins.co.uk>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 311
Lines: 9

> /usr/local/libexec clashes with /usr/local/lib for those of us who
> rely on file completion and typeahead to get things done. How
> difficult would it have been to choose a unique directory prefix?

I fully agree, but it is too late :-( `libexec' is in use for years,
mainly for Emacs and Fvwm.


    Werner

From groff-admin  Tue Oct 31 00:12:34 2000
Received: from unet.univie.ac.at (unet.univie.ac.at [131.130.230.7])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id AAA17404
	for <groff@ffii.org>; Tue, 31 Oct 2000 00:12:33 +0100
Received: from rigel.univie.ac.at (root@unet2-118.univie.ac.at [131.130.232.118])
	by unet.univie.ac.at (8.9.3/8.9.3) with ESMTP id AAA95372;
	Tue, 31 Oct 2000 00:18:14 +0100
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id OAA03021;
	Mon, 30 Oct 2000 14:40:19 +0100
Date: Mon, 30 Oct 2000 14:40:19 +0100 (CET)
Message-Id: <20001030.144019.112284620.wl@gnu.org>
To: Joerg.Freudenberger@nbg9.siemens.de
Cc: Ted.Harding@nessie.mcc.ac.uk, groff@ffii.org, eliz@is.elta.co.il
Subject: Re: [Groff] new groff directory structure
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <11DD9734B036D311947400805F0DD607BD8AE4@NBGV101A>
References: <11DD9734B036D311947400805F0DD607BD8AE4@NBGV101A>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 540
Lines: 16

> > What about other packages with a couple of versioned binaries
> > (contrary to Emacs and gcc which basically only have a single
> > binary)?
> 
> gcc has a sortiment of wrapper binaries ( cc, c++, f77, gcc, g++,
> g77 ... ) which are usualy in /bin. (configure with PREFIX=/ !!)
> The real work is done by a bundle of binaries in
> /lib/gcc-lib/<arch>/<version>.

But usually you use just one program and not, say, five in a pipe!  It
is horribly to say the hyphothetical command

  gcc -V 2.95.2 xxx | cpp -V 2.95.2 | ...


    Werner

From groff-admin  Tue Oct 31 09:25:03 2000
Received: from unet.univie.ac.at (unet.univie.ac.at [131.130.230.7])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA05233
	for <groff@ffii.org>; Tue, 31 Oct 2000 09:25:03 +0100
Received: from rigel.univie.ac.at (root@uvo1-52.univie.ac.at [131.130.231.52])
	by unet.univie.ac.at (8.9.3/8.9.3) with ESMTP id JAA63510;
	Tue, 31 Oct 2000 09:31:03 +0100
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id BAA03558;
	Tue, 31 Oct 2000 01:46:52 +0100
Date: Tue, 31 Oct 2000 01:46:52 +0100 (CET)
Message-Id: <20001031.014652.45748905.wl@gnu.org>
To: eliz@is.elta.co.il
Cc: Ted.Harding@nessie.mcc.ac.uk, groff@ffii.org
Subject: Re: [Groff] new groff directory structure
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <Pine.SUN.3.91.1001030192753.15587H-100000@is>
References: <20001030.101136.124290560.wl@gnu.org>
	<Pine.SUN.3.91.1001030192753.15587H-100000@is>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1894
Lines: 45

> > > > Maybe groff (the program) shall check how it is called.  If
> > > > arg[0] is, say, groff-1.16.1, it takes the `-1.16.1' postfix
> > > > and appends it to all programs it is calling, i.e.,
> > > > pic-1.16.1, grotty-1.16.1, etc.
> > > 
> > > I do NOT like the idea of groff checking how it is called. For
> > > one thing, it complicates groff for unnecessary reasons.
> 
> Actually, modifying the program's behavior depending on argv[0] is
> against the current GNU coding standards.  I think standards.texi in
> its latest versions mentions this.

Thanks for the info.  So my suggestion is bad.

> > For MSDOS (and Windows?) it might be necessary to replace the
> > scripts with small binaries since batch files don't support pipes,
> > AFAIK.
> 
> This depends on the implementation of the Groff pipeline and of the
> library functions that are called when the pipeline is run by the
> groff driver.  For example, the DJGPP port of Groff will have no
> redirection-related problems with batch files, because it calls the
> shell to run each batch in sequence, and redirects the standard
> handles of the shell.
> 
> There is one problem with batch files, though: that they don't take
> a variable number of arguments too well.  So my vote would be
> against scripts and batch files.

OK.  The planned wrapper programs will be extremely simple: Just
evaluating the GROFF_VERSION environment variable, building a path
with it, and calling a binary using that path.

BTW, do you have code for a similar wrapper program which works on
MS-DOS also?  I'm not too familiar with pipes...

> However, there's a difference between GCC and Groff: GCC users never
> invoke cc1 or cc1plus from the command line.  So they don't need to
> be on your $PATH.  With Groff, this won't work, I think.

So I want to have GROFF_VERSION.  If not set, use a default value
given at compile time.


    Werner

From groff-admin  Tue Oct 31 09:39:06 2000
Received: from mailgw3.netvision.net.il (mailgw3.netvision.net.il [194.90.1.11])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id JAA05288
	for <groff@ffii.org>; Tue, 31 Oct 2000 09:38:59 +0100
Received: from is.elta.co.il (is.elta.co.il [199.203.121.2])
	by mailgw3.netvision.net.il (8.9.3/8.9.3) with ESMTP id KAA14596;
	Tue, 31 Oct 2000 10:43:58 +0200 (IST)
Received: from is (is [199.203.121.2])
	by is.elta.co.il (8.9.3/8.8.8) with SMTP id KAA24748;
	Tue, 31 Oct 2000 10:43:43 +0200 (IST)
Date: Tue, 31 Oct 2000 10:43:43 +0200 (IST)
From: Eli Zaretskii <eliz@is.elta.co.il>
X-Sender: eliz@is
To: Werner LEMBERG <wl@gnu.org>
cc: Ted.Harding@nessie.mcc.ac.uk, groff@ffii.org
Subject: Re: [Groff] new groff directory structure
In-Reply-To: <20001031.014652.45748905.wl@gnu.org>
Message-ID: <Pine.SUN.3.91.1001031103137.24533H@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 1178
Lines: 30

On Tue, 31 Oct 2000, Werner LEMBERG wrote:

> OK.  The planned wrapper programs will be extremely simple: Just
> evaluating the GROFF_VERSION environment variable, building a path
> with it, and calling a binary using that path.

This will work, but I think it would be a good idea to have this indirect 
invocation be a configure-time option.  Running an extra program wastes 
system resources, which on some systems might be significant.  OTOH, I 
expect this feature to be rarely used, mainly by developers who want to 
have more than a single version operable at the same tyime.  IMHO we 
should not punish the majority for the needs of the few.

> BTW, do you have code for a similar wrapper program which works on
> MS-DOS also?  I'm not too familiar with pipes...

I'm not sure I understand what pipes have got to do with this.  Isn't 
the following enough for the wrapper to do what you want?

	char *version = getenv("GROFF_VERSION");

	sprintf (cmdbuf, "%s%s/%s %s", "/usr/local/lib/",
		 version ? version : GROFF_VERSION,
		 program_name,
		 rest_arguments);

	return system (cmdbuf);

(assuming rest_arguments holds all of the argv[1]...argv[argc-1] as a 
string.)

From groff-admin  Tue Oct 31 12:22:43 2000
Received: from unet.univie.ac.at (unet.univie.ac.at [131.130.230.7])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id MAA08496
	for <groff@ffii.org>; Tue, 31 Oct 2000 12:22:43 +0100
Received: from rigel.univie.ac.at (root@unet3-165.univie.ac.at [131.130.232.165])
	by unet.univie.ac.at (8.9.3/8.9.3) with ESMTP id MAA42714;
	Tue, 31 Oct 2000 12:28:45 +0100
Received: from localhost (sx0005@localhost [127.0.0.1])
	by rigel.univie.ac.at (8.9.3/8.8.8) with ESMTP id MAA04045;
	Tue, 31 Oct 2000 12:30:18 +0100
Date: Tue, 31 Oct 2000 12:30:18 +0100 (CET)
Message-Id: <20001031.123018.35659655.wl@gnu.org>
To: eliz@is.elta.co.il
Cc: Ted.Harding@nessie.mcc.ac.uk, groff@ffii.org
Subject: Re: [Groff] new groff directory structure
From: Werner LEMBERG <wl@gnu.org>
In-Reply-To: <Pine.SUN.3.91.1001031103137.24533H@is>
References: <20001031.014652.45748905.wl@gnu.org>
	<Pine.SUN.3.91.1001031103137.24533H@is>
X-Mailer: Mew version 1.95b54 on Emacs 20.6 / Mule 4.0 (HANANOEN)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 658
Lines: 16

> > OK.  The planned wrapper programs will be extremely simple: Just
> > evaluating the GROFF_VERSION environment variable, building a path
> > with it, and calling a binary using that path.
> 
> This will work, but I think it would be a good idea to have this
> indirect invocation be a configure-time option.  Running an extra
> program wastes system resources, which on some systems might be
> significant.

What about exec* (resp. spawn*)?  I thought that these functions
simply overlay the parent process (we don't need to return), avoiding
such a waste.  The only penalty would be minimal slower execution.
Please correct me if I'm wrong.


    Werner

From groff-admin  Tue Oct 31 12:47:45 2000
Received: from mailgw1.netvision.net.il (mailgw1.netvision.net.il [194.90.1.14])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id MAA08721
	for <groff@ffii.org>; Tue, 31 Oct 2000 12:47:43 +0100
Received: from is.elta.co.il (is.elta.co.il [199.203.121.2])
	by mailgw1.netvision.net.il (8.9.3/8.9.3) with ESMTP id NAA16727;
	Tue, 31 Oct 2000 13:53:46 +0200 (IST)
Received: from is (is [199.203.121.2])
	by is.elta.co.il (8.9.3/8.8.8) with SMTP id NAA26265;
	Tue, 31 Oct 2000 13:52:36 +0200 (IST)
Date: Tue, 31 Oct 2000 13:52:36 +0200 (IST)
From: Eli Zaretskii <eliz@is.elta.co.il>
X-Sender: eliz@is
To: Werner LEMBERG <wl@gnu.org>
cc: Ted.Harding@nessie.mcc.ac.uk, groff@ffii.org
Subject: Re: [Groff] new groff directory structure
In-Reply-To: <20001031.123018.35659655.wl@gnu.org>
Message-ID: <Pine.SUN.3.91.1001031134857.26066C-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 855
Lines: 18

On Tue, 31 Oct 2000, Werner LEMBERG wrote:

> > This will work, but I think it would be a good idea to have this
> > indirect invocation be a configure-time option.  Running an extra
> > program wastes system resources, which on some systems might be
> > significant.
> 
> What about exec* (resp. spawn*)?  I thought that these functions
> simply overlay the parent process

Not on DOS/Windows: spawn* don't overlay the current process (they aren't 
supposed to, by design), and exec* are implemented on several systems as 
a call to spawn* that doesn't return to the caller.

Am I wrong in assuming that this feature is not normally required by an 
average user (on all systems, not only on DOS/Windows)?  If I'm not 
wrong, why not make it an optional behavior?  I think TeX/Web2c doesn't 
support such multi-version installations, to take one example.

From groff-admin  Tue Oct 31 13:17:20 2000
Received: from mail1.svr.pol.co.uk (mail1.svr.pol.co.uk [195.92.193.18])
	by genba.ffii.org (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id NAA08912
	for <groff@ffii.org>; Tue, 31 Oct 2000 13:17:20 +0100
Received: from [195.92.198.123] (helo=mail17.svr.pol.co.uk)
	by mail1.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13qaRp-0000pJ-00
	for groff@ffii.org; Tue, 31 Oct 2000 12:23:25 +0000
Received: from modem-50.aldaron.dialup.pol.co.uk ([62.136.101.178] helo=inputplus.demon.co.uk)
	by mail17.svr.pol.co.uk with esmtp (Exim 3.13 #0)
	id 13qaRo-0002b4-00
	for groff@ffii.org; Tue, 31 Oct 2000 12:23:24 +0000
Received: from inputplus (ralph@localhost [127.0.0.1])
	by inputplus.demon.co.uk (8.8.7/8.8.7) with ESMTP id MAA07610
	for <groff@ffii.org>; Tue, 31 Oct 2000 12:23:22 GMT
Message-Id: <200010311223.MAA07610@inputplus.demon.co.uk>
To: groff@ffii.org
Subject: Re: [Groff] new groff directory structure 
In-Reply-To: Message from Eli Zaretskii <eliz@is.elta.co.il> 
   of "Tue, 31 Oct 2000 13:52:36 +0200." <Pine.SUN.3.91.1001031134857.26066C-100000@is> 
Date: Tue, 31 Oct 2000 12:23:22 +0000
From: Ralph Corderoy <ralph@inputplus.demon.co.uk>
Sender: groff-admin@ffii.org
Errors-To: groff-admin@ffii.org
X-BeenThere: groff@ffii.org
X-Mailman-Version: 2.0beta2.local2
Precedence: bulk
List-Id: GNU troff discussion list <groff.ffii.org>
Status: O
Content-Length: 795
Lines: 24

Hi,

> Am I wrong in assuming that this feature is not normally required by
> an average user (on all systems, not only on DOS/Windows)?  If I'm
> not wrong, why not make it an optional behavior?  I think TeX/Web2c
> doesn't support such multi-version installations, to take one
> example.

Nor do other multi-program packages, like GhostScript.

If I wanted to have multiple versions installed and switch between them
then I'd arrange for separate installation directories and then use
PATH to switch between them.  I'd expect them to follow PATH by
default.

I could do this either with a simple wrapper script, or by making it
easy to edit my PATH to contain the required directory.

Having every package provide this facility in differing ways doesn't
seem the `right thing' IMHO.


Ralph.


