Re: [Groff] Re: preconv supported encodings

[Pedro A . López-Valencia]

On 1/3/06, D. E. Evans wrote:
>Bruno Haible wrote:
>
>This list contains no CPxxx encodings, in particular no WINDOWS-
>encodings.  Microsoft continues to extend these encodings over
>and over again, with the result that, say, a text written today
>in CP950 on a Windows-XP machine is not readable as CP950 on an
>earlier version of the same OS. For this reason, the use of these
>encodings for manpages would be suboptimal.
>
> Windows is on Unicode now, anyways.  Stick with Unicode, and
> ignore the rest of the Windows encodings.

This is unrealistic to say the least. The fact that MS Windows NT uses
UCS2-LE for its  internal representation of text and in the NTFS
filesystem filename tables, (resorting to ugly hacks to let you
believe you can actually hande files with a CJK name without trouble
in an Western European setup untill you actually try it, believe me I
have), has absolutely nothing to do with end user working
environments.

MS uses CP encodings for end-user interaction and that won't be
changing in the forseeable future, AFAIK. If you can point me to the
MSDN article that proves me wrong, I'll galdly eat my hat ;-).

--
Pedro A. López-Valencia


___
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/listinfo/groff

Re: [Groff] mapping of glyphs to Unicode

[Pedro A . López-Valencia]

On 2/15/06, Michail Vidiassov <[EMAIL PROTECTED]> wrote:

>
> Oh woe is me! I live under Mac OS X or Windows XP,
>   bdf fonts are off-limits to me :(
> Maybe I have to use FontForge to convert bdf to bitmap truetype.
>
> Anyway, widely using a glyph present in just 1 font family seems risky.
>
> Can you list more fonts?
>

May be these truetype embeded bitmap fonts can help?
http://sourceforge.jp/projects/mplus-fonts/.

--
Pedro A. López-Valencia
Better to remain silent and let people believe you are a fool than
open your mouth and show to the world you are one.


___
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/listinfo/groff

Re: [Groff] Re: Bug in mm macro package

[Pedro A. López-Valencia]

On 2/18/06, Werner LEMBERG <[EMAIL PROTECTED]> wrote:

[snip]
>
> This is not correct.  The example DESC file for a PostScript printer
> given in the Troff User's Manual (CSTR 54), section 23, is as follows.
>
>   position 1 = R
>   position 2 = I
>   position 3 = B
>   position 4 = BI
>   ...
>   position 10 = S
>
> So we already have a good reason to use font names instead of font
> positions: Using `S' instead of `BI' on position 4 gives, well, funny
> results.
>
> I propose to install a modified .fp in the mm macros which calls .ftr
> (to translate the font name) if the user specifies a font position in
> the range 1-4.
>
>   .als fp-old fp

[snip]

>
> Comments?

I agree with you; we need the flexibility to be able to puch groff
into the document processing backend (where it is not presently) on
the one hand, and we should not hold on to the past so tightly that we
stultify the posibility of evolution on the other.

The use of numbers for font positions in troff is a historical
artifact due to the fact that it was cause it was written to drive the
Wang C/A/T phototypesetter exclusively, which only had four hardwired
font positions, 0->R, 1->I, 2->B, 4->S for four different typeface
*tapes*. That is no longer the case, the C/A/Ts still in existence, if
there is any, are used as doorstops and Wang is long gone and buried.
Later ditroff versions inherited the limitations, but only because it
would have meant a full rewrite.

--
Pedro A. López-Valencia


___
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/listinfo/groff

Fwd: [Groff] Re: groff + mp

[Pedro A. López-Valencia]

On 2/21/06, Michail Vidiassov wrote:

>
> On Tue, 14 Feb 2006, Werner LEMBERG wrote:
>
>> Michail Vidiassov wrote:

> >> BTW.  I think the latter support was broken netertheless.  Can you
> >> check?)
> >
> > I can't.  I don't have an old troff running.
>
> What is the real state of the troff land?
> What are the other players beyond groff?

There is two remaining versions of troff avaialble, afaik. Sun
Microsystems ships an incomplete troff based in licensed code from DWB
3.2; t doesn't include pic. There is another troff, the one in Plan9,
which has been ported to POSIX OSs as part of the "Plan 9 from User
Space" project (). It is a derivative of
DWB 3.4 and it uses UTF-8 text representation internally. Although it
is a source distribution, still belongs to Lucent and uses a license
approved by the OSI. (Yet it is a pain in the neck to use.)


___
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/listinfo/groff

Re: [Groff] Possible zsh contention in pdfroff (Re: [autoconf] ensure a VPATH build)

[Pedro A. López-Valencia]

On 2/21/06, Keith MARSHALL <[EMAIL PROTECTED]> wrote:

> Given the complete absence of related of bug reports, either:
>
> 1) This isn't true (which seems unlikely), or (more likely)
>
> 2) No one has attempted to build groff >= 1.19.2, or to use pdfroff,
>on any system where /bin/sh is linked to zsh.
>
> Can anyone either confirm or refute the hypothesis that there may be
> a zsh related contention in pdfroff?

I haven't used zsh as my default shell in a fair while, but I can
confirm that I did have autoconf hanging under zsh with pre 1.19.1 CVS
snapshots under Cygwin (using Cygwin's pty emulation, btw).


___
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/listinfo/groff

Re: [Groff] unicode support - where to compose?

[Pedro A. López-Valencia]

On 2/25/06, Larry Kollar wrote:
>
> How much work would it be to hook into whatever system-wide printer
> drivers there may be?

A lot.

> This is something I've been pondering for a
> while, and I don't even have a clue how gtroff figures out glyph
> widths (is there some communication between gtroff & the
> postprocessor?). It would be nice to have access to all the TrueType
> and PostScript fonts installed on the system without having to build
> metric files and so forth.
>
> If I were to poke at this, where would I start?

I've given a lot of thought to this as well, but I'm not the coder
kind ;-) There is already a possible path: fontconfig and freetype2. I
wonder if they could be used as backend for a utility that extracted
all the metrics information for a truetype font and dumped a properly
formed AFM. I know that can be done with freetype1 (as in ttftoafm and
ttftot42) but that library is obsolete...


___
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/listinfo/groff

Re: [Groff] Adding fonts to groff -- instructions?

[Pedro A. López-Valencia]

On 3/13/06, Robert Marks wrote:
> Ted suggests adding the font to /font/devps/DESC (see below).
> What is the purpose of this?
> I haven't done it and things seem to work, but perhaps I'm missing something.
>

PS printers and software RIPs (namely, ghostscript as it is the best
know and more widespread) have a limited repertoire of known fonts,
the (in)famous "LaserWriter 35), be it in ROM or disk. Everything else
must be downloaded; be it to the printer disk, RIP library directories
and configured accordingly; or downloaded with the print job. The DESC
file makes sure the latter happens with groff output.


___
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/listinfo/groff

Re: [Groff] Why does tty-char.tmac not represent meaning?

[Pedro A . López-Valencia]

  3. The Spanish inverted exclamation and question marks are
 represented by the same characters in upright position.
 While that will probably look strange to Spanish readers,
 i can think of no better way, and i think everybody will
 understand it.  The current "i" and "c" is unintelligible.

Yes, as a native Spanish speaker, I agree. Nice application of the principle of 
least surprise, I should add.
 

Re: [Groff] Bitstream vera sans

[Pedro A . López-Valencia]

On 9/08/2017 9:11 a.m., John Gardner wrote:


[too long]


You've e-mailed me directly, so the rest of the mailing list won't 
hear you (assuming you haven't BCCed Mkkel in your reply). I've added 
the list's address for you. =)




Do accept my apologies for the direct message. I hadn't my email client 
handy and used the web interface. I'll stick to my guns on my little 
font rant: Please don't recommend font editors even to extract metrics 
information as the sole act of opening the font will cause enough 
rounding errors to give you typesetting errors later on. One should 
recommend best practices to a beginner. :-)


The command line tools mentioned have a very long track of correctness 
and accuracy, that fontforge is barely starting to have lately (I've 
used it since it came out). TBH, these days the only font editor I would 
recommend is RoboFont, very commercial, expensive and macos only. But 
the core elements are free and libre (not GNU, though): fonttools, 
RoboFab, defcon and ufo2fdk; all in github as is the vogue. There is a 
tool based on these libraries, all written in python, that dices and 
slices a font to extract the exact AFM that corresponds to the binary 
blob. Unfortunately it is part of RoboFab, as far as I recall. I've been 
meaning to write a free replacement but I've been side-tracked for too long.


On CrossFont, I've never felt the need to use more than the trial 
version TBH. And yes, before version 5 it was horrible.