July 2004 - ntg-pdftex - NTG Mailing Lists

Re: backslashes in \pdfoutline
by Martin Schröder 14 Jul '04

14 Jul '04

On 2004-07-14 10:01:03 -0400, Karl Berry wrote: > I've come to believe that the status should remain quo, and all that's > desirable is to add the warning and fix the sorting. Spending time > implementing autoescape for the benefit of some hypothetical future > macro packages seems like a very low priority feature. Agreed. And even the warning and sorting won't go in 1.20a. Best regards Martin PS: Please let's keep ntg-pdftex at least in the cc. -- Martin Schröder, ms(a)artcom-gmbh.de ArtCom GmbH, Lise-Meitner-Str 5, 28359 Bremen, Germany Voice +49 421 20419-44 / Fax +49 421 20419-10 http://www.artcom-gmbh.de

1 0

Re: backslashes in \pdfoutline
by Hans Hagen 14 Jul '04

14 Jul '04

Karl Berry wrote: > pdftex pipes destinations names through unchanged > >Yes, and many other things as well (having just read through the new >pdftex manual, as Hartmut knows :). Trying to escape all of them seems >more problematic to me now. There are probably more important things >for pdftex development to work on. > > detecting forbidden things: unbalanced unescaped braces and a lonely > backslash at the string end. > > it is dangerous to start adding such features esp since i do some cleanup already far better is to have two primitives \escapedpdf{....} \escapedps{....} \pdfsomething{...... \escapedpdf{....} ...} Hans

1 0

[tex-implementors] MetaTeX project pre-announcement
by Taco Hoekwater 14 Jul '04

14 Jul '04

Hi everybody, I am happy to announce the birth of a new project called MetaTeX. The goal of the MetaTeX project is to create a GPL-ed TeX/MetaPost distribution for POSIX compliant systems, in which the amount of distributed files is minimized while the speed of the TeX 'make' process maximized. This will be done by joining all of the needed functionality into a single executable, and putting all of the texmf input files into a single archive file, eventually resulting in a distribution that (in binary state) will comprise only three actual files: an executable, a data archive, and an options file. As a side-effect, standalone packages of C library-versions of pdfetex, mpost and tex{exec,make} will be created. I am posting this pre- announcement in the hope that people who are interested in this distribution (and, possibly, even want to help) will join our mailing-list. See the website for more information on the project, and how to subscribe. http://www.metatex.org Please do not reply to this message in the group you found it in, since any reply you give will likely be considered off-topic, but feel free to forward this message to people you believe may be interested. Greetings and happy TeX-ing, Taco Hoekwater _______________________________________________ tex-implementors mailing list postmaster(a)tug.org http://tug.org/mailman/listinfo/tex-implementors

1 0

MetaTeX project pre-announcement
by Taco Hoekwater 14 Jul '04

14 Jul '04

Hi everybody, I am happy to announce the birth of a new project called MetaTeX. The goal of the MetaTeX project is to create a GPL-ed TeX/MetaPost distribution for POSIX compliant systems, in which the amount of distributed files is minimized while the speed of the TeX 'make' process maximized. This will be done by joining all of the needed functionality into a single executable, and putting all of the texmf input files into a single archive file, eventually resulting in a distribution that (in binary state) will comprise only three actual files: an executable, a data archive, and an options file. As a side-effect, standalone packages of C library-versions of pdfetex, mpost and tex{exec,make} will be created. I am posting this pre- announcement in the hope that people who are interested in this distribution (and, possibly, even want to help) will join our mailing-list. See the website for more information on the project, and how to subscribe. http://www.metatex.org Please do not reply to this message in the group you found it in, since any reply you give will likely be considered off-topic, but feel free to forward this message to people you believe may be interested. Greetings and happy TeX-ing, Taco Hoekwater

1 0

Re: [NTG-pdftex] Re: pdftex: unrecognized option `-output-format=pdf'
by Akira Kakuto 14 Jul '04

14 Jul '04

> in texmfmp.c, just add the three marked lines: > > #endif /* MP */ > + #if defined(pdfTeX) || defined(pdfeTeX) || defined(pdfxTeX) > + { "output-format", 1, 0, 0 }, > + #endif /* pdfTeX or pdfeTeX or pdfxTeX */ > { 0, 0, 0, 0 } }; > > It overruns the initex setting, but it can be overrun by \pdfoutput (not > the other way around, as it may be practical for such an option?). I'm sorry I did not notice that --output-format was thrown out of getopt option list. I think the option is useful in some cases. I hope that the above three lines are revived. Best regards Akira

1 0

pdftex: unrecognized option `-output-format=pdf'
by Hartmut Henkel 14 Jul '04

14 Jul '04

Option -output-format is part of the -help menu, but it seems not to work in rc4/teTeX 2.96.5 (error before even the substring "dvi" or "pdf" is checked); it's at least missing from the getopt option list. Should it be pulled in or thrown out? Regards, Hartmut

1 1

[ pdftex-Patches-91 ] \tagcode patch
by noreply＠sarovar.org 13 Jul '04

13 Jul '04

Patches item #91, was opened at 2004-07-13 10:58 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=91&group_id=106 Category: None Group: None Status: Open Resolution: None Priority: 3 Submitted By: Martin Schröder (oneiros) Assigned to: The Thanh Han (hanthethanh) Summary: \tagcode patch Initial Comment: Patch from Taco Hoekwater for an extra primitive named \tagcode, allowing read and write access to a character's char_tag info. The associated internal command is assign_font_int, with the new modifier value tag_code. More info at http://www.metatex.org/web/tagcode.html ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=91&group_id=106

1 0

\tagcode patch
by Taco Hoekwater 13 Jul '04

13 Jul '04

Hi all, I have created a small patch file that I expect some of us will consider useful: It allows the querying and removal of ligature/kerning/charlist information from within the TeX macro language. See: http://www.metatex.org/web/tagcode.html for more information -- groeten, Taco

1 0

Re: [NTG-pdftex] pdftex -output-comment=STRING
by Akira Kakuto 12 Jul '04

12 Jul '04

> pdftex --help gives > > -output-comment=STRING use STRING for PDF file comment instead of date > > but the manpage says that "In DVI mode, use string for the DVI file > comment instead of the date. This option is ignored in PDF mode." > > And i couldn't find any trace of STRING in the PDF file, but it's in the > DVI file. So it seems this would be more correct: > > -output-comment=STRING use STRING for DVI file comment instead of date I send an example patch: Best regards Akira ------------------------------------------------ --- texmfmp-help.h.orig Mon Jun 07 19:39:40 2004 +++ texmfmp-help.h Sun May 09 12:53:26 2004 @@ -31,7 +31,7 @@ " most commonly `aleph'.", "", " Alternatively, if the first non-option argument begins with a backslash,", - " interpret all non-option arguments as a line of e-Omega input.", + " interpret all non-option arguments as a line of Aleph input.", "", " Alternatively, if the first non-option argument begins with a &, the", " next word is taken as the FMT to read, overriding all else. Any", @@ -360,7 +360,9 @@ " the bits of NUMBER", "[-no]-mktex=FMT disable/enable mktexFMT generation (FMT=tex/tfm)", "-mltex enable MLTeX extensions such as \\charsubdef", - "-output-comment=STRING use STRING for PDF file comment instead of date", + "-output-comment=STRING use STRING for DVI file comment instead of date", + " in the case of DVI output (no effect for PDF)", + "-output-format=FORMAT use FORMAT for job output; FORMAT is `dvi' or `pdf'", "[-no]-parse-first-line disable/enable parsing of the first line of the", " input file", "-progname=STRING set program (and fmt) name to STRING", @@ -416,7 +418,9 @@ " the bits of NUMBER", "[-no]-mktex=FMT disable/enable mktexFMT generation (FMT=tex/tfm)", "-mltex enable MLTeX extensions such as \\charsubdef", - "-output-comment=STRING use STRING for PDF file comment instead of date", + "-output-comment=STRING use STRING for DVI file comment instead of date", + " in the case of DVI output (no effect for PDF)", + "-output-format=FORMAT use FORMAT for job output; FORMAT is `dvi' or `pdf'", "[-no]-parse-first-line disable/enable parsing of the first line of the", " input file", "-progname=STRING set program (and fmt) name to STRING", @@ -443,10 +447,10 @@ " Any remaining COMMANDS are processed as pdfxTeX input, after TEXNAME is read.", " If the first line of TEXNAME is %&FMT, and FMT is an existing .fmt file,", " use it. Else use `NAME.fmt', where NAME is the program invocation name,", - " most commonly `pdfetex'.", + " most commonly `pdfxtex'.", "", " Alternatively, if the first non-option argument begins with a backslash,", - " interpret all non-option arguments as a line of pdfeTeX input.", + " interpret all non-option arguments as a line of pdfxTeX input.", "", " Alternatively, if the first non-option argument begins with a &, the", " next word is taken as the FMT to read, overriding all else. Any", @@ -472,7 +476,9 @@ " the bits of NUMBER", "[-no]-mktex=FMT disable/enable mktexFMT generation (FMT=tex/tfm)", "-mltex enable MLTeX extensions such as \\charsubdef", - "-output-comment=STRING use STRING for PDF file comment instead of date", + "-output-comment=STRING use STRING for DVI file comment instead of date", + " in the case of DVI output (no effect for PDF)", + "-output-format=FORMAT use FORMAT for job output; FORMAT is `dvi' or `pdf'", "[-no]-parse-first-line disable/enable parsing of the first line of the", " input file", "-progname=STRING set program (and fmt) name to STRING",

2 1

Re: backslashes in \pdfoutline
by Martin Schröder 11 Jul '04

11 Jul '04

On 2004-07-10 20:39:06 -0400, Karl Berry wrote: > Hi Martin, > > As you may have noticed, I reported an issue on the pdftex lists a few > days ago regarding catcode 12 backslashes in \pdfoutline et al. > Although Hartmut and Heiko and I have discussed it some, and Hartmut has > written some patches, I haven't seen any definitive decisions on what we > want to do, and I want to fix texinfo.tex accordingly. > > (It comes up in practice with Texinfo because in Texinfo, unlike most > (all?) other common formats, backslash is supposed to be a "normal" > character that can be used in cross-reference targets.) > > In short, suppose we have: > > \catcode`\/=0 /catcode`\\=12 > /pdfcompresslevel=0 > /pdfdest name {xyz\} hello from xyz /vfill/eject > /pdfoutline goto name{xyz\}{to xyz\} > > This produces an invalid PDF file, because the \ escapes the > end-of-string ) delimiter in the output: > > << /S /GoTo /D (xyz\) >> > ... > (to xyz\) > > The choices seem to be: > > 1) do nothing, and leave it up to the document's TeX code to double the > backslashes in all cases. This is the current situation, de facto. > > 2) have pdftex escape the backslashes in the target names only, so that > it would produce (xyz\\) and (to xyz\). > > 3) have pdftex escape the backslashes in the outline text only, so that > it would produce (xyz\) and (to xyz\\). > > 4) have pdftex escape the backslashes in all situations, so that > it would produce (xyz\\) and (to xyz\\). Karl, 4 would be the logical choice for pdf. > Escaping the backslashes implies that TeX code cannot produce PDF string > escapes. This does not seem like a serious limitation to me, since TeX > can write any byte value (and pdftex "should" escape as necessary into > \nnn octal notation in string constants -- I have not tested this). I will. > Not escaping the backslashes implies that it's possible to write TeX > documents which produce invalid PDF. This seems bad, however, if this > is the case with other pdftex commands also, the argument loses its force. We'll have to check this for consistency. Best regards Martin PS: I'm cc'ing the ntg-pdftex list so that Hans and Thanh and others are included. -- Martin Schröder, ms(a)artcom-gmbh.de ArtCom GmbH, Lise-Meitner-Str 5, 28359 Bremen, Germany Voice +49 421 20419-44 / Fax +49 421 20419-10 http://www.artcom-gmbh.de

1 0