- ntg-pdftex - NTG Mailing Lists

[ pdftex-Bugs-378 ] Math chars and cmbright fail with pdfeTeX, Version 3.141592-1.30.0-rc2-2.2 (Web2C 7.5.5)
by noreply＠sarovar.org 06 Jul '05

06 Jul '05

Bugs item #378, was opened at 2005-07-05 17:17 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=378&group_id=106 Category: Fonts Group: v1.21a Status: Open Resolution: None Priority: 5 Submitted By: Peter Dyballa (ballapete) Assigned to: Nobody (None) Summary: Math chars and cmbright fail with pdfeTeX, Version 3.141592-1.30.0-rc2-2.2 (Web2C 7.5.5) Initial Comment: That's the document: \documentclass[12pt]{article} \usepackage{cmbright} %\usepackage{eurofont} \usepackage[T1]{fontenc} %\usepackage{lmodern} \usepackage[?]{amsmath} \begin{document} This is a test document for testing purposes in cmbright. It's worth maybe 0,02\,\texteuro. \section{Introcution} Let us do some math! % \[ \forall x \in \mathbf{R}, x^2 \geq 0 \] % NOT ok \[ x , 0 \]%\in \mathbf{R} x^2 \geq % OK % \begin{equation} % \forall x \in \mathbf{R}, x^2 \geq 0 % NOT ok % \end{equation} % \begin{equation} % x = r^2 + \sum_{0}^{100} y^2 % NOT ok % \end{equation} \begin{equation} % {\partial\over\partial x} f(x,y) = {\partial\over\partial y} f(x,y) = 0 % OK! \frac{\partial}{\partial x} f(x,y) = \frac{\partial}{\partial y} f(x,y) = 0 % OK! \end{equation} \end{document} Failure message: .... (./cmbright.aux) (/usr/local/teTeX/share/texmf.tetex/tex/latex/ base/ts1cmr.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ts1cmbr.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ omlcmbrm.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ omscmbrs.fd) [1{/usr/local/teTeX/share/texmf.local/fonts/map/pdftex/updmap/ pdftex.map}] (./cmbright.aux) ) </var/tmp/texfonts/pk/ljfour/public/cmbright/ cmbr10.720pk>xputenv.c:79: failed assertion `old_item' Abort No PS Type 1 fonts, only MF based. ---------------------------------------------------------------------- >Comment By: Hartmut Henkel (hhenkel) Date: 2005-07-06 08:21 Message: Logged In: YES user_id=929 I also can't reproduce the bug; have sent Pete an xputenv.c with a few diagnostic printf's, which he has forwarded to Gerben (who will soon be in holidays...). Regards, Hartmut ---------------------------------------------------------------------- Comment By: Martin Schröder (oneiros) Date: 2005-07-06 03:08 Message: Logged In: YES user_id=421 Argl. The comment got lost. :-( Your example is without eurofont, yet your log shows that you use eurofont. Without eurofont the example works perfectly for me with 1.21a and 1.30.0-rc2 ... ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=378&group_id=106

1 0

[ pdftex-Bugs-378 ] Math chars and cmbright fail with pdfeTeX, Version 3.141592-1.30.0-rc2-2.2 (Web2C 7.5.5)
by noreply＠sarovar.org 06 Jul '05

06 Jul '05

Bugs item #378, was opened at 2005-07-05 17:17 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=378&group_id=106 >Category: Fonts >Group: v1.21a Status: Open Resolution: None Priority: 5 Submitted By: Peter Dyballa (ballapete) Assigned to: Nobody (None) Summary: Math chars and cmbright fail with pdfeTeX, Version 3.141592-1.30.0-rc2-2.2 (Web2C 7.5.5) Initial Comment: That's the document: \documentclass[12pt]{article} \usepackage{cmbright} %\usepackage{eurofont} \usepackage[T1]{fontenc} %\usepackage{lmodern} \usepackage[?]{amsmath} \begin{document} This is a test document for testing purposes in cmbright. It's worth maybe 0,02\,\texteuro. \section{Introcution} Let us do some math! % \[ \forall x \in \mathbf{R}, x^2 \geq 0 \] % NOT ok \[ x , 0 \]%\in \mathbf{R} x^2 \geq % OK % \begin{equation} % \forall x \in \mathbf{R}, x^2 \geq 0 % NOT ok % \end{equation} % \begin{equation} % x = r^2 + \sum_{0}^{100} y^2 % NOT ok % \end{equation} \begin{equation} % {\partial\over\partial x} f(x,y) = {\partial\over\partial y} f(x,y) = 0 % OK! \frac{\partial}{\partial x} f(x,y) = \frac{\partial}{\partial y} f(x,y) = 0 % OK! \end{equation} \end{document} Failure message: .... (./cmbright.aux) (/usr/local/teTeX/share/texmf.tetex/tex/latex/ base/ts1cmr.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ts1cmbr.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ omlcmbrm.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ omscmbrs.fd) [1{/usr/local/teTeX/share/texmf.local/fonts/map/pdftex/updmap/ pdftex.map}] (./cmbright.aux) ) </var/tmp/texfonts/pk/ljfour/public/cmbright/ cmbr10.720pk>xputenv.c:79: failed assertion `old_item' Abort No PS Type 1 fonts, only MF based. ---------------------------------------------------------------------- Comment By: Martin Schröder (oneiros) Date: 2005-07-06 03:08 Message: Logged In: YES user_id=421 Argl. The comment got lost. :-( Your example is without eurofont, yet your log shows that you use eurofont. Without eurofont the example works perfectly for me with 1.21a and 1.30.0-rc2 ... ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=378&group_id=106

1 0

[ pdftex-Bugs-378 ] Math chars and cmbright fail with pdfeTeX, Version 3.141592-1.30.0-rc2-2.2 (Web2C 7.5.5)
by noreply＠sarovar.org 06 Jul '05

06 Jul '05

Bugs item #378, was opened at 2005-07-05 17:17 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=378&group_id=106 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Peter Dyballa (ballapete) Assigned to: Nobody (None) Summary: Math chars and cmbright fail with pdfeTeX, Version 3.141592-1.30.0-rc2-2.2 (Web2C 7.5.5) Initial Comment: That's the document: \documentclass[12pt]{article} \usepackage{cmbright} %\usepackage{eurofont} \usepackage[T1]{fontenc} %\usepackage{lmodern} \usepackage[?]{amsmath} \begin{document} This is a test document for testing purposes in cmbright. It's worth maybe 0,02\,\texteuro. \section{Introcution} Let us do some math! % \[ \forall x \in \mathbf{R}, x^2 \geq 0 \] % NOT ok \[ x , 0 \]%\in \mathbf{R} x^2 \geq % OK % \begin{equation} % \forall x \in \mathbf{R}, x^2 \geq 0 % NOT ok % \end{equation} % \begin{equation} % x = r^2 + \sum_{0}^{100} y^2 % NOT ok % \end{equation} \begin{equation} % {\partial\over\partial x} f(x,y) = {\partial\over\partial y} f(x,y) = 0 % OK! \frac{\partial}{\partial x} f(x,y) = \frac{\partial}{\partial y} f(x,y) = 0 % OK! \end{equation} \end{document} Failure message: .... (./cmbright.aux) (/usr/local/teTeX/share/texmf.tetex/tex/latex/ base/ts1cmr.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ts1cmbr.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ omlcmbrm.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ omscmbrs.fd) [1{/usr/local/teTeX/share/texmf.local/fonts/map/pdftex/updmap/ pdftex.map}] (./cmbright.aux) ) </var/tmp/texfonts/pk/ljfour/public/cmbright/ cmbr10.720pk>xputenv.c:79: failed assertion `old_item' Abort No PS Type 1 fonts, only MF based. ---------------------------------------------------------------------- >Comment By: Martin Schröder (oneiros) Date: 2005-07-06 03:08 Message: Logged In: YES user_id=421 Argl. The comment got lost. :-( Your example is without eurofont, yet your log shows that you use eurofont. Without eurofont the example works perfectly for me with 1.21a and 1.30.0-rc2 ... ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=378&group_id=106

1 0

Re: Fwd: [NTG-pdftex] \ifprimitive and \ifinternal/\iffeature (design proposal)
by Taco Hoekwater 05 Jul '05

05 Jul '05

Bernd Raichle wrote: > Hi Taco, > > I'm not on the NTG pdfTeX list (Rolf has forwarded your mail to the > ExTeX Developers list), thus I am replying directly. It is up to you > to forward my mail to the list ... Added the ntg-pdftex list in CC (but I am not on extex-devel so this may get confusing). > > \ifprimitive <csname> > > > > True if the <csname> has the same meaning as it would in initex, > > (but gives an error if <csname> is currently undefined). > > Ok. What is the result for a non csname, incl. active characters? I > assume false. I was thinking about issuing an error message, but Hans has convinced me that returning false is more usable than an error. So it will be false, even though it is clearly a user error to ask for the primitive- ness of something that can not possibly be a primitive. > > \executeprimitive <csname> > > > > Executes the primitive that would be known as <csname> in initex, > > even if the <csname> is currently undefined or defined to mean > > something else. Raises an error if there is no matching internal > > command (of course). > > Is \executeprimitive expandable or non expandable? Or does it depend > on the "expandability" of <csname>? Either unexandable or dependant on the <csname>, but definately not fully expandable (mainly because of the can of worms in your example). > > These are quite straightforward, and I do not assume this needs > > much discussion. But the next ones are related to "if feature ...", > > and those are bit harder to define. > > > > I propose to make it possible to 'predefine' primitives in the > > executable. All program code is completely ready, except that the > > hash table entry is never created, making it impossible to use > > that primitive in initex without extra work. > > A similar effect would be possible if (pdf)TeX has something like name > spaces. The predefined primitives can then reside in the namespace > "internal" and can be made available by importing the symbols from the > internal to the current namespace. That is what Heiko hinted at as well, and here is how I responded: Heiko said: > It needs some thinking. > First thoughts: > * It looks like a first step to a real namespace implementation, > a (perhaps read-only) internal namespace, namespaces for extensions, > and user/package macros. Why not going further then? Taco replied: > TeX's current internal structure does not allow a clean solution, > so if we want true namespaces in TeX, we probably have to start > thinking about a successor to pdftex. Otherwise, we would need > to redesign the whole symbol table / token handling while trying > to remain 100% compatible. That sounds like a nightmare to me. > > These last 4 primitives I proposed are an attempt to come up with > something that works next month (as opposed to next year), but > you are right that it probably needs more thinking, anyway. Back to Bernd's current e-mail: > Mmmh, what is the result when existing csnames are "overwritten" by > \expose... or when the meaning of an internal csname is assigned to > another csname by \let and then obscured? In general, if a macro progammer wants to be irresponsible, I'm not going out of my way to stop them from doing so. However, some of the more blatant user errors are easy to catch and I therefore propose to make them illegal from the start: \exposeinternal does not accept already defined <csname>s and \obscureinternal always makes the <csname> undefined > Here are some test cases ... > > \def\internalcommand{Hi!} > \begingroup > \exposeinternal\internalcommand A totally new error message is issued. Possibly ! Currently defined control sequence. \internalcommand ->Hi! But in this case perhaps even a stronger restriction is needed: ! Not at top level. > \obscureinternal\internalcommand > \show\internalcommand *\show\internalcommand > \internalcommand=undefined. > \begingroup > \exposeinternal\internalcommand > \global\let\x=\internalcommand > \endgroup > \show\x *\show\x > \x=\internalcommand. This is why the prefix is "obscure" instead of "undefine" > \exposeinternal\internalcommand > \let\x=\internalcommand > \obscureinternal\internalcommand > \show\x > \x=\internalcommand. > \exposeinternal\internalcommand > \def\x{\internalcommand} > \obscureinternal\internalcommand > \show\x > \x=macro: ->\internalcommand. > \x ! Undefined control sequence. \x ->\internalcommand > > Best wishes, > -bernd

1 0

[ pdftex-Bugs-378 ] Math chars and cmbright fail with pdfeTeX, Version 3.141592-1.30.0-rc2-2.2 (Web2C 7.5.5)
by noreply＠sarovar.org 05 Jul '05

05 Jul '05

Bugs item #378, was opened at 2005-07-05 17:17 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=378&group_id=106 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Peter Dyballa (ballapete) Assigned to: Nobody (None) Summary: Math chars and cmbright fail with pdfeTeX, Version 3.141592-1.30.0-rc2-2.2 (Web2C 7.5.5) Initial Comment: That's the document: \documentclass[12pt]{article} \usepackage{cmbright} %\usepackage{eurofont} \usepackage[T1]{fontenc} %\usepackage{lmodern} \usepackage[?]{amsmath} \begin{document} This is a test document for testing purposes in cmbright. It's worth maybe 0,02\,\texteuro. \section{Introcution} Let us do some math! % \[ \forall x \in \mathbf{R}, x^2 \geq 0 \] % NOT ok \[ x , 0 \]%\in \mathbf{R} x^2 \geq % OK % \begin{equation} % \forall x \in \mathbf{R}, x^2 \geq 0 % NOT ok % \end{equation} % \begin{equation} % x = r^2 + \sum_{0}^{100} y^2 % NOT ok % \end{equation} \begin{equation} % {\partial\over\partial x} f(x,y) = {\partial\over\partial y} f(x,y) = 0 % OK! \frac{\partial}{\partial x} f(x,y) = \frac{\partial}{\partial y} f(x,y) = 0 % OK! \end{equation} \end{document} Failure message: .... (./cmbright.aux) (/usr/local/teTeX/share/texmf.tetex/tex/latex/ base/ts1cmr.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ts1cmbr.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ omlcmbrm.fd) (/usr/local/teTeX/share/texmf.tetex/tex/latex/cmbright/ omscmbrs.fd) [1{/usr/local/teTeX/share/texmf.local/fonts/map/pdftex/updmap/ pdftex.map}] (./cmbright.aux) ) </var/tmp/texfonts/pk/ljfour/public/cmbright/ cmbr10.720pk>xputenv.c:79: failed assertion `old_item' Abort No PS Type 1 fonts, only MF based. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=378&group_id=106

1 0

[ pdftex-Bugs-377 ] unqoute image filenames not only on windows
by noreply＠sarovar.org 05 Jul '05

05 Jul '05

Bugs item #377, was opened at 2005-07-02 21:43 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106 Category: Image inclusion Group: None Status: Open Resolution: None Priority: 5 Submitted By: Georg Baum (baum) Assigned to: Martin Schröder (oneiros) Summary: unqoute image filenames not only on windows Initial Comment: The file pdftex-1.30.0-rc1/src/texk/web2c/pdftexdir/writeimg.c contains in the function readfile() a code block to unqoute file names which is inside #ifdef WIN32 ... #endif. This code should be enabled on all platforms (or at least on linux), since quoted filenames (which are necessary if the name contains spaces) do not only work well on linux. Thanks, Georg ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-07-05 10:57 Message: Logged In: YES user_id=1276 Hello, > I manually copied and cited your message because I could > not find out how to do that correctly with the tracker > system :-( Also the tiny typein area and the order of postings is very annoying. Perhaps we should better continue the discussion on the ntg-pdftex mailing list. >> Now you want the limitation of windows that quotes >> cannot be used apply on the other systems? > > I want to trade in the use of quotes for the use of > spaces. > Since you can only have one (not really, see below) it is > from a user's point of view better to support spaces on > all > platforms and not spaces on windows and quotes on other > platforms. >From a user's point it is better to support all characters instead of randomly choosen ones. The quoting method makes things worse: * There is no method to detect, whether quoting is supported: \includegraphics{"abc def.png"} The graphics package does not know, whether the TeX compiler will look for {abc def.png}, {"abc def.png"}, or {"abc}. Now you cannot use *both* quotes *and* spaces. * A very long time, spaces are considered "fancy" in file names, it is asking for trouble, because they are not supported on DOS file systems. But nowadays, spaces become "normal" despite the trouble (also on windows systems). The difference in our problem descriptions seems to be that you want to have *spaces* as usable characters in filenames, my conclusion is that we must allow *any* character as usable. Otherwise we will have bug reports about quotes. * And if we have a solution with *all* characters, there will be no problems with different encodings including ucs-2 or ucs-4. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- Comment By: Georg Baum (baum) Date: 2005-07-04 18:55 Message: Logged In: YES user_id=2590 Hello, I manually copied and cited your message because I could not find out how to do that correctly with the tracker system :-( > I must disagree, "the same behaviour on all platforms" > can not mean to add the limitations of all platforms. No, that is not my intention. > Now you want the limitation of windows that quotes > cannot be used apply on the other systems? I want to trade in the use of quotes for the use of spaces. Since you can only have one (not really, see below) it is from a user's point of view better to support spaces on all platforms and not spaces on windows and quotes on other platforms. > There are operating systems that deal with 8+3 file > names only. Do you want to forbid long file names, > for example? Of course not. > Thus we are loosing both quotes and spaces. Spaces still > cannot be used in general. Other TeX engines > does not know about quoting. There is no secure way > to know, whether quoting is working. Thus code that > is intended for different systems must avoid spaces. And quotes (in the filename itself, not for quoting), so this is not really an argument to support them. Concerning other implementations: \includegraphics{foo.png} does not work with plain latex, but with pdflatex. The same would apply to spaces in filenames, so I see no big problem there. > And the current quoting scheme does not solve the > problem at all, it moves the problem from spaces to > quotes, spaces can still not be used and quotes are > now made unavailable. Also the quoting scheme does > not allow quoting of the quotes themselves. It does (although I am not sure whether that is intended): Only the first quote is stripped from two consecutive quotes, e.g. the string '"a""bc"' is transformed into 'a"bc'. > A solution would allow *all* characters. > The solution should be able to deal with utf-8 or > even ucs-2/4 encoded file names. Agreed. If you think that the quoting is so bad then why not remove it altogether and tell users to use \pdfximage? My problem with the current implementation is the inconsistency: If filenames with spaces work on windows if quoted, I expect that they work on linux, too, since the OS supports them. Of course I can't expect the same from filenames with quotes: They are not supported by windows, so they can't be supported by pdftex on windows. I hope I have made my point clear. I can live with it if you still don't agree, so I'll stop arguing now. Georg ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-07-04 16:07 Message: Logged In: YES user_id=1276 Hello, I must disagree, "the same behaviour on all platforms" can not mean to add the limitations of all platforms. Now you want the limitation of windows that quotes cannot be used apply on the other systems? There are operating systems that deal with 8+3 file names only. Do you want to forbid long file names, for example? Thus we are loosing both quotes and spaces. Spaces still cannot be used in general. Other TeX engines does not know about quoting. There is no secure way to know, whether quoting is working. Thus code that is intended for different systems must avoid spaces. And the current quoting scheme does not solve the problem at all, it moves the problem from spaces to quotes, spaces can still not be used and quotes are now made unavailable. Also the quoting scheme does not allow quoting of the quotes themselves. A solution would allow *all* characters. The solution should be able to deal with utf-8 or even ucs-2/4 encoded file names. That is already given by the syntax of \pdfximage. Thus new primitives \pdfinput, \pdfopenin, \pdfopenout with <general text> would allow any file name and a macro code can test, whether this feature is available by testing the command names. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- Comment By: Georg Baum (baum) Date: 2005-07-04 13:46 Message: Logged In: YES user_id=2590 I agree that the quote stripping is a hack and should eventually be replaced by something better. I do think however that the behaviour on all platforms should be the same if it is at all possible. Right now, filenames with quotes work on linux and not on windows, and quoted filenames with spaces work on windows but not on linux. Thanks for the hint to pdfximage, I did not know about that (coming from standard LaTeX and always using \includegraphics} Georg ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-07-04 13:13 Message: Logged In: YES user_id=1276 Hello, it is a very bad idea, quotes are perfectly legal in linux and other operating systems. Also image inclusion is done by \pdfximage that uses <general text>, by <general text>, text surrounded by curly braces (catcode 1 and 2) *all* possible file names can be given including quotes and spaces. We need rather new primitives for \input, \openin, and \openout to provide the file names as <general text>. The all file names can be given without the need of quoting at all. See also the discussion about \pdfinput in the ntg-pdftex mailing list. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106

1 0

[ pdftex-Bugs-377 ] unqoute image filenames not only on windows
by noreply＠sarovar.org 04 Jul '05

04 Jul '05

Bugs item #377, was opened at 2005-07-02 21:43 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106 Category: Image inclusion Group: None Status: Open Resolution: None Priority: 5 Submitted By: Georg Baum (baum) Assigned to: Martin Schröder (oneiros) Summary: unqoute image filenames not only on windows Initial Comment: The file pdftex-1.30.0-rc1/src/texk/web2c/pdftexdir/writeimg.c contains in the function readfile() a code block to unqoute file names which is inside #ifdef WIN32 ... #endif. This code should be enabled on all platforms (or at least on linux), since quoted filenames (which are necessary if the name contains spaces) do not only work well on linux. Thanks, Georg ---------------------------------------------------------------------- >Comment By: Georg Baum (baum) Date: 2005-07-04 18:55 Message: Logged In: YES user_id=2590 Hello, I manually copied and cited your message because I could not find out how to do that correctly with the tracker system :-( > I must disagree, "the same behaviour on all platforms" > can not mean to add the limitations of all platforms. No, that is not my intention. > Now you want the limitation of windows that quotes > cannot be used apply on the other systems? I want to trade in the use of quotes for the use of spaces. Since you can only have one (not really, see below) it is from a user's point of view better to support spaces on all platforms and not spaces on windows and quotes on other platforms. > There are operating systems that deal with 8+3 file > names only. Do you want to forbid long file names, > for example? Of course not. > Thus we are loosing both quotes and spaces. Spaces still > cannot be used in general. Other TeX engines > does not know about quoting. There is no secure way > to know, whether quoting is working. Thus code that > is intended for different systems must avoid spaces. And quotes (in the filename itself, not for quoting), so this is not really an argument to support them. Concerning other implementations: \includegraphics{foo.png} does not work with plain latex, but with pdflatex. The same would apply to spaces in filenames, so I see no big problem there. > And the current quoting scheme does not solve the > problem at all, it moves the problem from spaces to > quotes, spaces can still not be used and quotes are > now made unavailable. Also the quoting scheme does > not allow quoting of the quotes themselves. It does (although I am not sure whether that is intended): Only the first quote is stripped from two consecutive quotes, e.g. the string '"a""bc"' is transformed into 'a"bc'. > A solution would allow *all* characters. > The solution should be able to deal with utf-8 or > even ucs-2/4 encoded file names. Agreed. If you think that the quoting is so bad then why not remove it altogether and tell users to use \pdfximage? My problem with the current implementation is the inconsistency: If filenames with spaces work on windows if quoted, I expect that they work on linux, too, since the OS supports them. Of course I can't expect the same from filenames with quotes: They are not supported by windows, so they can't be supported by pdftex on windows. I hope I have made my point clear. I can live with it if you still don't agree, so I'll stop arguing now. Georg ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-07-04 16:07 Message: Logged In: YES user_id=1276 Hello, I must disagree, "the same behaviour on all platforms" can not mean to add the limitations of all platforms. Now you want the limitation of windows that quotes cannot be used apply on the other systems? There are operating systems that deal with 8+3 file names only. Do you want to forbid long file names, for example? Thus we are loosing both quotes and spaces. Spaces still cannot be used in general. Other TeX engines does not know about quoting. There is no secure way to know, whether quoting is working. Thus code that is intended for different systems must avoid spaces. And the current quoting scheme does not solve the problem at all, it moves the problem from spaces to quotes, spaces can still not be used and quotes are now made unavailable. Also the quoting scheme does not allow quoting of the quotes themselves. A solution would allow *all* characters. The solution should be able to deal with utf-8 or even ucs-2/4 encoded file names. That is already given by the syntax of \pdfximage. Thus new primitives \pdfinput, \pdfopenin, \pdfopenout with <general text> would allow any file name and a macro code can test, whether this feature is available by testing the command names. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- Comment By: Georg Baum (baum) Date: 2005-07-04 13:46 Message: Logged In: YES user_id=2590 I agree that the quote stripping is a hack and should eventually be replaced by something better. I do think however that the behaviour on all platforms should be the same if it is at all possible. Right now, filenames with quotes work on linux and not on windows, and quoted filenames with spaces work on windows but not on linux. Thanks for the hint to pdfximage, I did not know about that (coming from standard LaTeX and always using \includegraphics} Georg ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-07-04 13:13 Message: Logged In: YES user_id=1276 Hello, it is a very bad idea, quotes are perfectly legal in linux and other operating systems. Also image inclusion is done by \pdfximage that uses <general text>, by <general text>, text surrounded by curly braces (catcode 1 and 2) *all* possible file names can be given including quotes and spaces. We need rather new primitives for \input, \openin, and \openout to provide the file names as <general text>. The all file names can be given without the need of quoting at all. See also the discussion about \pdfinput in the ntg-pdftex mailing list. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106

1 0

Patch EscapeAndOther
by Heiko Oberdiek 04 Jul '05

04 Jul '05

Hello, I have uploaded the patch EscapeAndOther, #375. The patch is based on pdftex-1.30.0-rc1. It supersedes patch EscapeDivers.tar.bz2, #371. Inherited from patch EscapeDivers: Expandable: \pdfstrcmp, \pdfescapestring, \pdfescapename Added: \pdfescapehex, \pdfunescapehex Fix: \pdfescapename: '%' is also a delimiter that needs escaping. New in patch EscapeAndOther: * \pdfcreationdate * \pdffilemoddate * \pdffilesize * \pdfmdfivesum * \pdffiledump * \pdfshellescape * \pdfmatch and \pdflastmatch * Start with bug fix: quotes are legal in other operating systems than windows. Syntax: %% expandable commands: \pdfstrcmp <general text> <general text> \pdfescapestring <general text> \pdfescapename <general text> \pdfescapehex <general text> \pdfunescapehex <general text> \pdfcreationdate \pdffilemoddate <general text> \pdffilesize <general text> \pdfmdfivesum <file spec> | <general text> \pdfmatch <match options> <general text> <general text> <match options> := [icase] [subcount <number>] \pdflastmatch <number> %% read-only integers \pdfshellescape Common for the following primitives: * The result string is given by characters with catcode "other" (12), only the space has catcode "space" (10). The commands follow the tradition of \meaning, \string, ... * The argument <general text> is expanded before use. It follows the tradition of \special, \message, \pdfobj, ... Description: \pdfstrcmp{<a>}{} Compares two strings and returns the strings "0" if <a> equals "-1" if <a> is less than "1" if <a> is greater than Use example: \ifcase\pdfstrcmp{abc}{def}\relax \message{abc = def} \or \message{abc > def} \else \message{abc < def} \fi Alternative: Implementing as read-only integer, then the use in \ifcase, \ifnum, ... is more save: \ifcase\pdfstrcmp{xyz}{abc}1\or 2\else 3\fi expands to 3, as read-only integer it expands to the expected 2. \pdfescapestring{<a>} Escapes the string <a> that it can be used as PDF string. '(', ')', '\' are escaped along with the control and 8-bit characters. Use example: \pdfinfo{/Title (\pdfescapestring{...})} \special{ps: [ /Title (\pdfescapestring{...}) /DOCINFO pdfmark} Alternative: Perhaps 8-bit characters don't need escaping. Whitespace, especially newlines should be escaped, because of the use for latex/dvips to avoid recoding problems (<LF> -> <CR><LF>). \pdfescapename{<a>} Escapes the string <a> that it can be used as PDF string. Whitespace, delimiters, '#' are escaped along with the control and 8-bit characters, recommended by the spec. Use example: \pdfobj stream attr{/Type/EmbeddedFile% /Subtype/\pdfescapename{text/plain; charset=iso-8859-1}% ... } file {...} \pdfescapehex{<a>}, \pdfunescapehex{} String <a> is converted to uppercase hexadecimal representation, is converted back. Use example: \pdfescapehex{Hello} is converted to 48656C6C6F \pdfinfo{/Title <\pdfescapehex{Hello}>} Also it can be used to write strings in auxiliary files and later reread without worrying about catcodes, unmached curly braces. \pdfcreationdate It expands to the date string that pdfTeX uses in the info dict as default. Rationale: * It provides seconds and especially the time zone. * Setting of /M date for annotations. * Because of the complicate change file structure of the sources it is not easy to synchronize the creation date with \year, \month, \day and \time. Thus \pdfcreationdate can be used to set these registers to the same values. Example: \pdfcreationdate expands to D:20050625015605+02'00' \pdfannot ...{/Subtype /FileAttachment \M (\pdfcreationdate) ...} \pdffilemoddate{<file>} It expands to the modification date of <file> in the same format as \pdfcreationdate (PDF date format). On error it returns the empty string. Rationale: * File embedding: the date is shown in the attachment tab. * "Make feature": files can be compared, it can be checked, which file is newer. Example: pdfTeX does not support EPS files, epstopdf.sty converts them to PDF either always or, if the PDF variants do not exists. Both ways are not satifactory, either the time penalty is large or pdf files are embedded that are out of date. \pdffilemoddate solves this problem. Error handling, see \pdffilesize I don't have implemented a \pdfcreationdate because of portability issue: The "ctime" field of struct "stat" is interpreted differently among operating systems: * creation date, e.g. win * inode change time, e.g. unix \pdffilesize{<file>} It expands to the size of <file> as string. On error it returns the empty string. Rationale: * File embedding: the size is shown in the attachment tab. * Sometimes it is useful to know if a file has size "0" (failed conversions, ...). Error handling: Currently the empty string is silently returned. Alternatives: * Stop with error message. But what can the user do? Error recovery is easy: no information available, thus return nothing. * Warning message. But the primitives could be used for checks on file existence. * Return status in \pdfretval. I doubt a little whether this is really necessary. It is very easy to implement. But the documentation grows by a large list of error codes with its problems: Much to explain to the users. How they are assigned? \pdfmdfivesum{<abc>} or \pdfmdfivesum file {<file>} It calculates the md5 sum and converts it to uppercase hexadecimal format (same as \pdfescapehex). The syntax is a simplified \pdfobj: Either the data is given directly or in a file. Rationale: * File embedding: providing /CheckSum. * Also the md5 sums of auxiliary files could be stored and compared in order to display a rerun warning. (Of course, it can be possible that different files have the same checksum, but the same file does not have different checksums.) \pdfshellescape It is a read-only integer that is 1 if \write18 is enabled, 0 otherwise. Rationale: * It thought that \ifeof18 with \pdftexversion to implement a safe test for the \write18 feature. But I had to learn that I was wrong, see thread in comp.text.tex: "Confused about pstricks and pdftricks": mikTeX's pdfTeX does not implement \ifeof18. For implementing the test, only a obscure way via \pdftexbanner remains, but this way is not very reliable, the contents of \pdftexbanner is not well defined, it could be anything. Quotes in file names: * File name handling is quite chaotic. Quotes are removed by the scanner for \input, \open*. This behaviour is schizophrene: to solve a problem with spaces, quotes are now forbidden. * Inconsistencies: The syntax of \pdfobj and \pdfximage would allow any file name, but \pdfobj removes quotes, only \pdfximage uses the more intelligent way, it removes quotes for windows only. Thus the patch fixes: * append_to_name (used in pack_file_name): quotes are removed for windows only. This fixes \pdfobj. * utils.c: new function "makecfilename", used by \pdfximage (readimage), \pdffilesize, ... \pdfmatch [icase] [subcount <number>}] {<pattern>}{<string>} Implements pattern matching using the POSIX regex (a standard library at least in my linux). It returns the same values as \pdfstrcmp, but with the following semantics: -1: error case (invalid pattern, ...) 0: no match 1: match found Options: * icase: case insensitive matching * subcount: it sets the table size for found subpatterns. A number "-1" resets the table size to the start default. See the manual page regex.3 and regex.7. The implementation shows a possible interface to pattern matching in TeX. Therefore only the basics is implemented. Flags: * REG_EXTENDED is set in the implementation. * REG_ICASE: can be set by user. * other: not implemented. \pdflastmatch <number> The result of \pdfmatch is stored in an array. The entry "0" contains the match, the following entries submatches. The positions of the matches are also available. They are encoded in the following manner to avoid another primitive: <position> "->" <match string> "->" is used as separator in the tradition of \meaning. There exists macros for parsing the output of \meaning (e.g. in LaTeX: \strip@prefix). The position "-1" with an empty string indicates that this entry is not set. Example: \def\msg#{\immediate\write16 } \msg{\pdfmatch{(l+)o (W(o))}{Hello World}} \msg{\pdflastmatch0} \msg{\pdflastmatch1} \msg{\pdflastmatch2} \msg{\pdflastmatch3} \msg{\pdflastmatch4} Result: 1 2->llo Wo 2->ll 6->Wo 7->o -1-> Alternative: PCRE (Perl-compatible regular expressions) is far more powerful. More options, named subpattern, ... License for 0.4 was GPL compatible, since 0.5 it is BSD, current version is 0.6. The TeX interface could be changed in the following way: * Addition: \pdflastmatchbyname <general text> It extracts matches for named subpattern. * Options can be given by the same name as in the PCRE description: \pdfmatch anchored caseless ... {}{} For easier/faster scanning the options could be restricted to be given in sorted order. * Or options can be given by letters in any order in an additional argument: \pdfmatch{<pattern>}{<options>}{<string>} \pdfmatch{l+}{ai}{Hello World} The implementation could then use strchr to check, whether an option is set. Patch instructions for testing are given in the patch description at sarovar. Have fun Heiko <oberdiek(a)uni-freiburg.de> --

4 13

[ pdftex-Bugs-377 ] unqoute image filenames not only on windows
by noreply＠sarovar.org 04 Jul '05

04 Jul '05

Bugs item #377, was opened at 2005-07-02 21:43 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106 Category: Image inclusion Group: None Status: Open Resolution: None Priority: 5 Submitted By: Georg Baum (baum) Assigned to: Martin Schröder (oneiros) Summary: unqoute image filenames not only on windows Initial Comment: The file pdftex-1.30.0-rc1/src/texk/web2c/pdftexdir/writeimg.c contains in the function readfile() a code block to unqoute file names which is inside #ifdef WIN32 ... #endif. This code should be enabled on all platforms (or at least on linux), since quoted filenames (which are necessary if the name contains spaces) do not only work well on linux. Thanks, Georg ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-07-04 16:07 Message: Logged In: YES user_id=1276 Hello, I must disagree, "the same behaviour on all platforms" can not mean to add the limitations of all platforms. Now you want the limitation of windows that quotes cannot be used apply on the other systems? There are operating systems that deal with 8+3 file names only. Do you want to forbid long file names, for example? Thus we are loosing both quotes and spaces. Spaces still cannot be used in general. Other TeX engines does not know about quoting. There is no secure way to know, whether quoting is working. Thus code that is intended for different systems must avoid spaces. And the current quoting scheme does not solve the problem at all, it moves the problem from spaces to quotes, spaces can still not be used and quotes are now made unavailable. Also the quoting scheme does not allow quoting of the quotes themselves. A solution would allow *all* characters. The solution should be able to deal with utf-8 or even ucs-2/4 encoded file names. That is already given by the syntax of \pdfximage. Thus new primitives \pdfinput, \pdfopenin, \pdfopenout with <general text> would allow any file name and a macro code can test, whether this feature is available by testing the command names. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- Comment By: Georg Baum (baum) Date: 2005-07-04 13:46 Message: Logged In: YES user_id=2590 I agree that the quote stripping is a hack and should eventually be replaced by something better. I do think however that the behaviour on all platforms should be the same if it is at all possible. Right now, filenames with quotes work on linux and not on windows, and quoted filenames with spaces work on windows but not on linux. Thanks for the hint to pdfximage, I did not know about that (coming from standard LaTeX and always using \includegraphics} Georg ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-07-04 13:13 Message: Logged In: YES user_id=1276 Hello, it is a very bad idea, quotes are perfectly legal in linux and other operating systems. Also image inclusion is done by \pdfximage that uses <general text>, by <general text>, text surrounded by curly braces (catcode 1 and 2) *all* possible file names can be given including quotes and spaces. We need rather new primitives for \input, \openin, and \openout to provide the file names as <general text>. The all file names can be given without the need of quoting at all. See also the discussion about \pdfinput in the ntg-pdftex mailing list. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106

1 0

[ pdftex-Bugs-377 ] unqoute image filenames not only on windows
by noreply＠sarovar.org 04 Jul '05

04 Jul '05

Bugs item #377, was opened at 2005-07-02 21:43 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106 Category: Image inclusion Group: None Status: Open Resolution: None Priority: 5 Submitted By: Georg Baum (baum) Assigned to: Martin Schröder (oneiros) Summary: unqoute image filenames not only on windows Initial Comment: The file pdftex-1.30.0-rc1/src/texk/web2c/pdftexdir/writeimg.c contains in the function readfile() a code block to unqoute file names which is inside #ifdef WIN32 ... #endif. This code should be enabled on all platforms (or at least on linux), since quoted filenames (which are necessary if the name contains spaces) do not only work well on linux. Thanks, Georg ---------------------------------------------------------------------- >Comment By: Georg Baum (baum) Date: 2005-07-04 13:46 Message: Logged In: YES user_id=2590 I agree that the quote stripping is a hack and should eventually be replaced by something better. I do think however that the behaviour on all platforms should be the same if it is at all possible. Right now, filenames with quotes work on linux and not on windows, and quoted filenames with spaces work on windows but not on linux. Thanks for the hint to pdfximage, I did not know about that (coming from standard LaTeX and always using \includegraphics} Georg ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-07-04 13:13 Message: Logged In: YES user_id=1276 Hello, it is a very bad idea, quotes are perfectly legal in linux and other operating systems. Also image inclusion is done by \pdfximage that uses <general text>, by <general text>, text surrounded by curly braces (catcode 1 and 2) *all* possible file names can be given including quotes and spaces. We need rather new primitives for \input, \openin, and \openout to provide the file names as <general text>. The all file names can be given without the need of quoting at all. See also the discussion about \pdfinput in the ntg-pdftex mailing list. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106

1 0

input
by Hans Hagen 04 Jul '05

04 Jul '05

Hi, how about \pdfinput {filename} \pdfopenin {filename} \pdfopenout {filename} and maybe even one that accepts a hex encoded string \pdfinput hex {filename} \pdfopenin hex {filename} \pdfopenout hex {filename} Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

4 11

\pdfpagesresources?
by Pawel Jackowski 04 Jul '05

04 Jul '05

Hi there, what do you think about adding \pdfpagesresources primitive toks list? > Thanh: > the answer is simple: nobody ever asked for it, and I don't > need it for myself :). If considered it can be added without > problems. I agree we can live without \pdfpagesresources, as well as we could without \pdfpagesattr. Hovever, collecting named resources per document is much easier than per page. In many cases we don't need different /Resources dict for each page and globally assigned (identical) \pdfpageresources toks list is unnecessarily written into PDF output each time the page is shipped out. It makes a difference for big documents. Of course one can save some time and space by writing resources as objects and storing object rerefences in \pdfpageresources. But to make things cool I'd opt to add primitive \pdfpagesresources toks list. ? -- Pawe/l Jackowski P.Jackowski(a)gust.org.pl

3 6

[ pdftex-Bugs-377 ] unqoute image filenames not only on windows
by noreply＠sarovar.org 04 Jul '05

04 Jul '05

Bugs item #377, was opened at 2005-07-02 21:43 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106 Category: Image inclusion Group: None Status: Open Resolution: None Priority: 5 Submitted By: Georg Baum (baum) Assigned to: Martin Schröder (oneiros) Summary: unqoute image filenames not only on windows Initial Comment: The file pdftex-1.30.0-rc1/src/texk/web2c/pdftexdir/writeimg.c contains in the function readfile() a code block to unqoute file names which is inside #ifdef WIN32 ... #endif. This code should be enabled on all platforms (or at least on linux), since quoted filenames (which are necessary if the name contains spaces) do not only work well on linux. Thanks, Georg ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-07-04 13:13 Message: Logged In: YES user_id=1276 Hello, it is a very bad idea, quotes are perfectly legal in linux and other operating systems. Also image inclusion is done by \pdfximage that uses <general text>, by <general text>, text surrounded by curly braces (catcode 1 and 2) *all* possible file names can be given including quotes and spaces. We need rather new primitives for \input, \openin, and \openout to provide the file names as <general text>. The all file names can be given without the need of quoting at all. See also the discussion about \pdfinput in the ntg-pdftex mailing list. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106

1 0

[ pdftex-Patches-376 ] Uniform and Normal Distribution Pseudo-random numbers patch
by noreply＠sarovar.org 03 Jul '05

03 Jul '05

Patches item #376, was opened at 2005-06-27 16:45 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=376&group_id=106 Category: Primitives Group: v1.30.0 Status: Open Resolution: None Priority: 5 Submitted By: Taco Hoekwater (taco) Assigned to: Nobody (None) Summary: Uniform and Normal Distribution Pseudo-random numbers patch Initial Comment: See the patch file itself or the mailing list for actual usage and comments. Requested by Pawel. It is intented to be applied after tex.pch (as usual) ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=376&group_id=106

4 4

[ pdftex-Bugs-377 ] unqoute image filenames not only on windows
by noreply＠sarovar.org 02 Jul '05

02 Jul '05

Bugs item #377, was opened at 2005-07-02 21:43 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106 Category: Image inclusion Group: None Status: Open Resolution: None Priority: 5 Submitted By: Georg Baum (baum) Assigned to: Martin Schröder (oneiros) Summary: unqoute image filenames not only on windows Initial Comment: The file pdftex-1.30.0-rc1/src/texk/web2c/pdftexdir/writeimg.c contains in the function readfile() a code block to unqoute file names which is inside #ifdef WIN32 ... #endif. This code should be enabled on all platforms (or at least on linux), since quoted filenames (which are necessary if the name contains spaces) do not only work well on linux. Thanks, Georg ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=377&group_id=106

2 1

pdftex patches
by The Thanh Han 30 Jun '05

30 Jun '05

Hi, I was away for a few days and was very impressed by the patches you guys made during that period. This *is* cool, I am quite glad that pdftex is not depending on 1 ror 2 people but there is a rather stable group of developers. Martin, what is the schedule for a next release where all the pending patches should be in? Thanh

3 2

\pdflastvbreakpenalty
by Pawel Jackowski 30 Jun '05

30 Jun '05

Hi All, there is \pdflastvbreakpenalty command mentioned in the manual. I've checked the current and some previous versions of pdfTeX and this command is undefined. Removed looong time ago? Shall we remove commands like that from the documentation, or keep it with obsolate info? BR, -- Pawe/l Jackowski P.Jackowski(a)gust.org.pl

5 6

[ pdftex-Patches-375 ] Patch EscapeAndOther
by noreply＠sarovar.org 27 Jun '05

27 Jun '05

Patches item #375, was opened at 2005-06-25 03:42 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=375&group_id=106 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Heiko Oberdiek (oberdiek) Assigned to: Nobody (None) Summary: Patch EscapeAndOther Initial Comment: Hello, Subject: Patch EscapeAndOther Date: 2005/06/25 Author: Heiko Oberdiek The patch is based on pdftex-1.30.0-rc1. It supersedes patch EscapeDivers.tar.bz2 Inherited from patch EscapeDivers: Expandable: \pdfstrcmp, \pdfescapestring, \pdfescapename Added: \pdfescapehex, \pdfunescapehex Fix: \pdfescapename: '%' is also a delimiter that needs escaping. New in patch EscapeAndOther: * \pdfcreationdate * \pdffilemoddate * \pdffilesize * \pdfmdfivesum * \pdffiledump * \pdfshellescape * \pdfmatch and \pdflastmatch * Start with bug fix: quotes are legal in other operating systems than windows. The description is in file README and I also post it to the ntg-pdftex mailing list. Patched files: The main part is implemented in: * utils.c * pdfstrcmp.ch is added. The other files only contains small changes: * pdftoepdf.cc, epdf.h, ptexlib.h because of changed signature of convertStringToPDFString() * pdftex.defines adds new/deletes old entries of utils.c * pdftex.mk, pdfetex.mk: addition of pdfstrcmp.ch Patch is applied by patch -p0 <EscapeDivers.patch in the base directory that contains "src". Example, the current directory contains pdftex-1.30.0-rc1.zip and EscapeAndOther.patch, contained in EscapeAndOther.tar.bz2: unzip pdftex-1.30.0-rc1.zip cd pdftex-1.30.0-rc1 patch -p0 <../EscapeAndOther.patch ./Build Test files that test the conversion of all characters: test-escapename.tex test-escapestring.tex test-escapehex.tex test-unescapehex.tex Have fun Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=375&group_id=106

3 5

escaped spaces in \pdfescapestring
by Pawel Jackowski 23 Jun '05

23 Jun '05

Hi Heiko, what is the reason to escape spaces and newlines in \pdfescapestring? \def\mystr{A B^^JC} \tt %\pdfescapename{\mystr} % this is obsious \pdfescapestring{\mystr} % ? \end BR, -- Pawe/l Jackowski P.Jackowski(a)gust.org.pl

2 4

[ pdftex-Patches-374 ] noligatures patch
by noreply＠sarovar.org 23 Jun '05

23 Jun '05

Patches item #374, was opened at 2005-06-23 10:54 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=374&group_id=106 >Category: Primitives >Group: v1.30.0 Status: Open Resolution: None Priority: 5 Submitted By: Taco Hoekwater (taco) >Assigned to: The Thanh Han (hanthethanh) Summary: noligatures patch Initial Comment: Usage is quite straightforward: \noligatures \f disables ligatures for the loaded font \f. The (unexpandable) value of \noligatures \f is either 1, when the font has no ligatures, or 0, when the font *does* have ligatures. This patch is to be applied after tex.pch, but see the change file itself for a remark regarding MLTeX. As requested by Hans. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=374&group_id=106

1 0

[ pdftex-Patches-374 ] noligatures patch
by noreply＠sarovar.org 23 Jun '05

23 Jun '05

Patches item #374, was opened at 2005-06-23 10:54 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=374&group_id=106 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Taco Hoekwater (taco) Assigned to: Nobody (None) Summary: noligatures patch Initial Comment: Usage is quite straightforward: \noligatures \f disables ligatures for the loaded font \f. The (unexpandable) value of \noligatures \f is either 1, when the font has no ligatures, or 0, when the font *does* have ligatures. This patch is to be applied after tex.pch, but see the change file itself for a remark regarding MLTeX. As requested by Hans. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=374&group_id=106

1 0

\random?
by Pawel Jackowski 23 Jun '05

23 Jun '05

Hi All, a couple of months ago I was asked to implement a simple random number generator in TeX. It is not very hard for uniform distributions, but I couldn't find satisfactory way to make it reliable and expandable. The problem is that all generators I know (I'm not expirienced in that stuff) use recently generated random number to make another one. For TeX it means we need to make an assignment. OK, there is no many practical reasons for that (: I've found two however. * if \pdfuniquresname is in use, one may want to make his own named resources unique as well * when playing i.e. with page backgrounds or other TeX-based pictures, it might be nice to have a possibility to use random text size/color/position. How do you think? Taco? -- Pawe/l Jackowski P.Jackowski(a)gust.org.pl

3 6

[ pdftex-Patches-371 ] Expandable \pdfstrcmp, \pdfescape*, added \pdf(de)escapehex
by noreply＠sarovar.org 21 Jun '05

21 Jun '05

Patches item #371, was opened at 2005-06-20 17:11 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=371&group_id=106 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Heiko Oberdiek (oberdiek) Assigned to: Nobody (None) Summary: Expandable \pdfstrcmp, \pdfescape*, added \pdf(de)escapehex Initial Comment: Hello, the result of last weekend is a patch that makes \pdfstrcmp (based on Harmut's work) and the \pdfescape* commands expandable. Thus \pdflastescaped* are not needed anymore. I could remove the dependency on a limited buffer by copying the destination strings directly in the string pool above the source string. Also the patch adds the fixes of bug 369. Fourth I add \pdfescapehex and \pdfunescapehex that implements encoding and decoding for hexadecimal strings: * Strings can also be given as hex strings in angle brackets. * Safe method to store and read string data to auxiliary files without worrying about catcodes, unmachted curly braces, ... * Help in handling UTF-8/Unicode or other hex coding stuff. Patch is applied by patch <EscapeDivers.patch in directory src/texk/web2c/pdftexdir The Makefile needs modifying. After the lines $(srcdir)/pdftexdir/tex.pch the line $(srcdir)/pdftexdir/pdfstrcmp.ch is inserted. This should be done two times for pdftex and pdfetex. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- >Comment By: Heiko Oberdiek (oberdiek) Date: 2005-06-21 08:52 Message: Logged In: YES user_id=1276 Hello, the first patch missed the changes in pdftex.defines. Now the patch is reorganized: Patched files: * utils.c main part of the fix, especially function convertStringToPDFName(). * pdftoepdf.cc, epdf.h, ptexlib.h because of changed signature of convertStringToPDFString() * pdftex.defines adds new/deletes old entries of utils.c * pdfstrcmp.ch is added. * pdftex.mk, ../pdfetexdir/pdfetex.mk: addition of pdfstrcmp.ch Now the patch must be applied in scr/texk/web2c: patch -p0 <EscapeDivers.patch With the following commands I could compile pdf(e)tex: unzip pdftex-1.30.0-rc1.zip cd pdftex-1.30.0-rc1/src/texk/web2c patch -p0 <../../../../EscapeDivers.patch cd - ./Build Sorry for the inconvenience Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=371&group_id=106

1 0

[ pdftex-Bugs-369 ] Bug fix for \pdfescapename and \pdfescapestring
by noreply＠sarovar.org 21 Jun '05

21 Jun '05

Bugs item #369, was opened at 2005-06-18 19:00 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=369&group_id=106 Category: None Group: None >Status: Closed >Resolution: Accepted Priority: 5 Submitted By: Heiko Oberdiek (oberdiek) Assigned to: Nobody (None) Summary: Bug fix for \pdfescapename and \pdfescapestring Initial Comment: Hello, the attached patch is based on pdftex-1.30.0-rc1. It fixes the following problems with \pdfescapename and \pdfescapestring: * In both cases zero data bytes result in truncated strings, because the length is measured by "strlen". This is fixed by adding a length parameter to the string conversion functions, see utils.c: escapestr() and escapename(). * \pdfescapename: * delimiter character must be escaped, too. * zero byte is ignored, see specification. * documentation: specification added for convertStringToPDFName() in utils.c Patched files: * utils.c main part of the fix, especially function convertStringToPDFName(). * pdftoepdf.cc, epdf.h, ptexlib.h because of changed signature of convertStringToPDFString() Patch is applied by patch <EscapeNameString.patch in directory src/texk/web2c/pdftexdir Test files that test the conversion of all characters: test-escapename.tex test-escapestring.tex Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- >Comment By: Martin Schröder (oneiros) Date: 2005-06-21 00:17 Message: Logged In: YES user_id=421 Included in #371 ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=369&group_id=106

1 0

[ pdftex-Bugs-367 ] Using hypertex
by noreply＠sarovar.org 21 Jun '05

21 Jun '05

Bugs item #367, was opened at 2005-06-16 16:00 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=367&group_id=106 Category: None Group: v1.21a >Status: Closed >Resolution: Rejected Priority: 5 Submitted By: Aleks Kleyn (aleks_kleyn) Assigned to: Nobody (None) Summary: Using hypertex Initial Comment: This bug appeared after May or April installation of MikTeX. I use AmsArt class together with hyperTeX package. However when I click bookmark or any \ref in document I do not go to appropriate place in text. When I use latex.exe and then convert dvi to pdf I get pdf file where \ref works as I expect, but I cannot create this way bookmarks. ---------------------------------------------------------------------- >Comment By: Martin Schröder (oneiros) Date: 2005-06-21 00:02 Message: Logged In: YES user_id=421 I'm closing this, as it's most likely a hyperref bug. Anyway please file this with MikTeX if you are sure it's a bug; but please read http://www.chiark.greenend.org.uk/~sgtatham/bugs.html before bothering anybody again. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=367&group_id=106

1 0