- ntg-pdftex - NTG Mailing Lists

[ pdftex-Patches-368 ] make \pdfstrcmp expandable (experimental patch)
by noreply＠sarovar.org 20 Jun '05

20 Jun '05

Patches item #368, was opened at 2005-06-18 00:29 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=368&group_id=106 Category: Primitives Group: v1.30.0 >Status: Closed Resolution: None Priority: 5 Submitted By: Hartmut Henkel (hhenkel) Assigned to: Nobody (None) Summary: make \pdfstrcmp expandable (experimental patch) Initial Comment: just moved \pdfstrcmp over to procedure conv_toks, so that the result -1/0/1 is inserted into the input token list. Does this make sense? ---------------------------------------------------------------------- >Comment By: Hartmut Henkel (hhenkel) Date: 2005-06-20 22:09 Message: Logged In: YES user_id=929 Patch 368 can be closed as an improved version is now included in patch 371 by Heiko. Many thanks to him and Taco for debugging! Regards, Hartmut ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-06-20 17:02 Message: Logged In: YES user_id=1276 Hello, Taco I agree with your analysis, last weekend I found the same variables that need saving, too. Also I have a patch that also make \pdfescape* expandable. Thus \pdflastescaped* are not needed anymore. And I could remove the dependency on a limited buffer by copying the destination strings directly in the string pool above the source string. Third I added \pdfescapehex and \pdfunescapehex that adds encoding/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. I hope, I can attach the patch here in this thread (upload part is missing?. Otherwise I will make an new patch. Patch is applied by patch <EscapeDivers.patch in directory src/texk/web2c/pdftexdir The Makefile must be modified. After the lines $(srcdir)/pdftexdir/tex.pch the line $(srcdir)/pdftexdir/pdfstrcmp.ch is inserted. This is done two times for pdftex and pdfetex. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- Comment By: Taco Hoekwater (taco) Date: 2005-06-20 16:33 Message: Logged In: YES user_id=1608 It makes sense, but it is not 100% safe as-is. In an expanding context, lots of stuff can go one at the same time. Esp., scan_toks touches some global variables, so compare_strings should save the globals def_ref, scanner_status and warning _index to prevent stack corruption later on. For instance, this crashes: % process by initex \input plain \edef\alpha{\pdfstrcmp{alpha}{alph}} \message{\meaning\alpha} \end % I am attaching a change file that modifies compare_strings() , cf. scan_general_text() in eTeX. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=368&group_id=106

1 0

[ pdftex-Patches-371 ] Expandable \pdfstrcmp, \pdfescape*, added \pdf(de)escapehex
by noreply＠sarovar.org 20 Jun '05

20 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> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=371&group_id=106

1 0

[ pdftex-Patches-368 ] make \pdfstrcmp expandable (experimental patch)
by noreply＠sarovar.org 20 Jun '05

20 Jun '05

Patches item #368, was opened at 2005-06-18 00:29 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=368&group_id=106 Category: Primitives Group: v1.30.0 Status: Open Resolution: None Priority: 5 Submitted By: Hartmut Henkel (hhenkel) Assigned to: Nobody (None) Summary: make \pdfstrcmp expandable (experimental patch) Initial Comment: just moved \pdfstrcmp over to procedure conv_toks, so that the result -1/0/1 is inserted into the input token list. Does this make sense? ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-06-20 17:02 Message: Logged In: YES user_id=1276 Hello, Taco I agree with your analysis, last weekend I found the same variables that need saving, too. Also I have a patch that also make \pdfescape* expandable. Thus \pdflastescaped* are not needed anymore. And I could remove the dependency on a limited buffer by copying the destination strings directly in the string pool above the source string. Third I added \pdfescapehex and \pdfunescapehex that adds encoding/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. I hope, I can attach the patch here in this thread (upload part is missing?. Otherwise I will make an new patch. Patch is applied by patch <EscapeDivers.patch in directory src/texk/web2c/pdftexdir The Makefile must be modified. After the lines $(srcdir)/pdftexdir/tex.pch the line $(srcdir)/pdftexdir/pdfstrcmp.ch is inserted. This is done two times for pdftex and pdfetex. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- Comment By: Taco Hoekwater (taco) Date: 2005-06-20 16:33 Message: Logged In: YES user_id=1608 It makes sense, but it is not 100% safe as-is. In an expanding context, lots of stuff can go one at the same time. Esp., scan_toks touches some global variables, so compare_strings should save the globals def_ref, scanner_status and warning _index to prevent stack corruption later on. For instance, this crashes: % process by initex \input plain \edef\alpha{\pdfstrcmp{alpha}{alph}} \message{\meaning\alpha} \end % I am attaching a change file that modifies compare_strings() , cf. scan_general_text() in eTeX. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=368&group_id=106

1 0

[ pdftex-Patches-368 ] make \pdfstrcmp expandable (experimental patch)
by noreply＠sarovar.org 20 Jun '05

20 Jun '05

Patches item #368, was opened at 2005-06-18 00:29 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=368&group_id=106 Category: Primitives Group: v1.30.0 Status: Open Resolution: None Priority: 5 Submitted By: Hartmut Henkel (hhenkel) Assigned to: Nobody (None) Summary: make \pdfstrcmp expandable (experimental patch) Initial Comment: just moved \pdfstrcmp over to procedure conv_toks, so that the result -1/0/1 is inserted into the input token list. Does this make sense? ---------------------------------------------------------------------- >Comment By: Taco Hoekwater (taco) Date: 2005-06-20 16:33 Message: Logged In: YES user_id=1608 It makes sense, but it is not 100% safe as-is. In an expanding context, lots of stuff can go one at the same time. Esp., scan_toks touches some global variables, so compare_strings should save the globals def_ref, scanner_status and warning _index to prevent stack corruption later on. For instance, this crashes: % process by initex \input plain \edef\alpha{\pdfstrcmp{alpha}{alph}} \message{\meaning\alpha} \end % I am attaching a change file that modifies compare_strings() , cf. scan_general_text() in eTeX. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=368&group_id=106

1 0

[ pdftex-Patches-366 ] patch to grow pdf_mem dynamically
by noreply＠sarovar.org 20 Jun '05

20 Jun '05

Patches item #366, was opened at 2005-06-15 23:51 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=366&group_id=106 Category: Memory allocation Group: v1.30.0 Status: Open Resolution: None Priority: 5 Submitted By: Hartmut Henkel (hhenkel) Assigned to: The Thanh Han (hanthethanh) Summary: patch to grow pdf_mem dynamically Initial Comment: experimental patch, pdf_mem_size from texmf.cnf would be ignored then. 10000 < pdf_mem_size < 10000000 (arbitrary upper limit, was 524288) ---------------------------------------------------------------------- >Comment By: Taco Hoekwater (taco) Date: 2005-06-20 15:10 Message: Logged In: YES user_id=1608 I was not seriously suggesting to change the default value to 100, I just had to do something to get the new code doing stuff. It is not obvious to me what macros I should use to fill up pdf_mem quickly :-) ---------------------------------------------------------------------- Comment By: Nobody (None) Date: 2005-06-20 13:42 Message: Logged In: NO 100 is a bit low, hz etc need it, so maybe 2K makes more sense Hans ---------------------------------------------------------------------- Comment By: Taco Hoekwater (taco) Date: 2005-06-20 11:00 Message: Logged In: YES user_id=1608 This version looks smart. I applied it, but I do not have a lot of documents that need more than a few words of pdf_mem, so I changed the compile-time lower limit to 100. Works fine on my test documents. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=366&group_id=106

1 0

[ pdftex-Patches-366 ] patch to grow pdf_mem dynamically
by noreply＠sarovar.org 20 Jun '05

20 Jun '05

Patches item #366, was opened at 2005-06-16 03:21 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=366&group_id=106 Category: Memory allocation Group: v1.30.0 Status: Open Resolution: None Priority: 5 Submitted By: Hartmut Henkel (hhenkel) Assigned to: The Thanh Han (hanthethanh) Summary: patch to grow pdf_mem dynamically Initial Comment: experimental patch, pdf_mem_size from texmf.cnf would be ignored then. 10000 < pdf_mem_size < 10000000 (arbitrary upper limit, was 524288) ---------------------------------------------------------------------- Comment By: Nobody (None) Date: 2005-06-20 17:12 Message: Logged In: NO 100 is a bit low, hz etc need it, so maybe 2K makes more sense Hans ---------------------------------------------------------------------- Comment By: Taco Hoekwater (taco) Date: 2005-06-20 14:30 Message: Logged In: YES user_id=1608 This version looks smart. I applied it, but I do not have a lot of documents that need more than a few words of pdf_mem, so I changed the compile-time lower limit to 100. Works fine on my test documents. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=366&group_id=106

1 0

[ pdftex-Patches-366 ] patch to grow pdf_mem dynamically
by noreply＠sarovar.org 20 Jun '05

20 Jun '05

Patches item #366, was opened at 2005-06-15 23:51 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=366&group_id=106 Category: Memory allocation Group: v1.30.0 Status: Open Resolution: None Priority: 5 Submitted By: Hartmut Henkel (hhenkel) Assigned to: The Thanh Han (hanthethanh) Summary: patch to grow pdf_mem dynamically Initial Comment: experimental patch, pdf_mem_size from texmf.cnf would be ignored then. 10000 < pdf_mem_size < 10000000 (arbitrary upper limit, was 524288) ---------------------------------------------------------------------- >Comment By: Taco Hoekwater (taco) Date: 2005-06-20 11:00 Message: Logged In: YES user_id=1608 This version looks smart. I applied it, but I do not have a lot of documents that need more than a few words of pdf_mem, so I changed the compile-time lower limit to 100. Works fine on my test documents. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=366&group_id=106

1 0

[ pdftex-Bugs-369 ] Bug fix for \pdfescapename and \pdfescapestring
by noreply＠sarovar.org 18 Jun '05

18 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: Open Resolution: None 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> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=369&group_id=106

1 0

\pdfescapestring/convertStringToPDFString
by Heiko Oberdiek 18 Jun '05

18 Jun '05

Hello, what is the purpose of \pdfescapestring? It calls convertStringToPDFString of utils.c with a useless semantics: char *convertStringToPDFString (char *in) { static char pstrbuf[MAX_PSTRING_LEN]; char *out = pstrbuf; int lin = strlen (in); int i, j; char buf[4]; j = 0; for (i = 0; i < lin; i++) { check_buf(j + sizeof(buf), MAX_PSTRING_LEN); if ((unsigned char)in[i] < ' ') { /* convert control characters into hex */ sprintf (buf, "#%02x", (unsigned int)(unsigned char)in[i]); out[j++] = buf[0]; out[j++] = buf[1]; out[j++] = buf[2]; } else if ((in[i] == '(') || (in[i] == ')')) { /* escape paranthesis */ out[j++] = '\\'; out[j++] = in[i]; } else if (in[i] == '\\') { /* escape backslash */ out[j++] = '\\'; out[j++] = '\\'; } else { /* copy char :-) */ out[j++] = in[i]; } } out[j] = '\0'; return pstrbuf; } It mixes escaping for PDF *strings* with escaping for PDF *names*. '(', ')', '\\' are escaped for PDF strings, control characters are escaped for PDF names. Thus I suggest to change the control character escaping from hexadecimal #xx to octal \ooo: sprintf (buf, "\\%03o", (unsigned int)(unsigned char)in[i]); out[j++] = buf[0]; out[j++] = buf[1]; out[j++] = buf[2]; out[j++] = buf[3]; Then \pdfescapestring could be used for conversions to pdf strings. Also a \pdfescapename / convertStringToPDFName is useful. Here especially the delimiter characters must be escaped in the #xx manner. Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> --

4 11

[ pdftex-Patches-368 ] make \pdfstrcmp expandable (experimental patch)
by noreply＠sarovar.org 17 Jun '05

17 Jun '05

Patches item #368, was opened at 2005-06-18 00:29 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=368&group_id=106 Category: Primitives Group: v1.30.0 Status: Open Resolution: None Priority: 5 Submitted By: Hartmut Henkel (hhenkel) Assigned to: Nobody (None) Summary: make \pdfstrcmp expandable (experimental patch) Initial Comment: just moved \pdfstrcmp over to procedure conv_toks, so that the result -1/0/1 is inserted into the input token list. Does this make sense? ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=368&group_id=106

1 0

[ pdftex-Bugs-367 ] Using hypertex
by noreply＠sarovar.org 16 Jun '05

16 Jun '05

Bugs item #367, was opened at 2005-06-16 14:00 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=367&group_id=106 Category: None Group: v1.21a Status: Open Resolution: None 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. ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=367&group_id=106

1 0

[ pdftex-Bugs-362 ] pdfobj produces extra /Filter for object streams with own filter
by noreply＠sarovar.org 16 Jun '05

16 Jun '05

Bugs item #362, was opened at 2005-06-02 12:55 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=362&group_id=106 Category: None Group: v1.21a >Status: Closed >Resolution: Rejected Priority: 5 Submitted By: Jens Elstner (daguy) Assigned to: Nobody (None) Summary: pdfobj produces extra /Filter for object streams with own filter Initial Comment: When I use the following code with pdftex and activated compression \pdfcompresslevel=9 \pdfobj stream% attr{/Filter/ASCIIHexDecode/FunctionType 0/Encode[0 255]/BitsPerSample 8/Domain[0 1]/Size[2]/Range[0 1 0 1 0 1 0 1]/Decode[0 1 0 1 0 1 0 1]}% {00000000FFFFFFFF}% output tint function The generated pdf file contains the object as shown below 3 0 obj << /Filter/ASCIIHexDecode/FunctionType 0/Encode[0 255]/BitsPerSample 8/Domain[0 1]/Size[2]/Range[0 1 0 1 0 1 0 1]/Decode[0 1 0 1 0 1 0 1] /Length 17 /Filter /FlateDecode >> stream x┌30Ç'7'c endobj The compression added an extra /Filter line and produces an error or does not work the way it was intended anymore. Proposed fix for this would be to check the <attr- spec> for /Filter and omit any compression if a user filter function is present already. Yours sincerely, Jens Elstner ---------------------------------------------------------------------- >Comment By: Martin Schröder (oneiros) Date: 2005-06-16 14:47 Message: Logged In: YES user_id=421 This is a (mis)feature of pdftex; pdftex does not scan the <attr-spec>. ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-06-02 14:52 Message: Logged In: YES user_id=1276 Hello, not a fix, just a workaround: \begingroup \pdfcompresslevel=0 % then pdfTeX does not % add /Filter itself \immediate\pdfobj stream attr{/Filter/...}{...}\relax \endgroup Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=362&group_id=106

1 0

[ pdftex-Bugs-320 ] \textcolor
by noreply＠sarovar.org 16 Jun '05

16 Jun '05

Bugs item #320, was opened at 2005-03-28 05:52 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=320&group_id=106 Category: None Group: v1.21a Status: Open >Resolution: Postponed Priority: 5 Submitted By: Aleks Kleyn (aleks_kleyn) Assigned to: Nobody (None) Summary: \textcolor Initial Comment: I want to change color of text using comand \textcolor {blue}. However when text flows from one page to another the color resets to standard on the next page. Initially I beleived that this problem of pdf. However I tried today latex.exe and then converted dvi to pdf. I got effect that I expected: text had on both pages the same color. However using of pdflatex is more better because it prepares bookmarks for pdf. ---------------------------------------------------------------------- >Comment By: Martin Schröder (oneiros) Date: 2005-06-16 14:45 Message: Logged In: YES user_id=421 The cause is a missing colour stack of pdfTeX; this is a known problem (and more properly a feature request). ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=320&group_id=106

1 0

[ pdftex-Bugs-311 ] Dangling font objects with non-replacable fonts
by noreply＠sarovar.org 16 Jun '05

16 Jun '05

Bugs item #311, was opened at 2005-03-20 22:00 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=311&group_id=106 Category: PDF inclusion Group: v1.21a >Status: Closed >Resolution: Fixed Priority: 5 Submitted By: Hartmut Henkel (hhenkel) Assigned to: The Thanh Han (hanthethanh) Summary: Dangling font objects with non-replacable fonts Initial Comment: If a PDF file with a non-replacable font (e. g. a non-embedded standard-14 font, or a TrueType font) is included into another PDF file, there is an unused/dangling fontfile stream object generated. Details, rough patch, test files see attached patch.tgz. Regards, Hartmut ---------------------------------------------------------------------- Comment By: Martin Schröder (oneiros) Date: 2005-06-16 14:43 Message: Logged In: YES user_id=421 This has been fixed in 1.30.0 ---------------------------------------------------------------------- Comment By: Hartmut Henkel (hhenkel) Date: 2005-03-20 22:31 Message: Logged In: YES user_id=929 Ooops, problem affects only Type-1 fonts, as the copyFont() function else returns already before calling lookup_fontmap(). Patch unchanged. Regards, Hartmut ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=311&group_id=106

1 0

[ pdftex-Bugs-306 ] cmap: unicode ???? is mapped to multiple glyphs
by noreply＠sarovar.org 16 Jun '05

16 Jun '05

Bugs item #306, was opened at 2005-03-05 17:52 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=306&group_id=106 Category: Fonts Group: v1.21a >Status: Closed >Resolution: Fixed Priority: 5 Submitted By: Peter Dyballa (ballapete) Assigned to: Hartmut Henkel (hhenkel) Summary: cmap: unicode ???? is mapped to multiple glyphs Initial Comment: I created symlinks from the Lucida fonts that come with Java runtime and I created from them TFM files with ttf2tfm version 1.5, and also a map file fragment and FD files. When I access LucidaSans-Oblique pdfeTeX, Version 3.141592-1.21a-2.2 (Web2C 7.5.4) reports this: Error: pdflatex (file /usr/local/teTeX/share/texmf.local/fonts/truetype/ lucida/llsro8a.ttf): cmap: unicode 2010 is mapped to multiple glyphs ==> Fatal error occurred, the output PDF file is not finished! U+2010 is HYPHEN ---------------------------------------------------------------------- Comment By: Martin Schröder (oneiros) Date: 2005-06-16 14:43 Message: Logged In: YES user_id=421 This has been fixed in 1.30.0 ---------------------------------------------------------------------- Comment By: Martin Schröder (oneiros) Date: 2005-05-12 17:08 Message: Logged In: YES user_id=421 What's the state of this? ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=306&group_id=106

1 0

\pdfretval vs. expandable \pdfstrcmp
by Heiko Oberdiek 16 Jun '05

16 Jun '05

Hello, is there a reason why \pdfstrcmp is not expandable? An expandable variant could be used in expandable contexts: \ifcase\numexpr\pdfstrcmp{StringA}{StringB}+1\relax ...\or ...\or ...\fi Inserting into a sorted linear list could be done in O(n): \edef\mylist{\addtolist\mylist{foo}}% Now it is O(n^2) :-( Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> --

2 1

[ pdftex-Patches-366 ] patch to grow pdf_mem dynamically
by noreply＠sarovar.org 15 Jun '05

15 Jun '05

Patches item #366, was opened at 2005-06-15 23:51 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=366&group_id=106 Category: Memory allocation Group: v1.30.0 Status: Open Resolution: None Priority: 5 Submitted By: Hartmut Henkel (hhenkel) Assigned to: The Thanh Han (hanthethanh) Summary: patch to grow pdf_mem dynamically Initial Comment: experimental patch, pdf_mem_size from texmf.cnf would be ignored then. 10000 < pdf_mem_size < 10000000 (arbitrary upper limit, was 524288) ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=495&aid=366&group_id=106

1 0

Version numbering
by Martin Schröder 10 Jun '05

10 Jun '05

One point discussed at EuroTeX2005 was that the current versioning scheme is too confusing (when do we increase which part) and also that a real _number_ is be preferred. In the future I want to switch to a scheme much like OpenBSD/OpenSSH: - version numbers are like 1.30.0 A BB C - C is only > 0 for bug fixing releases - A.BB is increased with every new release by 0.1 - with 2.00 we switch to A.B.C, i.e. 2.0.0 (we can't do this now as it would confuse some people because 1.3.0 is "smaller" than 1.21a) Comments? Best regards Martin -- http://www.tm.oneiros.de

5 18

rc1
by The Thanh Han 07 Jun '05

07 Jun '05

Hi Martin, I don't plan to upload further snapshots, so can you please make a pretest release or some rc1? Thanh

2 1

[Fwd: Re: [NTG-context] Installing Truetype fonts in context.]
by Hans Hagen 05 Jun '05

05 Jun '05

Hi, ttf2afm ... who maintains that? it would be handy if this program provides a -version option currently it's hard to see what one has on the machine 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 -----------------------------------------------------------------

2 1

bug in XObject
by Vit Zyka 03 Jun '05

03 Jun '05

Hallo pdftexdevelopers, there is a bug in \pdfxform. See the next example: ----------------------- \pdfcompresslevel=0 \footline={\hfill}% \setbox0\vbox{\hsize8.75cm\noindent Ahoj} \pdfxform attr{/Matrix [1.2 0.25 0.3 1 0 0]} 0 \pdfrefxform\pdflastxform \end ------------------------- The result XObject looks like: ------------------------ 1 0 obj << /Type /XObject /Subtype /Form /Matrix [1.2 0.25 0.3 1 0 0] /BBox [0 0 248.031 8.856] /FormType 1 /Matrix [1 0 0 1 0 0] /Resources 6 0 R /Length 69 >> stream 1 0 0 1 0 -833.034 cm BT /F1 9.963 Tf 0 834.971 Td[(Aho)-55(j)]TJ ET endstream endobj ------------------------ See the double /Matrix entry. That results in ignoring transformation by Acrobat (R) v.6 and 7. (v.4 and v.5 is OK). I believe that it is (exceptionally) bug in pdftex not in Acrobat ;) vit

5 5

[ pdftex-Bugs-362 ] pdfobj produces extra /Filter for object streams with own filter
by noreply＠sarovar.org 02 Jun '05

02 Jun '05

Bugs item #362, was opened at 2005-06-02 12:55 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=362&group_id=106 Category: None Group: v1.21a Status: Open Resolution: None Priority: 5 Submitted By: Jens Elstner (daguy) Assigned to: Nobody (None) Summary: pdfobj produces extra /Filter for object streams with own filter Initial Comment: When I use the following code with pdftex and activated compression \pdfcompresslevel=9 \pdfobj stream% attr{/Filter/ASCIIHexDecode/FunctionType 0/Encode[0 255]/BitsPerSample 8/Domain[0 1]/Size[2]/Range[0 1 0 1 0 1 0 1]/Decode[0 1 0 1 0 1 0 1]}% {00000000FFFFFFFF}% output tint function The generated pdf file contains the object as shown below 3 0 obj << /Filter/ASCIIHexDecode/FunctionType 0/Encode[0 255]/BitsPerSample 8/Domain[0 1]/Size[2]/Range[0 1 0 1 0 1 0 1]/Decode[0 1 0 1 0 1 0 1] /Length 17 /Filter /FlateDecode >> stream x┌30Ç'7'c endobj The compression added an extra /Filter line and produces an error or does not work the way it was intended anymore. Proposed fix for this would be to check the <attr- spec> for /Filter and omit any compression if a user filter function is present already. Yours sincerely, Jens Elstner ---------------------------------------------------------------------- Comment By: Heiko Oberdiek (oberdiek) Date: 2005-06-02 14:52 Message: Logged In: YES user_id=1276 Hello, not a fix, just a workaround: \begingroup \pdfcompresslevel=0 % then pdfTeX does not % add /Filter itself \immediate\pdfobj stream attr{/Filter/...}{...}\relax \endgroup Yours sincerely Heiko <oberdiek(a)uni-freiburg.de> ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=362&group_id=106

1 0

[ pdftex-Bugs-362 ] pdfobj produces extra /Filter for object streams with own filter
by noreply＠sarovar.org 02 Jun '05

02 Jun '05

Bugs item #362, was opened at 2005-06-02 12:55 You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=362&group_id=106 Category: None Group: v1.21a Status: Open Resolution: None Priority: 5 Submitted By: Jens Elstner (daguy) Assigned to: Nobody (None) Summary: pdfobj produces extra /Filter for object streams with own filter Initial Comment: When I use the following code with pdftex and activated compression \pdfcompresslevel=9 \pdfobj stream% attr{/Filter/ASCIIHexDecode/FunctionType 0/Encode[0 255]/BitsPerSample 8/Domain[0 1]/Size[2]/Range[0 1 0 1 0 1 0 1]/Decode[0 1 0 1 0 1 0 1]}% {00000000FFFFFFFF}% output tint function The generated pdf file contains the object as shown below 3 0 obj << /Filter/ASCIIHexDecode/FunctionType 0/Encode[0 255]/BitsPerSample 8/Domain[0 1]/Size[2]/Range[0 1 0 1 0 1 0 1]/Decode[0 1 0 1 0 1 0 1] /Length 17 /Filter /FlateDecode >> stream x┌30Ç'7'c endobj The compression added an extra /Filter line and produces an error or does not work the way it was intended anymore. Proposed fix for this would be to check the <attr- spec> for /Filter and omit any compression if a user filter function is present already. Yours sincerely, Jens Elstner ---------------------------------------------------------------------- You can respond by visiting: http://sarovar.org/tracker/?func=detail&atid=493&aid=362&group_id=106

1 0

unwanted font replacement during inclussion
by Vit Zyka 02 Jun '05

02 Jun '05

Hi, I noticed the next strange behaviour. Let us create very simple file test.tex: ------------------ \font\bf=cmbx12 Normal text. \bf Bold text. \end ------------------ Now include this file by plain pdftex: ------------------ \pdfximage{test.pdf} \pdfrefximage\pdflastximage \end ------------------ Everything is correct. No let us try to insert the file by context: ------------------ \starttext %\font\bf=cmbx12 %Normal text. \bf Bold text. \pdfximage{test.pdf} \pdfrefximage\pdflastximage \stoptext ------------------ And that is!!! Text 'bold text' is typeset with totaly different font from cmbx12, so it is smashed by wrong metric! See the attached file. If I uncomment two lines, so the font cmbx12 is not only present in the included file, the result is correct. Where is the problem? Something mixing cm and lm? I though that included file comes with all font resources. Hmmmm, now I feel it is more pdftex-related then context-related question. vit (pdfetex v1.30) -- Vít Zýka --- http://typokvitek.com Automatic document typesetting Automaticka sazba dokumentu -- --

4 3

Re: [NTG-pdftex] inclusion of PDF with hz-alg expandeded font
by Vit Zyka 02 Jun '05

02 Jun '05

The Thanh Han wrote: > remember when I fixed it. Anyway, can you please try it with > the latest snapshot (djgpp is also available). If you need Jupiii! Yes, ~1.30 really fixes the problem. That is very good news. Thank you, Thanh, very much. (Also good news for me is, that pdftex preserves form objects with geometrical transformations during PDF inclusion; If someday it preserves also annotations it would be really amazing!) > help with getting the djgpp binary work, please feel free to Thanks to Akira Kakuto who is maintaining the fresh food in the http://www.fsci.fuk.kindai.ac.jp/kakuto/win32-ptex/web2c75.html I do not need translating. Vit

2 1