texexec and return codes ... bad
Hi all!
(Gerben, you are on the To since you were interested in the ctxfmtutil,
but it is bugged ..., see below)
texexec does not return a useful return code:
# texexec --nonstopmode --fast --make --aleph cont-en
...
loading : Context Font Macros / UNICODE Support
! Aleph capacity exceeded, sorry [number of strings=33165].
\setvalue ...xpandafter \def \csname #1\endcsname
l.275 \defineucharmapping{\s!default}
#1#2%
No pages of output.
Transcript written on cont-en.log.
TeXExec |
TeXExec | tex engine path: /etc/texmf/web2c/aleph
TeXExec |
TeXExec |
TeXExec | runtime: 3.104333
# echo $?
0
#
This is bad, very bad.... something we can do about it?
And, since I am at it: It does not honor the --nonstopmode:
# texexec --nonstopmode --fast --make --xetex cont-en
...
loading : Context Encoding Macros (pdf)
! Undefined control sequence.
l.63 \ifcase\pdfgentounicode
\else \expandafter \endinput \fi
?
and waiting there.
Is there a way to fix these two problems, please???
For Gerben: ctxfmtutil has two problems:
- it does not detect if a build was broken due to the first bug
- it hangs sometimes due to the second bug
Best wishes
Norbert
-------------------------------------------------------------------------------
Dr. Norbert Preining
Pling hi all! On Fre, 05 Jan 2007, Norbert Preining wrote:
texexec does not return a useful return code: # texexec --nonstopmode --fast --make --aleph cont-en ... l.275 \defineucharmapping{\s!default} #1#2% ... # echo $? 0
And, since I am at it: It does not honor the --nonstopmode: # texexec --nonstopmode --fast --make --xetex cont-en ... loading : Context Encoding Macros (pdf) ! Undefined control sequence. l.63 \ifcase\pdfgentounicode \else \expandafter \endinput \fi ?
and waiting there.
Any ideas ... suggestions ...?
Best wishes
Norbert
-------------------------------------------------------------------------------
Dr. Norbert Preining
Norbert Preining wrote:
Any ideas ... suggestions ...?
fixed in the beta already, i'll make a new current zip (has to do with changes in pdftex) 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 -----------------------------------------------------------------
On Die, 09 Jan 2007, Hans Hagen wrote:
fixed in the beta already, i'll make a new current zip (has to do with changes in pdftex)
One question: I tested the release 2007-01-09 and the xetex bug is gone,
but the return value of texexec still is 0, even when aleph breaks
making a cont-en format. Does this mean that the beta version you are
talking about is *not* the 2007-01-09 release, but the next release?
Thanks
Norbert
-------------------------------------------------------------------------------
Dr. Norbert Preining
Norbert Preining wrote:
On Die, 09 Jan 2007, Hans Hagen wrote:
fixed in the beta already, i'll make a new current zip (has to do with changes in pdftex)
One question: I tested the release 2007-01-09 and the xetex bug is gone, but the return value of texexec still is 0, even when aleph breaks making a cont-en format. Does this mean that the beta version you are talking about is *not* the 2007-01-09 release, but the next release?
Hans, the same problem occurs in texsync, and I assume anything that is started by texmfstart. I do not know what the proper ruby syntax is for perl's exit(), but in general, you should return "zero" for success, anything else for failure (normally programs use simply "one" in all cases). If you do not call exit() yourself and do not call die() either (whatever the ruby equivalent of that is), the default return value is zero == success. That can be an enormous pain in the butt when texexec etc. are used inside scripts|makefiles. Best, Taco
Taco Hoekwater wrote:
Norbert Preining wrote:
On Die, 09 Jan 2007, Hans Hagen wrote:
fixed in the beta already, i'll make a new current zip (has to do with changes in pdftex)
One question: I tested the release 2007-01-09 and the xetex bug is gone, but the return value of texexec still is 0, even when aleph breaks making a cont-en format. Does this mean that the beta version you are talking about is *not* the 2007-01-09 release, but the next release?
Hans, the same problem occurs in texsync, and I assume anything that is started by texmfstart.
I do not know what the proper ruby syntax is for perl's exit(), but in general, you should return "zero" for success, anything else for failure (normally programs use simply "one" in all cases).
If you do not call exit() yourself and do not call die() either (whatever the ruby equivalent of that is), the default return value is zero == success. That can be an enormous pain in the butt when texexec etc. are used inside scripts|makefiles.
the problem is that those scripting languages behave differently (and to be honest i never use return codes because i never write complex os dependent scripts but package everything in e.g. ruby so that i can run everywhere; return codes is alien teritory for me); taco and i need to figure it out somehow 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 -----------------------------------------------------------------
Taco Hoekwater wrote:
Norbert Preining wrote:
On Die, 09 Jan 2007, Hans Hagen wrote:
fixed in the beta already, i'll make a new current zip (has to do with changes in pdftex)
One question: I tested the release 2007-01-09 and the xetex bug is gone, but the return value of texexec still is 0, even when aleph breaks making a cont-en format. Does this mean that the beta version you are talking about is *not* the 2007-01-09 release, but the next release?
Hans, the same problem occurs in texsync, and I assume anything that is started by texmfstart.
I do not know what the proper ruby syntax is for perl's exit(), but in general, you should return "zero" for success, anything else for failure (normally programs use simply "one" in all cases).
If you do not call exit() yourself and do not call die() either (whatever the ruby equivalent of that is), the default return value is zero == success. That can be an enormous pain in the butt when texexec etc. are used inside scripts|makefiles.
ok, i found out there i already implemented some sort of error reporting and exit stuff but not for formats; i'll make a new zip (and keep my fingers crossed) 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 -----------------------------------------------------------------
Norbert Preining wrote:
# texexec --nonstopmode --fast --make --xetex cont-en ... loading : Context Encoding Macros (pdf) ! Undefined control sequence. l.63 \ifcase\pdfgentounicode \else \expandafter \endinput \fi ?
and waiting there.
i hope that the current 'current' is ok; btw, \pdfgentounicode is part of the latest pdftex (the one that goed on tex live) 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 -----------------------------------------------------------------
On Die, 09 Jan 2007, Hans Hagen wrote:
# texexec --nonstopmode --fast --make --xetex cont-en ... loading : Context Encoding Macros (pdf) ! Undefined control sequence. l.63 \ifcase\pdfgentounicode \else \expandafter \endinput \fi ?
and waiting there.
i hope that the current 'current' is ok; btw, \pdfgentounicode is part of the latest pdftex (the one that goed on tex live)
But do you *EXPECT* that everyone will have pdftex 1.40.1 installed? Is
it obligatory?
In Debian we wont have it for some time, because it takes some time to
create new packages for TeX Live 2006/7.
Shouldn't the usage of a new primitive somehow be shielded in case
someone uses pdftex 1.30.5?
Best wishes
Norbert
-------------------------------------------------------------------------------
Dr. Norbert Preining
participants (3)
-
Hans Hagen -
Norbert Preining -
Taco Hoekwater