Hello, I noticed that TeXExec has no "native" support for Gamma, in the sense that it doesn't change the "cont-" prefix to "gamma-" when the TeX executable is set to Omega. I did a patch to make this possible, but of course since there are no gamma-<language>.tex files, it doesn't work that simply. I therefore did a more extensive patch (it has to dump the format as cont-en and then rename it) but this is sub-optimal. It would be very easy to do if (recent) fpTeX and teTeX supported the --job-name option ... is this the case? If so, the patch can be done very easily. Anybody using fpTeX or teTeX know about this? -- Giuseppe "Oblomov" Bilotta
Giuseppe Bilotta
very easy to do if (recent) fpTeX and teTeX supported the --job-name option ... is this the case? If so, the patch can be done very easily. Anybody using fpTeX or teTeX know about this?
dunno, there is: --progname name Pretend to be program name. This affects both the format used and the search paths. perhaps this is what you are looking for? Never tested this, of course :) Patrick -- Mehr Buchstaben - mehr Spass
Saturday, October 19, 2002 Patrick Gundlach wrote:
PG> Giuseppe Bilotta
very easy to do if (recent) fpTeX and teTeX supported the --job-name option ... is this the case? If so, the patch can be done very easily. Anybody using fpTeX or teTeX know about this?
PG> dunno, there is: PG> --progname name PG> Pretend to be program name. This affects both the PG> format used and the search paths. PG> perhaps this is what you are looking for? Never tested this, of course :) Uh, no, --progname only changes the (memory) settings, default format and search paths. --job-name changes the \jobname primitive ... uhm, wait a sec, this gave me an idea. Be back soon :) -- Giuseppe "Oblomov" Bilotta
Saturday, October 19, 2002 Giuseppe Bilotta wrote: GB> Uh, no, --progname only changes the (memory) settings, default GB> format and search paths. --job-name changes the GB> \jobname primitive ... uhm, wait a sec, this gave me an idea. Be GB> back soon :) Nope, it's not possible to change the jobname otherwise :\ -- Giuseppe "Oblomov" Bilotta
Giuseppe Bilotta writes:
Saturday, October 19, 2002 Giuseppe Bilotta wrote:
GB> Uh, no, --progname only changes the (memory) settings, default GB> format and search paths. --job-name changes the GB> \jobname primitive ... uhm, wait a sec, this gave me an idea. Be GB> back soon :)
Nope, it's not possible to change the jobname otherwise :\
Two points: - up to web2c 7.3.8, you can use the --fmt=NAME option to set the name of the format to be dumped by initex (or iniomega). - from 7.3.9 onward, this will no longer work. Use --jobname=NAME instead. -- Olaf Weber (This space left blank for technical reasons.)
Thursday, October 24, 2002 Olaf Weber wrote: OW> Two points: OW> - up to web2c 7.3.8, you can use the --fmt=NAME option to set the name OW> of the format to be dumped by initex (or iniomega). OW> - from 7.3.9 onward, this will no longer work. Use --jobname=NAME OW> instead. --fmt in web2c selects the format to undump, doesn't it? --job-name in MiKTeX changes \jobname; what does --jobname in web2c 7.3.9 do? -- Giuseppe "Oblomov" Bilotta
Giuseppe Bilotta writes:
Thursday, October 24, 2002 Olaf Weber wrote:
OW> Two points:
OW> - up to web2c 7.3.8, you can use the --fmt=NAME option to set the name OW> of the format to be dumped by initex (or iniomega). OW> - from 7.3.9 onward, this will no longer work. Use --jobname=NAME OW> instead.
--fmt in web2c selects the format to undump, doesn't it?
Now it does just that. In 7.3.8 and earlier it also selected the format name to dump to -- in retrospect a rather gross hack.
--job-name in MiKTeX changes \jobname; what does --jobname in web2c 7.3.9 do?
It sets \jobname. -- Olaf Weber (This space left blank for technical reasons.)
Friday, October 25, 2002 Olaf Weber wrote:
--job-name in MiKTeX changes \jobname; what does --jobname in web2c 7.3.9 do?
OW> It sets \jobname. Goodie! Funny thing, MikTeX uses --job-name, while web2c uses --jobname ... one would wish both used the same :) -- Giuseppe "Oblomov" Bilotta
At 06:13 PM 10/19/2002 +0200, Giuseppe Bilotta wrote:
Hello,
I noticed that TeXExec has no "native" support for Gamma, in the sense that it doesn't change the "cont-" prefix to "gamma-" when the TeX executable is set to Omega.
I did a patch to make this possible, but of course since there are no gamma-<language>.tex files, it doesn't work that simply. I therefore did a more extensive patch (it has to dump the format as cont-en and then rename it) but this is sub-optimal. It would be very easy to do if (recent) fpTeX and teTeX supported the --job-name option ... is this the case? If so, the patch can be done very easily. Anybody using fpTeX or teTeX know about this?
--tex=binary --format=formatname personally i think that gamma should just use a cont-* file, but with an ofmt extension; where etex can be distinguished by suffix, omega cannot, which is a bad thing since now one gets format clashes; i didn't experiment with this, but it may be possible to set the texformat paths in texmf.cnf on a per progrma basis (omega.texformats => .../web2c/omega) Hans ------------------------------------------------------------------------- Hans Hagen | PRAGMA ADE | pragma@wxs.nl Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com ------------------------------------------------------------------------- information: http://www.pragma-ade.com/roadmap.pdf documentation: http://www.pragma-ade.com/showcase.pdf -------------------------------------------------------------------------
Hans Hagen writes:
personally i think that gamma should just use a cont-* file, but with an ofmt extension; where etex can be distinguished by suffix, omega cannot, which is a bad thing since now one gets format clashes;
This is an interesting idea, and one I'd be willing to impose on Omega.
i didn't experiment with this, but it may be possible to set the texformat paths in texmf.cnf on a per progrma basis (omega.texformats => .../web2c/omega)
This should be possible, and would be an alternative practice. -- Olaf Weber (This space left blank for technical reasons.)
At 08:40 PM 10/20/2002 +0200, Olaf Weber wrote:
Hans Hagen writes:
personally i think that gamma should just use a cont-* file, but with an ofmt extension; where etex can be distinguished by suffix, omega cannot, which is a bad thing since now one gets format clashes;
This is an interesting idea, and one I'd be willing to impose on Omega.
Thanks! It would help a lot.
i didn't experiment with this, but it may be possible to set the texformat paths in texmf.cnf on a per progrma basis (omega.texformats => .../web2c/omega)
This should be possible, and would be an alternative practice.
Right, but it would also need some coordination, Hans ------------------------------------------------------------------------- Hans Hagen | PRAGMA ADE | pragma@wxs.nl Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com ------------------------------------------------------------------------- information: http://www.pragma-ade.com/roadmap.pdf documentation: http://www.pragma-ade.com/showcase.pdf -------------------------------------------------------------------------
Hans Hagen writes:
At 08:40 PM 10/20/2002 +0200, Olaf Weber wrote:
Hans Hagen writes:
personally i think that gamma should just use a cont-* file, but with an ofmt extension; where etex can be distinguished by suffix, omega cannot, which is a bad thing since now one gets format clashes;
This is an interesting idea, and one I'd be willing to impose on Omega.
Thanks! It would help a lot.
Still, for my peace of mind I'd like to know a bit more of the background, and why getting Omega to use a different suffix will help so much. (And does this mean that pdftex and pdfetex should be modified as well?) -- Olaf Weber (This space left blank for technical reasons.)
Monday, October 21, 2002 Olaf Weber wrote: OW> Hans Hagen writes:
At 08:40 PM 10/20/2002 +0200, Olaf Weber wrote:
Hans Hagen writes:
personally i think that gamma should just use a cont-* file, but with an ofmt extension; where etex can be distinguished by suffix, omega cannot, which is a bad thing since now one gets format clashes;
This is an interesting idea, and one I'd be willing to impose on Omega.
Thanks! It would help a lot.
OW> Still, for my peace of mind I'd like to know a bit more of the OW> background, and why getting Omega to use a different suffix will help OW> so much. (And does this mean that pdftex and pdfetex should be OW> modified as well?) e-TeX already uses the .efmt extension when running in extended mode. The reason to use different extensions is to solve easily the problem of re-naming formats when running with different engines (pdfLaTeX vs Lamba vs eLaTeX vs LaTeX, for example). While it may be interesting to find new names for old formats under new engines, it's really unnecessary (for example ConTeXt has a preliminary support for Omega as its engine; the ConTeXt format built on Omega has to be called Gamma, or you can't have Knuth's or pdf-TeX *and* Omega formats at the same time). -- Giuseppe "Oblomov" Bilotta
At 11:19 PM 10/21/2002 +0200, Olaf Weber wrote:
Hans Hagen writes:
At 08:40 PM 10/20/2002 +0200, Olaf Weber wrote:
Hans Hagen writes:
personally i think that gamma should just use a cont-* file, but with an ofmt extension; where etex can be distinguished by suffix, omega cannot, which is a bad thing since now one gets format clashes;
This is an interesting idea, and one I'd be willing to impose on Omega.
Thanks! It would help a lot.
Still, for my peace of mind I'd like to know a bit more of the background, and why getting Omega to use a different suffix will help so much. (And does this mean that pdftex and pdfetex should be modified as well?)
it would make sense to give pdftex/pdfetex other extensions as well, just for the sake of consistency on the other hand, a user can generate a cont-en.efmt file with pdfetex which is compatible with cont-en.efmt by etex apart from the pdf support, simply because pdftex is an extension [in the same way, etex is compatible and could have the same suffix, unless in non compatible ** mode] omega is different and not compatible (and not that stable either), and so users would like to use their macro packages (+formats) for both systems, so having an cont-en.efmt for pdfetex alongside cont-en.oft for omega is a good solution or that, in that case texexec --program=omega yourfile would launch omega and omega would look for the oft file itself (users nowadays are not that aware of formats and asking them to generate them ...) so, to be perfect it should be: tex -> fmt etec -> efmt (already) pdftex -> pft pdfetex -> epmt omega -> oft eomega -> eoft nts -> nfmt (already) but since pdftex is compatible is less urgent an alternative is: web2c/pdftex/pdftex.fmt web2c/pdftex/pdfetex.fmt etc Sorry for the mess, but solutions like pdflatex versus latex versus tex makes users think that they are all different tex's while it's the same program with different macros, and pdflatex/latex is actually the same thing apart from the backend. For context, i would prefer to keep cont-en as the one and only name for the english interface version, independent of the texprogram used Hans ------------------------------------------------------------------------- Hans Hagen | PRAGMA ADE | pragma@wxs.nl Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com ------------------------------------------------------------------------- information: http://www.pragma-ade.com/roadmap.pdf documentation: http://www.pragma-ade.com/showcase.pdf -------------------------------------------------------------------------
Tuesday, October 22, 2002 Hans Hagen wrote: HH> tex -> fmt HH> etec -> efmt (already) HH> pdftex -> pft HH> pdfetex -> epmt HH> omega -> oft HH> eomega -> eoft HH> nts -> nfmt (already) HH> but since pdftex is compatible is less urgent since it's efmt and nfmt, wouldn't consistency call for pfmt, ofmt, eofmt, pefmt ? -- Giuseppe "Oblomov" Bilotta
Giuseppe Bilotta writes:
Tuesday, October 22, 2002 Hans Hagen wrote:
HH> tex -> fmt HH> etec -> efmt (already) HH> pdftex -> pft HH> pdfetex -> epmt HH> omega -> oft HH> eomega -> eoft HH> nts -> nfmt (already)
HH> but since pdftex is compatible is less urgent
since it's efmt and nfmt, wouldn't consistency call for pfmt, ofmt, eofmt, pefmt ?
Yes, but thanks to the immense success of 8+3 filenames, they could end up being truncated to pfm, ofm, eof, pef respectively. ofm is known to collide with another suffix. -- Olaf Weber (This space left blank for technical reasons.)
Thursday, October 24, 2002 Olaf Weber wrote:
since it's efmt and nfmt, wouldn't consistency call for pfmt, ofmt, eofmt, pefmt ?
OW> Yes, but thanks to the immense success of 8+3 filenames, they could OW> end up being truncated to pfm, ofm, eof, pef respectively. ofm is OW> known to collide with another suffix. Are there Omega implementations for DOS? I mean, I think we should stop worrying about this. If an DOS implementation is indeed done, they can change it with no big problems. -- Giuseppe "Oblomov" Bilotta
At 08:54 AM 10/25/2002 +0200, Giuseppe Bilotta wrote:
Thursday, October 24, 2002 Olaf Weber wrote:
since it's efmt and nfmt, wouldn't consistency call for pfmt, ofmt, eofmt, pefmt ?
OW> Yes, but thanks to the immense success of 8+3 filenames, they could OW> end up being truncated to pfm, ofm, eof, pef respectively. ofm is OW> known to collide with another suffix.
Are there Omega implementations for DOS?
I mean, I think we should stop worrying about this. If an DOS implementation is indeed done, they can change it with no big problems.
Isn't that the djgpp variant? what does it do with suffixes? Hans ------------------------------------------------------------------------- Hans Hagen | PRAGMA ADE | pragma@wxs.nl Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com ------------------------------------------------------------------------- information: http://www.pragma-ade.com/roadmap.pdf documentation: http://www.pragma-ade.com/showcase.pdf -------------------------------------------------------------------------
Giuseppe Bilotta writes:
Thursday, October 24, 2002 Olaf Weber wrote:
since it's efmt and nfmt, wouldn't consistency call for pfmt, ofmt, eofmt, pefmt ?
Yes, but thanks to the immense success of 8+3 filenames, they could end up being truncated to pfm, ofm, eof, pef respectively. ofm is known to collide with another suffix.
Are there Omega implementations for DOS?
Last time I checked, it compiled successfully using djgpp. Actually, Eli Zaretskii spent a lot of time making the web2c stuff configure and compile cleanly under djgpp. While some of the supporting files have not been updated recently, getting it to compile again should be comparatively easy. It simply truncates extensions to three characters.
I mean, I think we should stop worrying about this. If an DOS implementation is indeed done, they can change it with no big problems.
IIRC, ISO CDOM format also privileges .3 filenames; look at the various compatibility levels. -- Olaf Weber (This space left blank for technical reasons.)
Friday, October 25, 2002 Olaf Weber wrote:
Are there Omega implementations for DOS?
OW> Last time I checked, it compiled successfully using djgpp. Actually, OW> Eli Zaretskii spent a lot of time making the web2c stuff configure and OW> compile cleanly under djgpp. While some of the supporting files have OW> not been updated recently, getting it to compile again should be OW> comparatively easy. OW> It simply truncates extensions to three characters. In this case these such implementations could "system-dependently" change the format extension (instead of truncating, removing the "m", for example).
I mean, I think we should stop worrying about this. If an DOS implementation is indeed done, they can change it with no big problems.
OW> IIRC, ISO CDOM format also privileges .3 filenames; look at the OW> various compatibility levels. Hm. I'm starting to think that Hans Hagen idea to put each format under a separate subfolder of the \fmt directory is not that bad ... could be pass the suggestion on to the TDS committee? -- Giuseppe "Oblomov" Bilotta
At 06:48 PM 10/24/2002 +0200, you wrote:
Giuseppe Bilotta writes:
Tuesday, October 22, 2002 Hans Hagen wrote:
HH> tex -> fmt HH> etec -> efmt (already) HH> pdftex -> pft HH> pdfetex -> epmt HH> omega -> oft HH> eomega -> eoft HH> nts -> nfmt (already)
HH> but since pdftex is compatible is less urgent
since it's efmt and nfmt, wouldn't consistency call for pfmt, ofmt, eofmt, pefmt ?
Yes, but thanks to the immense success of 8+3 filenames, they could end up being truncated to pfm, ofm, eof, pef respectively. ofm is known to collide with another suffix.
ah, pfm is also taken already, (font metrics) Hans ------------------------------------------------------------------------- Hans Hagen | PRAGMA ADE | pragma@wxs.nl Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com ------------------------------------------------------------------------- information: http://www.pragma-ade.com/roadmap.pdf documentation: http://www.pragma-ade.com/showcase.pdf -------------------------------------------------------------------------
participants (4)
-
Giuseppe Bilotta -
Hans Hagen -
Olaf Weber -
Patrick Gundlach