A proposal for texexec

A proposal for texexec

[Slawek Zak]

I think that it would be very useful if the file produced by TeX
invoked from texexec was placed under other name than the final
(.ps/.pdf) file and renamed when the compilation is over (=atomic
update).

It is difficult to work with pdf/ps viewers "watching" modification
time of the resulting file as a sign for redisplay (say gv); as they
try to display the incomplete file and popup a window with error
messages.

I think that it would be trivial to implement (there already is
--result option for texexec).
--
In a world without walls and borders, who needs windows and gates?
Suavek Zak / PGP: finger://zaks@prioris.mini.pw.edu.pl

RE: A proposal for texexec

[Berend de Boer]

> I think that it would be trivial to implement (there already is
> --result option for texexec).

Or fix the viewers :-) They should attempt to open the file for exclusive
use or with deny write or something like that: if not possible the file is
not yet ready. These viewers are too greedy.

Groetjes,

Berend. (-:

Re: A proposal for texexec

[Slawek Zak]

"Berend de Boer" <berend@pobox.com> writes:

> > I think that it would be trivial to implement (there already is
> > --result option for texexec).
> 
> Or fix the viewers :-) They should attempt to open the file for exclusive
> use or with deny write or something like that: if not possible the file is
> not yet ready. These viewers are too greedy.

Well, this is a desired behaviour. When the file is recompiled (newer
than the it was) they try to reopen and redisplay it. Good thing. It
wouldn't break if the refresh happened only when you make the viewer
redisplay it's window :)
-- 
Violators can be fined, arrested or jailed for making ugly faces at a dog.
		[real standing law in Oklahoma, United States of America]
Suavek Zak / PGP: finger://zaks@prioris.mini.pw.edu.pl

Re: A proposal for texexec

[Hans Hagen]

At 07:31 PM 6/19/00 +0200, Slawek Zak wrote:
>"Berend de Boer" <berend@pobox.com> writes:
>
>> > I think that it would be trivial to implement (there already is
>> > --result option for texexec).
>> 
>> Or fix the viewers :-) They should attempt to open the file for exclusive
>> use or with deny write or something like that: if not possible the file is
>> not yet ready. These viewers are too greedy.
>
>Well, this is a desired behaviour. When the file is recompiled (newer
>than the it was) they try to reopen and redisplay it. Good thing. It
>wouldn't break if the refresh happened only when you make the viewer
>redisplay it's window :)

The solution would be to use a temp file for processing. However, the
problem is that in that case the jobname changes as well as all auxiliary
files. This is not a problem in a stand alone or final run, but not that
handy when debugging files, since in some aspects the filename is
important. So, it's not that trivial. I can write a complement to --result
(--temp) but I wonder if it wouldn't introduce other problems. 

Hans 
-------------------------------------------------------------------------
                                                  Hans Hagen | PRAGMA ADE
                      Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.nl
-------------------------------------------------------------------------

Re: A proposal for texexec

[Slawek Zak]

Hans Hagen <pragma@wxs.nl> writes:

> The solution would be to use a temp file for processing. However, the
> problem is that in that case the jobname changes as well as all auxiliary
> files. This is not a problem in a stand alone or final run, but not that
> handy when debugging files, since in some aspects the filename is
> important. So, it's not that trivial. I can write a complement to --result
> (--temp) but I wonder if it wouldn't introduce other problems. 

Hmm. It would be feasible if the name for temp file remained the
same (suffix). Then renaming only the resulting (pdf/ps/dvi) file
would do the trick.
-- 
hundred-and-one symptoms of being an internet addict:
186. You overstay in the office so you can have more time surfing the net.
Suavek Zak / PGP: finger://zaks@prioris.mini.pw.edu.pl

Re: A proposal for texexec

[Hans Hagen]

At 05:49 PM 6/18/00 +0200, Slawek Zak wrote:
>I think that it would be very useful if the file produced by TeX
>invoked from texexec was placed under other name than the final
>(.ps/.pdf) file and renamed when the compilation is over (=atomic
>update).
>
>It is difficult to work with pdf/ps viewers "watching" modification
>time of the resulting file as a sign for redisplay (say gv); as they
>try to display the incomplete file and popup a window with error
>messages.
>
>I think that it would be trivial to implement (there already is
>--result option for texexec).

I can make it an option, but first we need to solve a problem: it is (at
least here) impossible to rename a file that gs has opened; somehow
overwriting works, but renaming fails. Suggestions?

Hans
-------------------------------------------------------------------------
                                                  Hans Hagen | PRAGMA ADE
                      Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.nl
-------------------------------------------------------------------------