* problem in kpse.rb
@ 2006-09-17 2:23 Johan Sandblom
2006-09-17 3:13 ` Sanjoy Mahajan
0 siblings, 1 reply; 6+ messages in thread
From: Johan Sandblom @ 2006-09-17 2:23 UTC (permalink / raw)
Attempting to find out why this happened on linux
~$ texmfstart ctxtools --updatecontext
CtxTools | updating
CtxTools | unable to change to
I found that line 114 of kpse.rb can be changed from
path = run("--expand-path=\\\$#{varname}") rescue ''
to
path = run("--expand-path=\$#{varname}") rescue ''
which allows it to work, that is, the line should be the same under
windows and linux.
~$ ruby --version
ruby 1.8.4 (2005-12-24) [i486-linux]
Regards, Johan
--
Johan Sandblom N8, MRC, Karolinska sjh
t +46851776108 17176 Stockholm
m +46735521477 Sweden
"What is wanted is not the will to believe, but the
will to find out, which is the exact opposite"
- Bertrand Russell
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: problem in kpse.rb
2006-09-17 2:23 problem in kpse.rb Johan Sandblom
@ 2006-09-17 3:13 ` Sanjoy Mahajan
2006-09-17 7:45 ` Johan Sandblom
0 siblings, 1 reply; 6+ messages in thread
From: Sanjoy Mahajan @ 2006-09-17 3:13 UTC (permalink / raw)
> path = run("--expand-path=\\\$#{varname}") rescue ''
>
> to
>
> path = run("--expand-path=\$#{varname}") rescue ''
I also ran into this (a few weeks ago on the list) and the patch I
sent Hans after testing it on my Ubuntu Linux laptop was to use two
backslashes:
run("--expand-path=\\$#{varname}")
so that Unix sees the command as, for example,
kpsewhich --expand-path=\$TEXMFLOCAL
which prevents the shell from expanding TEXMFLOCAL as a shell or
environment variable, leaving it for kpse to find it in texmf.cnf.
Interesting that \$ also works. I guess the \ gets passed unchanged,
so the $ still gets procted from the shell.
By the way, the rescue '' doesn't do anything, since a failed run of
kpsewhich doesn't throw an exception. I think I have a patch for
that.
Hans, should I resend those patches?
-Sanjoy
`Never underestimate the evil of which men of power are capable.'
--Bertrand Russell, _War Crimes in Vietnam_, chapter 1.
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: problem in kpse.rb
2006-09-17 3:13 ` Sanjoy Mahajan
@ 2006-09-17 7:45 ` Johan Sandblom
2006-09-17 9:02 ` Taco Hoekwater
0 siblings, 1 reply; 6+ messages in thread
From: Johan Sandblom @ 2006-09-17 7:45 UTC (permalink / raw)
Disappointingly, over here it doesn't work with two backslashes, only
with one. Possibly because there are two texmf.cnf (minimal linux
distro) which give different answers:
./texmf/web2c/texmf.cnf, which claims
% A place for local additions to a "standard" texmf tree.
% This tree is not used for local configuration maintained by
% texconfig, it uses TEXMFCONFIG below.
TEXMFLOCAL = $SELFAUTOPARENT/../texmf-local
and
./texmf-local/web2c/texmf.cnf, which claims
TEXMFMAIN = $SELFAUTOPARENT/texmf
TEXMFLOCAL = $SELFAUTOPARENT/texmf-local
TEXMFFONTS = $SELFAUTOPARENT/texmf-fonts
Regards, Johan
2006/9/17, Sanjoy Mahajan <sanjoy@mrao.cam.ac.uk>:
> > path = run("--expand-path=\\\$#{varname}") rescue ''
> >
> > to
> >
> > path = run("--expand-path=\$#{varname}") rescue ''
>
> I also ran into this (a few weeks ago on the list) and the patch I
> sent Hans after testing it on my Ubuntu Linux laptop was to use two
> backslashes:
>
> run("--expand-path=\\$#{varname}")
>
> so that Unix sees the command as, for example,
>
> kpsewhich --expand-path=\$TEXMFLOCAL
>
> which prevents the shell from expanding TEXMFLOCAL as a shell or
> environment variable, leaving it for kpse to find it in texmf.cnf.
> Interesting that \$ also works. I guess the \ gets passed unchanged,
> so the $ still gets procted from the shell.
>
> By the way, the rescue '' doesn't do anything, since a failed run of
> kpsewhich doesn't throw an exception. I think I have a patch for
> that.
>
> Hans, should I resend those patches?
>
> -Sanjoy
>
> `Never underestimate the evil of which men of power are capable.'
> --Bertrand Russell, _War Crimes in Vietnam_, chapter 1.
> _______________________________________________
> ntg-context mailing list
> ntg-context@ntg.nl
> http://www.ntg.nl/mailman/listinfo/ntg-context
>
--
Johan Sandblom N8, MRC, Karolinska sjh
t +46851776108 17176 Stockholm
m +46735521477 Sweden
"What is wanted is not the will to believe, but the
will to find out, which is the exact opposite"
- Bertrand Russell
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: problem in kpse.rb
2006-09-17 7:45 ` Johan Sandblom
@ 2006-09-17 9:02 ` Taco Hoekwater
2006-09-17 9:12 ` Johan Sandblom
2006-09-17 14:11 ` Hans Hagen
0 siblings, 2 replies; 6+ messages in thread
From: Taco Hoekwater @ 2006-09-17 9:02 UTC (permalink / raw)
Johan Sandblom wrote:
>>I also ran into this (a few weeks ago on the list) and the patch I
>>sent Hans after testing it on my Ubuntu Linux laptop was to use two
>>backslashes:
>>
>> run("--expand-path=\\$#{varname}")
My suggestion would be:
run("--expand-path='$#{varname}'")
Taco
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: problem in kpse.rb
2006-09-17 9:02 ` Taco Hoekwater
@ 2006-09-17 9:12 ` Johan Sandblom
2006-09-17 14:11 ` Hans Hagen
1 sibling, 0 replies; 6+ messages in thread
From: Johan Sandblom @ 2006-09-17 9:12 UTC (permalink / raw)
Cc: mailing list for ConTeXt users
Yes, that works.
Johan
2006/9/17, Taco Hoekwater <taco@elvenkind.com>:
> Johan Sandblom wrote:
> >>I also ran into this (a few weeks ago on the list) and the patch I
> >>sent Hans after testing it on my Ubuntu Linux laptop was to use two
> >>backslashes:
> >>
> >> run("--expand-path=\\$#{varname}")
>
> My suggestion would be:
>
> run("--expand-path='$#{varname}'")
>
> Taco
>
--
Johan Sandblom N8, MRC, Karolinska sjh
t +46851776108 17176 Stockholm
m +46735521477 Sweden
"What is wanted is not the will to believe, but the
will to find out, which is the exact opposite"
- Bertrand Russell
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: problem in kpse.rb
2006-09-17 9:02 ` Taco Hoekwater
2006-09-17 9:12 ` Johan Sandblom
@ 2006-09-17 14:11 ` Hans Hagen
1 sibling, 0 replies; 6+ messages in thread
From: Hans Hagen @ 2006-09-17 14:11 UTC (permalink / raw)
Taco Hoekwater wrote:
> Johan Sandblom wrote:
>
>>> I also ran into this (a few weeks ago on the list) and the patch I
>>> sent Hans after testing it on my Ubuntu Linux laptop was to use two
>>> backslashes:
>>>
>>> run("--expand-path=\\$#{varname}")
>>>
>
> My suggestion would be:
>
> run("--expand-path='$#{varname}'")
>
ok
-----------------------------------------------------------------
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
-----------------------------------------------------------------
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2006-09-17 14:11 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2006-09-17 2:23 problem in kpse.rb Johan Sandblom
2006-09-17 3:13 ` Sanjoy Mahajan
2006-09-17 7:45 ` Johan Sandblom
2006-09-17 9:02 ` Taco Hoekwater
2006-09-17 9:12 ` Johan Sandblom
2006-09-17 14:11 ` Hans Hagen
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).