Re: TEXINPUTS

2011-05-10 Thread Pavel Sanda
Enrico Forestieri wrote: > > that was the second check you forced me to do :) > > Sorry? "doublecheck" in mails above. p

Re: TEXINPUTS

2011-05-10 Thread Enrico Forestieri
On Tue, May 10, 2011 at 02:07:39PM +0200, Pavel Sanda wrote: > Enrico Forestieri wrote: > > only an environment variable is > > added before launching converters. > > yep thats what i meant; went only smoothly though those parts > and saw new code around startscript/process fucntions. > looking a

Re: TEXINPUTS

2011-05-10 Thread Pavel Sanda
Enrico Forestieri wrote: > only an environment variable is > added before launching converters. yep thats what i meant; went only smoothly though those parts and saw new code around startscript/process fucntions. looking again, there is always path.empty()?, so it should be ok... that was the sec

Re: TEXINPUTS

2011-05-10 Thread Enrico Forestieri
On Tue, May 10, 2011 at 11:39:21AM +0200, Pavel Sanda wrote: > Enrico Forestieri wrote: > > > i would be more careful about the > > > systemcall part, it looks quite invasive and in very sensitive area of > > > code. > > > > Uh? What do you mean? > > to double check anything which touches Syste

Re: TEXINPUTS

2011-05-10 Thread Pavel Sanda
Enrico Forestieri wrote: > > i would be more careful about the > > systemcall part, it looks quite invasive and in very sensitive area of code. > > Uh? What do you mean? to double check anything which touches SystemCallXXX:: pavel

Re: TEXINPUTS

2011-05-10 Thread Enrico Forestieri
On Tue, May 10, 2011 at 11:06:17AM +0200, Pavel Sanda wrote: > Richard Heck wrote: > >> To conclude, note that even if the patch may seem large, it is really > >> simple and not invasive, such that it can be safely backported to branch. > >> > > What's the policy as regards preference changes? We

Re: TEXINPUTS

2011-05-10 Thread Pavel Sanda
Richard Heck wrote: >> To conclude, note that even if the patch may seem large, it is really >> simple and not invasive, such that it can be safely backported to branch. >> > What's the policy as regards preference changes? We don't need to add this > to prefs2prefs, or increase the preferences fo

Re: TEXINPUTS

2011-05-09 Thread Richard Heck
On 5/9/11 7:37 PM, Enrico Forestieri wrote: On Mon, May 09, 2011 at 10:39:18AM -0400, Richard Heck wrote: On 05/08/2011 02:45 PM, Enrico Forestieri wrote: The \input@path mechanism (already used by LyX) is not always sufficient and there are cases where a file can be found by LaTeX and also by

Re: TEXINPUTS

2011-05-09 Thread Enrico Forestieri
On Mon, May 09, 2011 at 10:39:18AM -0400, Richard Heck wrote: > On 05/08/2011 02:45 PM, Enrico Forestieri wrote: > > The \input@path mechanism (already used by LyX) is not always sufficient > > and there are cases where a file can be found by LaTeX and also by other > > ancillary programs (such as

Re: TEXINPUTS

2011-05-09 Thread Richard Heck
On 05/08/2011 02:45 PM, Enrico Forestieri wrote: > The \input@path mechanism (already used by LyX) is not always sufficient > and there are cases where a file can be found by LaTeX and also by other > ancillary programs (such as dvi previewers) only if the document dir is > placed in the TEXINPUTS