>>>>> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> On Wednesday 23 October 2002 3:54 pm, Andre Poenitz wrote:
>> On Wed, Oct 23, 2002 at 04:03:52PM +0200, Jean-Marc Lasgouttes
>> wrote: > It would be easier to make lyx search in this directory
>> too, > like what the following simple patch does
>>
>> Are you sure our friends from the Dark Side like the '/' path
>> separator in
>>
>> fname = i18nLibFileSearch("../../lyxdoc", arg, "lyx");
>>
>> If so, I see no problem with the change...
Angus> Och, it's a nasty hack for people who should know better. Why
Angus> are you hard-coding such stuff?
I think the links are an even nastier hack. Have you actually read the
script? It has lines like
FILES=`ls ${DIR}/*.*`
Maybe too much msdos programming lastly, no?
Also doing 'grep ".lyx"' does not strike me as the best way of getting
only files with extension .lyx.
What happens when files are removed/added to the lyxdoc repository?
Can one run the script several times? What happens then? Does
.cvsignore grow without bounds? And then when one does a commit, does
.cvsignore get commited to the main lyx repository?
Come on, this can't be serious.
To come back to the original question, I am hard-coding this because I
belive it will work in 90% of the cases, and I do not care about the
cases where it does not work since it is merely a convenience for the
developpers. Actually, I can think of one case where it would be a
problem: if we had a separate branch for documentation for 1.2.x. Then
one would have to check it out separately (under another name) and my
nice scheme breaks down. So what?
JMarc
