[pdf-devel] Re: srcinfo-extractor.pl

[gerel]

 > Date: Wed, 26 Nov 2008 21:39:04 +0100
 > From: address@hidden
 > 
 > Hello Gerel.
 > 
 > I am writing the MANIFEST.wiki files and I was not able to make
 > prmgt/srcinfo-extractor.pl to work properly.
 > 
 > I manually added several new directories where to scan for
 > MANIFEST.wiki files:
 > 
 > # push these by hand
 > push (@dirs, ""); # for TOPDIR
 > push (@dirs, "prmgt");
 > push (@dirs, "lib");
 > push (@dirs, "m4");
 > push (@dirs, "src");
 > push (@dirs, "src/base");
 > push (@dirs, "src/object");
 > push (@dirs, "src/document");
 > push (@dirs, "src/page");
 > push (@dirs, "utils");
 > push (@dirs, "torture");
 > 
 > but in the output I am only getting the contents of the MANIFEST.wiki
 > file in the TOPDIR.
 > 

Yes, that's right, it's because you are running it from the TOPDIR when the
intended idea is to run it from the "prmgt/" dir.
If you think that it should be run from the top dir just remove a dot from
the TOPDIR variable, i.e:
##
my $TOPDIR=".";
###


 > It is possible to rewrite the script to automatically scan for any
 > MANIFEST.wiki file without having to manually push them into @dirs?
 > Can you provide a patch?

Yes, I thought about that before. The problem I got stuck with was how deep to
traverse in the file hierarchy, that's why I prefered to leave the issue opened.
We could use a MAX_DEPTH value like 3 that would search for
"src/base/somedir/MANIFEST.wiki" at most.

What you think ?

If you agree, I'll send you a patch On-The-Fly. :-)

un abrazo Jose.
-gerel