On Mon, 07 Jul 2003 21:01:52 +1000 Gonzalo Servat <[EMAIL PROTECTED]> wrote: > > Yes, it does need about 5384384343 perl modules. It does come with a script > to find all its dependencies and install them for you using CPAN. I had no > trouble doing it this way.
I haven't been able to get CPAN working. It does some really odd things like try fetch, ncftp, wget, curl blah blah to get the modules, and it seems to fail to distinguish between the lack of the program and the failure to connect to cpan. Worse, I have no trust in it being able to get the particular version of, say, html::mason that RT requires. I've heard that CPAN will even do dubious things like upgrade core modules that you might depend on. Why does't RT just include the extras modules that are known to work with it in a RT specific @INC directory? I've tried to get the modules manually, but there is often NO correlation between the module name as stated by perl/RT, and the actual tar.gz on cpan. EVEN IF I got CPAN working to my satisfaction, I always like to bundle up work I do so that I can just do a reinstall by just untarring or rpm'ing. Dumping foreign modules into perl, even site_perl makes that problematical. > > To my question ... is it worth it? Is it really so much better > > than other request/issue/problem trackers? e.g. scarab.tigris.org > > for instance? Does anyone have any other suggestions? > > IMHO, it is worth it - specially for version 3. It's a huge improvement > over version 2. It does take a little bit of reading to get used to it's > permissions system but once you get the hang of it you'll love it. OK, I'll persevere a little more, but it's hurting. How about gnats? Anyone work with that? Matt -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
