[DRE-maint] sponsorship for libfastthread-ruby

[Lucas Nussbaum]

On 25/03/08 at 16:24 -0400, Micah Anderson wrote:
> > 2) The following email snippet was taken off of ruby-talk google group:
> > 
> > http://groups.google.com/group/ruby-talk-google/browse_thread/thread/7933e7e987dad1c3
> 
> Thanks for finding this, this is interesting information.
> 
> It strikes me as odd that libfastthread needs to be released to fix this
> problem, rather than the problem is fixed in ruby itself. I suppose
> there may be a delay in the release of a new version of ruby that would
> have this fix? If so, have you looked at maybe taking those fixes and
> applying them to the debian ruby package? It seems a little silly to
> have a package for a short period of time to fix this bug, and then have
> to remove it again. Like will this be fixed in ruby in a month, two
> months? Before Lenny releases?
> 
> Again, you may be aware of more than I am about this situation, and
> I'm interested to know. If it still does make sense to package
> libfastthread, I'd be happy to sponsor it (although it would be nice to
> know if the version I packaged has anything to offer to yours).

If there's a serious bug in the thread implementation shipped in
Debian's ruby1.8 package, it would be nice to file a bug against a
ruby1.8 package describing the bug. If that bug has been known for
nearly a year, but has not been fixed in the various patchlevel versions
of ruby that have been released since then, I wonder if the bug is
really that serious.

If you can describe the bug (with a test case) and provide a clean patch
that fixes it, I could also include the patch in the ruby1.8 package. We
don't need to wait for upstream to fix bugs ;)
-- 
| Lucas Nussbaum
| lucas at lucas-nussbaum.net   http://www.lucas-nussbaum.net/ |
| jabber: lucas at nussbaum.fr             GPG: 1024D/023B3F4F |