On Sunday 03 February 2008, Jonathan Rockway wrote:
> * On Sun, Feb 03 2008, Gabor Szabo wrote:
> > On Feb 3, 2008 12:01 PM, Shlomi Fish <[EMAIL PROTECTED]> wrote:
> >> And some people go on holiday for more than that. The question is: how
> >> long should we wait? There wasn't a new release of IO::Socket::INET6
> >> for over three years, and it has three pending bug reports. This
> >> probably indicates that the author is missing-in-action.
> >
> > Instead of this infighting why not just upload a development version
> > of the module with something like this in the pod:
> >
> > "This is an unnofficial version of module X::Y::Z till the original
> > module author reappears or till I get official maintainership of the the
> > module."
>
> Excellent idea. Another possibility is to release a subclass of the
> module that has your fixes in it. Then people can use
> IO::Socket::INET6::Shlomi, and forget about the brokenness in
> IO::Socket::INET6. When I::S::6 fixes the bug you can re-upload your
> module as a simple "package IO::Socket::INET6::Shlomi; use base
> "IO::Socket::INET6"; 1".
What my out-of-mainline modifications to IO::Socket::INET6 resolve is:
1. Some warnings due to clashes of imported symbols between Socket6 and
Socket. These occur when loading IO::Socket::INET6 due to something like:
<<<<<<<<<
use IO::Socket;
use Socket6;
>>>>>>>>>
2. Some tests in t/*.t that fail on certain conditions.
Neither of these fixes can be done using a separate subclass, because the
warnings will still occur when loading IO::Socket::INET6 and the tests will
still fail in the original distribution.
Regards,
Shlomi Fish
---------------------------------------------------------------------
Shlomi Fish [EMAIL PROTECTED]
Homepage: http://www.shlomifish.org/
I'm not an actor - I just play one on T.V.
