IDNA::Punycode has disappeared from CPAN.

URI mentions its URI::_punycode being based on IDNA::Punycode.

No packages depend on libidna-punycode-perl - candidate for removal?

 - Jonas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=
=qzK6
-----END PGP SIGNATURE-----

Hi Jonas,

Jonas Smedegaard wrote:

Indeed, I recently used URI::_punycode when I needed some punycode
stuff -- and was happy that it's in such a common module/package. I
though wondered and still wonder about that strange module name. The
underscore often means "internal". Anyone knows more? The POD doesn't
help with that question.

Popcon shows about 35-45 installations with around 5 activity votes.

IMHO that are too many installations for a removal without RC bug, but
likely a removal candidate if it catches a non-trivial RC bug in the
future.

		Regards, Axel

Hi Axel,

Quoting Axel Beckert (2014-12-24 16:58:38)
in turn is used by ihost and as_iri:

perl -MURI -E 'say URI->new("http://xn--99zt52a.example.org/%E2%80%AE")->as_iri'

I now updated <https://github.com/miyagawa/URI-Find-UTF8/issues/2>
reflecting that new discovery :-)

Fair enough - but I do believe it effectively means we take over as
upstream, and should then adapt packaging to reflect that - not
repackage as dpkg source format "3.0 (native)", but have Homepage, watch
file and debian/copyright source URLs point to our resources at Alioth.

It must've happened before among our thousands of packages - do we have
formalized some procedures for which actions we should remember to take,
and how we then monitor its health?  Do we e.g. push to CPAN and monitor
tkwalitee and cpantesters.org, also against Windows and Mac?

As a minimum, I guess it makes sense to keep this bug open, and 45 hits
on popcon is not enough for me to volunteer as new upstream for code I
see no need for, even if bugfree...


 - Jonas

Hi,

Jonas Smedegaard wrote:

I disagree. As long as there's no major issue for the package in Sid...

I think that's all totally overkill.

Yes. I though would lower the severity to minor.

And I see no need for that effort.

For me, that makes quite some difference.

		Regards, Axel

Quoting Axel Beckert (2014-12-24 18:25:12)

Which parts do you disagree with?

Do you disagree that code is dead upstream, and perhaps believe it is
simply some temporary hickup causing the code to not appear at CPAN?

If you agree that code is dead upstream, but disagree that we have taken
over?  Or only *how* we should take over?
really means.

I think we can agree at least to strip last part: Interface of URI is
not identical, even if it covers all of its functionality.


 - Jonas