Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug.
https://bugzilla.redhat.com/show_bug.cgi?id=741777
--- Comment #5 from Ralf Corsepius <rc040203(a)freenet.de> 2011-10-13 00:43:54 EDT
---
(In reply to comment #4)
Only a half-hearted objection - I don't care too much about
perl-Tk.
But Tk::MMutil is nothing more than a Tk-specific EU::MM extension. Nothing
explicitly requires or buildrequires it.
Right - Minor packaging bugs.
Only a couple of perl-Tk packages
(perl-Tk-ProgressBar-Mac and perl-Tk-TableMatrix) expect it to be there at
build time due to BR perl(Tk)
Thanks for pointing out these 2 packages.
Besides these 2, I haven't found any further packages using perl(Tk/MMutil),
yet (But I am still checking).
- and however you resolve this, both would need
to be updated as a result (unless you do nothing, of course).
Right - Such packages
will be facing FTBS's during the next FTBS round or the
next perl mass rebuild.
I think it would be slightly more obvious for builds to fail because
EU::MM is
missing rather than Tk::MMutil is missing. Pretty much everyone knows what to
do if EU::MM is missing at build time (just BR EU::MM). Missing Tk::MMutil
could easily lead to "WTF is Tk::MMutil?" <check on CPAN> "WTF?
It's part of
Tk, and perl-Tk is installed, but Tk::MMutil is missing!!".
Well, isn't
that the usual "devel<->runtime" complaint? Nothing to worry much
about, IMO.
Bottom line, yes, perl-Tk should strictly require EU::MM.
I do
not agree. No "mere run-time module" should require EU::MM nor should any
of these require perl-devel. In case of perl(Tk), it's only Tk::MMutil which
pulls in EU::MM and perl-devel.
That said, to me, splitting out Tk/MMutil, is just an ordinary
"devel<->runtime" package split with all the "usual issues"
attached.
--
Configure bugmail:
https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.