Overview
Loading...
Request History
plater created request
Removed unneeded libmlt-devel build requirement ready for the libmlt6 package.
cgiboudeaux declined request
plater reopened request
I'll put libmlt6 into multimedia:libs and luigi can do the rest.
plater revoked request
I think this change is fine.
kdenlive (still) does have these anyway:
BuildRequires: pkgconfig(mlt++) >= %{mlt_version}
BuildRequires: pkgconfig(mlt-framework) >= %{mlt_version}
Seems that libmlt-devel was just superfluous...
@Vogtinator, @cgiboudeaux, @luca_b, @wolfi323: review reminder
May I ask why you revoked this?
As I wrote, I think this change makes sense, I just wanted to give other maintainers a chance to comment.
And without it, both libmlt6-devel and libmlt(7)-devel will be pulled in (once the latter is in Factory), which shouldn't cause problems I suppose but is totally unnecessary too IMHO.
a series of events that occurred at the time of the libmlt saga have led me to believe that it would have been declined by the maintainer anyway. Although I made alios maintainer of libmlt6 I had to submit it myself.
It was originally declined because of the libmlt "dilemma" AIUI.
But that's resolved now it seems with the submission of libmlt6 to Factory.
I'd say please reopen this, as mentioned I think we should do this change anyway (even regardless of the libmlt situation).
Thank you.
And TBH, this SR originally confused me as well. I thought that kdenlive of course should need libmlt-devel. (so I can understand it was declined without comment)
But then I noticed that it's actually not needed due to the other BuildRequires, that's why I wrote my first comment.
"that it would have been declined by the maintainer anyway"
Oh, and another thing: I am a maintainer (of kdenlive) as well. There is not just one.
So I can accept this SR, but I cannot do it now as you revoked it.
This BuildRequires is a good way to check if the provides/obsoletes are correct (and they're not, the libmlt-devel package definition is in a
%if 0block)But libmlt(7)-devel doesn't provide "pkgconfig(mlt++)" or "pkgconfig(mlt-framework)" anyway.
And the only provides/obsoletes I see are for kdenlive5, which don't depend on libmlt either.
Also: "the libmlt-devel package definition is in a %if 0 block"
What are you talking about here?
I don't find that in kdenlive.spec... :-/
Ah, that's probably about libmlt6 then. ;-)
AFAICS, the package definition actually is outside the "%if 0" block.
But I still don't see at all how it would be related to this change.
And the two libmlt?-devel packages don't have any provides/obsoletes either AFAICS (so I don't see why that should be checked), and I don't see why they should have, considering that they can co-exist.
@dplater: I understand that you don't want to be part of this disussion.
I created a new SR now with the same change (SR#895324), but still feel free to reopen yours if you want (I will revoke mine then).
Ok, SR#895324 was accepted and forwarded to Factory.
So in the end there was just some misunderstanding here.
@dplater: Let me say thank you for all the great work you did over the years, and still do.
I think all openSUSE users (including me) do owe you a lot!
openSUSE wouldn't be what it is without you.