[Bug 527] Review request: mlt - Toolkit for broadcasters, video editors, media players, transcoders

[RPM Fusion Bugzilla]

http://bugzilla.rpmfusion.org/show_bug.cgi?id=527


Dan Dennedy <dan at dennedy.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dan at dennedy.org




--- Comment #61 from Dan Dennedy <dan at dennedy.org>  2009-05-08 19:41:03 ---
Dear Zarko and Orcan,

I am sorry if my comments offended you. It may be known to some that I am not
particularly fond of some Fedora policies. However, I have also expressed that
I do respect the good intention and effort of the Fedora project. The goal here
is to make kdenlive, which is receiving praises, readily available to Fedora
users. Allow me to propose that we not let us not let our emotions from getting
in the way of that. I really have been trying to address the issues raised ever
since this process started in Fedora proper.

Let me address the technical issues:

1) Calling exit() in a shared lib. 
The reason this exists is historical and somewhat "complicated.". The short
story is that libmiracle (where this exists), miracle, and some other parts are
legacy baggage from the context of kdenlive and other forthcoming apps (GTK+
OpenShot) simply wanting to use MLT core. Therefore, to address this and some
other issues, I have chosen to remove this baggage (libmiracle, miracle, and
some others) from the next release of mlt. The work is complete, and I hope to
finish testing and make a new release within a week.

2) Strange, mixed license comments in mlt++ source.
Yep, another thing I inherited and overlooked ever since. This is being
addressed as part of the fore-mentioned reorganization because that process
also involves folding mlt++ into mlt.

3) unused-direct-shlib-dependency (libmlt linking to libm) and plugins
underlinking libm.
I have not yet addressed it, but I will.

4) "A segfault is always a bug."
Of course, but I am not reproducing this on any of my systems. One of them (a
mirror of a few customers' system) runs F8 and is working. And kdenlive users
on other distros have no startup segfault like this. I am trying to focus on
the above issues and request help from Fedora users and devs to locate this bug
that seems to be triggered specifically in this environment and/or method of
build.


-- 
Configure bugmail: http://bugzilla.rpmfusion.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
You are the assignee for the bug.