MJPEG Tools Svn

Fixed: mquant predictor was being set from suggested quantization
and not the actualy quantization after possible incrmements to
avoid saturation.
Experimental: fixed some obvious oops in the look-ahead rate-controller
adjusted a ridiculous large gain in pass-1 and put in
some additional 'sanity' limits on rate-control feedback.

This commit was manufactured by cvs2svn to create tag 'y4mdenoise_v4_beta1'.

Finally fixed the earthquaking problem in dark/cloudy areas!

The problem was that the match-size-throttle was picking a bad match
before it ever had a chance to find a good match.

Now, the match-size-throttle isn't applied unless the
match-count-throttle is also exceeded, or the end of a line is reached.

This has led to a SERIOUS quality increase, without making y4mdenoise
any slower than it was before!

Fixed typos

Added the information from Roland Stadtmueller
for the manpage of mplex. So users have a idea how to work with the -d|--subpicture-delay option

Mplex support for subtitle streams from tcextract

Mplex support for subtitle streams from tcextract

Changed the package requirement from gtk 1.2 ot 2.0

It was posted in the tracker:
https://sourceforge.net/tracker/?func=detail&atid=105776&aid=2964966&group_id=5776

Detail by Bewo:
The dependency can be found in the file debian/control.
There is the build dependency libgtk1.2-dev listed.

This dependency has to be changed from libgtk1.2-dev to libgtk2.0-dev

Added the Patch from: Phillip Brink

This Patch fixes:
* Fix mjpegtool's #include <SDL/SDL.h>
http://www.libsdl.org/faq.php?action=listentries&category=2#19

Conforming to libsdl's recommendation fixes a compile failure in some build
scenarios too interesting for me to explain them. Please accept this patch
as an improvement to your buildsystem :-).

* Insert SDL_CFLAGS into Makefile.am

fixes compile errors for portage-multilib reported on 2010/03/03 by xming

Added the patch provided by Salah Coronya

It fixes a segmentation fault with jpeg-7 and above where dinfo.do_fancy_upsampling isn't set by default to FALSE anymore.

The original problemreport can be found here:
http://bugs.gentoo.org/show_bug.cgi?id=293919