MJPEG Tools Svn

Use HAVE_GETOPT_LONG instead of HAVE_GETLON

General build fixes and preparing for a new release candidate.
* remove .cvsignore from Makefile.am. .cvsignore is a CVS-file, which has no
function whatsoever in a release tarball. So it doesn't need an entry.
* start using noinst_HEADERS and include_HEADERS instead of adding headers
to the appname_SOURCES. We need to be careful with the headers that we
install, and we need to be clear about source requirements for each of
the binaries.
* Add installable include headers for libmpeg2encpp.so. Andrew, please double
check, I simply picked all the ones included in mpeg2enc.cc plus the ones
needed by them.
* Remove all CPU optimization/acceleration code from libmjpegutils.a and put
it in a separate non-installed lib (libcpuaccel.a).
* Prepare all stuff for using libtool. This makes the build easier. However,
it doesn't work yet because it tries gcc instead of nasm on .s sources.
GStreamer doesn't have this, so I need to double check this.
* RPM building is broken. See bug #88417 in RedHat bugzilla.
* Some more...
* Make distcheck now works.

Now using correct *decode* frame rate for calculating the
H:M:S:F timestamp labels on GOPs.

Added new framework for special-purpose syntax constraints for working
around oddities in particular players.
Tidied up header generation so it is all in one place.

Introduce stripe parallelism to make it easier to split encoding phases / backup
in encoding.
Add buffer of output with support for backing up

Remove debugging mjpeg_info I left in inadvertently...

This commit was manufactured by cvs2svn to create branch 'MPEG_DEVEL'.

remove lingering cruft (HAVE_STDINT_H) from pgmtoy4m's prior existence outside mjpegtools

clean up, remove unused cruft from mjpeg_types.h and win32defs.h, HAVE_STDBOOL_H goes away since none of the C code uses it (and C++ has it builtin). mjpeg_types.h is a shadow of its former self ;)

using 0 and 1 obviates the need for stdbool/bool - the HAVE_STDBOOL_H can go away now.