| Age | Commit message (Collapse) | Author |
|---|
|
-- Move some generic checks out of m4/input.m4 and back into configure.ac
-- Remove ioctl_request.m4 and inline it in configure.ac
-- Fix compilation of internal libcdio and libvcd stuff -- now works for Darwin
-- Add a switch for enabling/disabling dvb support. Disabled by default for
all platforms except for Linux. Only allowed to be used on Linux.
-- Disable vcd support on all platforms except for FreeBSD, Linux, and Solaris,
because src/input/input_vcd.c is only supported on those platforms even
though libcdio and libvcd will compile on other platforms
-- Clean up some summary.m4 stuff
|
|
-- Create ACX_PACKAGE_CHECK to remove duplicate code for aalib, arts, dvdnav,
and libfame. Remove aa.m4, arts.m4, dvdnav.m4, and libfame.m4
-- ACX_PACKAGE_CHECK does not require compiling and running a test program.
Stole version code from autogen.sh
|
|
|
|
with versions of autoconf that do not have AC_PROG_OBJC. Pull missing bits
from autoconf 2.61, guard with m4_ifdef, and forcibly m4_include m4/objc.m4
into configure.ac, because aclocal won't pick it up with the m4_ifdef present.
|
|
|
|
It seems that even embedded in a shell if-statement, AC_PATH_XTRA still runs
before PKG_CHECK_MODULES, even if PKG_CHECK_MODULES comes first. There's
something funky going on inside autoconf wrt AC_PATH_XTRA. I know that the
intended behavior is to do the PKG_CHECK_MODULES first, but no matter what I
do, it never actually works that way. I give up, and so here's the code
written to reflect what _actually_ happens in that order that it happens.
|
|
-- Move x11.m4 back into configure.ac, because AC_PATH_XTRA was behaving oddly
inside a macro. That code is now cleaned up enough so that it's not too
painful having it there.
-- Move opengl.m4 inline into video_out.m4
-- Fix x11 checks to work on Darwin, including XShm checks, Xv checks, OpenGL,
and GLU checks.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
-- Make a pass at cleaning up contrib/ Makefile.am files
-- Rename many AM_CONDITIONALS for consistency and give them better names to
better indicate what they mean
-- Remove some dead input plugin tests
-- Clean up AM_PATH_DVDNAV
|
|
|
|
|
|
|
|
Move the definition of it to a much more appropriate location.
|
|
instead of building from top_srcdir. Also, use make -C instead of cd && make
|
|
|
|
If AC_SYS_LARGEFILES defines _FILE_OFFSET_BITS=64 or _LARGEFILE_SOURCE, they'll
only be in config.h. Add them to CPPFLAGS as well.
If AC_FUNC_FSEEKO defines _LARGE_FILES, add it to CPPFLAGS as well.
Do all of this for the benefit of those things that do not #include config.h
|
|
This is a transitional macro that's normally defined by the system includes if
_GNU_SOURCE is defined, which it will be because of AC_GNU_SOURCE. It's only
used by GNU libc. I cannot find anything in any of the code built by these
Makefiles that uses any of the types, structures, or functions that defining
this macro produces.
|
|
|
|
|
|
-- Create programs.m4 to hold program related checks
-- Move the always inline check macro (currently dead) into attributes.m4
-- Rename the remainder of _xine.m4 to types.m4, because there's just type
related checks left.
--HG--
rename : m4/_xine.m4 => m4/types.m4
|
|
|
|
-- Split input plugin checks out of configure.ac into m4/input.m4
-- Move m4/vcd.m4 (previously split out from configure.ac) into m4/input.m4
-- Fold m4/dl.m4 into configure.ac
-- Various miscellaneous cleanups throughout configure.ac and m4/*.m4
|
|
|
|
|
|
|
|
|
|
-- Revisit NLS stuff. Not much changed, remove the REVISIT tag
-- Change XDG_BASEDIR_CFLAGS to XDG_BASEDIR_CPPFLAGS, because it only includes
a -I, which is a CPPFLAGS option, not CFLAGS
-- Move XDG stuff up to library checks
-- Cleanup and move zlib checks
-- Move X11 checks up to checks for system services
-- Break audio out plugin stuff into m4/audio_out.m4
-- Break video out plugin stuff into m4/video_out.m4
-- Move libdir name stuff into m4/libFLAC.m4 since that's the only thing that
uses it.
|
|
|
|
|
|
|
|
|
|
|
|
-- Removed m4/optimizations.m4, and along with it, AC_OPTIMIZATIONS.
-- Stripped down, cleaned up, and merged the former together with other compiler
characteristic checks.
-- Do not set any optimization flags into CFLAGS. Update all Makefile.am's with
proper AM_CFLAGS, AM_CPPFLAGS, AM_LDFLAGS, etc. to set up optimization flags,
usually from DEFAULT_OCFLAGS.
-- Start cleaning up CFLAGS/CPPFLAGS stuff in Makefile.am's all over the place.
-- Correct a number of places where CFLAGS itself was being mangled in
Makefile to on-the-fly adjust optimizations to work around compiler bugs.
This stuff is now done correctly.
-- The run of automake from autogen.sh is now clean of warnings.
-- Cleaned out some (now) dead macros from m4/_xine.m4
-- Mac OS X intel builds out-of-the-box now -- dropped optimization on
post/deinterlace/plugins/kdetv-greedyh to O1_CFLAGS.
-- OBJCFLAGS is now getting set correctly everywhere that it needs to be
-- Various other miscellaneous cleanups all over
|
|
|
|
-- Move Windows port checks to an earlier point, but after AC_PROG_CC and
AC_PROG_EGREP are done, because it depends on both of them. Also
rename SYS to WIN32_SYS, because the former is way too generic and prone
to clashes from elsewhere.
-- Removed some dead automake work-arounds that should no longer be necessary
with the new automake 1.9 requirement
-- For Mac OS X universal builds, don't set CFLAGS/OBJCFLAGS/LDFLAGS so early
with multiple -arch flags. Set them at the end from UNIVERSAL_CFLAGS and
UNIVERSAL_LDFLAGS.
-- Move set of INCLUDES down to final set of CFLAGS/OBJCFLAGS/LDFLAGS.
|
|
|
|
-- Begin re-ordering tests as appropriate to match the recommended autoconf
layout
-- Remove dead tests, replacing some tests with standard existing autoconf
tests added in 2.59 (or possibly earlier)
-- Use $host_or_hostalias in more places instead of $host
-- Set OBJCFLAGS when CFLAGS is set just before producing output
|
|
-- Pull AC_PROG_OBJC and AC_PROG_OBJCPP from autoconf 2.61, but use the local
versions only if they're not defined by the version of autoconf in use.
-- Always call AC_PROG_OBJC since there's no harm in doing so. It'll fail most
of the time, but it's cleaner than adding yet another $host_os check
-- Remove -DCONFIG_DARWIN from CPPFLAGS since it was added specifically for
ffmpeg when ffmpeg was built using xine's built infrastructure rather than
its own. It's now defined properly where its needed elsewhere.
-- Set host_or_hostalias early rather than later in the middle of xv checking
since a bunch of other stuff depends on it too.
|
|
- Removed debug related targets from all Makefiles. Debug builds are now fully
controlled by --enable-debug passed to configure
- Disable optimizations by default when configuring with --enable-debug
- Added --enable-profiling for building profile builds. This has problems with
optimizations currently, but that'll get taken care of over time.
- Initialize ASFLAGS/CFLAGS/CPPFLAGS/OBJCFLAGS/LDFLAGS early so that autoconf
defaults don't come into play.
- Added some additional commentary to configure.ac
|
|
-- Execute version.sh from configure.ac to get version information. Update the
release version in this file now instead of configure.ac.
-- Clean up a bunch of cruft hanging around and consolidate version information
as much as possible.
-- Move autoconf/autoheader/automake initialization together.
-- Use AM_MAINTAINER_MODE
|
|
|
|
|
|
Xv drivers are permitted to return a bigger image than we asked for, to
work around hardware constraints. If this happens, 1.1.6's video_out_xv
cannot deinterlace properly. Fix this by deinterlacing based on the
width of the Xv image; the Xv driver will discard the extra horizontal data.
Without this patch, you get a barber-pole effect if the input video is
interlaced *and* the Xv driver chooses to round up the width of the
requested image.
--
Simon Farnsworth
|
