On Fri, 25 Feb 2022, Andreas Rheinhardt wrote:
Martin Storsjö:
On Fri, 25 Feb 2022, Michael Niedermayer wrote:
On Wed, Feb 23, 2022 at 04:29:43PM +0200, Martin Storsjö wrote:
This avoids including version.h in all source files, avoiding
unnecessary rebuilds when the version number is bumped. Only
version_major.h is included by the main header, which defines
availability of e.g. FF_API_* macros, and which is bumped much
less often.
---
fftools/cmdutils.c | 1 +
fftools/ffmpeg.c | 1 +
fftools/ffprobe.c | 1 +
libavcodec/Makefile | 1 +
libavcodec/aacenc.c | 1 +
libavcodec/avcodec.c | 1 +
libavcodec/avcodec.h | 2 +-
libavcodec/codec.h | 2 +-
libavcodec/dpxenc.c | 1 +
libavcodec/j2kenc.c | 1 +
libavcodec/libvorbisenc.c | 1 +
libavcodec/mjpegenc_common.c | 1 +
libavcodec/mpeg4videoenc.c | 1 +
libavcodec/options_table.h | 2 +-
libavcodec/packet.h | 2 +-
libavcodec/pthread_frame.c | 2 +-
libavcodec/tiffenc.c | 1 +
libavcodec/vaapi_encode_h264.c | 1 +
libavcodec/version.h | 28 ++---------------
libavcodec/version_major.h | 55 ++++++++++++++++++++++++++++++++++
libavformat/movenc.c | 1 +
21 files changed, 76 insertions(+), 31 deletions(-)
create mode 100644 libavcodec/version_major.h
not sure i missed some patch but this seems not to build
CC libavcodec/vc2enc.o
libavcodec/vc2enc.c: In function ‘vc2_encode_frame’:
libavcodec/vc2enc.c:957:48: error: ‘LIBAVCODEC_IDENT’ undeclared
(first use in this function); did you mean ‘LIBAVUTIL_IDENT’?
const char *aux_data = bitexact ? "Lavc" : LIBAVCODEC_IDENT;
^~~~~~~~~~~~~~~~
LIBAVUTIL_IDENT
libavcodec/vc2enc.c:957:48: note: each undeclared identifier is
reported only once for each function it appears in
ffbuild/common.mak:78: recipe for target 'libavcodec/vc2enc.o' failed
make: *** [libavcodec/vc2enc.o] Error 1
make: Target 'all' not remade because of errors.
This patch requires adjustments after
155cd6baa49797d57f5b42276eeed3f5408ef3f7 was pushed (it should apply and
build fine on a version slightly older than that), readding an include
of "version.h", I didn't want to spam the list with a full new round of
patches just because of that, but if you want to, I can send my latest
local revision of the patchset.
Sorry for this. My aim was actually to not just remove headers if it
compiles fine without them, but to only include headers that provide
nothing that is used. But apparently I messed it up with
155cd6baa49797d57f5b42276eeed3f5408ef3f7.
No worries, I guess I should have spotted it myself while reviewing that
patch...
// Martin
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
