On Tue, 28 Jun 2016, Anton Khirnov wrote:
Quoting Vittorio Giovara (2016-06-27 17:51:50)
On Mon, Jun 27, 2016 at 11:05 AM, Hendrik Leppkes <[email protected]> wrote:
On Mon, Jun 27, 2016 at 4:22 PM, Diego Biurrun <[email protected]> wrote:
It is also used by the Dirac parsing code.
---
configure | 15 ++++++++-------
libavcodec/Makefile | 1 +
libavcodec/mpeg12data.c | 20 --------------------
libavcodec/mpeg12framerate.c | 39 +++++++++++++++++++++++++++++++++++++++
4 files changed, 48 insertions(+), 27 deletions(-)
create mode 100644 libavcodec/mpeg12framerate.c
diff --git a/configure b/configure
index 1a58a4c..cc649ac 100755
--- a/configure
+++ b/configure
@@ -1705,6 +1705,7 @@ CONFIG_EXTRA="
lzf
me_cmp
mpeg_er
+ mpeg12framerate
mpegaudio
mpegaudiodsp
mpegvideo
Do we really need config entries for 50 line data table files, instead
of making the relevant parts in the Makefile reference those files?
Seems like a lot of noise to create features for single-file
"features" that are just shared tables.
I'd tend to agree, moving the tables to a separate file is okay, but
I'd just add the .o to the Makefile rather than adding an element to
configure.
+1 from me, a "subsystem" for a single short table is really silly.
+2 from me as well; keep in mind that the configure dependency handling
subsystem IIRC has got an O(n^2) complexity, so needlessly increasing the
number of components is to be avoided.
// Martin
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
