Commit graph

17815 commits

Author SHA1 Message Date
Mark Nauwelaerts
4a7f899f6e flacparse: fix parsing with unknown framesizes
Fixes #631814 (mostly).
2011-04-08 18:07:10 +01:00
Sebastian Dröge
c3678a5384 flacparse: Simplify frame header parsing by using lookup tables
Based on a patch by Felipe Contreras.

See bug #631200.
2011-04-08 18:07:10 +01:00
Sebastian Dröge
572366e179 flacparse: Don't parse the complete FLAC frames but only look for valid frame headers
Thanks to Felipe Contreras for the suggestion. This is partially
based on his patches and makes flacparse more than 3.5 times faster.

Looking for valid frame headers is unlikely to give false positives
because every frame header is at least 9 bytes long, contains a
14 bit sync code and a 8 bit checksum over the first 8 bytes.

Fixes bug #631200.
2011-04-08 18:07:10 +01:00
Sebastian Dröge
bfdc132ad0 flacparse: Really post tags only after the initial newsegment event
The first newsegment event will be send by the first call to
gst_base_parse_push_buffer() if necessary, posting the tags
before that is not a good idea. Instead do it from the
GstBaseParse::pre_push_buffer vfunc.
2011-04-08 18:07:09 +01:00
Tim-Philipp Müller
6b71b65289 Revert "baseparse: add skip property"
This reverts commit b5a3d60363.

Reverting this for now, since no one really seems to remember why this
property exists or what it could possibly be good for. It seems to have
been in the original mp3parse since the beginning of time and was back-
ported from there.
2011-04-08 18:07:09 +01:00
Sebastian Dröge
872b49efa7 flacparse: Fix uninitialized variable compiler warnings
These warnings are wrong, the variables are only used if they were
initialized by the bit reader.
2011-04-08 18:07:09 +01:00
Felipe Contreras
70d5346ee8 flacparse: fix picture parsing
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2011-04-08 18:07:09 +01:00
Sebastian Dröge
36f48450b1 flacparse: Push tags before the header buffers are pushed 2011-04-08 18:07:09 +01:00
Felipe Contreras
eaa6f789b2 flacparse: trivial caps fix
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2011-04-08 18:07:09 +01:00
Sebastian Dröge
4cc901ed96 audioparser: Let the format string agree with the parameters to fix compiler warning 2011-04-08 18:07:09 +01:00
Sebastian Dröge
6f04047f0c ac3parse: Use unchecked versions of the bitreader get functions
We didn't check the return values anyway...
2011-04-08 18:07:09 +01:00
Arun Raghavan
8ef15e6247 baseparse: Fix debug output
We lose the reference to the buffer after gst_pad_push(), so the debug
print should happen before.

https://bugzilla.gnome.org/show_bug.cgi?id=622276
2011-04-08 18:07:09 +01:00
Mark Nauwelaerts
fbbc09e47d audioparsers: add flacparse unit test
... and tweak parser test helper in the process.
2011-04-08 18:07:09 +01:00
Mark Nauwelaerts
0f5f06d5fa baseparse: support reverse playback
... in pull mode or upstream driven.
2011-04-08 18:07:09 +01:00
Mark Nauwelaerts
304b3c78dd baseparse: remove done TODOs and update documentation 2011-04-08 18:07:09 +01:00
Mark Nauwelaerts
1a54736255 baseparse: use determined seekability in answering SEEKING query 2011-04-08 18:07:09 +01:00
Mark Nauwelaerts
3f02dbd3ab baseparse: add skip property 2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
f8ec616d9d audioparsers: add ac3parse and mpegaudioparse unit test 2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
dd140c7b4d mpegaudioparse: initial version
... adequately equivalent to mp3parse, so lets boldly set it
to higher rank.
2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
2d482deabe aacparse: set minimum frame size at _start
... rather than one time at _init.
2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
edb495646d audioparsers: refactor existing unit tests using common helper 2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
589c455ccb baseparse: use _set_frame_props to configure frame lead_in and lead_out
... provided a corresponding decoder with sufficient leading and following
frames to carry out full decoding for a particular segment.
2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
f218b984cc baseparse: use _set_duration to configure duration update interval
... as it logically belongs there as one or the other; either subclass
can provide a duration, or an estimate must be made (reguarly updated).
2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
1f7359d388 baseparse: localize use of provided fps information 2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
5dd8a489d5 baseparse: seek table and accurate seek support 2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
5a69aa4903 baseparse: proper and more extended segment and seek handling
That is, loop pause handling, segment seek support, newsegment for gaps, etc
2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
f16146df62 baseparse: add index support 2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
261b7ec15f baseparse: refactor state reset 2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
2c81b3bf77 baseparse: prevent indefinite resyncing 2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
de189e327f baseparse: specific EOS handling if no output so far 2011-04-08 18:07:08 +01:00
Mark Nauwelaerts
0707c189db baseparse: adjust _set_frame_prop documentation and set default as claimed 2011-04-08 18:07:07 +01:00
Mark Nauwelaerts
723059c688 baseparse: fix bitrate copy-and-paste and update heuristic 2011-04-08 18:07:07 +01:00
Mark Nauwelaerts
fea135b3de baseparse: post duration message if average bitrates is updated 2011-04-08 18:07:07 +01:00
Mark Nauwelaerts
0198d6320b baseparse: remove is_seekable vmethod and use a set_seek instead
Seekability, like duration, etc is unlikely to change (frequently), and
the default assumption covers most cases, so let subclass set when needed.
At the same time, allow subclass to indicate if it has seek-metadata (table)
available, and possibly have it provide an average bitrate.
2011-04-08 18:07:07 +01:00
Mark Nauwelaerts
10611c2826 ac3parse: remove redundant default is_seekable 2011-04-08 18:07:07 +01:00
Mark Nauwelaerts
30be03004e baseparse: add another hook for subclass prior to pushing buffer
... and allow subclass to perform custom segment clipping, or to
emit tags or messages at this time.
2011-04-08 18:07:07 +01:00
Mark Nauwelaerts
dc6bd085ed baseparse: 0 converts to 0 by default 2011-04-08 18:07:07 +01:00
Mark Nauwelaerts
0e60101a28 basepase: refactor conversion using helper function and export default convert 2011-04-08 18:07:07 +01:00
Mark Nauwelaerts
a84eee533f baseparse: streamline query handling 2011-04-08 18:07:07 +01:00
Mark Nauwelaerts
1c65b8287d baseparse: cleanup struct and remove unused member 2011-04-08 18:07:07 +01:00
Mark Nauwelaerts
952b27a140 audioparsers: increase ranks to enable auto-plugging
Because we can, and should, have some shakedown testing before having
these make it into -good later on ...
2011-04-08 18:07:07 +01:00
Arun Raghavan
3448e83c73 baseparse: Allow chaining of subclass event handlers
This allows the child class to chain its event handler with
GstBaseParse, so that subclasses don't have to duplicate all the default
event handling logic.

https://bugzilla.gnome.org/show_bug.cgi?id=622276
2011-04-08 18:07:07 +01:00
Sebastian Dröge
8df3c8477a baseparse: Don't use GST_FLOW_IS_FATAL()
Also don't post an error message for UNEXPECTED and do it
for NOT_LINKED.
2011-04-08 18:07:06 +01:00
Mark Nauwelaerts
7831661c40 baseparse: non-TIME seek event is simply not handled 2011-04-08 18:07:06 +01:00
Mark Nauwelaerts
835473b792 baseparse: fix seek event ref handling 2011-04-08 18:07:06 +01:00
Mark Nauwelaerts
baefba9748 baseparse: prevent arithmetic overflows in pull mode buffer cache handling 2011-04-08 18:07:06 +01:00
Mark Nauwelaerts
a459494097 baseparse: fix seek handling
Allow a few more seek event type combinations, and really use the result
of gst_segment_set_seek to perform the seek.  Also add some debug.
2011-04-08 18:07:06 +01:00
Edward Hervey
6bd1be78bf check: Don't re-declare 'GList *buffers' in the tests
It's an external which lives in gstcheck.c. Redeclaring it makes some
compilers/architectures think the 'buffers' in the individual tests are
a different symbol... and therefore we end up comparing holodecks with
oranges.
2011-04-08 18:07:06 +01:00
Arun Raghavan
f7e228edb0 baseparse: Don't emit bitrate tags too early
We wait to parse a minimum number of frames (10, arbitrarily) before
emiting bitrate tags so that our early estimates are not wildly
inaccurate for streams that start with a silence. If the stream ends
before that, we just emit the tags anyway.

While it _would_ be nicer to be specify the threshold to start pushing
the tags in terms of duration, this would introduce more complexity than
this merits.

https://bugzilla.gnome.org/show_bug.cgi?id=614991
2011-04-08 18:07:06 +01:00
Sebastian Dröge
00b0444618 flacparse: Optionally check the overall frame checksums too before accepting a frame as valid
This is optional because it's a quite expensive operation and it's very
unlikely that a non-frame is detected as frame after the header CRC check
and checking all bits for valid values. The overall frame checksums are
mainly useful to detect inconsistencies in the encoded payload.
2011-04-08 18:07:06 +01:00