In ID3 v2.3 compressed frames will have a 4-byte data length indicator
after the frame header to indicate the size of the decompressed data.
This integer is unlikely to be a sync-safe integer for v2.3 tags,
only in v2.4 it's sync-safe.
Reversing the unsynchronisation seems to work slightly differently
for ID3 v2.3 tags and v2.4 tags: v2.3 tags don't have syncsafe frame
sizes in the frame header, so the unsynchronisation is applied to
the whole frame data including all the frame headers. v2.4 frames
have sync-safe sizes, however, so the unsynchronisation only needs
to be applied to the actual frame data, and it seems that's what's
being done as well. So we need to undo the unsynchronisation on a
per-frame basis for v2.4 tags for things to work properly.
Fixes extraction of coverart/images from APIC frames in ID3 v2.4
tags (#588148).
Add unit test for this as well.
Whenever we see a gap, we flush the temporary packets (but not the adapter). If we
had some data temporarily stored it will be outputted (the sound will sound a bit
garbled... but that's how it sounds on MacOSX :)
Reverse-engineered by comparing:
* A rtp hinted file provided by DarwinStreamingServer
* The output procued by DSS for that same file
Also used various streaming sources available on the internet to fine-tune
the code.
The header/codec_data extraction methods are from FFMpeg (LGPL).
Use some of the SDP attributes when they are present to specify the output
dimension and framerate. This allows us to receive jpeg frames larger than
2040 width/height.
Fixes#564437
sizeof("foo") includes the string's NUL-terminator in the size returned,
but we're writing strings here with an explicit size at the beginning
and no NUL-terminator. In most cases using sizeof("foo") as length in
memcpy is not harmful, but it is where the string goes right at the
end of our buffer to write, since we don't allocate space for that
NUL terminator.
These filters use information from previous frames to
generate the current frame and a caps change will make
the effect start from the beginning again.
This produces a water ripple effect on the video input,
based on motion or a rain drop algorithm.
Kindly relicensed to LGPL2+ by Kentaro Fukuchi <fukuchi@megaui.net>.
Fixes bug #588695.
This combines the StreakTV and BaltanTV filters from the
effectv project.
Kindly relicensed to LGPL2+ by Kentaro Fukuchi <fukuchi@megaui.net>.
Fixes bug #588368.
This filter adds a radiation-like motion blur effect
to the video stream.
Kindly relicensed to LGPL2+ by Kentaro Fukuchi <fukuchi@megaui.net>.
Fixes bug #588359.
This filter binarizes input frames and combines them with various
optical pattern.
Kindly relicensed to LGPL2+ by Kentaro Fukuchi <fukuchi@megaui.net>.
Fixes bug #588349.
When blending a source layer with an alpha of 'a' on top of another
destination layer we take the sum of:
* 'a' percent of the source layer
* (100 - 'a') percent of the destination layer (the remainder)
Once buffering has started (with an mdat atom), continue buffering
until moov atom is reached, which handles cases with multiple
mdat atoms. Also keep adapter/offset better in sync with upstream
and fix some debug statements. Fixes#587426.
This reverts commit 5503a59a57.
Reverting this since it causes regressions with a lot of sample files
I have, all of which worked fine with the last -good release (#586891).
Whenever we alloc something based on a user-supplied size, we should
really use g_try_new(), otherwise we can easily be made to abort by
passing a ridiculously large number to us for allocing. Fixes
problems with some fuzzed files.
Check the possibly 64-bit atom size more carefully before casting it
to an int and passing it to gst_pad_pull_range(), otherwise we might
end up pulling 0 bytes, getting an empty buffer as requested and
dereferencing not available data whilst thinking we actually asked
for and got 0x1000000000000 bytes. Similar fix for push mode operation
where neededbytes ends up being 0 bytes, which makes us assert. Fixes
crash with broken or fuzzed file (NB #122378).
Fix the caps to include the depth (instead of width twice) in the caps of
audio/x-raw-int.
Fix negotiation to not only copy the rate/channels of the first structure.
Don't call gst_avi_demux_src_convert() for each single index entry. Not
only do we already have the pointer to the stream context, we also know
the formats we want to convert from and to already, so we may just as
well use optimised conversion routines that bypass some of the checks
and lookups made in gst_avi_demux_src_convert().
Include the header from where we include all the system headers with the
socket stuff before we try to define EAI_ADDRFAMILY ourselves, otherwise
we define it ourselves and then get a compiler warning if a system header
defines it as well without guarding against it being defined already.
Don't leak buffers when resyncing to a keyframe.
Avoid leaking buffers when exiting the loop on error conditions.
Add some more debug info.
Fixes#585911
The previous patch to add support for additional sample formats possibly
introduced a reentrancy bug: a variable used for a loop index was declared
static. This patch fixes that, and also adds a "/* *INDENT-ON* */" annotation
following the macro block. (I don't know what the annotation is for, but the
adder, where I copied this from, has it).
Setting it to a value<16 would cause crashes before because
current_plane was set to the old number of planes-1. Also
fix calculations for non-2^n planes values.
The diff is a signed integer, not an unsigned one of course.
In modes other than GST_DEINTERLACE_ALL every frame has twice the
duration of the field duration.
When there are less timestamps that there are samples, fill up the sample table
with the last know timestamp. This situation can happen when the last sample
does not decode and doesn't need a timestamp. We however calculate the total
track length using the last sample timestamp so we need to have something
sensible in there.
Fixes#585056
in24 samples are normally big-endian but an enda box can change this to
little-endian. Recurse into the in24 box and find the enda box so that we get
the endianness right.
Fixes#582515
gst_adapter_take_buffer doesn't allow buffer to be empty.
Simply skip any part where the content is empty. Don't
create a pad for it either.
See #582169
Original commit message from CVS:
* gst/y4m/gsty4mencode.c: (gst_y4m_encode_init),
(gst_y4m_encode_setcaps):
* tests/check/elements/y4menc.c: (GST_START_TEST):
Plug some leaks; try to make build bot happy again.
Original commit message from CVS:
* ext/dv/gstdvdec.c:
* gst/subparse/gstsubparse.c: (parse_mdvdsub):
* gst/y4m/gsty4mencode.c: (gst_y4mencode_sinkconnect):
I'm a bad boy. using /1001. to force C to do float division
and not integer division (as it did in my last commit)
Thanks to David I. Lehn for pointing this mistake.
Original commit message from CVS:
* ext/dv/gstdvdec.c:
* ext/libfame/gstlibfame.c:
* gst/subparse/gstsubparse.c: (parse_mdvdsub):
* gst/y4m/gsty4mencode.c: (gst_y4mencode_sinkconnect):
replace framerate aproximations by their real value
(24000/1001, 30000/1001, 60000/1001)
Finish fixing bug #164049
Original commit message from CVS:
a52dec: Use a debug category, Output timestamps correctly
Emit tag info, Handle events, tell liba52dec about cpu
capabilities so it can use MMX etc.
dvdec: Fix a crasher accessing invalid memory
dvdnavsrc:Some support for byte-format seeking.
Small fixes for still frames and menu button overlays
mpeg2dec: Use a debug category. Adjust the report level of several items to
LOG. Call mpeg2_custom_fbuf to mark our buffers as 'custom buffers'
so it doesn't lose the GstBuffer pointer
navseek: Add the navseek debug element for seeking back and forth in a
video stream using arrow keys.
mpeg2subt:Pretty much a complete rewrite. Now a loopbased element. May still
require work to properly synchronise subtitle buffers.
mpegdemux:
dvddemux: Don't attempt to create subbuffers of size 0
Reduce a couple of error outputs to warnings.
y4mencode:Output the y4m frame header correctly
Original commit message from CVS:
Remove all config.h includes from header files, add it to each source file and remove duplicate config.h includes from several source files
Original commit message from CVS:
New mimetypes gone into effect today - this commit changes all old mimetypes over to the new mimetypes spec as described in the previous commit's document. Note: some plugins will break, some pipelines will break, expect HEAD to be broken or at least not 100% working for a few days, but don't forget to report bugs
Original commit message from CVS:
compatibility fix for new GST_DEBUG stuff.
Includes fixes for missing includes for config.h and unistd.h
I only ensured for plugins I can build that they work, so if some of them are still broken, you gotta fix them yourselves unfortunately.
Original commit message from CVS:
another batch of connect->link fixes
please let me know about issues
and please refrain of making them yourself, so that I don't spend double
the time resolving conflicts
Original commit message from CVS:
* a hack to work around intltool's brokenness
* a current check for mpeg2dec
* details->klass reorganizations
* an element browser that uses details->klass
* separated cdxa parse out from the avi directory
Original commit message from CVS:
* removal of //-style comments
* don't link plugins to core libs -- the versioning is done internally to the plugins with the plugin_info struct,
and symbol resolution is lazy, so we can always know if a plugin can be loaded by the plugin_info data. in theory.
Original commit message from CVS:
s/@GST_PLUGIN_LDFLAGS@/$(GST_PLUGIN_LDFLAGS)/
@-substitued variables variables are defined as make variables automagically,
and this gives the user the freedom to say make GST_PLUGIN_LDFLAGS=-myflag
For this add a "mode" property that defaults to "interlaced" for now as
most decoders/demuxers don't properly set the "interlaced" field on the
caps yet.
If this property is set to "auto" the element will work in passthrough
mode unless the caps contain the "interlaced" field.
Original commit message from CVS:
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace2_class_init), (gst_deinterlace2_init),
(gst_deinterlace2_set_property), (gst_deinterlace2_get_property):
Bring properties into this century.
Original commit message from CVS:
* gst/deinterlace2/tvtime/tomsmocomp.c:
(gst_deinterlace_method_tomsmocomp_class_init):
Fix unused variable compiler warning when not building
X86 assembly.
Original commit message from CVS:
* gst/dccp/gstdccp.c:
* gst/dccp/gstdccpclientsrc.c:
Fix compilation on Solaris by including filio.h as needed.
* gst/deinterlace2/tvtime/tomsmocomp/TomsMoCompAll.inc:
* gst/deinterlace2/tvtime/tomsmocomp/TomsMoCompAll2.inc:
Fix compilation with Forte - apparently it hates concatenating a
macro argument that starts with an underscore??
Original commit message from CVS:
* gst/deinterlace2/tvtime/tomsmocomp/SearchLoopBottom.inc:
* gst/deinterlace2/tvtime/tomsmocomp/SearchLoopTop.inc:
* gst/deinterlace2/tvtime/tomsmocomp/StrangeBob.inc:
* gst/deinterlace2/tvtime/tomsmocomp/WierdBob.inc:
Unroll the loop to handle two bytes at once. This should give
a small speedup and makes it possible to handle chroma and luma
different which is needed later.
Original commit message from CVS:
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace_method_class_init):
* gst/deinterlace2/gstdeinterlace2.h:
* gst/deinterlace2/tvtime/tomsmocomp.c:
(gst_deinterlace_method_tomsmocomp_class_init):
* gst/deinterlace2/tvtime/tomsmocomp/SearchLoopBottom.inc:
* gst/deinterlace2/tvtime/tomsmocomp/SearchLoopTop.inc:
* gst/deinterlace2/tvtime/tomsmocomp/StrangeBob.inc:
* gst/deinterlace2/tvtime/tomsmocomp/TomsMoCompAll.inc:
* gst/deinterlace2/tvtime/tomsmocomp/TomsMoCompAll2.inc:
* gst/deinterlace2/tvtime/tomsmocomp/WierdBob.inc:
* gst/deinterlace2/tvtime/tomsmocomp/tomsmocompmacros.h:
First part of the C implementation of the tomsmocomp deinterlacing
algorithm. This only supports search-effort=0 currently, is painfully
slow and needs some cleanup later when all search-effort settings
are implemented in C.
Original commit message from CVS:
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace_simple_method_interpolate_scanline),
(gst_deinterlace_simple_method_copy_scanline),
(gst_deinterlace_simple_method_deinterlace_frame):
* gst/deinterlace2/tvtime/greedy.c: (deinterlace_frame_di_greedy):
* gst/deinterlace2/tvtime/greedyh.c:
(deinterlace_frame_di_greedyh):
* gst/deinterlace2/tvtime/scalerbob.c:
(deinterlace_scanline_scaler_bob):
* gst/deinterlace2/tvtime/tomsmocomp.c: (Fieldcopy):
* gst/deinterlace2/tvtime/weave.c: (deinterlace_scanline_weave),
(copy_scanline):
* gst/deinterlace2/tvtime/weavebff.c: (deinterlace_scanline_weave),
(copy_scanline):
* gst/deinterlace2/tvtime/weavetff.c: (deinterlace_scanline_weave),
(copy_scanline):
Use oil_memcpy() instead of memcpy() as it's faster for the sizes that
are usually used here.
Original commit message from CVS:
* gst/deinterlace2/tvtime/vfir.c: (deinterlace_line_c),
(deinterlace_line_mmx), (gst_deinterlace_method_vfir_class_init):
Implement the VFIR deinterlacing method as simple method.
Original commit message from CVS:
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace_simple_method_interpolate_scanline),
(gst_deinterlace_simple_method_copy_scanline),
(gst_deinterlace_simple_method_deinterlace_frame),
(gst_deinterlace_simple_method_class_init),
(gst_deinterlace_simple_method_init):
* gst/deinterlace2/gstdeinterlace2.h:
Add a GstDeinterlaceSimpleMethod subclass of GstDeinterlaceMethod that
can be used by simple deinterlacing methods. They only have to provide
a function for interpolating a scanline or copying a scanline.
Original commit message from CVS:
* gst/deinterlace2/gstdeinterlace2.c: (gst_deinterlace2_chain):
Respect the latency of the deinterlacing algorithm for the timestamps
of every buffer.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedyh.asm:
* gst/deinterlace2/tvtime/tomsmocomp/SearchLoopBottom.inc:
Add the MMX registers to the clobbered registers only if __MMX__ is
defined.
Original commit message from CVS:
* gst/deinterlace2/Makefile.am:
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace2_methods_get_type), (gst_deinterlace2_set_method),
(gst_deinterlace2_class_init):
Enable tomsmocomp again as the C port will be ready for the next
release.
Original commit message from CVS:
* gst/deinterlace2/gstdeinterlace2.c: (gst_deinterlace2_init),
(gst_greatest_common_divisor), (gst_fraction_double),
(gst_deinterlace2_getcaps), (gst_deinterlace2_setcaps):
Don't use proxy_getcaps() but implement our own getcaps() function
that doubles/halfs the framerate if all fields should be sent out.
Original commit message from CVS:
* configure.ac:
* gst/deinterlace2/Makefile.am:
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace2_methods_get_type), (gst_deinterlace2_set_method),
(gst_deinterlace2_class_init), (gst_deinterlace2_init):
* gst/deinterlace2/gstdeinterlace2.h:
* gst/deinterlace2/tvtime/greedy.c:
(gst_deinterlace_method_greedy_l_class_init):
* gst/deinterlace2/tvtime/greedyh.c:
(gst_deinterlace_method_greedy_h_class_init):
* gst/deinterlace2/tvtime/vfir.c:
(gst_deinterlace_method_vfir_class_init):
Disable the tomsmocomp algorithm for this release as it's buggy
and has no C implementation yet.
Build the deinterlace2 plugin on all architectures but still mark it
as experimental.
Build the x86 inline assembly only if GCC inline assembly is supported
and only on x86 or amd64. Fixes bug #543286.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedyh.asm:
Always use the C implementation if width is not a multiple of 4. The
assembly optimized version only handle this and calling the C
implementation for the remaining part doesn't work because it needs
previous calculations.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedyh.asm:
* gst/deinterlace2/tvtime/greedyh.c:
* gst/deinterlace2/tvtime/greedyhmacros.h:
Some cleanup, use 3DNOW instead of TDNOW in macros.
* gst/deinterlace2/tvtime/tomsmocomp.c:
(gst_deinterlace_method_tomsmocomp_class_init):
* gst/deinterlace2/tvtime/tomsmocomp/TomsMoCompAll.inc:
* gst/deinterlace2/tvtime/tomsmocomp/tomsmocompmacros.h:
The SSE method in fact only needs MMXEXT, declare it as such.
Original commit message from CVS:
* ext/spc/gstspc.c: (spc_setup):
* gst/deinterlace2/tvtime/tomsmocomp/SearchLoopTop.inc:
Don't use declarations after statements in the remaining code.
Original commit message from CVS:
* gst/deinterlace2/tvtime/tomsmocomp/TomsMoCompAll2.inc:
Mark internal processing functions as static inline for quite some
speedup as they're used only once and need to get many local variables
passed as parameter.
Original commit message from CVS:
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace_method_get_latency),
(gst_deinterlace2_set_method), (gst_deinterlace2_class_init),
(gst_deinterlace2_push_history), (gst_deinterlace2_chain),
(gst_deinterlace2_setcaps), (gst_deinterlace2_src_query):
* gst/deinterlace2/gstdeinterlace2.h:
Include latency of the method in the returned latency.
Fix outputting of all fields, i.e. doubling of the framerate.
Original commit message from CVS:
* gst/deinterlace2/Makefile.am:
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace_method_class_init), (gst_deinterlace_method_init),
(gst_deinterlace_method_deinterlace_frame),
(gst_deinterlace_method_get_fields_required),
(gst_deinterlace2_methods_get_type), (_do_init),
(gst_deinterlace2_set_method), (gst_deinterlace2_class_init),
(gst_deinterlace2_child_proxy_get_child_by_index),
(gst_deinterlace2_child_proxy_get_children_count),
(gst_deinterlace2_child_proxy_interface_init),
(gst_deinterlace2_init), (gst_deinterlace2_finalize),
(gst_deinterlace2_chain), (gst_deinterlace2_src_query):
* gst/deinterlace2/gstdeinterlace2.h:
* gst/deinterlace2/tvtime/greedy.c:
(deinterlace_greedy_packed422_scanline_c),
(deinterlace_greedy_packed422_scanline_mmx),
(deinterlace_greedy_packed422_scanline_mmxext),
(deinterlace_frame_di_greedy),
(gst_deinterlace_method_greedy_l_set_property),
(gst_deinterlace_method_greedy_l_get_property),
(gst_deinterlace_method_greedy_l_class_init),
(gst_deinterlace_method_greedy_l_init):
* gst/deinterlace2/tvtime/greedyh.asm:
* gst/deinterlace2/tvtime/greedyh.c: (greedyDScaler_C),
(deinterlace_frame_di_greedyh),
(gst_deinterlace_method_greedy_h_set_property),
(gst_deinterlace_method_greedy_h_get_property),
(gst_deinterlace_method_greedy_h_class_init),
(gst_deinterlace_method_greedy_h_init):
* gst/deinterlace2/tvtime/greedyh.h:
* gst/deinterlace2/tvtime/plugins.h:
* gst/deinterlace2/tvtime/tomsmocomp.c:
(gst_deinterlace_method_tomsmocomp_set_property),
(gst_deinterlace_method_tomsmocomp_get_property),
(gst_deinterlace_method_tomsmocomp_class_init),
(gst_deinterlace_method_tomsmocomp_init):
* gst/deinterlace2/tvtime/tomsmocomp.h:
* gst/deinterlace2/tvtime/tomsmocomp/TomsMoCompAll.inc:
* gst/deinterlace2/tvtime/vfir.c: (deinterlace_frame_vfir),
(gst_deinterlace_method_vfir_class_init),
(gst_deinterlace_method_vfir_init):
Use a GstObject subtype for the deinterlacing methods and export
the different settings for each deinterlacing method via GObject
properties.
Implement GstChildProxy interface to allow access to the used
deinterlacing method and to allow adjusting the different settings.
Move global variables of the tomsmocomp deinterlacing method into
function local variables to make it possible to use this deinterlacing
method from different instances.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedyh.asm:
Support widths that are not a multiply of 4 when using the assembly
optimized greedyh implementations.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedyh.c:
(deinterlace_frame_di_greedyh):
Only build the assembly optimized implementations on x86.
Original commit message from CVS:
* gst/deinterlace2/Makefile.am:
* gst/deinterlace2/tvtime/tomsmocomp.c: (tomsmocomp_init),
(tomsmocomp_filter_mmx), (tomsmocomp_filter_3dnow),
(tomsmocomp_filter_sse), (deinterlace_frame_di_tomsmocomp):
* gst/deinterlace2/tvtime/tomsmocomp.h:
Remove useless file and mark everything possible as static.
* gst/deinterlace2/tvtime/greedy.c:
* gst/deinterlace2/tvtime/greedyh.c:
Use "_stdint.h" instead of <stdint.h>.
Original commit message from CVS:
* gst/deinterlace2/Makefile.am:
* gst/deinterlace2/gstdeinterlace2.c: (gst_deinterlace2_init):
* gst/deinterlace2/tvtime/greedy.c: (deinterlace_frame_di_greedy):
* gst/deinterlace2/tvtime/greedyh.c:
(deinterlace_frame_di_greedyh):
* gst/deinterlace2/tvtime/speedtools.h:
* gst/deinterlace2/tvtime/speedy.c:
* gst/deinterlace2/tvtime/speedy.h:
* gst/deinterlace2/tvtime/tomsmocomp.c: (Fieldcopy):
* gst/deinterlace2/tvtime/tomsmocomp/TomsMoCompAll.inc:
* gst/deinterlace2/tvtime/vfir.c: (deinterlace_frame_vfir):
Get rid of speedy.[ch] as we don't use most of it's code anyway
and it doesn't seem to be relicensed to LGPL. Use memcpy() instead
of the speedy memcpy everywhere instead.
* gst/deinterlace2/gstdeinterlace2.h:
Remove many unused declarations.
Original commit message from CVS:
* gst/deinterlace2/gstdeinterlace2.c: (gst_deinterlace2_src_query):
Divide latency be 2 to convert from fields to frames.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedy.c:
(deinterlace_greedy_packed422_scanline_c),
(deinterlace_greedy_packed422_scanline_mmx),
(deinterlace_greedy_packed422_scanline_mmxext),
(deinterlace_frame_di_greedy):
Don't use scanlines function from gstdeinterlace2 as it's
not appropiate for this method. Instead implement deinterlace_frame
function by taking the one from greedyh.
* gst/deinterlace2/tvtime/greedyh.c: (greedyDScaler_C):
Small fix for the C implementation.
* gst/deinterlace2/tvtime/vfir.c: (deinterlace_frame_vfir):
Don't use the scanlines function from gstdeinterlace2 as it's only
used for this method and will be removed. Instead implement
deinterlace_frame function and make it a bit more efficient.
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace2_class_init), (gst_deinterlace2_set_method),
(gst_deinterlace2_push_history), (gst_deinterlace2_chain),
(gst_deinterlace2_setcaps), (gst_deinterlace2_sink_event),
(gst_deinterlace2_change_state), (gst_deinterlace2_src_event),
(gst_deinterlace2_src_query):
Fix coding style and remove scanlines function as it's unused now.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedyh.asm:
* gst/deinterlace2/tvtime/greedyh.c: (greedyDScaler_C),
(deinterlace_frame_di_greedyh), (dscaler_greedyh_get_method):
* gst/deinterlace2/tvtime/greedyhmacros.h:
Add a C implementation for the greedyh deinterlacing method, clean
up the code a bit and mark the SSE version as MMXEXT as it doesn't
require any SSE instructions.
Original commit message from CVS:
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace2_set_property), (gst_deinterlace2_chain),
(gst_deinterlace2_setcaps):
If we're outputting all fields the framerate has to be doubled.
Set duration on the outgoing buffers.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedy.c:
(deinterlace_greedy_packed422_scanline_mmx),
(deinterlace_greedy_packed422_scanline_mmxext):
Optimize MMX/MMXEXT implementations a bit by requiring two less
memory accesses and fix the workaround for the missing right shift
on bytes to unset the highest bit of every byte.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedy.c:
(deinterlace_greedy_packed422_scanline_mmxext):
Remove sfence instruction as it's not needed and actually is an SSE
instruction.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedy.c:
(deinterlace_greedy_packed422_scanline_mmx),
(deinterlace_greedy_packed422_scanline):
Add plain MMX implementation for the greedyl method.
Original commit message from CVS:
* gst/deinterlace2/Makefile.am:
Move the assembly includes to noinst_HEADERS where they belong.
* gst/deinterlace2/tvtime/vfir.c: (deinterlace_line_c),
(deinterlace_line_mmx):
Fix C and MMX implementations a bit more.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedy.c:
(deinterlace_greedy_packed422_scanline_c),
(deinterlace_greedy_packed422_scanline_mmxext),
(deinterlace_greedy_packed422_scanline):
Fix the C implementation to produce correct results and optimize the
MMXEXT implementation.
Handle odd widths and don't read over array boundaries in the MMXEXT
implementation.
* gst/deinterlace2/tvtime/vfir.c: (deinterlace_line_c),
(deinterlace_line_mmx), (deinterlace_scanline_vfir):
Fix a small rounding bug in the MMX implementation, the MMX
implementation doesn't actually need MMXEXT instructions so don't mark
it as such.
Handle odd widths in both implementations.
Original commit message from CVS:
* gst/deinterlace2/tvtime/greedy.c:
(deinterlace_greedy_packed422_scanline_sse),
(deinterlace_greedy_packed422_scanline_c),
(deinterlace_greedy_packed422_scanline):
Implement a C version of the greedy low motion algorithm and mark the
assembly optimized version as SSE as it uses SSE instructions
additional to MMX instructions.
Original commit message from CVS:
* gst/deinterlace2/tvtime/vfir.c: (deinterlace_line_mmxext),
(deinterlace_line_c), (deinterlace_scanline_vfir):
Make it possible to use the vfir method on X86 CPUs without MMXEXT too
but use the MMXEXT optimized code whenever possible.
Original commit message from CVS:
* gst/deinterlace2/gstdeinterlace2.c:
(gst_deinterlace2_class_init), (gst_deinterlace2_init),
(gst_deinterlace2_reset_history), (gst_deinterlace2_reset),
(gst_deinterlace2_finalize), (gst_deinterlace2_chain),
(gst_deinterlace2_sink_event), (gst_deinterlace2_change_state),
(gst_deinterlace2_src_query):
* gst/deinterlace2/gstdeinterlace2.h:
Reset element state on PAUSED->READY properly, don't leak any buffers
when finalizing, allocate buffers with gst_pad_alloc_buffer() and
properly return flow returns from gst_pad_push() instead of ignoring them.
Original commit message from CVS:
* configure.ac:
* gst/deinterlace2/Makefile.am:
* gst/deinterlace2/tvtime/greedyh.asm:
* gst/deinterlace2/tvtime/tomsmocomp/SearchLoopBottom.inc:
Fix compilation on generic x86/amd64 and include deinterlace2 in the
build system. Because of several bugs it's still enabled only
by --enable-experimental.
We shouldn't register a new GstTag for every unknown tag
we find as this might lead to conflicts and also those
tags are essentially unknown.
Add mappings for some known tags and also convert string
dates to GDate, as found in many FLV files.
Add support for ECMA arrays in script tags. This fixes
seeking on some files that have the seek table stored
inside an ECMA array instead of the normal array.
Original commit message from CVS:
2008-11-24 Julien Moutte <julien@fluendo.com>
* gst/flv/gstflvdemux.c: (gst_flv_demux_find_offset),
(gst_flv_demux_handle_seek_push),
(gst_flv_demux_handle_seek_pull):
Fix non key unit seeking by always going to the previous
keyframe. Mark
the discont flag when we've moved in the file.
* gst/flv/gstflvparse.c: (gst_flv_parse_audio_negotiate): MP3
streams
are parsed already, makes autoplugged pipelines shorter.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_cleanup),
(gst_flv_demux_loop), (gst_flv_demux_handle_seek_push),
(gst_flv_demux_handle_seek_pull), (gst_flv_demux_sink_event),
(gst_flv_demux_dispose), (gst_flv_demux_init):
* gst/flv/gstflvdemux.h:
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video), (gst_flv_parse_tag_timestamp):
Put the GstSegment directly into the instance struct instead of
allocating and free'ing it again.
Push tags already if only one pad was added, no need to wait for
the second one.
When generating our index set has_video and has_audio if we find
video or audio in case the FLV header has incorrect data.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_chain),
(gst_flv_demux_pull_tag), (gst_flv_demux_pull_header),
(gst_flv_demux_create_index):
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_script),
(gst_flv_parse_tag_audio), (gst_flv_parse_tag_video),
(gst_flv_parse_tag_timestamp), (gst_flv_parse_tag_type),
(gst_flv_parse_header):
* gst/flv/gstflvparse.h:
Don't memcpy() all data we want to push downstream, instead just
create subbuffers and push them downstream.
Fix some minor memory leaks.
Original commit message from CVS:
* gst/flv/Makefile.am:
Fix (non-critical) syntax error and add all required CFLAGS and LIBS.
* gst/flv/gstflvparse.c: (FLV_GET_STRING),
(gst_flv_parse_metadata_item), (gst_flv_parse_tag_script),
(gst_flv_parse_tag_audio), (gst_flv_parse_tag_video),
(gst_flv_parse_tag_timestamp), (gst_flv_parse_tag_type):
Rewrite the script tag parsing to make sure we don't try to read
more data than we have. Also use GST_READ_UINT24_BE directly and
fix some minor memory leaks.
This should make all crashes on fuzzed FLV files disappear.
Original commit message from CVS:
* gst/flv/gstflvparse.c: (FLV_GET_STRING),
(gst_flv_parse_tag_audio), (gst_flv_parse_tag_video),
(gst_flv_parse_tag_type), (gst_flv_parse_header):
Properly check everywhere that we have enough data to parse and
don't read outside the allocated memory region.
Original commit message from CVS:
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video):
If the caps change during playback and negotiation fails error out
instead of trying to continue.
Original commit message from CVS:
* gst/flv/gstflvmux.c: (gst_flv_mux_audio_pad_setcaps),
(gst_flv_mux_request_new_pad), (gst_flv_mux_write_buffer),
(gst_flv_mux_collected):
* gst/flv/gstflvmux.h:
* gst/flv/gstflvparse.c: (gst_flv_parse_audio_negotiate):
Add support for Speex audio and allow buffers without valid
timestamp in the muxer.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_loop),
(gst_flv_demux_find_offset), (gst_flv_demux_handle_seek_push),
(gst_flv_demux_handle_seek_pull):
Don't post an error message on the bus if sending EOS downstream
didn't work. Fixes bug #550454.
Fix seek event handling to look at the flags of the seek event
instead of assuming some random flags, don't send segment-start
messages when operating in push mode and push seek events upstream
if we couldn't handle them.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_create_index),
(gst_flv_demux_loop):
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_script),
(gst_flv_parse_tag_audio), (gst_flv_parse_tag_video),
(gst_flv_parse_tag_timestamp):
* gst/flv/gstflvparse.h:
In pull mode we create our own index before doing anything else
and don't use the index provided by some files (which are more than
often incorrect and cause failed seeks).
For push mode we still use the index provided by the file and extend it
while doing the playback.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_push_src_event),
(gst_flv_demux_loop), (gst_flv_demux_handle_seek_pull),
(gst_flv_demux_sink_event):
Instead of using gst_pad_event_default() use a small
gst_pad_push_event() wrapper that only does what we want and is much
more simple.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_change_state),
(gst_flv_demux_set_index), (gst_flv_demux_init):
* gst/flv/gstflvdemux.h:
If our index was created by the element and not provided from the
outside we should destroy it when starting a new stream to get
all old entries removed.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_pull_range):
Improve debugging a bit when pulling a buffer from upstream fails.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_cleanup),
(gst_flv_demux_handle_seek_pull), (gst_flv_demux_dispose):
* gst/flv/gstflvdemux.h:
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video):
Close the currently playing segment from the streaming thread
instead of the thread where the seek event is handled.
Original commit message from CVS:
* gst/flv/gstflvmux.c: (gst_flv_mux_audio_pad_setcaps),
(gst_flv_mux_write_buffer):
Don't set video_codec to the value that actually should go
into audio codec, otherwise we create invalid files.
Fixes bug #556564.
Original commit message from CVS:
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video):
Use gst_pad_alloc_buffer_and_set_caps() to make sure we get
a buffer with caps that we can work with (i.e. the pad's caps).
Add non-keyframe video frames to the index too but without the
keyframe flag.
Add audio frames to the index only if we have no video stream.
Original commit message from CVS:
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video):
Create pads from the pad templates, use fixed caps on them
and only activate them after the caps are set.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_loop):
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_timestamp):
* gst/flv/gstflvparse.h:
Get an approximate duration of the file by looking at the timestamp
of the last tag in pull mode. If we get (maybe better) duration from
metadata later we'll use that instead.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_pull_range),
(gst_flv_demux_pull_tag), (gst_flv_demux_pull_header):
Refactor _pull_range() logic with checks into a seperate function
to make things a bit more readable.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_chain),
(gst_flv_demux_base_init):
Use gst_element_class_set_details_simple().
If we get GST_FLOW_NOT_LINKED in the parse loop but at least
one of the pads is linked continue the loop.
Original commit message from CVS:
* gst/flv/gstflvparse.c: (gst_flv_parse_audio_negotiate),
(gst_flv_parse_tag_audio), (gst_flv_parse_video_negotiate):
Correct caps for video codec id 5: It's On2 VP6 with alpha channel
which needs a different decoder and has different caps.
Add support for audio codec id 14, which is MP3 with 8kHz sampling
rate.
Fix endianness and signedness for raw audio codec ids.
Add support for alaw and mulaw audio.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_chain):
Go out of the parse loop as soon as we get an error instead
of parsing until the GstAdapter is empty.
Add some explanations about the header and tag size.
Don't print synchronizing message if everything is fine.
Original commit message from CVS:
* gst/flv/Makefile.am:
* gst/flv/gstflvdemux.c: (plugin_init):
* gst/flv/gstflvmux.c: (gst_flv_mux_base_init),
(gst_flv_mux_class_init), (gst_flv_mux_init),
(gst_flv_mux_finalize), (gst_flv_mux_reset),
(gst_flv_mux_handle_src_event), (gst_flv_mux_handle_sink_event),
(gst_flv_mux_video_pad_setcaps), (gst_flv_mux_audio_pad_setcaps),
(gst_flv_mux_request_new_pad), (gst_flv_mux_release_pad),
(gst_flv_mux_write_header), (gst_flv_mux_write_buffer),
(gst_flv_mux_collected), (gst_flv_mux_change_state):
* gst/flv/gstflvmux.h:
Add first version of a FLV muxer. The only missing feature is writing
of stream metadata.
Original commit message from CVS:
* gst/flv/gstflvparse.c: (gst_flv_parse_metadata_item),
(gst_flv_parse_tag_script):
Handle NULL returns from FLV_GET_STRING() more gracefully. Fixes
crash caused by a strlen on a NULL string (#527622).
Original commit message from CVS:
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video):
Don't strdup (and thus leak) codec name strings when passing
them to gst_tag_list_add().
Original commit message from CVS:
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video):
Fix list of supported and known codecs.
Emit tag with the codec name so it gets properly reported in totem and
other applications.
Original commit message from CVS:
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video):
Output segment with proper 'stop' value, makes flvdemux 100% compatible
with gnonlin.
Original commit message from CVS:
* gst/flv/gstflvparse.c:
Add mapping for Nellymoser ASAO audio codec.
(gst_flv_parse_tag_audio), (gst_flv_parse_tag_video): Make sure we
actually have data to read at the end of the tag. This avoids trying
to allocate negative buffers.
Original commit message from CVS:
2007-10-22 Julien MOUTTE <julien@moutte.net>
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video), (gst_flv_parse_tag_type): Don't
emit no-more-pads for single pad scenarios as the header
is definitely not reliable. We emit them for 2 pads scenarios
though to speed up media discovery.
Original commit message from CVS:
2007-09-27 Julien MOUTTE <julien@moutte.net>
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video): I got it wrong again, audio rate
was not detected correctly in all cases.
Original commit message from CVS:
2007-09-26 Julien MOUTTE <julien@moutte.net>
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video): codec_data is needed for every tag
not just the first one. (Fix a stupid bug i introduced without
testing)
Original commit message from CVS:
2007-09-26 Julien MOUTTE <julien@moutte.net>
* gst/flv/gstflvparse.c: (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video): Fix bit masks operations to be
sure we detect the codec_tags and sample rates correctly.
Fix raw audio caps generation.
Original commit message from CVS:
2007-08-27 Julien MOUTTE <julien@moutte.net>
* gst/flv/gstflvdemux.c: (gst_flv_demux_handle_seek_pull):
Make sure we initialize the seek result.
Original commit message from CVS:
2007-08-22 Julien MOUTTE <julien@moutte.net>
* gst/flv/gstflvdemux.c: (gst_flv_demux_cleanup),
(gst_flv_demux_pull_tag):
* gst/flv/gstflvdemux.h:
* gst/flv/gstflvparse.c: (gst_flv_parse_metadata_item),
(gst_flv_parse_tag_script), (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video): Handle pixel aspect ratio through
metadata tags like ASF does. Fluendo muxer supports this and
Flash players can support it as well this way.
Original commit message from CVS:
2007-08-22 Julien MOUTTE <julien@moutte.net>
* gst/flv/gstflvdemux.c: (gst_flv_demux_pull_tag):
* gst/flv/gstflvparse.c: (gst_flv_parse_metadata_item),
(gst_flv_parse_tag_script), (gst_flv_parse_tag_audio),
(gst_flv_parse_tag_video): Make sure we don't try filling up the
index if no times object was parsed. Fix the way we decide to
push
tags and emit no-more-pads. Fix some printf typing in debugging.
Original commit message from CVS:
* gst/flv/gstflvdemux.c: (gst_flv_demux_set_index),
(gst_flv_demux_get_index):
Fix locking and refcounting on the index.
Original commit message from CVS:
2007-08-14 Julien MOUTTE <julien@moutte.net>
* gst/flv/gstflvdemux.c: (gst_flv_demux_cleanup),
(gst_flv_demux_adapter_flush), (gst_flv_demux_chain),
(gst_flv_demux_pull_tag), (gst_flv_demux_do_seek),
(gst_flv_demux_handle_seek), (gst_flv_demux_sink_event),
(gst_flv_demux_src_event), (gst_flv_demux_query),
(gst_flv_demux_change_state), (gst_flv_demux_set_index),
(gst_flv_demux_get_index), (gst_flv_demux_dispose),
(gst_flv_demux_class_init): First method for seeking in pull
mode using the index built step by step or coming from metadata.
* gst/flv/gstflvdemux.h:
* gst/flv/gstflvparse.c: (FLV_GET_STRING),
(gst_flv_parse_metadata_item), (gst_flv_parse_tag_script),
(gst_flv_parse_tag_audio), (gst_flv_parse_tag_video): Parse
more metadata types and keyframes index.
Original commit message from CVS:
2007-07-19 Julien MOUTTE <julien@moutte.net>
* configure.ac:
* gst/flv/Makefile.am:
* gst/flv/gstflvdemux.c: (gst_flv_demux_flush),
(gst_flv_demux_cleanup), (gst_flv_demux_chain),
(gst_flv_demux_pull_tag), (gst_flv_demux_pull_header),
(gst_flv_demux_seek_to_prev_keyframe), (gst_flv_demux_loop),
(gst_flv_demux_sink_activate),
(gst_flv_demux_sink_activate_push),
(gst_flv_demux_sink_activate_pull), (gst_flv_demux_sink_event),
(gst_flv_demux_change_state), (gst_flv_demux_dispose),
(gst_flv_demux_base_init), (gst_flv_demux_class_init),
(gst_flv_demux_init), (plugin_init):
* gst/flv/gstflvdemux.h:
* gst/flv/gstflvparse.c: (FLV_GET_BEUI24), (FLV_GET_STRING),
(gst_flv_demux_query_types), (gst_flv_demux_query),
(gst_flv_parse_metadata_item), (gst_flv_parse_tag_script),
(gst_flv_parse_tag_audio), (gst_flv_parse_tag_video),
(gst_flv_parse_tag_type), (gst_flv_parse_header):
* gst/flv/gstflvparse.h: Adds a first draft of an FLV demuxer.
It does not do seeking yet, it supports pull and push mode so
YES
you can use it to play youtube videos directly from an HTTP uri.
Not so much testing done yet but it parses metadata, reply to
duration queries, etc...
First of all a keyframe seek should be done to the
keyframe right before the requested position and not
to the keyframe that is nearest to the requested position.
Use per track index arrays and use our new binary search function
from core to speed up the search.
Rewrite the quant table parsing to also handle multiple tables in one JPEG HDQ
segment.
Handle more jpeg types by keeping track of the tables used per component and
putting the used ones in the quant headers.
Read the timestamp of the incomming buffer before we push it in the adapter and
flush it out again as the buffer might be unreffed then and we read from invalid
memory.
Fixes#581444.
Don't require width/height on the caps. Use the SOF header to find width/height
and fall back to the caps if there is no SOF. Also use the SOF info to find the
subsampling and quantization tables used. This allows us to set the right type
value in the JPEG rtp header.
Deprecate the quality property, it's unused now and it was used wrongly before.
Always send full quant tables for now until we have some code to detect default
ones.
Fixes#580880
Use the width and the height from the payload headers and set them on the
output caps for added awesomeness.
Fix quant parsing, we need to check the type in the lower 6 bits.
Add first bits of caching quantization tables.
Server eof (e.g. connection closed) is announced as connection closed,
so better record state and act accordingly to prevent (read/write)
errors during subsequent teardown/cleanup sequences. #Fixes 580851.(c).
We didn't handle unsynchronization at all up to now, which might have
caused frames to not be extracted - esp. frames after an APIC picture
frame. Fixes#577468.
If the codec is actually something else (e.g. mjpeg) change the caps to
match when parsing the ESDS atom.
Also, for AAC, override rate and channels with correct values read from
ESDS, since the rate/channels values elsewhere are often wrong.
We implemented the AAL2 packing, add the encoding-name for those to the caps and
a property to force AAL2 decoding (always TRUE for now).
Implement RFC3551 unpacking for regular G726.
See #567140.
