summaryrefslogtreecommitdiffstats
path: root/gst
AgeCommit message (Collapse)AuthorFilesLines
2009-05-26dvdspu: Add a simple default colour table.Jan Schmidt1-0/+8
When we're not provided with a palette in advance, draw with a grey colour or two, instead of YUV green.
2009-05-26mpegdemux: Add sparse stream filling.Jan Schmidt2-0/+97
First stab at sending new-segment events to effect sparse stream updates.
2009-05-26mpegdemux: Add support for outputting sub-picture streams found in files.Jan Schmidt2-8/+32
Output subpicture streams when they are found on the private stream ID. Don't strip off the first byte of such packets when pushing.
2009-05-26mpegtsdemux: Add mapping for DVD and Bluray subpicture streams.Jan Schmidt4-5/+28
Add output subpicture pads for DVD (video/x-dvd-subpicture) and Bluray PGS (subpicture/x-pgs) streams. Remove an unused variable from the PES filter.
2009-05-26dvdspu: Don't accidentally lose the colour palette when flushing the SPUJan Schmidt1-9/+16
Fixes racy startup on DVDs where it sometimes gets entirely the wrong set of colours in the menus and subtitles.
2009-05-26rtpsource: byteswap the port from GstNetAddressWim Taymans1-2/+3
Since the port in GstNetAddress is in network order we might need to byteswap it before adding it to the source statistics.
2009-05-25gstpesfilter: Directly use gst_adapter_take_buffer().Edward Hervey1-7/+1
2009-05-25gstpesfilter: Don't peek the adapter if we don't have enough data.Edward Hervey1-3/+3
2009-05-25gstpesfilter: Don't skip private streams PES but push them out.Edward Hervey1-1/+3
The one thing we *DO* need to do for those streams is to skip all the PTS/DTS/Scrambling/DSM/extension/... handling.
2009-05-25mpegtsdemux: Ignore NULL packets as early as possible.Edward Hervey1-1/+7
This avoids: * creating a MpegTSStream structure for nothing * processing packet data for nothing
2009-05-25mpegtsdemux: Add mapping for HDV private streamsEdward Hervey2-0/+12
2009-05-25rtpbin: remove ptdemux ghostpadsWim Taymans1-0/+13
2009-05-22adpcmdec: Add new plugin for ms-adpcm decoding.Michael Smith2-0/+464
2009-05-22mpegvideoparse: Detect interlaced content and set it on outgoing caps.Edward Hervey3-0/+10
I also added the parsing of all the other bits in the sequence extension header in case we need it later.
2009-05-22rtpbin: add to new signal to remove SSRC padsWim Taymans1-15/+6
2009-05-22ssrcdemux: emit signal when pads are removedAli Sabil3-1/+80
Add action signal to clear an SSRC in the ssrc demuxer. Add signal to notify of removed ssrc. See #554839
2009-05-22rtpbin: use our ghostpads instead of its targetWim Taymans1-44/+46
Since we keep a reference to our ghostpads, we can use them to track sessions. This avoid us having to mess with the target of the ghostpad.
2009-05-22rtpbin: don't warn when getting request pads twiceWim Taymans1-52/+34
Allow getting the request pads multiple times, just return the previously created pads.
2009-05-22rtpsource: add RTP and RTCP source addressWim Taymans1-0/+48
Add the RTP and RTCP sender addresses in the stats structure.
2009-05-22rtpsession: reuse source code for SDESWim Taymans1-58/+2
Reuse the RTPSource object property instead of duplicating code.
2009-05-22rtpbin: set target state on new elementsWim Taymans1-20/+37
Set the state on newly added elements to the state of the parent. Add some debug info and do some cleanups
2009-05-22rtpbin: unref requests pads after releasingWim Taymans1-0/+4
2009-05-22mpegtsdemux: fix memleaks and refcountsWim Taymans1-4/+3
Use correct constants for PID_type so that we clear the right filter. provide_clock must return a ref to a clock.
2009-05-22id3tag: canonicalise function namesTim-Philipp Müller3-23/+23
2009-05-22id3tag: change GType to GstId3Tag so it doesn't conflict with the id3tag ↵Tim-Philipp Müller3-16/+16
plugin in -ugly
2009-05-22id3tag: register GType of the base class with a less generic nameTim-Philipp Müller1-1/+6
.. so we can easily move the base class into -base later without causing GType name conflicts.
2009-05-22rtpbin: Implement releasing the streamsOlivier Crête1-21/+53
See #561752
2009-05-22rtpbin: Keep jb signals handlerOlivier Crête1-3/+7
Keep the signal handlers so they can be disconnected at release time See #561752
2009-05-22rtpbin: use the right lock for the sessionsWim Taymans1-0/+4
Use the right lock when iterating the sessions.
2009-05-22rtpbin: Free session if request pads are releasedOlivier Crête1-3/+8
Free the session when all the request pads are released. Don't mess with the session list in free_session as it is called from a foreach on that list. Set the state of the upstream element to NULL first. See #561752
2009-05-22rtpbin: Implement relasing of the rtp recv padOlivier Crête1-2/+17
2009-05-22rtpbin: Implement releasing of rtp send padsOlivier Crête1-6/+25
2009-05-22rtpbin: Implement release of the recv rtcp padOlivier Crête1-3/+13
See #561752
2009-05-22rtpbin: Implement releasing of rtcp src padOlivier Crête1-2/+7
See #561752
2009-05-21sdpdemux: rework RTCP sending and RTP receivingWim Taymans2-7/+70
When we are dealing with multiast, create the udp src and sink elements pointing to the multicast addresses. When we are doing unicast, receive data on the local ports and don't send RTCP because we don't know where we have to send it. Fixes #583188
2009-05-21id3tag: Add new id3 tagging plugin, supports v1, v2.3, and v2.4.Michael Smith7-0/+2106
By default, does v1 and v2.3, but there are properties to select. Will hopefully replace id3mux, id3v2mux, in the not-too-distant future.
2009-05-20mxf: Fix frame_layout for non-interlaced formats. Fixes #583337Edward Hervey1-1/+1
2009-05-16fpsdisplaysink: Fix resetting of the sink in NULLJan Schmidt1-4/+12
Reset the fpsdisplaysink in NULL by removing the textoverlay if we created it. Fixes: #582633
2009-05-16Moved 'y4menc' from -bad to -goodJan Schmidt4-561/+0
2009-05-15[qtmux] Fixes segfault when adding a blob as first tag.Thiago Santos1-7/+5
Moves tags data initialization to the function that actually appends the tags to the list. Fixes #582702 Also fixes some style caught by the pre-commit hook.
2009-05-15mxfdemux: Use a RW lock to protect metadata and add all pads at once without ↵Sebastian Dröge2-70/+93
a lock held This makes it possible, among other things, to do a query in the pad-added callback. Fixes bug #582656.
2009-05-14dvdspu: Push update frame, if any, when entering stills.Jan Schmidt1-0/+7
When entering a still frame generates an updated buffer, make sure to push it out, otherwise we may not put a frame onscreen with a rendered button, causing raciness as to whether buttons get drawn or not when jumping back to the menu on some discs.
2009-05-13y4menc: change my emailWim Taymans1-1/+1
change my email to something more current See #580783
2009-05-13y4menc: don't strip timestampsWim Taymans1-2/+6
Fixes #582483
2009-05-13Moved 'deinterlace2' from -bad to -goodSebastian Dröge41-8726/+0
And remove old deinterlace plugin as deinterlace2 will be called deinterlace in -good.
2009-05-12Moved 'flv' from -bad to -goodSebastian Dröge7-3916/+0
2009-05-11dvdspu: Make the debugging output a bit clearer.Jan Schmidt1-5/+10
Display more info about custom DVD events in the debug messages.
2009-05-11dvdspu: Always push a frame at the start of a still frame, and fix a leak.Jan Schmidt1-10/+24
Make sure to push the frame for a still frame, with discont = true and timestamp=none, so that it gets displayed by the sink. Also, don't leak each rendered video frame during still menus.
2009-05-11dvdspu: Add a guard when we don't have any subpicture buffer to renderJan Schmidt1-0/+2
2009-05-11mpegtsdemux: Protect bitrate estimation against bogus values.Edward Hervey2-5/+18
If the estimated bitrate is lower than 188 bytes, there's most likely something completely wrong with the two samples. If that happens, force recalculation. Use guint64 for observation PCR, I saw cases where it would overflow.