summaryrefslogtreecommitdiffstats
path: root/gst
AgeCommit message (Collapse)AuthorFilesLines
2009-04-17deinterlace2: Reset history if DISCONT is set on the incoming bufferSebastian Dröge1-0/+5
2009-04-17deinterlace2: Fix timestamps for buffers with RFF flag setSebastian Dröge1-7/+5
2009-04-16mxf: Properly handle the new interlaced support from libgstvideo for (de)muxingSebastian Dröge1-1/+9
2009-04-16deinterlace2: Rename line_length to row_stride and remove output_strideSebastian Dröge9-58/+58
2009-04-16deinterlace2: Implement support for RFF and ONEFIELD buffer flagsSebastian Dröge1-24/+56
2009-04-16fpsdisplaysink: add a initial port of a sink with fps displayStefan Kost3-0/+592
This now works with a event probe. Needs some extra work.
2009-04-16camerabin: clean up and fix assertion fail when setting zoom in NULL stateLasse Laukkanen1-31/+53
2009-04-16camerabin: fix format matching when detecting allowed frame rateLasse Laukkanen1-15/+28
2009-04-16camerabin: don't lose pending state when changing resolutionLasse Laukkanen1-3/+12
2009-04-16camerabin: change img-done signal parameter from GString* to const gchar*Lasse Laukkanen3-34/+40
Don't allow setting filename via img-done signal parameter but force app use filename property. Don't stop capture when setting filename property. Update check unit test based on the change.
2009-04-15rtpmux: Add GST_DEBUG_FUNCPTRsOlivier Crête2-8/+11
2009-04-15rtpdtmfmux: Release locked pad on release_padOlivier Crête1-0/+18
Release the special pad if the pad is removed from the muxer.
2009-04-15rtpdtmfmux: Release special on pad disposeLaurent Glayal1-4/+11
Fixes #577690
2009-04-15jitterbuffer: prevent overflow in EOS estimationWim Taymans1-1/+1
Use a guint64 instead of a guint to hold a 64bit value to prevent completely bogues EOS estimation values due to overflows.
2009-04-15rtpbin: we should not provide a clockWim Taymans2-17/+0
There is no need to provide a clock.
2009-04-15jitterbuffer: more estimated EOS fixesWim Taymans1-6/+18
Do more accurate EOS estimate and guard against backward timestamps.
2009-04-15jitterbuffer: release lock before pushing EOSWim Taymans1-1/+1
Make sure we release the jitterbuffer lock before we start pushing out data because else we might deadlock.
2009-04-15deinterlace2: Move output buffer from the instance struct to a function ↵Sebastian Dröge5-37/+29
parameter
2009-04-15deinterlace2: Add initial support for automatic detection of the field orderSebastian Dröge2-5/+11
2009-04-15deinterlace2: Add support for YVYU colorspaceSebastian Dröge1-2/+4
This is the same as YUY2 with just Cr and Cb swapped. As we don't make a difference between them when deinterlacing this works.
2009-04-13mxfmux: Small cleanupSebastian Dröge1-12/+2
2009-04-07mxfdemux: Don't use invalid buffer offsets, instead assume offset == 0Sebastian Dröge1-1/+2
2009-04-07mxf: Fix comparisonSebastian Dröge1-2/+1
2009-04-07Use UL database for all metadata ULsSebastian Dröge3-595/+635
2009-04-07baseparse: Fix slightly broken buffer-in-segment check (aacparse, amrparse, ↵René Stadler3-3/+3
flacparse)
2009-04-05baseparse: Fix push mode seeking (aacparse, amrparse)René Stadler2-10/+24
Sending the flush-start event forward before taking the stream lock actually works, in contrast to deadlocking in downstream preroll_wait (hunk 1). After that we get the chain function being stuck in a busy loop. This is fixed by updating the minimum frame size inside the synchronization loop because the subclass asks for more data in this way (hunk 2). Finally, this leads to a very probable crash because the subclass can find a valid frame with a size greater than the currently available data in the adapter. This makes the subsequent gst_adapter_take_buffer call return NULL, which is not expected (hunk 3).
2009-04-02dvdspu: Add some debugJan Schmidt1-4/+10
Add a little bit of extra debug. Wrap a hardcoded #if 0 into a define instead.
2009-04-01legacyresample: fix negotiation so that upstream can actually fixate to ↵Tim-Philipp Müller1-5/+25
downstream's rate If one side has a preference for a particular sample rate or set of sample rates, we should honour this in the caps we advertise and transform to and from, so that elements actually know about the other side's sample rate preference and can negotiate to it if supported. Also add unit test for this.
2009-03-31baseparse: Delay newsegment as long as possible.Mark Nauwelaerts2-50/+68
If newsegment is sent (too) early, caps may not yet be fixed/set, and downstream may not have been linked.
2009-03-27rtpbin: add on_npt_stop signalWim Taymans4-14/+194
Add the on_npt_stop signal to rtpbin and rtpjitterbuffer to notify the application that the NPT stop position has been reached.
2009-03-27outputselector: reset state when going to READYWim Taymans1-5/+33
Reset the last-buffer, the pending pad and the segment when going to the READY state. Fixes #576712.
2009-03-26qtmux: fix reusing elementMark Nauwelaerts1-0/+7
State change to READY and then back to PAUSED should still provide the proper structures as are otherwise freshly available following a request_new_pad. Pointed out by Thiago Santos.
2009-03-26mxf: Fix compilation and compiler errors with GLib < 2.16.0Sebastian Dröge2-6/+6
2009-03-26mxf: Fix "cast to pointer type of different size" compiler warningSebastian Dröge1-11/+12
2009-03-26mxf: Rename forgotten function to fix unresolved symbols errorSebastian Dröge1-1/+1
2009-03-26mxf: Use mxf_u{l,uid}_is_equal() and friends instead of memcmp()Sebastian Dröge4-14/+16
2009-03-26mxf: Differentiate between UL and UUIDSebastian Dröge10-590/+742
2009-03-26mxf: Source files and #include cleanupSebastian Dröge31-786/+721
2009-03-26xdgmime: Use GIOs g_content_type_guess() if possibleSebastian Dröge2-17/+55
2009-03-26mxf: Use the UL database for some more ULsSebastian Dröge1-21/+5
2009-03-26mxf: Remove some GLib < 2.14 compatibility stuffSebastian Dröge1-8/+2
2009-03-26xdgmime: Link with winsock library to fix build with MinGWРуслан Ижбулатов1-1/+1
Partially fixes bug #573595.
2009-03-26dccp: Port DCCP plugin to MinGWРуслан Ижбулатов12-51/+123
Partially fixes bug #573595.
2009-03-26Merge branch 'mxfmux'Sebastian Dröge22-292/+6858
2009-03-26mxfmux: CleanupSebastian Dröge3-6/+6
2009-03-26mxfmux: Fix handling of buffers with more than one edit unit and EOS handlingSebastian Dröge4-180/+193
2009-03-24selector: merge the tagsWim Taymans1-8/+10
Merge the tags received on the input-selector sinkpads instead of only keeping the last one we saw.
2009-03-23aacparse: Fix busyloop when seeking. Fixes #575388René Stadler1-5/+0
The problem is that after a discont, set_min_frame_size(1024) is called when detect_stream returns FALSE. However, detect_stream calls check_adts_frame which sets the frame size on its own to something larger than 1024. This is the same situation as in the beginning, so the base class ends up calling check_valid_frame in an endless loop.
2009-03-23aacparse: Refactor check_valid_frame to expose broken codeRené Stadler1-8/+20
Just moving code around and removing an unhelpful/misleading comment.
2009-03-23qtmux: fix includes for lseekWim Taymans1-0/+5
--