Vincent Penquerc'h [Fri, 6 Jun 2014 11:18:49 +0000 (12:18 +0100)]
oggdemux: allow unset seek stop time in push mode
Tim-Philipp Müller [Wed, 11 Jun 2014 11:50:23 +0000 (12:50 +0100)]
docs: add streamsynchronizer to documentation
Tim-Philipp Müller [Wed, 11 Jun 2014 11:43:35 +0000 (12:43 +0100)]
docs: add playsink element to documentation
Tim-Philipp Müller [Wed, 11 Jun 2014 09:53:50 +0000 (10:53 +0100)]
docs: add navigation interface to docs
Thiago Santos [Tue, 10 Jun 2014 15:59:53 +0000 (12:59 -0300)]
appsrc: add send_event handler for flushing
Adds a send_event handling for allowing appsrc to flush its internal
data, allowing users to flush the pipeline without setting it to null.
https://bugzilla.gnome.org/show_bug.cgi?id=724231
Thiago Santos [Tue, 10 Jun 2014 00:05:00 +0000 (21:05 -0300)]
videoscale: vs_image: strides are a gsize
The strides that are set from the GstVideoInfo structs are
a gsize. Using an int can cause overflows when dealing with large
enough images
https://bugzilla.gnome.org/show_bug.cgi?id=731195
Thiago Santos [Mon, 9 Jun 2014 22:44:56 +0000 (19:44 -0300)]
video: avoid overflows when doing int operations for size
size is a gsize, so cast the operands to it to avoid overflows
and setting wrong value to the video size.
Includes tests.
https://bugzilla.gnome.org/show_bug.cgi?id=731195
Edward Hervey [Mon, 9 Jun 2014 08:53:03 +0000 (10:53 +0200)]
theoraenc: Remove unneeded check
running timestamps are guaranteed to be positive and valid since the
GstVideoEncoder base class will clip incoming buffers
CID #1139797
Edward Hervey [Mon, 9 Jun 2014 08:38:53 +0000 (10:38 +0200)]
vorbisenc: add missing va_end in variadic function
Coverity 1139944
Vincent Penquerc'h [Fri, 6 Jun 2014 09:35:31 +0000 (10:35 +0100)]
tests: fix uninitialized variable use in video decoder test
Sebastian Dröge [Thu, 5 Jun 2014 13:35:31 +0000 (15:35 +0200)]
uridecodebin: Also catch CODEC_NOT_FOUND errors and delay them until all decodebins are done
Sebastian Dröge [Wed, 4 Jun 2014 15:00:34 +0000 (17:00 +0200)]
uridecodebin: Ignore missing-plugin messages unless all decodebins post one
When playing RTSP streams there will be one decodebin per stream. If some of
them fail because of a missing plugin we should not fail completely but play
the supported streams at least.
https://bugzilla.gnome.org/show_bug.cgi?id=730868
Sebastian Dröge [Wed, 4 Jun 2014 12:14:14 +0000 (14:14 +0200)]
decodebin: Do async-done on expose errors too
Michael Olbrich [Tue, 20 May 2014 10:28:15 +0000 (12:28 +0200)]
dmabuf: fix checking mmap flags
A simple '&' is not sufficiant. With mmapping_flags == PROT_READ and
prot == PROT_READ|PROT_WRITE the check produces the wrong result.
Change the check to make sure that prot is a subset of mmapping_flags.
https://bugzilla.gnome.org/show_bug.cgi?id=730559
Vincent Penquerc'h [Tue, 3 Jun 2014 14:16:44 +0000 (15:16 +0100)]
alsasink: make gst-ident happy
Vincent Penquerc'h [Tue, 3 Jun 2014 14:10:33 +0000 (15:10 +0100)]
alsasink: fix occasional crash intersecting invalid values
When a pipeline using alsasink and push mode upstream fails
to preroll, the following state will be the case:
- A loop upstream will be PAUSED, pushing a first buffer
- alsasink will be READY, pending PAUSED, because async
On error, the pipeline will switch to NULL. alsasink is in
READY, so goes to NULL immediately. It zeroes its cached
caps. Meanwhile, the upstream loop can cause a caps query,
conccurent with the state change. This will use those cached
caps. If the zeroing happens between the NULL test and the
dereferencing, GStreamer will critical down in the GstValue
code.
Since it appears that such a gap between states (PAUSED
and pushing upstream, and NULL downstream) is expected, we
need to protect the read/write access to the cached caps.
This fixes the critical.
See https://bugzilla.gnome.org/show_bug.cgi?id=731121
Thibault Saunier [Mon, 14 Oct 2013 21:56:55 +0000 (18:56 -0300)]
videodecoder: Keep still meaningfull pending events on FLUSH_STOP
Only EOS and segment should be deleted in that case.
+ Add a testcase
https://bugzilla.gnome.org/show_bug.cgi?id=709868
Thibault Saunier [Mon, 14 Oct 2013 21:48:08 +0000 (18:48 -0300)]
audiodecoder: Keep still meaningfull pending events on FLUSH_STOP
Only EOS and segment should be deleted in that case.
https://bugzilla.gnome.org/show_bug.cgi?id=709868
Thibault Saunier [Mon, 14 Oct 2013 21:45:10 +0000 (18:45 -0300)]
videoencoder: Keep still meaningfull pending events on FLUSH_STOP
Only EOS and segment should be deleted in that case.
https://bugzilla.gnome.org/show_bug.cgi?id=709868
Thibault Saunier [Thu, 10 Oct 2013 21:50:17 +0000 (18:50 -0300)]
streamsplitter: Keep still meaningfull pending events on FLUSH_STOP
Only EOS and segment should be deleted in that case.
https://bugzilla.gnome.org/show_bug.cgi?id=709868
Thibault Saunier [Thu, 10 Oct 2013 21:48:47 +0000 (18:48 -0300)]
audioencoder: Keep still meaningfull pending events on FLUSH_STOP
Only EOS and segment should be deleted in that case.
https://bugzilla.gnome.org/show_bug.cgi?id=709868
Vincent Penquerc'h [Mon, 2 Jun 2014 11:40:27 +0000 (12:40 +0100)]
oggstream: consider all opus packets as "keyframes"
This lets oggdemux determine they are not delta units, and removes
spurious per packet warnings about being unable to determine the
packet's keyframeness.
Edward Hervey [Mon, 12 May 2014 15:13:50 +0000 (17:13 +0200)]
mikey: Free MikeyPayload in error cases
CID #1212136
Thiago Santos [Sun, 16 Mar 2014 17:27:30 +0000 (14:27 -0300)]
decodebin: aggregate buffering messages
Aggregate buffering messages to only post the lower value
to avoid setting pipeline to playing while any multiqueue
is still buffering.
There are 3 scenarios where the entries should be removed from
the list:
1) When decodebin is set to READY
2) When an element posts a 100% buffering (already implemented)
3) When a multiqueue is removed from decodebin.
For item 3 we don't need to handle it because this should only
happen when either 1 is hapenning or when it is playing a
chained file, for which number 2 should have happened for the
previous stream to finish
https://bugzilla.gnome.org/show_bug.cgi?id=726423
Philip Withnall [Wed, 28 May 2014 09:23:24 +0000 (10:23 +0100)]
audio: Add a missing precondition to gst_audio_format_from_string()
https://bugzilla.gnome.org/show_bug.cgi?id=730874
Thiago Santos [Mon, 26 May 2014 23:57:30 +0000 (20:57 -0300)]
tests: videodecoder: audiodecoder: add tests for eos after segment
Tests that pushing a buffer after the segment returns EOS
Thiago Santos [Tue, 27 May 2014 00:24:07 +0000 (21:24 -0300)]
videodecoder: actually return the push result in backwards playback
It was always returning _OK regardless of what downstream returned
Thiago Santos [Mon, 26 May 2014 15:44:48 +0000 (12:44 -0300)]
videodecoder: return EOS when segment is over
if a buffer is clipped by being completely out of segment, check if this
buffer is after the end of the segment and return EOS upstream
https://bugzilla.gnome.org/show_bug.cgi?id=709224
Thiago Santos [Mon, 26 May 2014 15:44:38 +0000 (12:44 -0300)]
audiodecoder: return EOS when segment is over
if a buffer is clipped by being completely out of segment, check if this
buffer is after the end of the segment and return EOS upstream
https://bugzilla.gnome.org/show_bug.cgi?id=709224
Thiago Santos [Mon, 26 May 2014 14:45:29 +0000 (11:45 -0300)]
oggdemux: use new gstutils helper GstFlowCombiner
Fixes the handling of GST_FLOW_EOS by using the helper object
from gstutils that does the correct combination of flow returns.
https://bugzilla.gnome.org/show_bug.cgi?id=709224
Tim-Philipp Müller [Fri, 23 May 2014 18:21:35 +0000 (19:21 +0100)]
tools: play: use cubic volume factor when adjusting volume
This is more natural and better-suited for a playback application.
Sebastian Dröge [Wed, 21 May 2014 11:23:24 +0000 (13:23 +0200)]
Back to development
Sebastian Dröge [Wed, 21 May 2014 11:06:34 +0000 (13:06 +0200)]
Release 1.3.2
Sebastian Dröge [Wed, 21 May 2014 10:01:15 +0000 (12:01 +0200)]
Update .po files
Sebastian Dröge [Wed, 21 May 2014 08:50:56 +0000 (10:50 +0200)]
Automatic update of common submodule
From 211fa5f to 1f5d3c3
Sebastian Dröge [Wed, 21 May 2014 08:43:49 +0000 (10:43 +0200)]
video: And check comparison for real
Sebastian Dröge [Wed, 21 May 2014 08:40:32 +0000 (10:40 +0200)]
video: Fix broken comparison in unit test
libs/video.c:540:50: error: comparison of constant 2 with boolean expression is always false
[-Werror,-Wtautological-constant-out-of-range-compare]
&& !GST_VIDEO_INFO_N_PLANES (&vinfo) > 2) {
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^ ~
Wim Taymans [Tue, 20 May 2014 13:59:53 +0000 (15:59 +0200)]
rtsp-transport: clarify port usage
Comment in the docs what the client_port and server_port fields are used
for in TCP mode (if the application wants to set those values).
Michael Olbrich [Tue, 20 May 2014 09:18:56 +0000 (11:18 +0200)]
dmabuf: share the mapping with shared copies of the memory
With lots of shared memory instances (e.g. created by a RTP payloader) the
overhead of duplicating the file descriptor and creating extra mappings is
significant. To avoid this, the parent memory maps the whole region and the
shared copies just reuse the same mapping.
https://bugzilla.gnome.org/show_bug.cgi?id=730441
Göran Jönsson [Mon, 19 May 2014 11:28:52 +0000 (13:28 +0200)]
rtspconnection: Add read source on write socket.
Add a read source on write socket when lost tunnel.
To be able to detect when clint closes get channel.
This is already done in gst_rtsp_source_dispatch_write but
only when the queue is empty.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=730368
Sebastian Dröge [Tue, 20 May 2014 07:48:56 +0000 (09:48 +0200)]
playsink: Always take the playsink lock when adding or removing pad probes
Otherwise we might end up inside the callback without having stored
the probe id... then try to remove that probe (not!) from the callback
and wait forever for the pad to unblock.
Vincent Penquerc'h [Mon, 19 May 2014 12:57:41 +0000 (13:57 +0100)]
alsasink: pass correct error to g_strerror
The error we get is a negated errno.
While there, fix a couple typos in messages.
Sebastian Dröge [Mon, 19 May 2014 09:17:33 +0000 (11:17 +0200)]
gst-play: Free playlist_file string if only printing the version
Anuj Jaiswal [Tue, 13 May 2014 08:08:20 +0000 (14:08 +0600)]
audio_sink and video_sink leakage fixed
https://bugzilla.gnome.org/show_bug.cgi?id=730010
Edward Hervey [Tue, 13 May 2014 09:51:55 +0000 (11:51 +0200)]
rtspconnection: Don't use argument for local storage
By re-using the uri argument for storing local data, we could end up in
a situation where we would free uri ... which would actually be the
string passed in argument.
Instead explicitely use a local variable. Fixes double-free issues.
CID #1212176
Nicolas Dufresne [Mon, 12 May 2014 17:18:50 +0000 (13:18 -0400)]
video-info: Also check the stride and offset are equal
gst_video_info_is_equal() was not checking if stride and offset
had changed.
https://bugzilla.gnome.org/show_bug.cgi?id=729896
Edward Hervey [Mon, 12 May 2014 15:17:07 +0000 (17:17 +0200)]
videodecoder: Free data after removing it from the list
While it wouldn't have caused any failures (g_list_remove doesn't dereference
the provided pointer), it does make the code cleaner.
CID #1212174
Edward Hervey [Mon, 12 May 2014 15:15:17 +0000 (17:15 +0200)]
mikey: Actually replace payload ...
This function is intented to replace the payload, let's actually do that
instead of putting back the same (freed) payload
CID #1212175
Edward Hervey [Mon, 12 May 2014 15:13:50 +0000 (17:13 +0200)]
mikey: Free MikeyPayload in error cases
CID #1212135
CID #1212136
CID #1212137
CID #1212138
Thibault Saunier [Sat, 10 May 2014 21:50:44 +0000 (23:50 +0200)]
pango: Do not try to add a feature to a caps features ANY
It does not makes sense and asserts
Tim-Philipp Müller [Fri, 9 May 2014 14:32:18 +0000 (15:32 +0100)]
tag: xmp: fix leaks in error code paths
CID 1212133
Göran Jönsson [Tue, 6 May 2014 09:12:19 +0000 (11:12 +0200)]
rtspconnection: Reset control_stream.
Reset control_stream when gst_rtsp_connection_close.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=729632
Nicolas Dufresne [Tue, 15 Apr 2014 18:51:46 +0000 (14:51 -0400)]
videodecoder: Retry setting configuration with modified config
Buffer pool set_config() may return FALSE if requested configuration needed small
changes. Reget the config and try setting it again. This ensure we have a configured
pool if possible.
Wim Taymans [Thu, 8 May 2014 15:10:26 +0000 (17:10 +0200)]
uridecodebin: use downloadbuffer for download buffering
Use the new downloadbuffer element to implement the download buffering
feature
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=680183
Luis de Bethencourt [Tue, 6 May 2014 17:01:32 +0000 (13:01 -0400)]
oggmux: push eos event when empty pad data
If gst_ogg_mux_queue_pads returns NULL it means we are at EOS, because we get a
NULL buffer and this function never sets bestpad.
https://bugzilla.gnome.org/show_bug.cgi?id=729315
Руслан Ижбулатов [Tue, 6 May 2014 08:07:38 +0000 (08:07 +0000)]
configure: Use X11 detection macro from common
https://bugzilla.gnome.org/show_bug.cgi?id=729621
Tim-Philipp Müller [Tue, 6 May 2014 06:51:11 +0000 (07:51 +0100)]
examples: playback-test: fix crashes when setting buffer-size
playbin's buffer-size property takes a gint, not a gint64,
so only pass the bits expected to the vararg function, or
the terminator might not be found, leading to crashes, esp.
with negative numbers.
Spotted by Ravi Kiran K N <ravi.kiran@samsung.com>
https://bugzilla.gnome.org/show_bug.cgi?id=729617
Tim-Philipp Müller [Tue, 6 May 2014 06:50:16 +0000 (07:50 +0100)]
examples: fix indentation of playback-test
Tim-Philipp Müller [Tue, 6 May 2014 07:13:24 +0000 (08:13 +0100)]
Revert "playback-test: Set buffer-size only for non-negative size"
This reverts commit
07a637e2847d56d0f2b0c0ac9095bf37dd324e26.
Ravi Kiran K N [Tue, 6 May 2014 06:01:18 +0000 (11:31 +0530)]
playback-test: Set buffer-size only for non-negative size
https://bugzilla.gnome.org/show_bug.cgi?id=729617
Luis de Bethencourt [Tue, 6 May 2014 03:29:44 +0000 (23:29 -0400)]
win32: Update defs file
commit
622007e7db7e3d32bf8e04e673e057897b646220 added the function
gst_discoverer_info_get_missing_elements_installer_details (). It needs to be
added to the defs file.
Руслан Ижбулатов [Sun, 4 May 2014 15:54:54 +0000 (15:54 +0000)]
rtsp: Link to ws2_32 on Windows
Needed for getsockname and setsockopt
https://bugzilla.gnome.org/show_bug.cgi?id=729514
Руслан Ижбулатов [Sun, 4 May 2014 15:54:06 +0000 (15:54 +0000)]
Make X11 detection more precise
Don't be content with just X11/Xlib.h, check for X11/XKBlib.h as well.
This prevents false positives (for example, from partial X11 headers
installed by tcl/tk).
https://bugzilla.gnome.org/show_bug.cgi?id=729513
Руслан Ижбулатов [Sun, 4 May 2014 15:57:35 +0000 (15:57 +0000)]
tests: fix printf format compiler warning in playback test on win32
https://bugzilla.gnome.org/show_bug.cgi?id=729515
Tim-Philipp Müller [Sun, 4 May 2014 17:14:54 +0000 (18:14 +0100)]
Add new unit test binary to .gitignore
Thibault Saunier [Tue, 14 Jan 2014 14:39:55 +0000 (15:39 +0100)]
discoverer: Add APIs to simply get installer details for missing plugins
Currently the API is far from optimal and the user has to work around
our badly defined API to simply install missing plugins.
API:
new:
gst_discoverer_info_get_missing_elements_installer_details
deprecated:
gst_discoverer_info_get_misc
gst_discoverer_stream_info_get_misc
https://bugzilla.gnome.org/show_bug.cgi?id=720596
Sebastian Dröge [Sat, 3 May 2014 18:48:27 +0000 (20:48 +0200)]
Back to development
Sebastian Dröge [Sat, 3 May 2014 16:57:38 +0000 (18:57 +0200)]
textoverlay: Link unit test with the local version of the library, not an installed one
Sebastian Dröge [Sat, 3 May 2014 15:50:10 +0000 (17:50 +0200)]
Release 1.3.1
Sebastian Dröge [Sat, 3 May 2014 15:48:04 +0000 (17:48 +0200)]
Update .po files
Sebastian Dröge [Sat, 3 May 2014 15:22:10 +0000 (17:22 +0200)]
po: Update translations
Olivier Crête [Fri, 2 May 2014 23:09:59 +0000 (19:09 -0400)]
rtpbasepayload: Implement reconfigure event & renegotiation without subclass
Implement the reconfigure event, also do correct downstream caps negotiation
if the subclass doesn't implementy set_caps.
https://bugzilla.gnome.org/show_bug.cgi?id=725361
Olivier Crête [Fri, 2 May 2014 23:09:44 +0000 (19:09 -0400)]
tests/check/libs/rtpbasepayload.c: Run gst-indent
https://bugzilla.gnome.org/show_bug.cgi?id=725361
Sebastian Dröge [Sat, 3 May 2014 08:14:51 +0000 (10:14 +0200)]
Automatic update of common submodule
From bcb1518 to 211fa5f
Olivier Crête [Fri, 2 May 2014 22:30:16 +0000 (18:30 -0400)]
rtpbasepayload: Save the PT after fixating
Tim-Philipp Müller [Fri, 2 May 2014 18:36:34 +0000 (19:36 +0100)]
rtspdefs: remove outdated comments
Vincent Penquerc'h [Fri, 2 May 2014 14:09:35 +0000 (15:09 +0100)]
rtpbuffer: avoid underflow in size calculation
Thiago Santos [Thu, 1 May 2014 22:31:09 +0000 (19:31 -0300)]
videodecoder: do not parse caps for not using it
Saving some cpu
John Bassett [Fri, 3 Jan 2014 10:06:22 +0000 (11:06 +0100)]
rtpbasepayload: restrict initial random sequence number to be <= 32767
In order to prevent SRTP roll over counter issues the initial sequence
number is restricted to <= 32767. This is recommended by RFC 4568 section 6.4.
Sebastian Dröge [Thu, 1 May 2014 13:11:04 +0000 (15:11 +0200)]
sdp: Add some more gobject-introspection annotations for bindings
https://bugzilla.gnome.org/show_bug.cgi?id=729123
Sebastian Dröge [Thu, 1 May 2014 11:15:57 +0000 (13:15 +0200)]
playbin: Don't block on non-serialized events
https://bugzilla.gnome.org/show_bug.cgi?id=729321
Sebastian Dröge [Thu, 1 May 2014 11:08:24 +0000 (13:08 +0200)]
playsink: Don't block on non-serialized events
https://bugzilla.gnome.org/show_bug.cgi?id=729321
Sebastian Dröge [Thu, 1 May 2014 11:06:53 +0000 (13:06 +0200)]
playsinkconvertbin: Don't block on non-serialized events
https://bugzilla.gnome.org/show_bug.cgi?id=729321
Sebastian Dröge [Thu, 1 May 2014 11:05:05 +0000 (13:05 +0200)]
subtitleoverlay: Don't block on non-serialized events
https://bugzilla.gnome.org/show_bug.cgi?id=729321
Vincent Penquerc'h [Wed, 30 Apr 2014 10:06:27 +0000 (11:06 +0100)]
rtcpbuffer: check claimed data size against available size
Coverity 1208773
Göran Jönsson [Wed, 23 Apr 2014 06:06:36 +0000 (08:06 +0200)]
rtspconnection: Empty queue when flush.
Empty the watchs queue when calling
gst_rtsp_watch_set_flushing with flushing variabel is TRUE.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=728772
Ognyan Tonchev [Sun, 16 Mar 2014 15:09:36 +0000 (16:09 +0100)]
rtspconnection: Add more tests
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=728907
Luis de Bethencourt [Tue, 29 Apr 2014 14:15:47 +0000 (10:15 -0400)]
videotestsrc: fix undefined behaviour of left-shift
With a small type for the color values being left-shifted, the result is
undefined and it could potentially overflow.
https://bugzilla.gnome.org/show_bug.cgi?id=729195
Tim-Philipp Müller [Tue, 29 Apr 2014 09:59:02 +0000 (10:59 +0100)]
win32: fix export files again
Revert unintended parts of
d8a0927930a87a2eb60d4c98cb3fea8aed911b27
Christian Fredrik Kalager Schaller [Tue, 29 Apr 2014 09:39:18 +0000 (11:39 +0200)]
Add mikey.h file
Haakon Sporsheim [Tue, 29 Apr 2014 07:58:21 +0000 (09:58 +0200)]
audiodecoder: Make caps writable before fixating
https://bugzilla.gnome.org/show_bug.cgi?id=729114
Sebastian Dröge [Tue, 29 Apr 2014 07:54:18 +0000 (09:54 +0200)]
sdpmessage: Add array length annotation to gst_sdp_message_parse_buffer
https://bugzilla.gnome.org/show_bug.cgi?id=729123
Stian Selnes [Tue, 29 Apr 2014 06:46:02 +0000 (08:46 +0200)]
rtpbuffer: fix memory leak when gst_rtp_buffer_map fails
Make sure rtp->data[3] is set before jumping to error path.
https://bugzilla.gnome.org/show_bug.cgi?id=729117
Ravi Kiran K N [Mon, 28 Apr 2014 13:17:06 +0000 (18:47 +0530)]
gst-play: add option to supply media files from playlist file
https://bugzilla.gnome.org/show_bug.cgi?id=728845
Tim-Philipp Müller [Sat, 26 Apr 2014 23:49:01 +0000 (00:49 +0100)]
giobasesink: we mustn't change the format of a query response
Not even in the DEFAULT case. That's bad 0.10 behaviour, no caller
is ever going to check the format of the response.
Tim-Philipp Müller [Sat, 26 Apr 2014 23:25:16 +0000 (00:25 +0100)]
playbin: add nick for soft colorbalance play flag to fix gst-inspect
Fix gst-inspect-1.0 playbin criticals when printing the
flags, which was caused by a missing nick name for one
of the flags.
Tim-Philipp Müller [Sat, 26 Apr 2014 22:26:09 +0000 (23:26 +0100)]
docs: remove outdated and pointless 'Last reviewed' lines from docs
They are very confusing for people, and more often than not
also just not very accurate. Seeing 'last reviewed: 2005' in
your docs is not very confidence-inspiring. Let's just remove
those comments.
Sebastian Dröge [Fri, 25 Apr 2014 15:32:59 +0000 (17:32 +0200)]
giobasesink: Implement handling of the SEEKING query
Edward Hervey [Fri, 25 Apr 2014 09:30:37 +0000 (11:30 +0200)]
audiodecoder: Plug caps leaks
We were returning in various places without unreffing the caps, and
we were also leaking (overwriting) the caps we got from _get_current_caps()
Spotted by Haakon Sporsheim in #gstreamer
Sebastian Dröge [Tue, 22 Apr 2014 16:28:10 +0000 (18:28 +0200)]
audioresample: Don't left-shift into the sign bit, instead use unsigned integers