Make subroutines for encoding and decoding arrays of 16-bit big-endian
integers, and in the process fix a bug: When decoding an odd number of
bytes from be16, the least significant byte of the last int16 in the
array was properly taken to be zero instead of actually being read
(since it's outside the array). However, when encoding an odd number
of bytes, the least significant byte of the last int16 in the array
was written to the output as-is instead of being taken to be zero;
thus, we encoded one byte more than we should. This was probably not
harmful, and the value was dropped at decoding anyway; nevertheless,
writing a constant zero is the safe thing to do, and this patch does
so.
R=aluebs@webrtc.org, bjornv@webrtc.org, henrik.lundin@webrtc.org, turaj@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/28569004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7527 4adac7df-926f-26a2-2b94-8c16560cd09d
The problem was that if the output sample rate is changed such from one
where no resampling is needed to a rate that requires resampling, the
first output from the resampler will contain an onset period. The
solution provided in this CL is to keep a copy of the last output frame
in ACM, and if the resampler is engaged, it will be primed with this
old frame before resampling the current frame.
BUG=3919
R=bjornv@webrtc.org, turaj@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/27729004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7479 4adac7df-926f-26a2-2b94-8c16560cd09d
This CL implements a new unit test. The test is designed to trigger
a problem in ACM where switching the desired output frequency creates
a short discontinuity in the output audio. The problem itself is not
solved in this CL, but the failing test is disabled for now.
BUG=3919
R=kwiberg@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/23019004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7443 4adac7df-926f-26a2-2b94-8c16560cd09d
Libvpx now supports GN and this CL turns on compiling it.
I also introduced an executable target 'webrtc_tests'
that depends on all in WeBRTC + tests in order to get a full
linking step executed (since we've seen link problems for GN
when rolling WebRTC into Chromium).
I also converted a few test targets and made a GN file for
third_party/gflags.
BUG=3441
TESTED=Trybots + full Chromium build with a symlinked src/third_party/webrtc
dir to a workspace wit this CL applied.
R=brettw@chromium.orgTBR=niklas.enbom@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/25569004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7344 4adac7df-926f-26a2-2b94-8c16560cd09d
r7049 added some unnecessary casts ("return 0" -> "return static_cast<uint16_t>(0)"). r7123 converted these to "return 0u". The original impetus for this was to eliminate type conversion warnings. However, the 'u's are unnecessary; Visual Studio can return "0" from a function returning an unsigned value without producing a warning. The real reason for the original warnings was that the code was returning -1 from a function returning an unsigned value, which does need a cast; the -1s were eliminated before the above two revisions landed.
Also reverse the order of some conditionals to prevent possible underflow.
While the underflow wouldn't have changed the behavior of the code, it's easier
to reason about the code when such underflow can't happen, and possibly safer
against future modifications as well.
BUG=3663
TEST=none
R=tina.legrand@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/22599004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7296 4adac7df-926f-26a2-2b94-8c16560cd09d
This time reverts the Cl that actually broke the tests. Got the wrong rev before. :/
BUG=3520
TESTED=Locally with CHECKOUT_SOURCE_ROOT=`pwd` build/android/test_runner.py gtest -s modules_unittests --gtest_filter=AcmReceiverBitExactness.8kHzOutput --verbose --isolate-file-path=webrtc/modules/modules_unittests.isolate
TBR=henrik.lundin@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/26579004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7264 4adac7df-926f-26a2-2b94-8c16560cd09d
In order to maintain test coverage for the old API (AudioCodingModule)
during the transition period, the old test was copied to
AcmReceiverTestOldApi.
Modified and extended AudioCoding and the implementation to make the
test compile and run.
Created a converter method from new to old config struct
BUG=3520
R=kwiberg@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/31409004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7259 4adac7df-926f-26a2-2b94-8c16560cd09d
The converted tests are:
AcmIsacMtTest
AcmReceiverBitExactness
AcmSenderBitExactness
AudioCodingModuleMtTest
AudioCodingModuleTest
In order to maintain test coverage for the old API (AudioCodingModule)
during the transition period, the old tests were copied and given the
suffix OldApi:
AcmIsacMtTestOldApi
AcmReceiverBitExactnessOldApi
AcmSenderBitExactnessOldApi
AudioCodingModuleMtTestOldApi
AudioCodingModuleTestOldApi
BUG=3520
R=kwiberg@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/23719004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7258 4adac7df-926f-26a2-2b94-8c16560cd09d
Before this change it could happen that a large jump in timestamp (a
jump not correlated to wall-clock change) caused the audio to go silent
without recovering. The reason was that all incoming packets after the
jump were considered too old compared to the last decoded packet, and
were deleted. With CL changes two things:
1. If the only available packet in the buffer is an old packet, NetEq
will do Expand instead of immediate reset. This is to avoid that one
late packet triggers a reset.
2. Old packets are discarded only when the decision to decode a packet
has been taken. This is to allow the buffer to grow and eventually
flush if no decodable packet has been found for some time.
This CL also includes a new unit test for this situation.
BUG=3785
R=minyue@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/22709004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7255 4adac7df-926f-26a2-2b94-8c16560cd09d
RTPStream, and NetEq as such. Also mark all other virtual overrides in the same
files.
This will make further changes to these classes safer by ensuring that the
compile breaks if the base class changes and not all overrides are fixed.
This also deletes ACMTest.cc, which existed solely to define ~ACMTest(), which
was marked pure virtual in the header. (Pure virtual destructors still need a
definition.) Because there is another pure virtual method in this class, the
class is already abstract, so there's no benefit to making the desturctor pure.
Making it non-pure allows removing the separate source file.
BUG=none
TEST=none
R=henrik.lundin@webrtc.org, niklas.enbom@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/29389004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7144 4adac7df-926f-26a2-2b94-8c16560cd09d
