NetEq: Rectify the implementation of PacketBuffer::DiscardOldPackets

NetEq: Rectify the implementation of PacketBuffer::DiscardOldPackets

The implementation of this method did not follow the description in
the method comment. It was supposed to delete all packets in a range
[A, B], but if at least one packet in the buffer had a timestamp lower
than A, then no packets at all were discarded. This is now fixed.

BUG=webrtc:7937
Review-Url: https://codereview.webrtc.org/2969123003
Cr-Commit-Position: refs/heads/master@{#18903}
Committed: https://chromium.googlesource.com/external/webrtc/+/63d146b743fb2a4ee637dc59c5ab0a77d7ee1864

[hlundin-webrtc, 2017-07-05 11:39:26]

henrik.lundin@webrtc.org changed reviewers:
+ minyue@webrtc.org

[hlundin-webrtc, 2017-07-05 11:39:27]

PTAL.

[minyue-webrtc, 2017-07-05 12:11:08]

lgtm % nits

https://codereview.webrtc.org/2969123003/diff/40001/webrtc/modules/audio_codi...
File webrtc/modules/audio_coding/neteq/packet_buffer_unittest.cc (right):

https://codereview.webrtc.org/2969123003/diff/40001/webrtc/modules/audio_codi...
webrtc/modules/audio_coding/neteq/packet_buffer_unittest.cc:322: //
DiscardNextPacket call.
nice, another way would be avoid the for-loop. It is not needed to test
DiscardNextPacket so many times.

https://codereview.webrtc.org/2969123003/diff/40001/webrtc/modules/audio_codi...
webrtc/modules/audio_coding/neteq/packet_buffer_unittest.cc:330:
EXPECT_CALL(check, Call(i));
why not arbitrary number

https://codereview.webrtc.org/2969123003/diff/40001/webrtc/modules/audio_codi...
webrtc/modules/audio_coding/neteq/packet_buffer_unittest.cc:337: // This will
all remaining packets but one. The oldest packet is older than
This will "discard" all remaining ...

https://codereview.webrtc.org/2969123003/diff/40001/webrtc/modules/audio_codi...
webrtc/modules/audio_coding/neteq/packet_buffer_unittest.cc:339:
EXPECT_CALL(mock_stats, PacketsDiscarded(kRemainingPackets - 1));
define 1 as kSkipPackets

[hlundin-webrtc, 2017-07-05 13:42:35]

Thanks! I will CQ the new patch set.

https://codereview.webrtc.org/2969123003/diff/40001/webrtc/modules/audio_codi...
File webrtc/modules/audio_coding/neteq/packet_buffer_unittest.cc (right):

https://codereview.webrtc.org/2969123003/diff/40001/webrtc/modules/audio_codi...
webrtc/modules/audio_coding/neteq/packet_buffer_unittest.cc:322: //
DiscardNextPacket call.
On 2017/07/05 12:11:08, minyue-webrtc wrote:
> nice, another way would be avoid the for-loop. It is not needed to test
> DiscardNextPacket so many times.

True. But I still think I would like to have the InSequence at least.

https://codereview.webrtc.org/2969123003/diff/40001/webrtc/modules/audio_codi...
webrtc/modules/audio_coding/neteq/packet_buffer_unittest.cc:330:
EXPECT_CALL(check, Call(i));
On 2017/07/05 12:11:08, minyue-webrtc wrote:
> why not arbitrary number

Indexing the calls to the check function makes the printouts in case of error
more readable.

https://codereview.webrtc.org/2969123003/diff/40001/webrtc/modules/audio_codi...
webrtc/modules/audio_coding/neteq/packet_buffer_unittest.cc:337: // This will
all remaining packets but one. The oldest packet is older than
On 2017/07/05 12:11:08, minyue-webrtc wrote:
> This will "discard" all remaining ...

Done.

https://codereview.webrtc.org/2969123003/diff/40001/webrtc/modules/audio_codi...
webrtc/modules/audio_coding/neteq/packet_buffer_unittest.cc:339:
EXPECT_CALL(mock_stats, PacketsDiscarded(kRemainingPackets - 1));
On 2017/07/05 12:11:08, minyue-webrtc wrote:
> define 1 as kSkipPackets

Done.

[hlundin-webrtc, 2017-07-05 13:42:40]

The CQ bit was checked by henrik.lundin@webrtc.org

[hlundin-webrtc, 2017-07-05 13:42:40]

The patchset sent to the CQ was uploaded after l-g-t-m from minyue@webrtc.org
Link to the patchset: https://codereview.webrtc.org/2969123003/#ps60001 (title:
"Updates after review")

[commit-bot: I haz the power, 2017-07-05 13:42:45]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-07-05 14:03:34]

CQ is committing da patch.

Bot data: {"patchset_id": 60001, "attempt_start_ts": 1499262160252770,
"parent_rev": "440ea8cdffa2558500b1abf57e4faaafe1bac038", "commit_rev":
"63d146b743fb2a4ee637dc59c5ab0a77d7ee1864"}

[commit-bot: I haz the power, 2017-07-05 14:03:38]

Description was changed from

==========
NetEq: Rectify the implementation of PacketBuffer::DiscardOldPackets

The implementation of this method did not follow the description in
the method comment. It was supposed to delete all packets in a range
[A, B], but if at least one packet in the buffer had a timestamp lower
than A, then no packets at all were discarded. This is now fixed.

BUG=webrtc:7937
==========

to

==========
NetEq: Rectify the implementation of PacketBuffer::DiscardOldPackets

The implementation of this method did not follow the description in
the method comment. It was supposed to delete all packets in a range
[A, B], but if at least one packet in the buffer had a timestamp lower
than A, then no packets at all were discarded. This is now fixed.

BUG=webrtc:7937

Review-Url: https://codereview.webrtc.org/2969123003
Cr-Commit-Position: refs/heads/master@{#18903}
Committed:
https://chromium.googlesource.com/external/webrtc/+/63d146b743fb2a4ee637dc59c...
==========

[commit-bot: I haz the power, 2017-07-05 14:03:39]

Committed patchset #4 (id:60001) as
https://chromium.googlesource.com/external/webrtc/+/63d146b743fb2a4ee637dc59c...