This test was missing, which made me believe that it wasn't supported as
the handover state only included SSN and not MID. But when adding tests,
I saw that the current implementation used the SSN field to handover the
MID information for ordered streams which is sufficient given the 32 bit
type used for that (SSNs are only 16 bits).
For unordered streams, there is no need to handover any state there are
no expected next MID for unordered streams (they can be received in any
order).
So, adding tests and removing the handover state I just added.
Bug: webrtc:41481008
Change-Id: If1799cb1def5bd9f585a87cff6d835f4a9053b4f
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/370121
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Victor Boivie <boivie@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#43495}
The same as https://webrtc-review.googlesource.com/c/src/+/331340, but
for interleaved messages.
This avoids inserting into maps where possible, and also fixes a bug
when the payload was accidentally copied unintentionally -
crbug.com/365594101.
Bug: chromium:365594101
Change-Id: Iaeaa97b0cf3a26ada9afc61f2545760b7ab4c731
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/363960
Reviewed-by: Florent Castelli <orphis@webrtc.org>
Commit-Queue: Victor Boivie <boivie@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#43099}
This is the receive-side part of supporting what is frequently called
"ndata", but actually RFC8260 - "User Message Interleaving".
This CL adds a new ReassemblyStreams implementation that can assemble
I-DATA chunks and process I-FORWARD-TSN for partial reliability.
Bug: webrtc:5696
Change-Id: I3cfbea62e7b6c02fbd3f51b43ba3fb7863cf0f88
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/218506
Commit-Queue: Victor Boivie <boivie@webrtc.org>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37128}
