b1ae10b172
We already support decoding of the x-mt line. This change adds the a=x-mt line to the SDP offer. This is not a backward compatible change for media transport (because of the changes in pre-shared key handling) 1) if media transport is enabled, and SDES is enabled, generate the media transport offer. 2) if media transport generated the offer, add that offer to the x-mt line. 3) in order to create media transport, require an x-mt line (backward incompatible). The way it works is that 1) PeerConnection, on the offerer, asks jsep transport for the configuration of the media transport. 2) Tentative media transport is created in JsepTransportController when that happens. 3) SessionDescription will include configuration from this tentative media transport. 4) When the LocalDescription is set on the offerer, the tentative media transport is promoted to the real media transport. Caveats: - now we really only support MaxBundle. In the previous implementations, two media transports were briefly created in some tests, and the second one was destroyed shortly after instantiation. - we, for now, enforce SDES. In the future, whether SDES is used will be refactored out of the peer connection. In the future (on the callee) we should ignore 'is_media_transport' setting. If Offer contains x-mt, media transport should be used (if the factory is present). However, we need to decide how to negotiate media transport for data channels vs data transport for media (x-mt line at this point doesn't differentiate the two, so we still need to use app setting). This change also removes the negotation of pre-shared key from the a=crypto line. Instead, media transport will have its own, 256bit key. Such key should be transported in the x-mt line. This makes the code much simpler, and simplifies the dependency / a=crypto lines parsing. Also, adds a proper test for the connection re-offer (on both sides: callee and caller). Before, it was possible that media transport could get recreated, based on the offer. The tests we had didn't test this scenario, and the loopback media factory didn't allow for such test. This change adds counts to that loopback media factory, and asserts that only 1 media transport is created, even when there is a re-offer. Bug: webrtc:9719 Change-Id: Ibd8739af90e914da40ab412454bba8e1529f5a01 Reviewed-on: https://webrtc-review.googlesource.com/c/125040 Reviewed-by: Bjorn Mellem <mellem@webrtc.org> Reviewed-by: Steve Anton <steveanton@webrtc.org> Commit-Queue: Peter Slatala <psla@webrtc.org> Cr-Commit-Position: refs/heads/master@{#26933}
How to write code in the api/ directory
Mostly, just follow the regular style guide, but:
- Note that
api/code is not exempt from the “.hand.ccfiles come in pairs” rule, so if you declare something inapi/path/to/foo.h, it should be defined inapi/path/to/foo.cc. - Headers in
api/should, if possible, not#includeheaders outsideapi/. It’s not always possible to avoid this, but be aware that it adds to a small mountain of technical debt that we’re trying to shrink. .ccfiles inapi/, on the other hand, are free to#includeheaders outsideapi/.
That is, the preferred way for api/ code to access non-api/ code is to call
it from a .cc file, so that users of our API headers won’t transitively
#include non-public headers.
For headers in api/ that need to refer to non-public types, forward
declarations are often a lesser evil than including non-public header files. The
usual rules still apply, though.
.cc files in api/ should preferably be kept reasonably small. If a
substantial implementation is needed, consider putting it with our non-public
code, and just call it from the api/ .cc file.