ref(sdp): Do not add recvonly SSRC for muted video tracks.
Currently, when a participant joins hidden/video muted, we add a
recvonly SSRC so that the client sends RTCP reports with that SSRC. This
SSRC doesn't have neither simulcast nor RTX enabled. When the user
decides to enable his video camera, we "enhance" this SSRC with params
for RTX and simulcast. However, Chrome does not take into account these
new params because of
https://bugs.chromium.org/p/webrtc/issues/detail?id=7555. By not adding
a recvonly SSRC, we activate the default behavior in Chrome which is to
send RTCP reports with SSRC 1. So, when the user decides to enable his
video camera, instead of updating/enhancing the existing SSRC, we
initalize the primary SSRC with the correct params.
fix(stats): lazy initialization for statistics Sets
Internet Explorer is throwing an error when doing a for...of over
Sets. This is because IE needs a Set polyfill, which may be applied
after lib-jitsi-meet has loaded. However, lib-jitsi-meet on load
creates two Sets, one for CallStats.fabrics and another for
Statistics.instances. When the polyfill is then applied, the two
existing Sets do not get new methods. The fix is to lazily
initialize the Sets to allow for polyfills to be applied.
fix: ensure container is defined while attaching a track
For muted tracks, it is possible for undefined to get pushed
into the track's internal containers reference. The fix is
to immediately set the internal container variable in .attach()
to the passed in container value.
Sending the bare jid, speeds call setup as otherwise jigasi needs to do a feature discovery finding the conference component address and use it to construct the room address to join.
Splits the participant connection status detection logic into two
separate flows one for P2P and one for JVB.
Fixes problem where user outside of last N would be reported as INACTIVE
when in P2P mode.
Added some improvements to the detection logic by adding more priority
for the video tracks frozen detection (if supported) over the last N.
When the browser supports this detection and it's playing video we
should not check the last N, because it's less reliable.
Will emit BEFORE_STATISTICS_DISPOSED event, before the last CallStats
fabric is terminated to make sure that the final logs batch is reported
correctly.
Also adds a check for CallStats instance to queue events when
the backend has been initialized, but there is no CallStats instance
available to report on.
* ref(CallStats): cleanup constructor
Changes CallStats constructor to not take the whole JingleSessionPC as
it only needs an alias and the TraceablePeerConnection instance.
Describes the arguments in JSDoc.
* ref(CallStats): rename var
Everything is callstats c'mon...
* ref(CallStats): remove _checkInitialize
The _checkInitialize was trying to workaround CallStats lib issue
without really checking for any specific type of error on whether or not
it makes sense to retry.
Also it depended on some internal fields of 'callStatsBackend' and was
binding 'initCallback' to the backend instead of CallStats instance
which made no sense (it means it's very likely this functionality was
broken anyway).
It would be hard to fix it in a clean way, because CallStats instance
fields would have to be stored in static variables in order to make
the initCallback work (called from '_checkInitialize').
We also need to have more than one CallStats instance running at the
same time, because of the P2P which makes things even more complex.
* fix(CallStats): do not catch 'sendApplicationLog'
Wrapping 'sendApplicationLog' in 'tryCatch' will result in an endless
loop, because it will be logged on the logger.error which then tries
to send the logs immediately again.
* ref(CallStats): do not use call on static
There's nothing more confusing that seeing 'this' in a static method.
Wow maybe these methods are not really static !?
* ref(stats): fix var name
* feat(stats): report P2P to CallStats
Will create new CallStats fabric for the P2P peerconnection in order to
log peer to peer connections.
Refactoring was required in the statistics and CallStats module to be
able to have more than one CallStats instance. Because each CallStats
fabric reports one peer connection now each CallStats will correspond to
one TraceablePeerConnection. CallStats instances are now stored in a Map
mapped by the TraceablePeerConnection.id field.
In order to be able to execute some global/static CallStats reporting
methods all Statistics instances also need to be stored in a static
field.
CallStats API backend(new callstats()) will be initialized only once for
the values provided in the first call to initializeBackend. It is not
possible to have more that one CallStats backend running at the same
time (at the time of this writing). If we would have a routine for
disposing global "Statistics" module we could try to cleanup static
reference and allow to initialize it with new values (but no such use
case yet).
* ref(CallStats): move initCallback
Since there is no alphabetic order preserved in this file anyway at
least place it closer to it's usage.
* ref(CallStats): remove tryCatch
Temporarily removes tryCatch to make the ES6 conversion easier.
* ref(CallStats): convert to ES6
* style(CallStats): fix indentation
* fix(statistics): use import for CallStats
* ref(CallStats): convert static methods
Some of the methods should not be static, because it only make sense
to call it when there is CallStats instance available.
* ref(CallStats): rename var
* doc(CallStats): remove misplaced comment
* chore(CallStats): remove invalid eslint comment
* fix(CallStats): undefined CallStats namespace
If no CallStats ID namespace option is provided the conference will be
reported without it.
* style(stats|CallStats): remove extra lines
* fix(CallStats): do not log error from tryCatch
GlobalOnErrorHandler calls logger.error anyway.
* fix(CallStats): cleanup tryCatch
If I understand correctly our initial intention with doing tryCatch was
to avoid crashing the whole app in case the CallStats backend would
crash. With this commit the tryCatch is done by wrapping original
backend instance methods or using explicit try catch block where
the method is called only from one place or a value needs to be returned
in case of a crash.
* ref(CallStats): make backend a static var
* fix(CallStats): invalid eslint comments
* ref(CallStats): use for..of
* ref(CallStats): fixes around REST args
* ref(CallStats): rename var
* ref(CallStats): reorder static methods
Also renamed some callbacks
* doc(CallStats): adds some docs
* ref(CallStats): make methods not static
Both 'sendDominantSpeakerEvent' and 'sendScreenSharingEvent' methods are
not really static as they require instance to be called.
* fix(CallStats): invalid key
* fix(CallStats): reduce amount of debug logs
* feat(p2p|CallStats): log hold/resume
Will put CallStats fabric for the JVB connection "on hold" while in p2p
mode.
* doc(CallStats): add FIXME
* doc(JitsiConference,CallStats): typos and renaming
When muted track was added to conference "mute" operation was
executed again which was not executing anything because the state
of the track was already muted.
If the platform supports it, it calls MediaStreamTrack._switchCamera in order to
switch the camera facing mode at a low level. It's currently only implemented
for React Native.
feat: add methods to check muti-mic support and stats collecting
For audio previewing, some browsers cannot have multiple local
audio streams at once. For audio output previewing, local
stats need to be collected to detect changes in volume level.
Pull Request #435
* Adds peer connection statuses (active/inactive/interrupted/restoring).
The peer connection used to be check with is connection active. Adding new states like active, inactive, when the connection was intentionally stopped by the bridge, interrupted due to network problem and restoring, which is the state of a peer which was inactive as being out of lastN and is now entering lastM and eventually will become active.
* Adds timers to track restoring status.
If restoring too long we set the connection status to interrupted. We save timestamps for all participants entering lastN and set the status for the connection to restoring for 5 seconds, or till it goes into active or exits lastN.
* Adds initial connectionStatus of newly created participant - active.
* Fixes some jsdocs.
* Uses symbols for strings and use of map, fixing comments on PR.
* Removes symbols.
* ref(JSPC): rollback doRenegotiate
Modify the code to the old strategy without shared 'doRenegotiate'
method.
* ref(JSPC): add/remove duplication
Removes duplication between add and remove remote stream methods.
* fix(JSPC): remove Timeout
Removes wait timeout for remote stream added/removed. It should not be
necessary given that the task executes on the modification queue and
the initial offer/answer should execute before
'source-add'/'source-remove'. Jingle 'session-invite'/'session-accept'
is sent, before any other notifications.
* feat(JSPC): verify SSRC changes
Will print an error if there was change to local SSRCs where it should
not happen (video mute on browsers where video stream is disposed on
mute).
* fix(JSPC): not always renegotiate
When adding/removing tracks as mute/unmute it only makes sense to
renegotiate if the initial offer/answer cycle has already been executed.
* ref(JSPC): remove duplication
Removes duplication around add/remove as unmute/mute.
* ref(JingleSessionPC): add local tracks with offer/answer
Refactor the current code to add local tracks together with initial
offer/answer to make this atomic/single operation on
the modificationQueue.
This is required to be able to get rid of 'delayedIceCandidates' list
and to execute 'addIceCandidate' task on the queue. If local track
addition is not bundled with initial offer it will often happen that ICE
candidates are queued, before the offer/answer which is queued only once
the local tracks task is done.
* ref(JSPC): use queue for ICE candidates
Refactors the code to get rid of 'delayedIceCandidates' buffer and
use the modification queue to synchronise with the initial offer/answer.
* doc(JingleSessionPC): improve docs
* ref(p2p): use SDP 'inactive' for JVB
This commit gets rid of detach/attach logic and
replaces it with SDP media direction modification.
Now after switching to P2P instead of detaching local
tracks from the JVB peerconnection only the media direction
will be changed to 'inactive'. This will suspend media
transfer on the JVB connection.
It's meant to mitigate Chrome issue where it would randomly
choose to stop sending audio stream, after we modify it's
local SSRC.
* ref(SdpConsistency): cleanup 'recvonly'
Checking on 'recvonly' is no longer reliable way for detecting
whether video media is recvonly, because it can be set to
'inactive' at any point.
Added a check to make sure that the recvonly SSRC is not generated
more than once from '_createOfferOrAnswer'.
* log TPC instance
Logs the TraceablePeerConnection from LocalSdpMunger and
SdpConsistency as part of log messages. Given that there can be
2 TPCs at the same time it was hard to tell for which connection
a log message was printed.
* fix(SdpConsistency): depend on 'recvonly' direction
When createAnswer/createOffer returns SDP it will contain 'recvonly'
direction to reflect current media stream state even though we set
it to 'sendrecv' previously. Using 'hasAnyTracksOfType(VIDEO)' can
not be used to reliably detect 'recvonly' direction when doing video
mute, because muted video track is still in the TPC's map.
* doc(LocalSdpMunger): update description
LocalSdpMunger is currently used only for muted local video tracks.
* fix(P2P): enable P2P on Firefox
After 'detach/attach' is gone it seems that P2P works fine with Firefox.
* ref(TPC): simplify expression
* ref(SdpConsistency): do not pass whole TPC
There's no need to pass whole TPC instance if the only thing needed is
a logging prefix.
* Adds video sip gw handling.
Parsing presence and availability of the service. Adding possibility to start/stop a video sip gw session.
* Fix comments.
* Adds more jsdoc about create session and renames import to use the filename is imports.
The first time a lastN was received it was ignored because the old lastN was
null, handle that case properly.
Furthermore, use default function parameters to avoid a conditional.
damencho
Saúl Ibarra Corretgé
bgrozev
George Politis
Leonard Kim
Iñaki Baz Castillo
Saúl Ibarra Corretgé
Lyubo Marinov
Aaron van Meerten
paweldomas
yanas
hristoterezov