This switch input can be long and is not useful

Gitlab: #272
Change-Id: I53363f820e15431af55dcf0610451efc84a41d02

This reverts commit 42fe874c.

Reason for revert: <crashed on iOS>

Change-Id: I9eb71b16f43d0d77a763f1764a78cfc985763073

Change-Id: Iafca382f1b76ae3c77fc3613315c0f664647bdf6

Change-Id: Iaf3504ac0dcb0ccb9f924700c16f472e6ba0d61b

Change-Id: Ifa4bc41febb2149abd14fe8f4f30bb8e93bf9bb2

Change-Id: I75a1c7725059df1de153c57c565e954ee7537a98

Change-Id: Ia36712d1f7342bcffd8c47a7bcb3c2490b3762b1

Sébastien Blin authored 4 years ago and Adrien Béraud committed 4 years ago

Change-Id: I0de5e7483fd3fbef2dc767af1c44440f5cc33053

Change-Id: If1a026fd7ad474129a95c3fb4434cd0b3b598be5

The shutdown() are useless because the broken pipe error must trigger
a shutdown itself and this can cause some locks when the shutdown
callback is triggered when the socket is used to read/write data

Change-Id: If6c0cd481bfd8e00b57a30c4fb6d350756bff951

Ming Rui Zhang authored 4 years ago and Adrien Béraud committed 4 years ago

Change-Id: I2a4a3aa1fc4fbfea771556bf7f652aee335a66ce

Sébastien Blin authored 4 years ago and Adrien Béraud committed 4 years ago

Any client can try to answer a call multiple times. For example if dring
is running with the --auto-answer flag and the client set auto answer on
the daemon. This cause weird behavior like rotation not taken into account,
etc.

So, don't call answerCall if already answered.

Change-Id: Ibb967787aaf83e86a98fba7192810e86975304f1

This fork is deprecated and not used anymore on proxies
Also use official repo for http-parser

Change-Id: I2dcccb5616171ce20de17e805e77d72a8ebd167d

Cyrille Béraud authored 4 years ago and Pierre Lespagnol committed 4 years ago

Change-Id: Ie3cfe63ed6128aa294abab10e667142e36d44cb5

This reverts commit b4833f39.

Change-Id: I5368327b91ab6e4c7dcadba0ac4184174b573e9e

Change-Id: I735c0747045160a7a05fe075bd2be6c549589a6d

Change-Id: Ie6b67935036116b12aa5f8330e6a1a9e6a10679e

And some clean

Change-Id: I0e7a607b1749e608f0911f37065a0695df2d45b5

Because of the ConnectionManager, the daemon is now able to send
direct messages such as typing indications and read status. This
can be also used to move current features and improve it.

This patch is a first step to improve vCard syncing. The idea is
to send the vCard directly when a peer is connected. To avoid to
send the vCard everytime, a state is written in the cache directory.

In the future, a method to update the vCard as soon as the client
updates its profile will be possible.

Change-Id: I3801cbe04b3441f67ea6685aa2047e5a545958e5

Change-Id: I0d41bd1a8e55371748d36a1b5c3654ec03f6fd5b

This patch aims to improve the conference management for the host.
Now, the host is able to switch between 3 conferences layout:
1. The grid view (actual one) where all participants are shown
at the same height/width
2. The One big/Other in small which show one participant bigger than
the others
3. One participant in big

The daemon's API got two new methods:
+ setConferenceLayout() to switch between these layouts
+ setActiveParticipant() used in the 2 last layouts.

Change-Id: I3c16569e24d1b63331ffe9d79e35790a6ac47a0c

Change-Id: I2bf4ba82b72963109f1ffc7f959fc928abb0bbe7

Change-Id: I610f85d22ff06e715af2a8866dd4f135e0a98afe

Vladimir Stoiakin authored 4 years ago and Adrien Béraud committed 4 years ago

Change-Id: Ib5505badbef60a8c0832495589728de298f83607

Vladimir Stoiakin authored 4 years ago and Adrien Béraud committed 4 years ago

Change-Id: I321bdff614f1a90c5745cbedbebb89f0e0a1494f

Change-Id: I94c88a504ae2a98043be8fd201f95227002d99c4

Change-Id: Iffb7c8b4be4f084575e5e1a7bc6a02c86cb07fda

Change-Id: Iec1bb9a287d97dff8cc4342fe5ebf7d2238f10b4

Change-Id: I67971c024245e95e870bfed320db160a9142ff39

Change-Id: I990948900098449e37338bc6eed4ae4c4e110249

Change-Id: Ic1f124297a2f5cc7f8409e3ff064a444ebe04a1a

Change-Id: I5da233b3792215626fd82e827509ef95528e5be8

Instead of letting the TURN session destroy itself when finished, the current
code is overriding the user_data attached to the session. This is not needed
because the turn session will be stopped when the timer will complete and
this can cause some crashes in pj_turn_session_on_rx_pkt2 because the data
is erased while not protected via turn_sock->grp_lock() (mostly in turn_sock.c,
on_data_read)

Change-Id: Ia4a12917711b5534b372cb3a26e9141e5393c569
Gitlab: #299

The relay candidate was declared as suceeded even if not in the valid list.
Because of this, some negotiations were successful but no valid candidates
were given to the app.

Moreover, the connection timeout checks were using the wrong timer. This
avoids to cancel successful negotiations to check a connection timeout.

Finally, some clean is done with this patch

Change-Id: Iacfbf3b9f5faafe1bd05f66157485226761c30bf
Gitlab: #256

if onTlsStateChange is called during destroy, callbacks can change
between the check and the actual call.

Change-Id: I8528205c25da76281e7e6bc9fdb3b9c0fbf596b2
Gitlab: #234

Change-Id: Ie24e29fc97f1f83cc0148c1c75788cef8f1008df

Change-Id: I936dc5577014ac551b5f8726a3d3b12be46c091f

The same TID is used for all outgoing transfer (because it's the same file).
So we should use a vector of transfer not a single outgoing transfer

Change-Id: I3b492dc0198903672540a4365bfd32d6458970fd
Gitlab: #259

Change-Id: I4863d9d781634e3a15d75b030eaa8a0d08925f69
Gitlab: #246

Change-Id: If047cc1e32a5cb04c5b6a5341c416be6db3ee7e4