- 02 Aug, 2018 1 commit
-
-
Philippe Gorley authored
Adds a way to get input parameters. Simplifies usage of simple filters. Adds const where useful. Change-Id: I768d2f3e4e46e791e10530a1c3010607fecc42da
-
- 31 Jul, 2018 1 commit
-
-
Adrien Béraud authored
Change-Id: I0b6602938b931d3bdf06cdd1e96635c8781a9663
-
- 30 Jul, 2018 7 commits
-
-
Andreas Traczyk authored
Change-Id: I02061e1319499e694817000e6184d1735c67261f
-
Philippe Gorley authored
If the PCM signed 16 decoders cannot be found, skip the test Change-Id: If7ddf9778465415232e59f43b88ba17bf3bf7dbd Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
Adrien Béraud authored
Change-Id: I74b2c33e010aeff66e3bf346ca9db2c98d140448
-
Adrien Béraud authored
Change-Id: I4ba5e23ae4bd40c8520f8e4a5a59583616b2e8aa Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
Philippe Gorley authored
FFmpeg's RTP muxer does not support GSM, while Ubuntu does not package iLBC. Removes the old configure switches. Removes package detection from contrib. Keeps contrib rules in case GSM is later readded. Change-Id: I705a521e2baf34c757c22a72e04cdf27233b5a7b
-
Philippe Gorley authored
Makes both ringtone and dtmf tones shared pointers instead of raw pointers to avoid race conditions and use-after-free's. Change-Id: Ib8dc12c17905ca9bf0bc9dc01667b3194be70927
-
Philippe Gorley authored
This is redundant, the information can already be found in the AVStream object and is set by avcodec_parameters_to_context. Let FFmpeg take care of these details. Change-Id: I199de87a852caeef34047e4f906ccd21dcb696c4 Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
- 27 Jul, 2018 3 commits
-
-
Adrien Béraud authored
Change-Id: Ie34463b6ecbd04d37552dedbb07771f968d34061 Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
Adrien Béraud authored
requestKeyframe is stateless and failure is not critical. Change-Id: Iced5b13496e07153e7352d316c4cb5b06a9aab89 Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
Philippe Gorley authored
Moves the AudioBuffer to AVFrame conversion logic outside of the MediaEncoder, so the latter only deals with AVFrame. Does not copy the buffer, the resulting AVFrame will simply point to the existing buffer. This is the first in a series of patches attempting to rid Ring of its RingBuffer/RingBufferPool system. Change-Id: I17c9fcdb96dc945c3eeace647303339d558bcf38 Reviewed-by:
Adrien Béraud <adrien.beraud@savoirfairelinux.com>
-
- 26 Jul, 2018 2 commits
-
-
Adrien Béraud authored
Change-Id: I790780a6de8e4bf76177960a84527fd4eb6378f1 Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
Hugo Lefeuvre authored
Change-Id: Ifbc5f3792a7019dcfa8249eb30afe2a64cbeb57c Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
- 23 Jul, 2018 1 commit
-
-
Philippe Gorley authored
FFmpeg's av_image_fill_black supports more pixel formats. Works around casting int* to ptrdiff_t* on platforms where ptrdiff_t is not the same size as int by recreating the linesize array. Change-Id: I2825b18c6acc921a3cd3bb1398565df0aa657153
-
- 20 Jul, 2018 2 commits
-
-
Andreas Traczyk authored
Change-Id: Ib78e459812dddc3d25dab7998f7ea5b9da07a758 Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
Andreas Traczyk authored
As described in RFC 2462, an ipv6 address can have several states and some should be ignored when initializing an ICE session: In fact DEPRECATED address SHOULD NOT be used in new communications. Also, a TENTATIVE address is not ready to be used too. So, this patch ignores this kind of addresses when initializing a new connection. Change-Id: I95637286eb68451352fd7289a49ddc3cdf5ac383 Gitlab: #29Reviewed-by:
Andreas Traczyk <andreas.traczyk@savoirfairelinux.com>
-
- 18 Jul, 2018 2 commits
-
-
Philippe Gorley authored
Change-Id: I9b6a97a7f60a4533f24a48c4638b47973c54b0f5 Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
Hugo Lefeuvre authored
setAccountDetails emits AccountDetailsChanged, not AccountsChanged. Change-Id: Ic0af35d10ab7050f648b0dece0975f59370d077c
-
- 12 Jul, 2018 4 commits
-
-
Philippe Gorley authored
Moves the recorder up one level to the VideoInput, VideoReceiveThread and AudioReceiveThread, instead of the MediaDecoder (there's no equivalent to the VideoInput in the audio layer). Emits the RecordPlaybackStopped when the recording is stopped, so the client can sync its recording state with the daemon, in case the daemon stops recording by itself (rather than user intervention). Change-Id: I743b080cb354273ec074fec51caf2a4328fc1c58
-
Andreas Traczyk authored
- adds headers for msc compliance - removes C99-style casting of a list initializer in a C++ file Change-Id: Id77a3945c060b26d6a1bc5920e8461cc7310b309 Reviewed-by:
Philippe Gorley <philippe.gorley@savoirfairelinux.com>
-
Andreas Traczyk authored
Change-Id: I280be53a1246686dd6aba3be81f9937eee37ce27
-
Andreas Traczyk authored
Change-Id: Idcce4c27fba8d6680ae451afdc07b8ea2b72a471
-
- 11 Jul, 2018 1 commit
-
-
Sébastien Blin authored
a CANCEL message takes two parameters: 1. The peer device 2. The transfer id to cancel Only one was given in some cases Change-Id: I4f73b3d64be90bdb4d90fcb6bb935eb460840df5 Gitlab: #18
-
- 10 Jul, 2018 5 commits
-
-
Philippe Gorley authored
It better explains what the method does, as the recorder is not started at that moment). Change-Id: Ib536600b663e9ee188257b3b166f497f5a2b2c1e Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
Philippe Gorley authored
MediaRecorder no longer uses VideoSender (encoder) as a source for its local video and will directly use VideoInput (decoder), as it is earlier in the pipeline. This is the first step to support switching inputs while recording. Change-Id: Ia163efa3b20a349a93fc7b05213ec5e00de1704e Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
Philippe Gorley authored
Lowercase pins are valid, and easier to enter on a phone's smaller keyboard. Also makes it so some characters that can be difficult to tell apart from each other in uppercase easier to differentiate. Change-Id: I7aee29ceef0fbc69c046e0bb7ced486de2a3d4a4
-
Philippe Gorley authored
Fixes some recorder timing issues. Change-Id: Ia6f5d016a35dbe41ab08f01cd9506f318d120cfe
-
Adrien Béraud authored
Change-Id: I506dca0e5c4bb58b0c639219d4a4224f85fb2611 Reviewed-by:
Philippe Gorley <philippe.gorley@savoirfairelinux.com>
-
- 07 Jul, 2018 1 commit
-
-
Adrien Béraud authored
This reverts commit 6c77382c. Reason for revert: causes crashes in some conditions Change-Id: Ie6ce94a9893b08bcc553b67e043d6c6d7b38acfa
-
- 06 Jul, 2018 3 commits
-
-
Sébastien Blin authored
If a migration launched by the daemon fails when decrypting the account archive, it's because the password is incorrect. The client should do the migration itself. So, the daemon should informs this client with ERROR_NEED_MIGRATION Change-Id: If4871004bf6e09fedcb957b024f50cb723298001 Reviewed-by:
Andreas Traczyk <andreas.traczyk@savoirfairelinux.com>
-
Philippe Gorley authored
Overloads << operator to more easily stringify a MediaStream. Change-Id: I701b85fb7b98a43d336f4cb9a3fe1ab419e04827 Reviewed-by:
Sebastien Blin <sebastien.blin@savoirfairelinux.com>
-
Sébastien Blin authored
Currently, the daemon lie a bit about sendTextMessage. Indeed, it takes a map (string, string) like {{datatype: msg}} but the datatype is ignored. It can be problematic if a client wants to manipulate a custom datatype because the daemon will replace the datatype by "plain/text". Now, an ImMessage can have a datatype which permits to not lie in the api and manipulate the datatype the user wants. Change-Id: I1ec5a34082337570bf43231078fc894304b00182 Reviewed-by:
Philippe Gorley <philippe.gorley@savoirfairelinux.com>
-
- 05 Jul, 2018 3 commits
-
-
Philippe Gorley authored
Adds a Timer object to easily benchmark code. Should be used only for debugging code, so a warning is issued if the header is included. Change-Id: I89bc5454fc0c182a50b14cafd205ee116b4a167e
-
Philippe Gorley authored
Also fixes a regression where recorder instance couldn't be used more than once per call. Change-Id: Ia883e77a93ae3fcab7ead04f6001ddba1c1396ea
-
Philippe Gorley authored
Change-Id: I3a8f56aa4d93f73bd1290534e682f7d5f5025c5a
-
- 04 Jul, 2018 2 commits
-
-
Philippe Gorley authored
Adds a way to wait until the thread finishes its work before joining. Change-Id: I7080a4515ce6c1158c74c6eb8973a13a4b097419
-
Philippe Gorley authored
Change-Id: I35eaa5a339d5b4b6cb25fb7b12d6006cbf3202a5
-
- 03 Jul, 2018 2 commits
-
-
Hugo Lefeuvre authored
This commit adds a new startRecording method to the Recordable class. This method can be used to start a recording and takes advantage of MediaRecorder's new API (no call to setRecordingPath). Also, startRecording is meant to be passed the path directly instead of using initRecFilename which is removed in this commit. In addition to these changes we modify the Recordable class to operate on a single recorder_ object created at construction time instead of creating/destroying it in toggleRecording and stopRecording. Change-Id: Icf170e2a0ba5bbbb34afa37a54cf5edc76c8338d Reviewed-by:
Philippe Gorley <philippe.gorley@savoirfairelinux.com>
-
Hugo Lefeuvre authored
Whenever startRecording() is called and path_ is not empty (new API) the start time variable is left uninitialized. This commit addresses this issue. Change-Id: I9020e91695c6803d881967999d868d4ba4fd1c72 Reviewed-by:
Philippe Gorley <philippe.gorley@savoirfairelinux.com>
-