GitLab

Turns out it was only used once in the whole tree.

Avoid an infinite loop for this case too (only 3 tries to connect, then abort)

This improves readability a lot and prevent us from reinventing the wheel.

misc/objects.c: Don't rely on vlc_object_destroy() to destroy objects, but expects vlc_object_release to do it when the refcount goes to zero.
* Meaning, that when created objects gets a refcount to 1.
* Destroying is instantaneous and we don't have to poll for a few secondss or so to wait until the object's refcount reach 0.
* We now track vlc_object_t's mem leaks when libvlc_global is released (Hard error for now, so they don't get unoticed)
* We fail hard if an object is released with a refcount of 1 without being detached from its parent and its children, to make sure such cases don't go unoticed.
(make test or make check still pass after that one. VLC is known to leak one object when no module is loaded, this must be fixed).

Missing some cases that I could not test.
Also ffmpeg/chroma.c is locked, so I can but leave it broken.
Fix is for the remaining modules is obvious and the issue is easy to detect.

  not all video. enables to restream/transcode rtsp/h264 streams.
  Should also partly fix #397, it seems that most of h264 streams are
  with packetization-mode=1 eg. in decoding order (thou it should be
  checked here, and FIXME is for that).

- only start GET_PARAMETER keepalive-thread if x-asf is setted eg. when
  server is wms, DSS 5.5.4 & 5.5.5 doesn't like GET_PARAMETER at all and
  closes the connection when getting that, and they seem to use
  rtcp-data for liveness. Kasenna uses that KA in useragent-string.

input/input.c: Define DEMUX_CAN_SEEK and correctly set "seekable" accordingly. (This could introduce regression such as seekable item no more seekable, but I should have greped enough...).

Fix segfault when Connect() returned failure and fix memleaks. Make sure live555 modules exits while in Connect() if vlc is asked to stop.

SET_PARAMETER is simply *wrong*. Servers are not supposed to support except for proprietary extensions (meaning we cannot use for just any random server).
Even GET_PARAMETER is hardly good, as it is not mandatory to implement either, but it is a defined way to ping (which SET_PARAMETER ain't).
In any case, the only proper way to ping a stream is RTCP feedback.
When it comes to the control session, better check OPTIONS Allow: header for GET_PARAMETER.

I really wonder how live555 can be so low-level though - smells like we are trying to be too intelligent here.

on Glen Gray patch.
(The code is commented until it is confirmed to compil/work.)

Fix hang in live555 when sending RTSP OPTIONS by letting readSocket() timeout after 5 seconds. Added reconnect case to modules/demux/live555.cpp (without backing off scenario).

Patch by Glen Gray: Enable keepalive for Kasenna servers, they parse the User Agent string and check for _KA appended to the User Agent name.

Patch by Glen Gray: RTSP keep alive fix. If the server doesn't handle RTCP-RR packets and requires specific GET_PARAMETER commands (such as Kasenna) then the session times out when vlc is paused because the current version only sends the info when Demux() is called. Demux() is only called when data is received. This patch fixes this issue with a boolean that is set during PAUSED state that allows the timer thread to send the GET_PARAMTER messages instead and thus keeping the RTSP session alive.