GitLab

core: create two variables at initialization (for var_AddCallback to get a higher chance to succeed!)

For the moment, it is only deactivated when using input_Read
and input_Preparse (it was making blocking vlc at start-up).

It is unused and not correctly implemented (!b_block case).

The pause must be done when the fifo of the decoder gets empty otherwise
the pause occurs too soon (especially with large caching).

It avoids to quit the pause state.

p_input->b_eof cannot be true at this stage.

Added a new option, play-and-pause. When this option is set, the input thread is automatically paused when each item reaches the end of the file.

This option is intended to make it easier to use VLC during live
productions featuring instructors and speakers working
with American Sign Language (ASL), or really any Sign Language, where
it is common for a lecturer to refer to video clips during a
presentation. The automatic pausing provides a smooth pause for the
video operator to switch the projection system back to the camera
with the live speaker once a clip has finished playing.
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>

This try to avoid vlc_object_find() as much as possible.
This is conservative, because where there is no associated parent input, we'll try to find in certain cases the parent input. This will probably be
 removed later on. Because yes, there is not necessarily a parent input for access and demux, especially if created from stream_UrlNew().

Deeper code can cope with too overly large values.
We need to accept seemingly too far start times because the length is
typically not known at startup (and hence set to zero).

 It will allows to control the moment at which the meta are updated
to fix the delay introduced by timeshift and buffering.
 It will also simplify the way meta data are read by demuxers.

It would be the proper thing to do to convert real time access_demux
to work without pf_demux callback (it includes oss,alsa,rtp,screen
at least).

 It allows to decrease the number of input thread wakeup and to
improve its reactiveness at user commands in case of data starvation.

Yes it's invalid, but there's little point in rejecting it

file:// URI syntax on Windows features a leading slash before the drive
letter (file:///C:/Foobar/...). This is irrelevant on WinCE, that only
has a root directory.

The repeat was done before buffered data were consumed.

Patch by ozvald (see #2733)

I'm not sure that VLC can't still leak memory in this case...

Brought to you by M2X. Needless to mention that this is complete 100%
crap. Depending cases, it's not either ignoring errors (and e.g. leaking
memory which is very stupid in OOM situations), or worse, introducing
bugs in places where realloc() was used correctly.

And exactly one day after I WARNED EXPLICITLY AGAINST DOING THIS.

This reverts commit ec5d8611.