It's already computed by the core.

Vertically swapping the position of the video keeps the place "height".
The vertical position use the inverted position of the bottom position
rather than the top position of the video.

Vertically swapping the position of the video keeps the place "height".
The vertical position use the inverted position of the bottom position
rather than the top position of the video.

Only for display modules that render regions separately.

We already require these display modules to be able to place and scale/stretch the video picture.

Steve Lhomme authored 1 year ago and Jean-Baptiste Kempf committed 1 year ago

The display module should not need to modify them.

Steve Lhomme authored 1 year ago and Jean-Baptiste Kempf committed 1 year ago

It contains only the useful fields from subpicture_t.

A display module should not need to know about channels, order, ephemer, fade,
etc.

Steve Lhomme authored 1 year ago and Jean-Baptiste Kempf committed 1 year ago

Similar to the base opengl display module.

When we create the window we know if it's for a projection or not.
That's all we need to initialize the sensors.

The code is disabled for display module not using the sensor (no projection).
The code is in C++ so we don't want to pull it when we don't need to.

No need to pass a whole vout_display_t.

CommonWindowInit() is not called in UWP builds, but the initial
CommonPlacePicture() needs to be done.

direct3d11 already does its own in Direct3D11Open().
Do it earlier in direct3d9 to be closer to the CommonWindowInit() call.

We don't need to have them separated, they are closely tied.

We can just fill an opaque event_thread_t pointer.

Steve Lhomme authored 2 years ago and Jean-Baptiste Kempf committed 2 years ago

We do not handle the reset_pictures callback so we can't return an error
here. We never returned one anyway.

Take only a struct vout_display_size rather than a full
struct vout_display_cfg_t. No functional changes.

This puts all the video size and placement properties back together as
vout_display_cfg_t originally was before other parameters were added.

No functional changes.

Rémi Denis-Courmont authored 2 years ago and Hugo Beauzée-Luyssen committed 2 years ago

The baseline windowing code has not been specific to the video output
for a while.

Alexandre Janniaux authored 3 years ago and Jean-Baptiste Kempf committed 2 years ago

The display code is better suited to execute Swap() than the
vout_display_opengl object, since:
 - It actually knows whether it's an on-screen or off-screen
   implementation.
 - It could add code to render more things between swapping.
 - In particular, it could render, but discard the rendering instead of
   swapping depending on the time spent or GPU state/chrono.

The API is not used anymore, and with the removal of the reference
counting pattern in objects, not required.

Indeed, the OpenGL providers are neither multi-thread nor reentrant
so it needs a locked stated when multiple clients needs to use it,
which will also needs to be reference counted across the clients, so
the whole reference counting can be done there.

In addition, if the context was given from module to other module, like
in the filter chain, the video contexts forwarded in those chains would
already be doing the reference counting. As a side note, there would
probably be few reasons to implement such forwarding currently anyway
since it would means that the beginning of the chain would lose usage
of the context, which also encompass a pool of output picture too
currently.

Lyndon Brown authored 5 years ago and Hugo Beauzée-Luyssen committed 3 years ago

see e967f81f.

note, this does **not** affect cat-based module selection items
(of which there are just three in use by the core), since that
mechanism uses subcats not cats.

Romain Vimont authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

This prepares to expose output size changes to filters in a generic way.

This avoids to expose the exact same module in different plugins.

Moreover, the "draw" filter is necessary to the OpenGL filter engine,
but is not directly related to any vout, so it was odd to include it in
vout plugins.

It's already set as a read-only pointer in vd->cfg before calling Open().

Romain Vimont authored 3 years ago and Jean-Baptiste Kempf committed 3 years ago

Benefits of *not* using designated initializers:
 - we can't forget a callback
 - it's the same in C and C++ (<20)

Benefits of using designated initializers:
 - the order of callbacks may not be wrong
 - we can read immediately which function is assigned to which callback
   (especially if some are NULL)
 - an optional callback may be added without modifying all the modules
 - we can grep a callback name to find its assignments easily
 - it is consistent with others xxx_operations in the codebase

Romain Vimont authored 3 years ago and Romain Vimont committed 3 years ago

The concrete sys type is local to each module.

Only 4 modules actually use the Reset Pictures feature to create a new filter
pipeline when they can't handle the source anymore. It's still faster than
closing the module and probing for a new display module.

In the future it may be called directly from the thread generating the event to
lower the latency.

Romain Vimont authored 4 years ago and Alexandre Janniaux committed 4 years ago

Implement a filter which just draws the input picture to the output.

This will be useful for "blend" filters which needs something to blend
with, even if they are used as the first filter.

Co-authored-by: Alexandre Janniaux <ajanni@videolabs.io>
Signed-off-by: Alexandre Janniaux <ajanni@videolabs.io>

Romain Vimont authored 4 years ago and Alexandre Janniaux committed 4 years ago

Load OpenGL filters from modules. Convert the existing renderer to a
module.

Co-authored-by: Alexandre Janniaux <ajanni@videolabs.io>
Signed-off-by: Alexandre Janniaux <ajanni@videolabs.io>

vd->cfg is only used in the vout thread so we don't need a local copy. We don't
need to modify it either.

This will prevent display modules from modifying it.

Only osys->source can be modified and sent to the display.

This is what other callers are doing. The format is only set when there are no
errors.