Prince Gupta authored 3 years ago and Jean-Baptiste Kempf committed 3 years ago

also move non-content item out off Control.contentItem

Prince Gupta authored 3 years ago and Jean-Baptiste Kempf committed 3 years ago

restore windows after initialization of CSDWin32EventHandler

ref #26104

Prince Gupta authored 3 years ago and Jean-Baptiste Kempf committed 3 years ago

fixes blank window visible at startup
ref #26104

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

There is no reason to compute the same offset for every fragment.

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

This code had been duplicated by error by
7fc16277.

Prince Gupta authored 3 years ago and Jean-Baptiste Kempf committed 3 years ago

fixes warning i.e cannot read property 'width' of null

Fixes #17029.

Fatih Uzunoğlu authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

QtQuick views destroy the delegate when it goes out of
boundaries set by cacheBuffer property. Since drag
items are "connected" to the delegate, the drag event
can not be handled properly.

Setting attached ListView.delayRemove property of the
delegate when drag is active prevents it getting
destroyed.

Steve Lhomme authored 3 years ago and Rémi Denis-Courmont committed 3 years ago

There was a ming32-only check to use LoadLibrary for this call. It was not
supposed to be done on any mingw64 builds. But the define used was only defined
on mingw64 for 64-bits builds.

Fixes LibVLCSharp#295

Patch from upstream master.

Alexandre Janniaux authored 3 years ago and Rémi Denis-Courmont committed 3 years ago

The device needs to be released in the offscreen implementation case but
it's not possible to differentiate with the on-screen implementation
that gets the window provided. Having the device available everywhere is
easier than providing callbacks to allocate the device for now.

Fix NULL dereference with window for on-screen implementation.

Regression from f483bdc9.

Fixes #26122

Alaric Senat authored 3 years ago and Pierre Lamot committed 3 years ago

Fix a regression introduced by ec78dfa6
where the lock was forgotten.

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

The input picture may contain padding, for two reasons:
 1. the rectangle defined by the picture_t offset and visible width and
    height may not cover the whole picture size;
 2. if the OpenGL API does not support NPOT textures, textures are
    created with power-of-two dimensions.

As a consequence, for sampling, the coordinates (between 0 and 1 in both
dimensions) must be transformed to take the padding into account.

These transformations were computed and performed per-plane. Therefore,
there were _n_ "TexCoordsMaps" matrices for a picture with _n_ planes.

However, these transformations are necessarily the same for all planes:
 1. the picture_t offset/sizes are defined for the whole picture (then
    scaled appropriately for each plane);
 2. since the ratio between different planes of the same pictures is
    always a power of two (e.g. in YUV 4:2:0, the size of U and V planes
    are half the size of the Y plane in both dimensions), this property
    holds even if we align the dimensions of each plane texture to the
    next power-of-two.

Therefore, use a single transformation matrix.

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

Multiply two vec2 directly instead of multiplying their components
separately.

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

We request kCVPixelFormatType_32BGRA which is BGRA, not RGB32.

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

CMTime is **not** equivalent to vlc_tick_t. In particular, the timescale
doesn't match. So it should be converted before usage.

In addition, GET_TIME is supposed to return where the access is
currently reading, which should also match the timestamps of the output
frames from the access.

Fixes #26101

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

AVCaptureVideoDataOutputSampleBufferDelegate is implemented by
VLCAVDecompressedVideoOutput but wasn't signalled, leading to warnings
when compiling.

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

ARC is not supported when storing NSObject in a structure. But it is
well-supported when storing a NSObject as (__bridge_retained void*) into
the p_sys until (__bridge_tranfer) releases it.

It highly simplify the whole handling of ARC objects and avoid using
CFRelease/CFRetain in the code, unifying with other Objective-C modules
like VLCVideoUIView or VLCOpenGLES2VideoView.

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

...instead of using custom debug strings

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

Set the framerate to 1 / framerate_base, avoiding the following warning
for a 30fps capture stream:

    rawvideo decoder warning: invalid frame rate 0/30000, using 25 fps
    instead

Marvin Scholz authored 3 years ago and Hugo Beauzée-Luyssen committed 3 years ago

Add a property that returns the NSImage that is currently used as
application icon for VLC. This is not always the same as
NSImageNameApplicationIcon due to easter eggs like the Christmas cone.

François Cartegnie authored 3 years ago and Jean-Baptiste Kempf committed 3 years ago

refs #26126