fix(present): black video flashes and other bugs linked to Qt (original PR)

[PeculiarProgrammer]

This pull request fixes the screen going black after the video ends in versions of PySide6 v6.5.3 and newer. This allows for other errors (like #315) that require a newer version of PySide6 to also be fixed.

Fixes Issue

Closes #315

Description

I was attempting to get manim-slides to work with the new versions of PySide6, as to stop the Device loss detected in Present() error, and found that at the end of the video, a frame of size (-1, -1) would attempt to appear. So, to prevent this, I added a requirement that the frame should only be displayed if it is the same size as the frame before it. Upon adding, the video no longer disappeared once it ended.

Check List

Check all the applicable boxes:

• I understand that my contributions needs to pass the checks;
• If I created new functions / methods, I documented them and add type hints;
• If I modified already existing code, I updated the documentation accordingly;
• The title of my pull request is a short description of the requested changes.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 80.54%. Comparing base (c3e1aa0) to head (3b23c77).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #464      +/-   ##
==========================================
+ Coverage   80.41%   80.54%   +0.12%     
==========================================
  Files          22       22              
  Lines        1828     1835       +7     
==========================================
+ Hits         1470     1478       +8     
+ Misses        358      357       -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[jeertmans]

Hi @PeculiarProgrammer, thanks for your great work! I would be very happy to have a solution to this problem :)

Small question for now: why not just check if the size is not (-1,-1)? Checking if the frame size changes might not be compatible with manual window resizing, no?

[PeculiarProgrammer]

I was considering making it check for (-1, -1), however, I thought that it was possible for frames of other sizes to also be dead. It could just be unnecessary future-proofing, but at the worst, it would mean one frame is skipped if the frame changes size. Plus, I have found, at least in my testing, that the frame size stays constant despite adjusting the size of the window.

While testing to see if the frame size changes with window size changes, I also found that dead frames have a format of Invalid, while normal frames have that of NV12. Implementing a check to make sure the frame is not of format Invalid is probably optimal then. I'm not by my computer now, but I'll make that change once I am.

[jeertmans]

Thanks for your analysis! I think I would rather prefer checking for the format validity then, rather for the size as it does not convey any meaningful information (or is probably a consequence of the format being good or not).

[PeculiarProgrammer]

I swapped the reliance to be on pixel format, so it should be good now.

[jeertmans]

Oops, I didn't want to push on my main branch... trying to fix that asap

[jeertmans]

Sorry but I could not re-open your PR, especially as I have no more write access to your repository :/

I re-use your commits and opened a new PR, see #464.

Note: next time, please use a separate branch (instead of main) for pull-requests, as it makes it harder for me to make changes.