OpenCV 4.10.0 - Numpy 2.0

[ilya-lavrenov]

Checklist

• Used a personal fork of the feedstock to propose changes
• Bumped the build number (if the version is unchanged)
• Reset the build number to 0 (if the version changed)
• Re-rendered with the latest conda-smithy (Use the phrase @conda-forge-admin, please rerender in a comment in this PR for automated rerendering)
• Ensured the license file is being packaged.

[conda-forge-webservices]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

[ilya-lavrenov]

@conda-forge-admin, please rerender

[hmaarrfk]

Do we want to drop qt5 builds

[h-vetinari]

Do we want to drop qt5 builds

Fine by me per se. What's the status on the overall qt6 migration? How many feedstocks (if any) are still qt5-only? I'd kinda like to drop qt5 in general...

[hmaarrfk]

Spyder is in general PyQt (not sure if they handle 6) and require webengine. Which we don't have for Qt6.

However. All platforms have a "headless" option. So it should be co-installable, just without the GUI parts on opencv.

[h-vetinari]

Let's move this discussion to a wider forum...

[h-vetinari]

The situation here is exceptional in many ways - the outputs are named differently (qt-main vs qt6-main) to support parallel installation, and so we cannot migrate as usual (until we publish a qt-main==6.x).

[hmaarrfk]

The situation here is exceptional in many ways

qt is in fact, an exceptional package!

[hmaarrfk]

So the unique thing here is that users can still install opencv + spyder.

They just won't be able to get:

• Spyder + OpenCV + HighGUI for OpenCV 4.10

They owuld get

• Spyder + OpenCV Headless 4.10
  or
• Spyder + OpenCV 4.9 + HighGUI

I think that is ok.....

[jakirkham]

Worth noting that this also contains NumPy 2 support

[hmaarrfk]

ah, then maybe the right thing is to pull in the numpy 2.0 migration and "do one last qt5 build".

I "promise it will be the last one".

[hmaarrfk]

@ilya-lavrenov could you pull in the numpy 2.0 migration from:
0df0cee#diff-514275248012b4feffb0d4ca055d87a939663091311ef79db51d7e4a7dad3ed2

it should build in a way that is compatible with numpy 1.X and 2.X giving this the last hurra!

[ilya-lavrenov]

@conda-forge-admin, please rerender

[github-actions]

Hi! This is the friendly automated conda-forge-webservice.

I tried to rerender for you but ran into some issues. Please check the output logs of the latest webservices GitHub actions workflow run for errors. You can also ping conda-forge/core for further assistance or you can try rerendeing locally.

This message was generated by GitHub actions workflow run https://github.com/conda-forge/opencv-feedstock/actions/runs/9364092311.

[ilya-lavrenov]

@conda-forge-admin, please rerender

[ilya-lavrenov]

Looks like numpy 2.0 is not supported by OpenCV.
They use Numpy API directly during OpenCV Python API implementation.

[traversaro]

Looks like numpy 2.0 is not supported by OpenCV.
They use Numpy API directly during OpenCV Python API implementation.

According to opencv/opencv#25455 opencv 4.10 should support numpy 2.

[hmaarrfk]

There is probably something off in our windows installation script.

[hmaarrfk]

Back when I started to get involved in CF, one of my major contributions was a refactor to this installation script.
https://github.com/ilya-lavrenov/opencv-feedstock/blob/opencv-4.10.0/recipe/bld.bat#L117

there is probably alot left over from the python 2.7/3 days that can be removed.

If you do end up attempting a cleanup Ilya, please skip most of the jobs by adding something like:

skip: true   # [py!=310]
skip: true   # [ffmpeg == 6]

and troubleshoot with that.

[github-actions]

Hi! This is the friendly conda-forge automerge bot!

I considered the following status checks when analyzing this PR:

• linter: passed
• azure: passed

Thus the PR was passing and merged! Have a great day!

[jakirkham]

Congratulations all! 🥳

Thanks for all your hard work on getting this one done! 👏