Thanks for looking into the patch :)

>I think we can assume that the same MIT license in the root directory of
>the nixpkgs repository also applies to the unmarked files in the
>repository.

Ok.

>I suppose the purpose of this glversion.txt is to avoid reproducibility
>issues?  I see it was added as a "pretty ugly hack" in the nixpkgs
>repository [1].

Yes, exactly. I also noticed, that it doesn't matter what the VENDOR and
RENDERER variables contain, it seems that it doesn't get used anyway.

>(Interestingly enough, this text file doesn't seem to
>cause me any issues, even though my computer doesn't support OpenGL 3.0,
>so I guess it's fine.)

I guess it would cause issues once a program tries to use OpenGL
extensions which are not supported on a given computer, otherwise it
shouldn't make any difference. As i understand it this file is supposed
to contain a minimum of OpenGL extensions which are widely available, so
that this (hopefully™) won't happen.

>At the top of the makefile patch, could you add a brief explanation of
>the purpose of the patch and/or link to the source of the patch?  It
>makes the patch much more maintainable, and you can find examples of
>this throughout the gnu/packages/patches subdirectory in the Guix
>repository.

Done.

>There's a minor parenthesis accounting mistake on the lines above.  If
>you delete the last right parenthesis, the indentation issue should be
>fixed and the code should compile.

snap, it seems i made a mistake in magit while adding this line...

>I almost missed this note on my first review!  Please place the package
>definition somewhere in the file so that it leaves the file in some sort
>of alphabetical order.

Yes, after closer inspection of that file i found the order - everything
was mixed at the end so i didn't see it.

>I'd name the above phase "patch-makefiles" since you've patched multiple
>files, but that's just me. ;-)

Makes sense 😁.

So i hope i have addressed all things you mentioned.


Malte