Added area kwarg to mask.to_surface

[Matiiss]

Close #661
(I don't think any of that renaming mentioned there is going to happen, so, adding the area kwarg is about all there's left to it)

[MyreMylar]

Should the size of the area rectangle also be used to determine the size of the final surface created here - where no surface is provided?

It feels a bit odd to effectively ask for a chunk of a mask to be turned into a surface and then get more than that chunk back.

[MyreMylar]

Going to approve this, despite my slight quibbles about functionality as I think it will work either way. How it currently works is more similar to the .blit() version of the area parameter -but it does feel more logical to produce a cropped mask surface if we aren't explicitly supplying a surface.

[Matiiss]

Going to approve this, despite my slight quibbles about functionality as I think it will work either way. How it currently works is more similar to the .blit() version of the area parameter -but it does feel more logical to produce a cropped mask surface if we aren't explicitly supplying a surface.

I did use .blit()'s implementation as a reference, it made sense for this to be consistent with that at the time.

[zoldalma999]

Sorry for the late response, but I agree with Myre here. It would be better to use the rect's size for creating the surface. Otherwise looks good to me!

[Matiiss]

Some TODOs for myself:

• fix the docs (area rect doesn't affect the destination)
• add tests for a non-solid (un)setsurface to ensure that they are offset properly as well since their (0, 0) should align with the (0, 0) of the original mask

Other than that, please do check my C code, I went through tons of iterations on this, so something might have got lost or I might have missed something.