[akx]

Note

Stacked on #9908.
That PR ought to be reviewed and merged first and I'll happily rebase this.

Refs #9898 (not "fixes", since I can't be sure this ends up fixing OP's issues).

What's moving?

Some infrastructure:

• PILLOW_TEST_SNAPSHOTS can now be set to create, overwrite or update, so assert_image_similar_tofile and assert_image_equal_tofile can update the golden image to what was passed in; this is similar to --snapshot-update for Syrupy or INSTA_UPDATE for Rust's Insta.
• A test based on Kerning woes #9898's example code, golden snapshots updated as they improve over the patches.

These are the actual fixes; see #9908 for notes on the clean-ups:

• Passing in a bytestring where some characters had the high bit set would render mojibake/tofu/.NOTDEF (whatever you want to call it) when using the basic renderer. Very corner-case, but also the fix is just a proper (unsigned char) cast.
• Fractional variable font axis values were truncated; no more!
• When using basic font layout (which is the case for the default font added in Added TrueType default font to allow for different sizes #7354), advance values were 64× off due to type confusion (see Document and clarify _imagingft #9908's description). No more!
• Finally (the last 2/3 commits), glyph positioning within a single call is now subpixel and not pixel-snap-hinted (when available), leading to more accurate rendering.

[akx]

I'll rebase this again after #9908 gets merged since this builds on it 👍

[akx]

Rebased on the freshly rebased ft-clarify (#9908).

[akx]

Rebased on the once more freshly rebased ft-clarify, now with release notes targeting version 13.1.0.

[radarhere]

PILLOW_TEST_SNAPSHOTS can now be set to create, overwrite or update, so assert_image_similar_tofile and assert_image_equal_tofile can update the golden image to what was passed in; this is similar to --snapshot-update for Syrupy or INSTA_UPDATE for Rust's Insta.

I'm not sold on the need for this. It's rare that many test images need to be updated at once. For most scenarios, if you're writing tests, you can simply im.save() the individual image.

OVERWRITE seems dangerous. The reason for assert_image_similar is that some images are rendered differently in different environments. We run the risk of receiving PRs where someone has resaved all of the assert_image_similar images on their machine and blindly committed them.

[akx]

I'm not sold on the need for this. It's rare that many test images need to be updated at once.

This is a PR where I needed to do just that, and doing it manually seemed like a chore, since there were lots of affected tests, scattered throughout.

OVERWRITE seems dangerous. We run the risk of receiving PRs where someone has resaved all of the assert_image_similar images on their machine and blindly committed them.

Sure, and such a PR would probably be tagged in review, right, just like any change..? Either way I'm not convinced there's a huge risk of inadvertent usage, as you'd have to find the envvar, turn it on, then notice that "huh, a lot of files have changed, wonder why, ah well, I'll commit them anyway", and make a PR with all of them?

Either way, I can rip out the commit that adds that switch if you like.