Antony Lee
Antony Lee
Perhaps it wasn't. I only wanted to point out that "someone intended to test with -OO", "we're not actually testing with -OO" and "tests are failing with -OO" :-)
Now fixed; the test failure is unrelated.
@rcomer Perhaps we can decide whether we want to move forward with #14943 (or something along these lines) -- I don't really want to spend time arguing for it anymore....
OK, I rebased and reopened #14943, let's discuss that over there.
> keep self.figure with the current semantics, but discourate (deprecate?) it's use in favor of get_figure() - if needed turn it into a property. FWIW deprecating `.figure` would be extremely...
From a quick look, the relevant functions for cairo's "toy" font APIs are _cairo_{ft,quartz,win32}_font_face_create_for_toy, which do appear to have some (extremely limited) support for style/weight. Now we just need to...
I think the family name is too specific (from what I remember looking at the cairo code; haven't re-checked). OTOH as the cairo docs say themselves, trying to render text...
I rediscovered this issue in #28456, requesting support for `ax.set_xticks( **kwargs)` I didn't notice that `plt.xticks` actually already implements this possibility. On that thread @tacaswell argued (as above) to further...
> I'm a bit worried about having two ways to do the same thing. It's probably not going to happen, but I guess if we only wanted one way to...
> There's a conflict of interest: In a world, where ticks can change e.g. because of zooming, you can't have both options to be available: > > * I want...