Remove *math* parameter of various mathtext internal APIs. #22507
Conversation
a0135d5 to
bba4a0a
Compare
|
How does this affect the possibility to support |
|
I think it's OK? Your implementation at #22173 should not be affected too much by this, as long as it doesn't perform the same dash-to-unicodeminus replacement as is now done by symbol(). |
bba4a0a to
f708792
Compare
|
OK! Just was a bit worried that the long dash was already in the string, so that Not sure I will finish #22173 (especially since I read that you have one), but we'll see. |
|
I don't have an implementation of Replacment to unicode minus occurs within |
|
OK! Just have some memory of reading that you had a branch hanging around (but I may be wrong and either I didn't read it or it was someone else). Ahh, I see now. Primarily read the description, which caused my doubt. |
|
I think I gave it a try at some point, but am not sure it's really ready either, so let's try to get yours in. |
|
The old image had a non-unicode minus (due to #21687); this patch changes the image. But we try to avoid adding new (or modified) baseline images if possible (due to repo size concerns); here the test was really to check what are the ticks so now I just do that instead. |
|
Not sure if all the failures are related.... |
|
Should be good to go now. |
The *math* parameter is passed through many layers of the call stack but is ultimately only used for a single purpose: deciding whether to replace the ASCII hyphen by a (longer) unicode minus. Instead of doing that, just do the substitution at the parsing stage. In particular, this fixes problematic unicode minus support with the "cm" fontset. This patch also reverts a significant part of 52003e4, as LogFormatters no longer need to pass unicode minuses in mathtext -- everything gets converted by mathtext. Likewise, this change also invalidates the test_log_scales baseline image (old, buggy wrt. unicode minus); replace it by a test that the drawn ticks are as expected (which was the intent in 90c1aa3).
The math parameter is passed through many layers of the call stack
but is ultimately only used for a single purpose: deciding whether to
replace the ASCII hyphen by a (longer) unicode minus. Instead of doing
that, just do the substitution at the parsing stage. In particular,
this fixes problematic unicode minus support with the "cm" fontset.
This patch also reverts a significant part of 52003e4, as LogFormatters
no longer need to pass unicode minuses in mathtext -- everything gets
converted by mathtext. Likewise, this change also invalidates the
test_log_scales baseline image (old, buggy wrt. unicode minus); replace
it by a test that the drawn ticks are as expected (which was the intent
in 90c1aa3).
Closes #21687.
PR Summary
PR Checklist
Tests and Styling
pytestpasses).flake8-docstringsand runflake8 --docstring-convention=all).Documentation
doc/users/next_whats_new/(follow instructions in README.rst there).doc/api/next_api_changes/(follow instructions in README.rst there).