pango: update to 1.50.2

[cinerea0]

Testing the changes

• I tested the changes in this PR: YES

[ericonr]

I've found the musl test failure. Our glib is misconfigured due to a lack of CFLAGS (upstream's fault, I'm not sure how it went undetected before).

It should be using the USE_XLOCALE code path in gstrfuncs.c, but it isn't, so it's using the g_ascii_formatd implementation, which has a mistaken g_return_val_if_fail (format[0] == '%', NULL); for some reason. It fails, because the format being passed is " %f" (note the space).

I'm not sure why the test exists, but trying to fix glib first.

[ericonr]

My mistake, glib isn't misconfigured, musl doesn't have strtoll_l or strtoull_l.

So possibly patching out the assert is enough, will check.

[ericonr]

the code is 20y old, I'm curious how it was never caught before :P

[ericonr]

Am trying again.

[cinerea0]

Temporarily marking as draft so I can test 1.50.2 fully.

[cinerea0]

@ericonr The checks are failing again.

[ericonr]

Oof, these are different failures...

[ericonr]

So uh, tests work here :p

[ram02z]

This pango update broke text rendering on swaybar.

[ericonr]

@ram02z can you please open an issue and share screenshots?

I don't think this has been reported to sway upstream yet :/

[cinerea0]

@ram02z Also, could you check if the update in #34725 fixes your problem?

[ram02z]

@ram02z can you please open an issue and share screenshots?

I don't think this has been reported to sway upstream yet :/

Sorry, it was i3status-rs, not swaybar.
Seems to be this greshake/i3status-rust#1367

@ram02z Also, could you check if the update in #34725 fixes your problem?

Yeah, pango 1.50.3 fixes the bug.