https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #34 from Akira TAGOH ---
(In reply to Oleg Oshmyan from comment #33)
> > we can't guess a language from a character coverage completely,
> > particularly if it is all-in-one font.
>
> Yes; that's why glyph-coverage-based
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #33 from Oleg Oshmyan ---
> we can't guess a language from a character coverage completely, particularly
> if it is all-in-one font.
Yes; that's why glyph-coverage-based language lists aren't as useful as
human-set fonts.conf
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #32 from Akira TAGOH ---
(In reply to Oleg Oshmyan from comment #31)
> No, I mean fonts that are designed to support a single language. A font
> _can_ provide good support for everything at once via GSUB, but such global
> support
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #31 from Oleg Oshmyan ---
> they are assigned to the same codepoint but have different shape with GSUB.
> fontconfig doesn't get involved with the rendering thing at all. it is out of
> the scope.
No, I mean fonts that are
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #30 from Akira TAGOH ---
(In reply to Oleg Oshmyan from comment #28)
> Ah, you must mean the other kind of lang object, the one Fontconfig extracts
> from each font. I'm actually unsure how it does this; isn't it still built
>
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #29 from Oleg Oshmyan ---
Seeing the __libass_delimiter magic though, I suspect this isn't quite that
easy.
--
You are receiving this mail because:
You are on the CC list for the bug.
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #28 from Oleg Oshmyan ---
Ah, you must mean the other kind of lang object, the one Fontconfig extracts
from each font. I'm actually unsure how it does this; isn't it still built from
the fonts' built-in code page or Unicode
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #27 from Oleg Oshmyan ---
Why do you keep mentioning English? libass is looking for a font to display
numbers in. There's no English involved.
> How about simply taking a look at "lang" for a font?
It's empty. You've removed it.
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #26 from Akira TAGOH ---
(In reply to Oleg Oshmyan from comment #24)
> The list returned by Fontconfig is ordered by priority. libass uses the
> first (highest-priority) font that satisfies the request. The choice of
> priority
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #25 from Oleg Oshmyan ---
What libass could, in fact, do better here is it could better match the size of
the primary font (Noto Sans Arabic) used for the numerals and the fallback font
used for the Latin, so that the apparent
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #24 from Oleg Oshmyan ---
> > Fontconfig doesn't remove the font from the complete font list seen by
> > libass
>
> No, not exactly. What libass does is to obtain *all* the font list for
> sans-serif no matter what coverage they
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #23 from Akira TAGOH ---
Hmm, this issue seems not happening with google-noto-sans-arabic-fonts instead
of google-noto-sans-arabic-vf-fonts.
libass doesn't support variable fonts well then.
--
You are receiving this mail
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #22 from Akira TAGOH ---
(In reply to Łukasz Patron from comment #20)
> (In reply to Akira TAGOH from comment #19)
> > > Yeah, the thing I noticed was that after Noto Sans update, KDE file
> > > picker rows got higher —
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #21 from Akira TAGOH ---
Created attachment 2021107
--> https://bugzilla.redhat.com/attachment.cgi?id=2021107=edit
screenshot for freetype code
> FYI I believe Fontconfig allows multiple language tags to be specified per
>
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #20 from Łukasz Patron ---
(In reply to Akira TAGOH from comment #19)
> > Yeah, the thing I noticed was that after Noto Sans update, KDE file picker
> > rows got higher — https://imgur.com/a/6dBRYLg (before; after).
>
>
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #19 from Akira TAGOH ---
> Yeah, the thing I noticed was that after Noto Sans update, KDE file picker
> rows got higher — https://imgur.com/a/6dBRYLg (before; after).
Apparently it is different issue.
--
You are receiving
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #18 from Oleg Oshmyan ---
> but the problem is that Noto Sans Arabic isn't for ar only. this was why I
> removed it there.
FYI I believe Fontconfig allows multiple language tags to be specified per
font. I suspect this is
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #17 from Łukasz Patron ---
(In reply to Oleg Oshmyan from comment #16)
> libass uses Fontconfig's FcConfigSubstitute to convert mpv's "sans-serif"
> font name to a real font:
> https://github.com/libass/libass/blob/
>
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
Oleg Oshmyan changed:
What|Removed |Added
CC||chor...@inbox.lv
--- Comment #16 from
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #15 from Akira TAGOH ---
Well, not that easy to fix. It seems that they use FC_CHARSET to check the
coverage and select a font (see
https://github.com/libass/libass/blob/master/libass/ass_fontselect.c#L722 and
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #14 from Łukasz Patron ---
(In reply to Łukasz Patron from comment #13)
> (In reply to Akira TAGOH from comment #12)
> > Sorry, correctly libass. see
> > https://github.com/libass/libass/blob/
> >
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #13 from Łukasz Patron ---
(In reply to Akira TAGOH from comment #12)
> Sorry, correctly libass. see
> https://github.com/libass/libass/blob/
> 649a7c2e1fc6f4188ea1a89968560715800b883d/libass/ass_fontconfig.c#L209
>
> > Are they
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
Akira TAGOH changed:
What|Removed |Added
Component|mpv |libass
CC|
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #11 from Łukasz Patron ---
(In reply to Akira TAGOH from comment #10)
> Well, I'd say this is mpv issue actually. I see they intentionally ignore
> lang object in a fontconfig cache and process fonts in their own way.
> Picking up
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
Akira TAGOH changed:
What|Removed |Added
Assignee|ta...@redhat.com|ngomp...@gmail.com
CC|
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #9 from Łukasz Patron ---
(In reply to Łukasz Patron from comment #8)
> (In reply to Akira TAGOH from comment #6)
> > That is weird. Noto Sans Arabic doesn't have "en" covarage. I can't
> > reproduce this except running with
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #8 from Łukasz Patron ---
(In reply to Akira TAGOH from comment #6)
> That is weird. Noto Sans Arabic doesn't have "en" covarage. I can't
> reproduce this except running with --osd-font="Noto Sans Arabic". Do you
> have any your
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #7 from Akira TAGOH ---
> FWIW Noto Sans Arabic doesn't have alphabet glyphs. apparently libass is
> falling back to reuder alphabets. though Noto Sans and Noto Sans Arabic has
> same outline for numeric characters. If mpv use
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #6 from Akira TAGOH ---
That is weird. Noto Sans Arabic doesn't have "en" covarage. I can't reproduce
this except running with --osd-font="Noto Sans Arabic". Do you have any your
own config? how can I reproduce this?
I tried
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
Łukasz Patron changed:
What|Removed |Added
Flags|needinfo?(priv@gmail.co |
|m)
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
Akira TAGOH changed:
What|Removed |Added
Flags||needinfo?(priv@gmail.co
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
Red Hat One Jira (issues.redhat.com) changed:
What|Removed |Added
Link ID||Red Hat Issue
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
--- Comment #3 from Łukasz Patron ---
(In reply to Łukasz Patron from comment #2)
> (In reply to faustian from comment #1)
> > Comparing to last version 56-google-noto-sans-arabic-vf.conf lacks language
> > match test.
> >
> > diff
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
Łukasz Patron changed:
What|Removed |Added
Doc Type|--- |If docs needed, set a value
---
https://bugzilla.redhat.com/show_bug.cgi?id=2262410
faust...@go2.pl changed:
What|Removed |Added
CC||faust...@go2.pl
--- Comment #1 from
35 matches
Mail list logo
