[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#907999: fonts-noto-cjk: Unable to use system fonts in Firefox after #907048



Package: fonts-noto-cjk
Version: 1:20170601+repack1-3
Severity: normal

Dear Maintainer,

I think the changes made to the config is to aggressive. After this
change Firefox cannot use other system fonts other than Noto CJK TC
when my locale is set to zh_TW.

What Firefox currently does is, it constructs a pattern like
'{FamilyToFind}, -moz-sentinel' where -moz-sentinel should never be
substituted. Then it substitute the pattern recursively until it finds
all font family or the sentinel.

With the config change, even -moz-sentinel will be replaced by Noto
CJK TC so effectively all system fonts in CSS will only match Noto CJK
TC. To reproduce, load this snippet in Firefox and inspect the font
used (replace DejaVu Sans with any system font you installed):

  data:text/html,<div style="font-family:DejaVu Sans">Hello</div>

One can argue that this is a bug in Firefox but my point is using
fontconifg matching list editing to control global fonts ordering is
error prone. What if some fonts installed config sorted later also
prepend? I think we should have a better mechanism to sort system
fonts globally and allow our user to customize. As currently is very
hard for users to customize the font order with user config without
guessing or try-n-error the rule installed by font packages.

Kanru

-- System Information:
Debian Release: buster/sid
  APT prefers unstable-debug
  APT policy: (500, 'unstable-debug'), (500, 'unstable'), (1, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 4.17.0-3-amd64 (SMP w/8 CPU cores)
Locale: LANG=zh_TW.UTF-8, LC_CTYPE=zh_TW.UTF-8 (charmap=UTF-8), LANGUAGE=zh_TW.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

fonts-noto-cjk depends on no packages.

fonts-noto-cjk recommends no packages.

Versions of packages fonts-noto-cjk suggests:
ii  fonts-noto-cjk-extra  1:20170601+repack1-3

-- no debconf information


Reply to: