Keep Latin families first and read only the tagged language for CJK variants

default/fontconfig/conf.avail/50-omarchy.conf

@@ -1,20 +1,10 @@
 <?xml version="1.0"?>
 <!DOCTYPE fontconfig SYSTEM "fonts.dtd">
 <fontconfig>
-  <!-- CJK: give language-tagged text the Noto CJK variant that matches its
+  <!-- CJK: fold the language tag onto one tag per Noto CJK variant -->
-       language. noto-fonts-cjk ships all five, Han glyph shapes differ between
-       them, and the generic assigns below would otherwise hand tagged CJK text
-       to families with no Han coverage, leaving the variant to a charset-scan
-       lottery. Chinese arrives under many tags (W3C recommends the zh-Hans and
-       zh-Hant script forms), so the tags are first folded onto one tag per
-       variant. Order is load-bearing: a bare zh and a bare zh-hant satisfy
-       the contains test of every one of their extensions, so both are pinned
-       by exact matches before the extension folds run. Cantonese under its
-       own primary tag folds to Hong Kong forms, except explicitly
-       Simplified Cantonese, which folds to SC. -->
 
   <match target="pattern">
-    <test name="lang" compare="eq">
+    <test name="lang" qual="first" compare="eq">
       <string>zh</string>
     </test>
     <edit name="lang" mode="assign">
@@ -23,7 +13,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="eq">
+    <test name="lang" qual="first" compare="eq">
       <string>zh-hant</string>
     </test>
     <edit name="lang" mode="assign">
@@ -32,7 +22,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-hant-hk</string>
     </test>
     <edit name="lang" mode="assign">
@@ -41,7 +31,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-hant-mo</string>
     </test>
     <edit name="lang" mode="assign">
@@ -50,7 +40,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-hant</string>
     </test>
     <edit name="lang" mode="assign">
@@ -59,7 +49,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-hans</string>
     </test>
     <edit name="lang" mode="assign">
@@ -68,7 +58,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-sg</string>
     </test>
     <edit name="lang" mode="assign">
@@ -77,7 +67,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-mo</string>
     </test>
     <edit name="lang" mode="assign">
@@ -86,7 +76,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="eq">
+    <test name="lang" qual="first" compare="eq">
       <string>yue</string>
     </test>
     <edit name="lang" mode="assign">
@@ -95,7 +85,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>yue-hans</string>
     </test>
     <edit name="lang" mode="assign">
@@ -104,7 +94,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>yue-cn</string>
     </test>
     <edit name="lang" mode="assign">
@@ -113,7 +103,7 @@
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>yue</string>
     </test>
     <edit name="lang" mode="assign">
@@ -121,189 +111,184 @@
     </edit>
   </match>
 
-
+  <!-- Then append that variant after the generic family it falls back from -->
-  <!-- Each variant is then prepended for its tag, ahead of the assigns below
-       because these test the generic names those assigns replace. Unlike the
-       Arabic rule further down, these carry a family test, so the prepend
-       lands just ahead of the matched generic and a concrete family the app
-       asked for stays in front of the variant. -->
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-cn</string>
     </test>
     <test name="family">
       <string>sans-serif</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Sans CJK SC</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-cn</string>
     </test>
     <test name="family">
       <string>serif</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Serif CJK SC</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-cn</string>
     </test>
     <test name="family">
       <string>monospace</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Sans Mono CJK SC</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-hk</string>
     </test>
     <test name="family">
       <string>sans-serif</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Sans CJK HK</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-hk</string>
     </test>
     <test name="family">
       <string>serif</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Serif CJK HK</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-hk</string>
     </test>
     <test name="family">
       <string>monospace</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Sans Mono CJK HK</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-tw</string>
     </test>
     <test name="family">
       <string>sans-serif</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Sans CJK TC</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-tw</string>
     </test>
     <test name="family">
       <string>serif</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Serif CJK TC</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>zh-tw</string>
     </test>
     <test name="family">
       <string>monospace</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Sans Mono CJK TC</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>ja</string>
     </test>
     <test name="family">
       <string>sans-serif</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Sans CJK JP</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>ja</string>
     </test>
     <test name="family">
       <string>serif</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Serif CJK JP</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>ja</string>
     </test>
     <test name="family">
       <string>monospace</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Sans Mono CJK JP</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>ko</string>
     </test>
     <test name="family">
       <string>sans-serif</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Sans CJK KR</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>ko</string>
     </test>
     <test name="family">
       <string>serif</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Serif CJK KR</string>
     </edit>
   </match>
 
   <match target="pattern">
-    <test name="lang" compare="contains">
+    <test name="lang" qual="first" compare="contains">
       <string>ko</string>
     </test>
     <test name="family">
       <string>monospace</string>
     </test>
-    <edit name="family" mode="prepend" binding="strong">
+    <edit name="family" mode="append" binding="strong">
       <string>Noto Sans Mono CJK KR</string>
     </edit>
   </match>

plans/omarchycn.md

@@ -16,7 +16,7 @@ Progress source of truth for the `omarchycn` branches of `omarchy`, `omarchy-iso
 - [x] T0 Branches and this list
 - [x] T1 China package sources (PRD 1): region state, `omarchy-region` reader, `default/region/china/pacman.sed`, applied by `omarchy-refresh-pacman` and `install/post-install/pacman.sh`; tests
 - [x] T2 Region choice in the installer (PRD 2): `omarchy_prompt_region` in `setup-form.sh`, ISO configurator screen, orchestrator passes `OMARCHY_REGION`, `install/config/region.sh` records it
-- [ ] T3 Input method (PRD 3): rime-ice-installer patch and PKGBUILD in omarchy-pkgs, IME packages in `omarchy-other.packages`, install-time package add, user step runs the installer from bundled assets, ISO build bundles assets and the package
+- [x] T3 Input method (PRD 3): rime-ice-installer patch and PKGBUILD in omarchy-pkgs, IME packages in `omarchy-other.packages`, install-time package add, user step runs the installer from bundled assets, ISO build bundles assets and the package
 - [ ] T4 CJK fontconfig fix (PRD 4): rebuild PR 9693 on the current base
 - [ ] T5 AI (PRD 5): Dim and Kimi Code for every region (rebuild PR 9694), Claude Code third-party provider setup shown only in China
 - [ ] T6 Developer mirrors (PRD 6): npm, PyPI/uv, Go, rustup, crates.io, RubyGems, Maven via system drop-ins under `default/region/china/`

test/shell.d/fontconfig-cjk-test.sh

@@ -0,0 +1,70 @@
+#!/bin/bash
+
+set -euo pipefail
+
+source "$(cd -- "$(dirname -- "${BASH_SOURCE[0]}")" && pwd)/base-test.sh"
+
+if ! command -v fc-match >/dev/null || [[ $(fc-list : file) != *NotoSansCJK-Regular.ttc* ]]; then
+  skip "fontconfig CJK variants need fc-match and noto-fonts-cjk"
+  exit 0
+fi
+
+test_tmp=$(mktemp -d)
+trap 'rm -rf "$test_tmp"' EXIT
+
+{
+  echo '<?xml version="1.0"?><!DOCTYPE fontconfig SYSTEM "fonts.dtd"><fontconfig>'
+  echo "<dir>/usr/share/fonts</dir><cachedir>$test_tmp/cache</cachedir>"
+  for conf in /etc/fonts/conf.d/*.conf; do
+    if [[ ${conf##*/} == "50-omarchy.conf" ]]; then
+      echo "<include>$ROOT/default/fontconfig/conf.avail/50-omarchy.conf</include>"
+    else
+      echo "<include>$conf</include>"
+    fi
+  done
+  echo '</fontconfig>'
+} >"$test_tmp/fonts.conf"
+export FONTCONFIG_FILE="$test_tmp/fonts.conf"
+fc-cache >/dev/null
+
+declare -A latin=([sans-serif]="Liberation Sans" [serif]="Liberation Serif" [monospace]="JetBrainsMono Nerd Font")
+declare -A cjk=([sans-serif]="Noto Sans CJK" [serif]="Noto Serif CJK" [monospace]="Noto Sans Mono CJK")
+declare -A variant=(
+  [zh]=SC [zh-cn]=SC [zh-sg]=SC [zh-hans]=SC [zh-hans-cn]=SC [zh-hans-hk]=SC
+  [zh-tw]=TC [zh-hant]=TC [zh-hant-tw]=TC
+  [zh-hk]=HK [zh-mo]=HK [zh-hant-hk]=HK [zh-hant-mo]=HK
+  [yue]=HK [yue-hk]=HK [yue-hant]=HK [yue-hant-hk]=HK [yue-hant-cn]=HK
+  [yue-hans]=SC [yue-hans-cn]=SC [yue-hans-hk]=SC [yue-cn]=SC
+  [ja]=JP [ja-jp]=JP [ko]=KR [ko-kr]=KR
+)
+
+family() { LANG=$1 fc-match -f '%{family[0]}' "$2"; }
+
+for locale in C.UTF-8 zh_CN.UTF-8; do
+  for tag in "${!variant[@]}"; do
+    for generic in sans-serif serif monospace; do
+      actual=$(family "$locale" "$generic:lang=$tag")
+      [[ $actual == "${latin[$generic]}" ]] || fail "$locale $generic:lang=$tag keeps Latin in ${latin[$generic]}" "$actual"
+      actual=$(family "$locale" "$generic:lang=$tag:charset=9aa8")
+      expected="${cjk[$generic]} ${variant[$tag]}"
+      [[ $actual == "$expected" ]] || fail "$locale $generic:lang=$tag sets Han in $expected" "$actual"
+    done
+  done
+  pass "$locale: tagged CJK keeps Omarchy's Latin families and sets Han in the tag's variant"
+done
+
+for generic in sans-serif serif monospace; do
+  actual=$(family zh_CN.UTF-8 "$generic")
+  [[ $actual == "${latin[$generic]}" ]] || fail "zh_CN untagged $generic keeps ${latin[$generic]}" "$actual"
+  actual=$(family zh_CN.UTF-8 "$generic:charset=9aa8")
+  [[ $actual == "${cjk[$generic]} SC" ]] || fail "zh_CN untagged $generic sets Han in SC" "$actual"
+done
+pass "a Chinese locale keeps Omarchy's Latin families and sets untagged Han in SC"
+
+[[ $(LANG=C.UTF-8 fc-pattern --config "sans-serif:lang=ja" family) == *'"Noto Sans CJK JP"(s)'* ]] ||
+  fail "lang=ja binds its variant strongly, so the prefix check below can see one"
+for tag in jam kok; do
+  actual=$(LANG=C.UTF-8 fc-pattern --config "sans-serif:lang=$tag" family)
+  [[ ! $actual =~ CJK\ [A-Z]+\"\(s\) ]] || fail "lang=$tag is not taken for Japanese or Korean" "$actual"
+done
+pass "languages that only share a prefix with ja or ko get no CJK variant"