Apply @kidayasuo's proposal

The proposal at:
w3c/csswg-drafts#9503 (comment)

Author: kojiishi

auto-spacing.txt

@@ -586,26 +586,26 @@
 2DD8..2DDE     ; N  # N   ETHIOPIC SYLLABLE GYA..ETHIOPIC SYLLABLE GYO
 2DE0..2DFF     ; N  # N   COMBINING CYRILLIC LETTER BE..COMBINING CYRILLIC LETTER IOTIFIED BIG YUS
 2E2F           ; N  # N   VERTICAL TILDE
-2E80..2E99     ; W  # W   CJK RADICAL REPEAT..CJK RADICAL RAP
-2E9B..2EF3     ; W  # W   CJK RADICAL CHOKE..CJK RADICAL C-SIMPLIFIED TURTLE
-2F00..2FD5     ; W  # W   KANGXI RADICAL ONE..KANGXI RADICAL FLUTE
+2E80..2E99     ; O  # W   CJK RADICAL REPEAT..CJK RADICAL RAP
+2E9B..2EF3     ; O  # W   CJK RADICAL CHOKE..CJK RADICAL C-SIMPLIFIED TURTLE
+2F00..2FD5     ; O  # W   KANGXI RADICAL ONE..KANGXI RADICAL FLUTE
 2FF0..2FFB     ; O  # W   IDEOGRAPHIC DESCRIPTION CHARACTER LEFT TO RIGHT..IDEOGRAPHIC DESCRIPTION CHARACTER OVERLAID
 3000           ; O  # F   IDEOGRAPHIC SPACE
 3001..3004     ; O  # W   IDEOGRAPHIC COMMA..JAPANESE INDUSTRIAL STANDARD SYMBOL
-3005           ; W  # W   IDEOGRAPHIC ITERATION MARK
-3006           ; N  # W   IDEOGRAPHIC CLOSING MARK
-3007           ; W  # W   IDEOGRAPHIC NUMBER ZERO
-3008..3020     ; O  # W   LEFT ANGLE BRACKET..POSTAL MARK FACE
-3021..3029     ; W  # W   HANGZHOU NUMERAL ONE..HANGZHOU NUMERAL NINE
+3005           ; N  # W   IDEOGRAPHIC ITERATION MARK
+3006           ; W  # W   IDEOGRAPHIC CLOSING MARK
+3007..3029     ; O  # W   IDEOGRAPHIC NUMBER ZERO..HANGZHOU NUMERAL NINE
 302A..302F     ; N  # W   IDEOGRAPHIC LEVEL TONE MARK..HANGUL DOUBLE DOT TONE MARK
 3030           ; O  # W   WAVY DASH
-3031..3035     ; N  # W   VERTICAL KANA REPEAT MARK..VERTICAL KANA REPEAT MARK LOWER HALF
-3036..3037     ; O  # W   CIRCLED POSTAL MARK..IDEOGRAPHIC TELEGRAPH LINE FEED SEPARATOR SYMBOL
-3038..303B     ; W  # W   HANGZHOU NUMERAL TEN..VERTICAL IDEOGRAPHIC ITERATION MARK
-303C           ; N  # W   MASU MARK
+3031..3035     ; W  # W   VERTICAL KANA REPEAT MARK..VERTICAL KANA REPEAT MARK LOWER HALF
+3036..303A     ; O  # W   CIRCLED POSTAL MARK..HANGZHOU NUMERAL THIRTY
+303B           ; N  # W   VERTICAL IDEOGRAPHIC ITERATION MARK
+303C           ; W  # W   MASU MARK
 303D..303E     ; O  # W   PART ALTERNATION MARK..IDEOGRAPHIC VARIATION INDICATOR
 3041..3096     ; W  # W   HIRAGANA LETTER SMALL A..HIRAGANA LETTER SMALL KE
-3099..309F     ; W  # W   COMBINING KATAKANA-HIRAGANA VOICED SOUND MARK..HIRAGANA DIGRAPH YORI
+3099..309A     ; N  # W   COMBINING KATAKANA-HIRAGANA VOICED SOUND MARK..COMBINING KATAKANA-HIRAGANA SEMI-VOICED SOUND MARK
+309B..309C     ; O  # W   KATAKANA-HIRAGANA VOICED SOUND MARK..KATAKANA-HIRAGANA SEMI-VOICED SOUND MARK
+309D..309F     ; W  # W   HIRAGANA ITERATION MARK..HIRAGANA DIGRAPH YORI
 30A0           ; O  # W   KATAKANA-HIRAGANA DOUBLE HYPHEN
 30A1..30FA     ; W  # W   KATAKANA LETTER SMALL A..KATAKANA LETTER VO
 30FB           ; O  # W   KATAKANA MIDDLE DOT
@@ -614,7 +614,7 @@
 3131..318E     ; N  # W   HANGUL LETTER KIYEOK..HANGUL LETTER ARAEAE
 3190..319F     ; O  # W   IDEOGRAPHIC ANNOTATION LINKING MARK..IDEOGRAPHIC ANNOTATION MAN MARK
 31A0..31BF     ; N  # W   BOPOMOFO LETTER BU..BOPOMOFO LETTER AH
-31C0..31E3     ; W  # W   CJK STROKE T..CJK STROKE Q
+31C0..31E3     ; O  # W   CJK STROKE T..CJK STROKE Q
 31F0..31FF     ; W  # W   KATAKANA LETTER SMALL KU..KATAKANA LETTER SMALL RO
 3200..321E     ; O  # W   PARENTHESIZED HANGUL KIYEOK..PARENTHESIZED KOREAN CHARACTER O HU
 3220..3247     ; O  # W   PARENTHESIZED IDEOGRAPH ONE..CIRCLED IDEOGRAPH KOTO
@@ -890,19 +890,20 @@ FFFD           ; O  # A   REPLACEMENT CHARACTER
 16F8F..16F9F   ; N  # N   MIAO TONE RIGHT..MIAO LETTER REFORMED TONE-8
 16FE0..16FE1   ; N  # W   TANGUT ITERATION MARK..NUSHU ITERATION MARK
 16FE2          ; O  # W   OLD CHINESE HOOK MARK
-16FE3..16FE4   ; N  # W   OLD CHINESE ITERATION MARK..KHITAN SMALL SCRIPT FILLER
-16FF0..16FF1   ; W  # W   VIETNAMESE ALTERNATE READING MARK CA..VIETNAMESE ALTERNATE READING MARK NHAY
-17000..187F7   ; N  # W   ..
-18800..18CD5   ; N  # W   TANGUT COMPONENT-001..KHITAN SMALL SCRIPT CHARACTER-18CD5
-18D00..18D08   ; N  # W   ..
+16FE3          ; W  # W   OLD CHINESE ITERATION MARK
+16FE4          ; N  # W   KHITAN SMALL SCRIPT FILLER
+16FF0..16FF1   ; N  # W   VIETNAMESE ALTERNATE READING MARK CA..VIETNAMESE ALTERNATE READING MARK NHAY
+17000..187F7   ; W  # W   ..
+18800..18CD5   ; W  # W   TANGUT COMPONENT-001..KHITAN SMALL SCRIPT CHARACTER-18CD5
+18D00..18D08   ; W  # W   ..
 1AFF0..1AFF3   ; O  # W   KATAKANA LETTER MINNAN TONE-2..KATAKANA LETTER MINNAN TONE-5
 1AFF5..1AFFB   ; O  # W   KATAKANA LETTER MINNAN TONE-7..KATAKANA LETTER MINNAN NASALIZED TONE-5
 1AFFD..1AFFE   ; O  # W   KATAKANA LETTER MINNAN NASALIZED TONE-7..KATAKANA LETTER MINNAN NASALIZED TONE-8
-1B000..1B11E   ; N  # W   KATAKANA LETTER ARCHAIC E..HENTAIGANA LETTER N-MU-MO-2
+1B000..1B11E   ; W  # W   KATAKANA LETTER ARCHAIC E..HENTAIGANA LETTER N-MU-MO-2
 1B11F..1B122   ; O  # W   HIRAGANA LETTER ARCHAIC WU..KATAKANA LETTER ARCHAIC WU
-1B150..1B152   ; N  # W   HIRAGANA LETTER SMALL WI..HIRAGANA LETTER SMALL WO
-1B164..1B167   ; N  # W   KATAKANA LETTER SMALL WI..KATAKANA LETTER SMALL N
-1B170..1B2FB   ; N  # W   NUSHU CHARACTER-1B170..NUSHU CHARACTER-1B2FB
+1B150..1B152   ; W  # W   HIRAGANA LETTER SMALL WI..HIRAGANA LETTER SMALL WO
+1B164..1B167   ; W  # W   KATAKANA LETTER SMALL WI..KATAKANA LETTER SMALL N
+1B170..1B2FB   ; W  # W   NUSHU CHARACTER-1B170..NUSHU CHARACTER-1B2FB
 1BC00..1BC6A   ; N  # N   DUPLOYAN LETTER H..DUPLOYAN LETTER VOCALIC M
 1BC70..1BC7C   ; N  # N   DUPLOYAN AFFIX LEFT HORIZONTAL SECANT..DUPLOYAN AFFIX ATTACHED TANGENT HOOK
 1BC80..1BC88   ; N  # N   DUPLOYAN AFFIX HIGH ACUTE..DUPLOYAN AFFIX HIGH VERTICAL

src/auto-spacing.py

@@ -8,13 +8,12 @@
 class AutoSpacing(object):
 
     def __init__(self) -> None:
-        # https://drafts.csswg.org/css-text-4/#text-spacing-classes
-        ideographs = UnicodeSet(r'[[:sc=Han:]]')
-        ideographs.addAll(UnicodeSet(r'[[\u3041-\u30FF]-[:P:]]'))
-        ideographs.addAll(UnicodeSet(r'[[\u31C0-\u31FF]]'))
+        # https://github.com/w3c/csswg-drafts/pull/9503#discussion_r1374477268
+        ideographs = UnicodeSet(
+            r'[[[:sc=Hiragana:][:sc=Katakana:][:sc=Common:][:ideographic:]&[:gc=L:]&[:ea=W:]]' +
+            r'[[:gc=L:][:gc=Nl:]&[:gc=So:]&[:sc=Hani:]]]')
 
-        letters_numerals = UnicodeSet(r'[[:L:][:M:][:Nd:]]')
-        letters_numerals.removeAll(UnicodeSet(r'[[:ea=F:]]'))
+        letters_numerals = UnicodeSet(r'[[:L:][:M:][:Nd:]-[:ea=F:]]')
 
         self.ideographs = ideographs
         self.letters_numerals = letters_numerals