Simplify derivation of ambiguous

Jules-Bertholet · Jules-Bertholet · commit be1aa170cb40 · 2024-06-17T18:57:36.000-04:00
Use `Letter` general category instead of script and block.
Changes `ℓ` to narrow, matching common fonts
diff --git a/scripts/unicode.py b/scripts/unicode.py
@@ -18,7 +18,6 @@
 # - NormalizationTest.txt (for tests only)
 # - PropList.txt
 # - ReadMe.txt
-# - Scripts.txt
 # - UnicodeData.txt
 # - auxiliary/GraphemeBreakProperty.txt
 # - emoji/emoji-data.txt
@@ -430,22 +429,10 @@ def load_east_asian_widths() -> list[EastAsianWidth]:
             # Catch any leftover codepoints and assign them implicit Neutral/narrow width.
             width_map.append(EastAsianWidth.NARROW)
 
-    # Characters from alphabetic scripts are narrow
-    load_property(
-        "Scripts.txt",
-        r"(?:Latin|Greek|Cyrillic)",
-        lambda cp: (
-            operator.setitem(width_map, cp, EastAsianWidth.NARROW)
-            if width_map[cp] == EastAsianWidth.AMBIGUOUS
-            and not (0x2160 <= cp <= 0x217F)  # Roman numerals remain ambiguous
-            else None
-        ),
-    )
-
-    # Ambiguous `Modifier_Letter`s and `Modifier_Symbol`s are narrow
+    # Ambiguous `Letter`s and `Modifier_Symbol`s are narrow
     load_property(
         "extracted/DerivedGeneralCategory.txt",
-        r"(:?Lm|Sk)",
+        r"(:?Lu|Ll|Lt|Lm|Lo|Sk)",
         lambda cp: (
             operator.setitem(width_map, cp, EastAsianWidth.NARROW)
             if width_map[cp] == EastAsianWidth.AMBIGUOUS
diff --git a/src/lib.rs b/src/lib.rs
@@ -122,8 +122,7 @@
 //!       - Has an [`East_Asian_Width`] of [`Ambiguous`], or
 //!         has a canonical decomposition to an [`Ambiguous`] character followed by [`'\u{0338}'` COMBINING LONG SOLIDUS OVERLAY], or
 //!         is [`'\u{0387}'` GREEK ANO TELEIA](https://util.unicode.org/UnicodeJsps/character.jsp?a=0387), and
-//!       - Does not have a [`General_Category`] of `Modifier_Letter` or `Modifier_Symbol`, and
-//!       - Does not have a [`Script`] of `Latin`, `Greek`, or `Cyrillic`, or is a Roman numeral in the range `'\u{2160}'..='\u{217F}'`.
+//!       - Does not have a [`General_Category`] of `Letter` or `Modifier_Symbol`.
 //!    7. All other characters have width 1.
 //!
 //! [`'\u{0338}'` COMBINING LONG SOLIDUS OVERLAY]: https://util.unicode.org/UnicodeJsps/character.jsp?a=0338
diff --git a/src/tables.rs b/src/tables.rs
@@ -1877,7 +1877,7 @@ static WIDTH_LEAVES: Align32<[[u8; 32]; WIDTH_LEAVES_LEN]> = Align32([
     ],
     #[cfg(feature = "cjk")]
     [
-        0x95, 0x59, 0x59, 0x55, 0x95, 0x65, 0x55, 0x55, 0x69, 0x55, 0x55, 0x55, 0x55, 0x55, 0x55,
+        0x95, 0x59, 0x59, 0x55, 0x55, 0x65, 0x55, 0x55, 0x69, 0x55, 0x55, 0x55, 0x55, 0x55, 0x55,
         0x55, 0x55, 0x55, 0x55, 0x55, 0x95, 0x56, 0x95, 0x6A, 0xAA, 0xAA, 0xAA, 0x55, 0xAA, 0xAA,
         0x5A, 0x55,
     ],

Original file line number	Diff line number	Diff line change
`@@ -1877,7 +1877,7 @@ static WIDTH_LEAVES: Align32<[[u8; 32]; WIDTH_LEAVES_LEN]> = Align32([`
`1877`	`1877`	`],`
`1878`	`1878`	`#[cfg(feature = "cjk")]`
`1879`	`1879`	`[`
`1880`		`- 0x95, 0x59, 0x59, 0x55, 0x95, 0x65, 0x55, 0x55, 0x69, 0x55, 0x55, 0x55, 0x55, 0x55, 0x55,`
	`1880`	`+ 0x95, 0x59, 0x59, 0x55, 0x55, 0x65, 0x55, 0x55, 0x69, 0x55, 0x55, 0x55, 0x55, 0x55, 0x55,`
`1881`	`1881`	`0x55, 0x55, 0x55, 0x55, 0x55, 0x95, 0x56, 0x95, 0x6A, 0xAA, 0xAA, 0xAA, 0x55, 0xAA, 0xAA,`
`1882`	`1882`	`0x5A, 0x55,`
`1883`	`1883`	`],`