Assign width 3 to KHMER SIGN BEYYAL

See https://unicode.org/charts/nameslist/n_1780.html

Author: Jules-Bertholet

scripts/unicode.py

@@ -184,6 +184,9 @@ class WidthState(enum.IntEnum):
     WIDE = 0x1_0002
     "Two columns wide."
 
+    THREE = 0x1_0003
+    "Three columns wide."
+
     # \r\n
     LINE_FEED = 0b0000_0000_0000_0001
     "\\n (CRLF has width 1)"
@@ -341,6 +344,10 @@ def table_width(self) -> CharWidthInTable:
             case _:
                 return CharWidthInTable.SPECIAL
 
+    def is_carried(self) -> bool:
+        "Whether this corresponds to a non-default `WidthInfo`."
+        return int(self) <= 0xFFFF
+
     def width_alone(self) -> int:
         "The width of a character with this type when it appears alone."
         match self:
@@ -357,6 +364,8 @@ def width_alone(self) -> int:
                 | WidthState.EMOJI_PRESENTATION
             ):
                 return 2
+            case WidthState.THREE:
+                return 3
             case _:
                 return 1
 
@@ -598,6 +607,7 @@ def load_width_maps() -> tuple[list[WidthState], list[WidthState]]:
         (alef_joining, WidthState.JOINING_GROUP_ALEF),
         (range(0x1780, 0x17A3), WidthState.KHMER_COENG_ELIGIBLE_LETTER),
         ([0x17A7, 0x17AB, 0x17AC, 0x17AF], WidthState.KHMER_COENG_ELIGIBLE_LETTER),
+        ([0x17D8], WidthState.THREE),
         ([0x1A10], WidthState.BUGINESE_LETTER_YA),
         (range(0x2D31, 0x2D66), WidthState.TIFINAGH_CONSONANT),
         ([0x2D6F], WidthState.TIFINAGH_CONSONANT),
@@ -1196,7 +1206,11 @@ def lookup_fns(
         s += f"            '\\u{{{lo:X}}}'"
         if hi != lo:
             s += f"..='\\u{{{hi:X}}}'"
-        s += f" => ({width.width_alone()}, WidthInfo::{str(width.name)}),\n"
+        if width.is_carried():
+            width_info = width.name
+        else:
+            width_info = "DEFAULT"
+        s += f" => ({width.width_alone()}, WidthInfo::{width_info}),\n"
 
     s += f"""            _ => (2, WidthInfo::EMOJI_PRESENTATION),
         }}
@@ -1531,7 +1545,7 @@ def emit_module(
         )
 
         for variant in WidthState:
-            if variant.table_width() == CharWidthInTable.SPECIAL:
+            if variant.is_carried():
                 if variant.is_cjk_only():
                     module.write('    #[cfg(feature = "cjk")]\n')
                 module.write(
@@ -1925,7 +1939,7 @@ def emit_module(
         test_width_variants = []
         test_width_variants_cjk = []
         for variant in WidthState:
-            if variant.table_width() == CharWidthInTable.SPECIAL:
+            if variant.is_carried():
                 if not variant.is_cjk_only():
                     test_width_variants.append(variant)
                 if not variant.is_non_cjk_only():
@@ -2003,10 +2017,7 @@ def emit_module(
         )
 
         for variant in WidthState:
-            if (
-                variant.table_width() == CharWidthInTable.SPECIAL
-                and not variant.is_cjk_only()
-            ):
+            if variant.is_carried() and not variant.is_cjk_only():
                 module.write(f"        WidthInfo::{variant.name},\n")
 
         module.write(
@@ -2018,10 +2029,7 @@ def emit_module(
         )
 
         for variant in WidthState:
-            if (
-                variant.table_width() == CharWidthInTable.SPECIAL
-                and not variant.is_non_cjk_only()
-            ):
+            if variant.is_carried() and not variant.is_non_cjk_only():
                 module.write(f"        WidthInfo::{variant.name},\n")
 
         module.write(

src/lib.rs

@@ -88,7 +88,8 @@
 //! 2. In all other cases, the width of the string equals the sum of its character widths:
 //!    1. [`'\u{2D7F}'` TIFINAGH CONSONANT JOINER] has width 1 (outside of the ligatures described previously).
 //!    2. [`'\u{115F}'` HANGUL CHOSEONG FILLER](https://util.unicode.org/UnicodeJsps/character.jsp?a=115F) has width 2.
-//!    3. The following have width 0:
+//!    3. [`'\u{17D8}'` KHMER SIGN BEYYAL](https://util.unicode.org/UnicodeJsps/character.jsp?a=17D8) has width 3.
+//!    4. The following have width 0:
 //!       - [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BDefault_Ignorable_Code_Point%7D)
 //!         with the [`Default_Ignorable_Code_Point`] property.
 //!       - [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BGrapheme_Extend%7D)
@@ -111,15 +112,15 @@
 //!         - [`'\u{0891}'` PIASTRE MARK ABOVE](https://util.unicode.org/UnicodeJsps/character.jsp?a=0891), and
 //!         - [`'\u{08E2}'` DISPUTED END OF AYAH](https://util.unicode.org/UnicodeJsps/character.jsp?a=08E2).
 //!       - [`'\u{A8FA}'` DEVANAGARI CARET](https://util.unicode.org/UnicodeJsps/character.jsp?a=A8FA).
-//!    4. [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BEast_Asian_Width%3DF%7D%5Cp%7BEast_Asian_Width%3DW%7D)
+//!    5. [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BEast_Asian_Width%3DF%7D%5Cp%7BEast_Asian_Width%3DW%7D)
 //!       with an [`East_Asian_Width`] of [`Fullwidth`] or [`Wide`] have width 2.
-//!    5. Characters fulfilling all of the following conditions have width 2 in an East Asian context, and width 1 otherwise:
+//!    6. Characters fulfilling all of the following conditions have width 2 in an East Asian context, and width 1 otherwise:
 //!       - 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_Symbol`, and
 //!       - Does not have a [`Script`] of `Latin`, `Greek`, or `Cyrillic`, or is a Roman numeral in the range `'\u{2160}'..='\u{217F}'`.
-//!    6. All other characters have width 1.
+//!    7. All other characters have width 1.
 //!
 //! [`'\u{0338}'` COMBINING LONG SOLIDUS OVERLAY]: https://util.unicode.org/UnicodeJsps/character.jsp?a=0338
 //! [`'\u{2D7F}'` TIFINAGH CONSONANT JOINER]: https://util.unicode.org/UnicodeJsps/character.jsp?a=2D7F

src/tables.rs

@@ -161,6 +161,7 @@ fn lookup_width(c: char) -> (u8, WidthInfo) {
             '\u{5DC}' => (1, WidthInfo::HEBREW_LETTER_LAMED),
             '\u{622}'..='\u{882}' => (1, WidthInfo::JOINING_GROUP_ALEF),
             '\u{1780}'..='\u{17AF}' => (1, WidthInfo::KHMER_COENG_ELIGIBLE_LETTER),
+            '\u{17D8}' => (3, WidthInfo::DEFAULT),
             '\u{1A10}' => (1, WidthInfo::BUGINESE_LETTER_YA),
             '\u{2D31}'..='\u{2D6F}' => (1, WidthInfo::TIFINAGH_CONSONANT),
             '\u{A4FC}'..='\u{A4FD}' => (1, WidthInfo::LISU_TONE_LETTER_MYA_NA_JEU),
@@ -444,6 +445,7 @@ fn lookup_width_cjk(c: char) -> (u8, WidthInfo) {
             '\u{5DC}' => (1, WidthInfo::HEBREW_LETTER_LAMED),
             '\u{622}'..='\u{882}' => (1, WidthInfo::JOINING_GROUP_ALEF),
             '\u{1780}'..='\u{17AF}' => (1, WidthInfo::KHMER_COENG_ELIGIBLE_LETTER),
+            '\u{17D8}' => (3, WidthInfo::DEFAULT),
             '\u{1A10}' => (1, WidthInfo::BUGINESE_LETTER_YA),
             '\u{2D31}'..='\u{2D6F}' => (1, WidthInfo::TIFINAGH_CONSONANT),
             '\u{A4FC}'..='\u{A4FD}' => (1, WidthInfo::LISU_TONE_LETTER_MYA_NA_JEU),
@@ -1220,7 +1222,7 @@ static WIDTH_LEAVES: Align32<[[u8; 32]; WIDTH_LEAVES_LEN]> = Align32([
     ],
     [
         0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F, 0xD5, 0xD5, 0xD7, 0x55, 0x10, 0x00,
-        0x50, 0x55, 0x45, 0x01, 0x00, 0x00, 0x55, 0x55, 0x51, 0x55, 0x55, 0x55, 0x55, 0x55, 0x55,
+        0x50, 0x55, 0x45, 0x01, 0x00, 0x00, 0x55, 0x57, 0x51, 0x55, 0x55, 0x55, 0x55, 0x55, 0x55,
         0x55, 0x55,
     ],
     [

tests/tests.rs

@@ -409,6 +409,12 @@ fn test_khmer_coeng() {
     }
 }
 
+#[test]
+fn test_khmer_sign_beyyal() {
+    assert_width!("៘", 3, 3);
+    assert_width!("។ល។", 3, 3);
+}
+
 #[test]
 fn test_emoji_modifier() {
     assert_width!("\u{1F46A}", 2, 2);