Use built-in conversion function from unicode_script to convert script for harfbuzz (servo#38704)

Replaces a big match statement with a call to `unicode_script`'s
`Script::short_name` method which has the same big match statement. We
special case `Script::Unknown` because the `short_name` method returns
empty string for that variant, but harfbuzz represents it with `Zzzz`.

EDIT: `Zzzz` seems to be in the spec so I've sent a PR to
`unicode_script` to use it:
unicode-rs/unicode-script#23
EDIT: And unicode-rs/unicode-script#24 which
would allow us to remove this method entirely.

Signed-off-by: Nico Burns <[email protected]>

Author: nicoburns

components/fonts/shaper.rs

@@ -23,7 +23,7 @@ use harfbuzz_sys::{
     hb_font_destroy, hb_font_funcs_create, hb_font_funcs_set_glyph_h_advance_func,
     hb_font_funcs_set_nominal_glyph_func, hb_font_funcs_t, hb_font_set_funcs, hb_font_set_ppem,
     hb_font_set_scale, hb_font_t, hb_glyph_info_t, hb_glyph_position_t, hb_ot_layout_get_baseline,
-    hb_position_t, hb_shape, hb_tag_t,
+    hb_position_t, hb_script_from_iso15924_tag, hb_shape, hb_tag_t,
 };
 use log::debug;
 use num_traits::Zero;
@@ -201,169 +201,16 @@ impl Shaper {
     }
 }
 
-pub fn unicode_to_hb_script(script: unicode_script::Script) -> harfbuzz_sys::hb_script_t {
-    use harfbuzz_sys::*;
-    use unicode_script::Script::*;
-    match script {
-        Adlam => HB_SCRIPT_ADLAM,
-        Ahom => HB_SCRIPT_AHOM,
-        Anatolian_Hieroglyphs => HB_SCRIPT_ANATOLIAN_HIEROGLYPHS,
-        Arabic => HB_SCRIPT_ARABIC,
-        Armenian => HB_SCRIPT_ARMENIAN,
-        Avestan => HB_SCRIPT_AVESTAN,
-        Balinese => HB_SCRIPT_BALINESE,
-        Bamum => HB_SCRIPT_BAMUM,
-        Bassa_Vah => HB_SCRIPT_BASSA_VAH,
-        Batak => HB_SCRIPT_BATAK,
-        Bengali => HB_SCRIPT_BENGALI,
-        Bhaiksuki => HB_SCRIPT_BHAIKSUKI,
-        Bopomofo => HB_SCRIPT_BOPOMOFO,
-        Brahmi => HB_SCRIPT_BRAHMI,
-        Braille => HB_SCRIPT_BRAILLE,
-        Buginese => HB_SCRIPT_BUGINESE,
-        Buhid => HB_SCRIPT_BUHID,
-        Canadian_Aboriginal => HB_SCRIPT_CANADIAN_SYLLABICS,
-        Carian => HB_SCRIPT_CARIAN,
-        Caucasian_Albanian => HB_SCRIPT_CAUCASIAN_ALBANIAN,
-        Chakma => HB_SCRIPT_CHAKMA,
-        Cham => HB_SCRIPT_CHAM,
-        Cherokee => HB_SCRIPT_CHEROKEE,
-        Chorasmian => HB_SCRIPT_CHORASMIAN,
-        Common => HB_SCRIPT_COMMON,
-        Coptic => HB_SCRIPT_COPTIC,
-        Cuneiform => HB_SCRIPT_CUNEIFORM,
-        Cypriot => HB_SCRIPT_CYPRIOT,
-        Cyrillic => HB_SCRIPT_CYRILLIC,
-        Deseret => HB_SCRIPT_DESERET,
-        Devanagari => HB_SCRIPT_DEVANAGARI,
-        Dives_Akuru => HB_SCRIPT_DIVES_AKURU,
-        Dogra => HB_SCRIPT_DOGRA,
-        Duployan => HB_SCRIPT_DUPLOYAN,
-        Egyptian_Hieroglyphs => HB_SCRIPT_EGYPTIAN_HIEROGLYPHS,
-        Elbasan => HB_SCRIPT_ELBASAN,
-        Elymaic => HB_SCRIPT_ELYMAIC,
-        Ethiopic => HB_SCRIPT_ETHIOPIC,
-        Georgian => HB_SCRIPT_GEORGIAN,
-        Glagolitic => HB_SCRIPT_GLAGOLITIC,
-        Gothic => HB_SCRIPT_GOTHIC,
-        Grantha => HB_SCRIPT_GRANTHA,
-        Greek => HB_SCRIPT_GREEK,
-        Gujarati => HB_SCRIPT_GUJARATI,
-        Gunjala_Gondi => HB_SCRIPT_GUNJALA_GONDI,
-        Gurmukhi => HB_SCRIPT_GURMUKHI,
-        Han => HB_SCRIPT_HAN,
-        Hangul => HB_SCRIPT_HANGUL,
-        Hanifi_Rohingya => HB_SCRIPT_HANIFI_ROHINGYA,
-        Hanunoo => HB_SCRIPT_HANUNOO,
-        Hatran => HB_SCRIPT_HATRAN,
-        Hebrew => HB_SCRIPT_HEBREW,
-        Hiragana => HB_SCRIPT_HIRAGANA,
-        Imperial_Aramaic => HB_SCRIPT_IMPERIAL_ARAMAIC,
-        Inherited => HB_SCRIPT_INHERITED,
-        Inscriptional_Pahlavi => HB_SCRIPT_INSCRIPTIONAL_PAHLAVI,
-        Inscriptional_Parthian => HB_SCRIPT_INSCRIPTIONAL_PARTHIAN,
-        Javanese => HB_SCRIPT_JAVANESE,
-        Kaithi => HB_SCRIPT_KAITHI,
-        Kannada => HB_SCRIPT_KANNADA,
-        Katakana => HB_SCRIPT_KATAKANA,
-        Kayah_Li => HB_SCRIPT_KAYAH_LI,
-        Kharoshthi => HB_SCRIPT_KHAROSHTHI,
-        Khitan_Small_Script => HB_SCRIPT_KHITAN_SMALL_SCRIPT,
-        Khmer => HB_SCRIPT_KHMER,
-        Khojki => HB_SCRIPT_KHOJKI,
-        Khudawadi => HB_SCRIPT_KHUDAWADI,
-        Lao => HB_SCRIPT_LAO,
-        Latin => HB_SCRIPT_LATIN,
-        Lepcha => HB_SCRIPT_LEPCHA,
-        Limbu => HB_SCRIPT_LIMBU,
-        Linear_A => HB_SCRIPT_LINEAR_A,
-        Linear_B => HB_SCRIPT_LINEAR_B,
-        Lisu => HB_SCRIPT_LISU,
-        Lycian => HB_SCRIPT_LYCIAN,
-        Lydian => HB_SCRIPT_LYDIAN,
-        Mahajani => HB_SCRIPT_MAHAJANI,
-        Makasar => HB_SCRIPT_MAKASAR,
-        Malayalam => HB_SCRIPT_MALAYALAM,
-        Mandaic => HB_SCRIPT_MANDAIC,
-        Manichaean => HB_SCRIPT_MANICHAEAN,
-        Marchen => HB_SCRIPT_MARCHEN,
-        Masaram_Gondi => HB_SCRIPT_MASARAM_GONDI,
-        Medefaidrin => HB_SCRIPT_MEDEFAIDRIN,
-        Meetei_Mayek => HB_SCRIPT_MEETEI_MAYEK,
-        Mende_Kikakui => HB_SCRIPT_MENDE_KIKAKUI,
-        Meroitic_Cursive => HB_SCRIPT_MEROITIC_CURSIVE,
-        Meroitic_Hieroglyphs => HB_SCRIPT_MEROITIC_HIEROGLYPHS,
-        Miao => HB_SCRIPT_MIAO,
-        Modi => HB_SCRIPT_MODI,
-        Mongolian => HB_SCRIPT_MONGOLIAN,
-        Mro => HB_SCRIPT_MRO,
-        Multani => HB_SCRIPT_MULTANI,
-        Myanmar => HB_SCRIPT_MYANMAR,
-        Nabataean => HB_SCRIPT_NABATAEAN,
-        Nandinagari => HB_SCRIPT_NANDINAGARI,
-        New_Tai_Lue => HB_SCRIPT_NEW_TAI_LUE,
-        Newa => HB_SCRIPT_NEWA,
-        Nko => HB_SCRIPT_NKO,
-        Nushu => HB_SCRIPT_NUSHU,
-        Nyiakeng_Puachue_Hmong => HB_SCRIPT_NYIAKENG_PUACHUE_HMONG,
-        Ogham => HB_SCRIPT_OGHAM,
-        Ol_Chiki => HB_SCRIPT_OL_CHIKI,
-        Old_Hungarian => HB_SCRIPT_OLD_HUNGARIAN,
-        Old_Italic => HB_SCRIPT_OLD_ITALIC,
-        Old_North_Arabian => HB_SCRIPT_OLD_NORTH_ARABIAN,
-        Old_Permic => HB_SCRIPT_OLD_PERMIC,
-        Old_Persian => HB_SCRIPT_OLD_PERSIAN,
-        Old_Sogdian => HB_SCRIPT_OLD_SOGDIAN,
-        Old_South_Arabian => HB_SCRIPT_OLD_SOUTH_ARABIAN,
-        Old_Turkic => HB_SCRIPT_OLD_TURKIC,
-        Oriya => HB_SCRIPT_ORIYA,
-        Osage => HB_SCRIPT_OSAGE,
-        Osmanya => HB_SCRIPT_OSMANYA,
-        Pahawh_Hmong => HB_SCRIPT_PAHAWH_HMONG,
-        Palmyrene => HB_SCRIPT_PALMYRENE,
-        Pau_Cin_Hau => HB_SCRIPT_PAU_CIN_HAU,
-        Phags_Pa => HB_SCRIPT_PHAGS_PA,
-        Phoenician => HB_SCRIPT_PHOENICIAN,
-        Psalter_Pahlavi => HB_SCRIPT_PSALTER_PAHLAVI,
-        Rejang => HB_SCRIPT_REJANG,
-        Runic => HB_SCRIPT_RUNIC,
-        Samaritan => HB_SCRIPT_SAMARITAN,
-        Saurashtra => HB_SCRIPT_SAURASHTRA,
-        Sharada => HB_SCRIPT_SHARADA,
-        Shavian => HB_SCRIPT_SHAVIAN,
-        Siddham => HB_SCRIPT_SIDDHAM,
-        SignWriting => HB_SCRIPT_SIGNWRITING,
-        Sinhala => HB_SCRIPT_SINHALA,
-        Sogdian => HB_SCRIPT_SOGDIAN,
-        Sora_Sompeng => HB_SCRIPT_SORA_SOMPENG,
-        Soyombo => HB_SCRIPT_SOYOMBO,
-        Sundanese => HB_SCRIPT_SUNDANESE,
-        Syloti_Nagri => HB_SCRIPT_SYLOTI_NAGRI,
-        Syriac => HB_SCRIPT_SYRIAC,
-        Tagalog => HB_SCRIPT_TAGALOG,
-        Tagbanwa => HB_SCRIPT_TAGBANWA,
-        Tai_Le => HB_SCRIPT_TAI_LE,
-        Tai_Tham => HB_SCRIPT_TAI_THAM,
-        Tai_Viet => HB_SCRIPT_TAI_VIET,
-        Takri => HB_SCRIPT_TAKRI,
-        Tamil => HB_SCRIPT_TAMIL,
-        Tangut => HB_SCRIPT_TANGUT,
-        Telugu => HB_SCRIPT_TELUGU,
-        Thaana => HB_SCRIPT_THAANA,
-        Thai => HB_SCRIPT_THAI,
-        Tibetan => HB_SCRIPT_TIBETAN,
-        Tifinagh => HB_SCRIPT_TIFINAGH,
-        Tirhuta => HB_SCRIPT_TIRHUTA,
-        Ugaritic => HB_SCRIPT_UGARITIC,
-        Unknown => HB_SCRIPT_UNKNOWN,
-        Vai => HB_SCRIPT_VAI,
-        Warang_Citi => HB_SCRIPT_WARANG_CITI,
-        Wancho => HB_SCRIPT_WANCHO,
-        Yezidi => HB_SCRIPT_YEZIDI,
-        Yi => HB_SCRIPT_YI,
-        Zanabazar_Square => HB_SCRIPT_ZANABAZAR_SQUARE,
-        _ => HB_SCRIPT_UNKNOWN,
-    }
+pub fn unicode_script_to_iso15924_tag(script: unicode_script::Script) -> u32 {
+    let bytes: [u8; 4] = match script {
+        unicode_script::Script::Unknown => *b"Zzzz",
+        _ => {
+            let short_name = script.short_name();
+            short_name.as_bytes().try_into().unwrap()
+        },
+    };
+
+    u32::from_be_bytes(bytes)
 }
 
 impl Shaper {
@@ -381,7 +228,9 @@ impl Shaper {
                 },
             );
 
-            hb_buffer_set_script(hb_buffer, unicode_to_hb_script(options.script));
+            let script =
+                hb_script_from_iso15924_tag(unicode_script_to_iso15924_tag(options.script));
+            hb_buffer_set_script(hb_buffer, script);
 
             hb_buffer_add_utf8(
                 hb_buffer,