summaryrefslogtreecommitdiffstats
path: root/vendor/regex-syntax/src/unicode_tables/word_break.rs
diff options
context:
space:
mode:
Diffstat (limited to 'vendor/regex-syntax/src/unicode_tables/word_break.rs')
-rw-r--r--vendor/regex-syntax/src/unicode_tables/word_break.rs1060
1 files changed, 1060 insertions, 0 deletions
diff --git a/vendor/regex-syntax/src/unicode_tables/word_break.rs b/vendor/regex-syntax/src/unicode_tables/word_break.rs
new file mode 100644
index 000000000..bd23e00a8
--- /dev/null
+++ b/vendor/regex-syntax/src/unicode_tables/word_break.rs
@@ -0,0 +1,1060 @@
+// DO NOT EDIT THIS FILE. IT WAS AUTOMATICALLY GENERATED BY:
+//
+// ucd-generate word-break ucd-13.0.0 --chars
+//
+// Unicode version: 13.0.0.
+//
+// ucd-generate 0.2.8 is available on crates.io.
+
+pub const BY_NAME: &'static [(&'static str, &'static [(char, char)])] = &[
+ ("ALetter", ALETTER),
+ ("CR", CR),
+ ("Double_Quote", DOUBLE_QUOTE),
+ ("Extend", EXTEND),
+ ("ExtendNumLet", EXTENDNUMLET),
+ ("Format", FORMAT),
+ ("Hebrew_Letter", HEBREW_LETTER),
+ ("Katakana", KATAKANA),
+ ("LF", LF),
+ ("MidLetter", MIDLETTER),
+ ("MidNum", MIDNUM),
+ ("MidNumLet", MIDNUMLET),
+ ("Newline", NEWLINE),
+ ("Numeric", NUMERIC),
+ ("Regional_Indicator", REGIONAL_INDICATOR),
+ ("Single_Quote", SINGLE_QUOTE),
+ ("WSegSpace", WSEGSPACE),
+ ("ZWJ", ZWJ),
+];
+
+pub const ALETTER: &'static [(char, char)] = &[
+ ('A', 'Z'),
+ ('a', 'z'),
+ ('ª', 'ª'),
+ ('µ', 'µ'),
+ ('º', 'º'),
+ ('À', 'Ö'),
+ ('Ø', 'ö'),
+ ('ø', '˗'),
+ ('˞', '˿'),
+ ('Ͱ', 'ʹ'),
+ ('Ͷ', 'ͷ'),
+ ('ͺ', 'ͽ'),
+ ('Ϳ', 'Ϳ'),
+ ('Ά', 'Ά'),
+ ('Έ', 'Ί'),
+ ('Ό', 'Ό'),
+ ('Ύ', 'Ρ'),
+ ('Σ', 'ϵ'),
+ ('Ϸ', 'ҁ'),
+ ('Ҋ', 'ԯ'),
+ ('Ա', 'Ֆ'),
+ ('ՙ', '՜'),
+ ('՞', '՞'),
+ ('ՠ', 'ֈ'),
+ ('֊', '֊'),
+ ('׳', '׳'),
+ ('ؠ', 'ي'),
+ ('ٮ', 'ٯ'),
+ ('ٱ', 'ۓ'),
+ ('ە', 'ە'),
+ ('ۥ', 'ۦ'),
+ ('ۮ', 'ۯ'),
+ ('ۺ', 'ۼ'),
+ ('ۿ', 'ۿ'),
+ ('ܐ', 'ܐ'),
+ ('ܒ', 'ܯ'),
+ ('ݍ', 'ޥ'),
+ ('ޱ', 'ޱ'),
+ ('ߊ', 'ߪ'),
+ ('ߴ', 'ߵ'),
+ ('ߺ', 'ߺ'),
+ ('ࠀ', 'ࠕ'),
+ ('ࠚ', 'ࠚ'),
+ ('ࠤ', 'ࠤ'),
+ ('ࠨ', 'ࠨ'),
+ ('ࡀ', 'ࡘ'),
+ ('ࡠ', 'ࡪ'),
+ ('ࢠ', 'ࢴ'),
+ ('ࢶ', '\u{8c7}'),
+ ('ऄ', 'ह'),
+ ('ऽ', 'ऽ'),
+ ('ॐ', 'ॐ'),
+ ('क़', 'ॡ'),
+ ('ॱ', 'ঀ'),
+ ('অ', 'ঌ'),
+ ('এ', 'ঐ'),
+ ('ও', 'ন'),
+ ('প', 'র'),
+ ('ল', 'ল'),
+ ('শ', 'হ'),
+ ('ঽ', 'ঽ'),
+ ('ৎ', 'ৎ'),
+ ('ড়', 'ঢ়'),
+ ('য়', 'ৡ'),
+ ('ৰ', 'ৱ'),
+ ('ৼ', 'ৼ'),
+ ('ਅ', 'ਊ'),
+ ('ਏ', 'ਐ'),
+ ('ਓ', 'ਨ'),
+ ('ਪ', 'ਰ'),
+ ('ਲ', 'ਲ਼'),
+ ('ਵ', 'ਸ਼'),
+ ('ਸ', 'ਹ'),
+ ('ਖ਼', 'ੜ'),
+ ('ਫ਼', 'ਫ਼'),
+ ('ੲ', 'ੴ'),
+ ('અ', 'ઍ'),
+ ('એ', 'ઑ'),
+ ('ઓ', 'ન'),
+ ('પ', 'ર'),
+ ('લ', 'ળ'),
+ ('વ', 'હ'),
+ ('ઽ', 'ઽ'),
+ ('ૐ', 'ૐ'),
+ ('ૠ', 'ૡ'),
+ ('ૹ', 'ૹ'),
+ ('ଅ', 'ଌ'),
+ ('ଏ', 'ଐ'),
+ ('ଓ', 'ନ'),
+ ('ପ', 'ର'),
+ ('ଲ', 'ଳ'),
+ ('ଵ', 'ହ'),
+ ('ଽ', 'ଽ'),
+ ('ଡ଼', 'ଢ଼'),
+ ('ୟ', 'ୡ'),
+ ('ୱ', 'ୱ'),
+ ('ஃ', 'ஃ'),
+ ('அ', 'ஊ'),
+ ('எ', 'ஐ'),
+ ('ஒ', 'க'),
+ ('ங', 'ச'),
+ ('ஜ', 'ஜ'),
+ ('ஞ', 'ட'),
+ ('ண', 'த'),
+ ('ந', 'ப'),
+ ('ம', 'ஹ'),
+ ('ௐ', 'ௐ'),
+ ('అ', 'ఌ'),
+ ('ఎ', 'ఐ'),
+ ('ఒ', 'న'),
+ ('ప', 'హ'),
+ ('ఽ', 'ఽ'),
+ ('ౘ', 'ౚ'),
+ ('ౠ', 'ౡ'),
+ ('ಀ', 'ಀ'),
+ ('ಅ', 'ಌ'),
+ ('ಎ', 'ಐ'),
+ ('ಒ', 'ನ'),
+ ('ಪ', 'ಳ'),
+ ('ವ', 'ಹ'),
+ ('ಽ', 'ಽ'),
+ ('ೞ', 'ೞ'),
+ ('ೠ', 'ೡ'),
+ ('ೱ', 'ೲ'),
+ ('\u{d04}', 'ഌ'),
+ ('എ', 'ഐ'),
+ ('ഒ', 'ഺ'),
+ ('ഽ', 'ഽ'),
+ ('ൎ', 'ൎ'),
+ ('ൔ', 'ൖ'),
+ ('ൟ', 'ൡ'),
+ ('ൺ', 'ൿ'),
+ ('අ', 'ඖ'),
+ ('ක', 'න'),
+ ('ඳ', 'ර'),
+ ('ල', 'ල'),
+ ('ව', 'ෆ'),
+ ('ༀ', 'ༀ'),
+ ('ཀ', 'ཇ'),
+ ('ཉ', 'ཬ'),
+ ('ྈ', 'ྌ'),
+ ('Ⴀ', 'Ⴥ'),
+ ('Ⴧ', 'Ⴧ'),
+ ('Ⴭ', 'Ⴭ'),
+ ('ა', 'ჺ'),
+ ('ჼ', 'ቈ'),
+ ('ቊ', 'ቍ'),
+ ('ቐ', 'ቖ'),
+ ('ቘ', 'ቘ'),
+ ('ቚ', 'ቝ'),
+ ('በ', 'ኈ'),
+ ('ኊ', 'ኍ'),
+ ('ነ', 'ኰ'),
+ ('ኲ', 'ኵ'),
+ ('ኸ', 'ኾ'),
+ ('ዀ', 'ዀ'),
+ ('ዂ', 'ዅ'),
+ ('ወ', 'ዖ'),
+ ('ዘ', 'ጐ'),
+ ('ጒ', 'ጕ'),
+ ('ጘ', 'ፚ'),
+ ('ᎀ', 'ᎏ'),
+ ('Ꭰ', 'Ᏽ'),
+ ('ᏸ', 'ᏽ'),
+ ('ᐁ', 'ᙬ'),
+ ('ᙯ', 'ᙿ'),
+ ('ᚁ', 'ᚚ'),
+ ('ᚠ', 'ᛪ'),
+ ('ᛮ', 'ᛸ'),
+ ('ᜀ', 'ᜌ'),
+ ('ᜎ', 'ᜑ'),
+ ('ᜠ', 'ᜱ'),
+ ('ᝀ', 'ᝑ'),
+ ('ᝠ', 'ᝬ'),
+ ('ᝮ', 'ᝰ'),
+ ('ᠠ', 'ᡸ'),
+ ('ᢀ', 'ᢄ'),
+ ('ᢇ', 'ᢨ'),
+ ('ᢪ', 'ᢪ'),
+ ('ᢰ', 'ᣵ'),
+ ('ᤀ', 'ᤞ'),
+ ('ᨀ', 'ᨖ'),
+ ('ᬅ', 'ᬳ'),
+ ('ᭅ', 'ᭋ'),
+ ('ᮃ', 'ᮠ'),
+ ('ᮮ', 'ᮯ'),
+ ('ᮺ', 'ᯥ'),
+ ('ᰀ', 'ᰣ'),
+ ('ᱍ', 'ᱏ'),
+ ('ᱚ', 'ᱽ'),
+ ('ᲀ', 'ᲈ'),
+ ('Ა', 'Ჺ'),
+ ('Ჽ', 'Ჿ'),
+ ('ᳩ', 'ᳬ'),
+ ('ᳮ', 'ᳳ'),
+ ('ᳵ', 'ᳶ'),
+ ('ᳺ', 'ᳺ'),
+ ('ᴀ', 'ᶿ'),
+ ('Ḁ', 'ἕ'),
+ ('Ἐ', 'Ἕ'),
+ ('ἠ', 'ὅ'),
+ ('Ὀ', 'Ὅ'),
+ ('ὐ', 'ὗ'),
+ ('Ὑ', 'Ὑ'),
+ ('Ὓ', 'Ὓ'),
+ ('Ὕ', 'Ὕ'),
+ ('Ὗ', 'ώ'),
+ ('ᾀ', 'ᾴ'),
+ ('ᾶ', 'ᾼ'),
+ ('ι', 'ι'),
+ ('ῂ', 'ῄ'),
+ ('ῆ', 'ῌ'),
+ ('ῐ', 'ΐ'),
+ ('ῖ', 'Ί'),
+ ('ῠ', 'Ῥ'),
+ ('ῲ', 'ῴ'),
+ ('ῶ', 'ῼ'),
+ ('ⁱ', 'ⁱ'),
+ ('ⁿ', 'ⁿ'),
+ ('ₐ', 'ₜ'),
+ ('ℂ', 'ℂ'),
+ ('ℇ', 'ℇ'),
+ ('ℊ', 'ℓ'),
+ ('ℕ', 'ℕ'),
+ ('ℙ', 'ℝ'),
+ ('ℤ', 'ℤ'),
+ ('Ω', 'Ω'),
+ ('ℨ', 'ℨ'),
+ ('K', 'ℭ'),
+ ('ℯ', 'ℹ'),
+ ('ℼ', 'ℿ'),
+ ('ⅅ', 'ⅉ'),
+ ('ⅎ', 'ⅎ'),
+ ('Ⅰ', 'ↈ'),
+ ('Ⓐ', 'ⓩ'),
+ ('Ⰰ', 'Ⱞ'),
+ ('ⰰ', 'ⱞ'),
+ ('Ⱡ', 'ⳤ'),
+ ('Ⳬ', 'ⳮ'),
+ ('Ⳳ', 'ⳳ'),
+ ('ⴀ', 'ⴥ'),
+ ('ⴧ', 'ⴧ'),
+ ('ⴭ', 'ⴭ'),
+ ('ⴰ', 'ⵧ'),
+ ('ⵯ', 'ⵯ'),
+ ('ⶀ', 'ⶖ'),
+ ('ⶠ', 'ⶦ'),
+ ('ⶨ', 'ⶮ'),
+ ('ⶰ', 'ⶶ'),
+ ('ⶸ', 'ⶾ'),
+ ('ⷀ', 'ⷆ'),
+ ('ⷈ', 'ⷎ'),
+ ('ⷐ', 'ⷖ'),
+ ('ⷘ', 'ⷞ'),
+ ('ⸯ', 'ⸯ'),
+ ('々', '々'),
+ ('〻', '〼'),
+ ('ㄅ', 'ㄯ'),
+ ('ㄱ', 'ㆎ'),
+ ('ㆠ', '\u{31bf}'),
+ ('ꀀ', 'ꒌ'),
+ ('ꓐ', 'ꓽ'),
+ ('ꔀ', 'ꘌ'),
+ ('ꘐ', 'ꘟ'),
+ ('ꘪ', 'ꘫ'),
+ ('Ꙁ', 'ꙮ'),
+ ('ꙿ', 'ꚝ'),
+ ('ꚠ', 'ꛯ'),
+ ('꜈', 'ꞿ'),
+ ('Ꟃ', '\u{a7ca}'),
+ ('\u{a7f5}', 'ꠁ'),
+ ('ꠃ', 'ꠅ'),
+ ('ꠇ', 'ꠊ'),
+ ('ꠌ', 'ꠢ'),
+ ('ꡀ', 'ꡳ'),
+ ('ꢂ', 'ꢳ'),
+ ('ꣲ', 'ꣷ'),
+ ('ꣻ', 'ꣻ'),
+ ('ꣽ', 'ꣾ'),
+ ('ꤊ', 'ꤥ'),
+ ('ꤰ', 'ꥆ'),
+ ('ꥠ', 'ꥼ'),
+ ('ꦄ', 'ꦲ'),
+ ('ꧏ', 'ꧏ'),
+ ('ꨀ', 'ꨨ'),
+ ('ꩀ', 'ꩂ'),
+ ('ꩄ', 'ꩋ'),
+ ('ꫠ', 'ꫪ'),
+ ('ꫲ', 'ꫴ'),
+ ('ꬁ', 'ꬆ'),
+ ('ꬉ', 'ꬎ'),
+ ('ꬑ', 'ꬖ'),
+ ('ꬠ', 'ꬦ'),
+ ('ꬨ', 'ꬮ'),
+ ('ꬰ', '\u{ab69}'),
+ ('ꭰ', 'ꯢ'),
+ ('가', '힣'),
+ ('ힰ', 'ퟆ'),
+ ('ퟋ', 'ퟻ'),
+ ('ff', 'st'),
+ ('ﬓ', 'ﬗ'),
+ ('ﭐ', 'ﮱ'),
+ ('ﯓ', 'ﴽ'),
+ ('ﵐ', 'ﶏ'),
+ ('ﶒ', 'ﷇ'),
+ ('ﷰ', 'ﷻ'),
+ ('ﹰ', 'ﹴ'),
+ ('ﹶ', 'ﻼ'),
+ ('A', 'Z'),
+ ('a', 'z'),
+ ('ᅠ', 'ᄒ'),
+ ('ᅡ', 'ᅦ'),
+ ('ᅧ', 'ᅬ'),
+ ('ᅭ', 'ᅲ'),
+ ('ᅳ', 'ᅵ'),
+ ('𐀀', '𐀋'),
+ ('𐀍', '𐀦'),
+ ('𐀨', '𐀺'),
+ ('𐀼', '𐀽'),
+ ('𐀿', '𐁍'),
+ ('𐁐', '𐁝'),
+ ('𐂀', '𐃺'),
+ ('𐅀', '𐅴'),
+ ('𐊀', '𐊜'),
+ ('𐊠', '𐋐'),
+ ('𐌀', '𐌟'),
+ ('𐌭', '𐍊'),
+ ('𐍐', '𐍵'),
+ ('𐎀', '𐎝'),
+ ('𐎠', '𐏃'),
+ ('𐏈', '𐏏'),
+ ('𐏑', '𐏕'),
+ ('𐐀', '𐒝'),
+ ('𐒰', '𐓓'),
+ ('𐓘', '𐓻'),
+ ('𐔀', '𐔧'),
+ ('𐔰', '𐕣'),
+ ('𐘀', '𐜶'),
+ ('𐝀', '𐝕'),
+ ('𐝠', '𐝧'),
+ ('𐠀', '𐠅'),
+ ('𐠈', '𐠈'),
+ ('𐠊', '𐠵'),
+ ('𐠷', '𐠸'),
+ ('𐠼', '𐠼'),
+ ('𐠿', '𐡕'),
+ ('𐡠', '𐡶'),
+ ('𐢀', '𐢞'),
+ ('𐣠', '𐣲'),
+ ('𐣴', '𐣵'),
+ ('𐤀', '𐤕'),
+ ('𐤠', '𐤹'),
+ ('𐦀', '𐦷'),
+ ('𐦾', '𐦿'),
+ ('𐨀', '𐨀'),
+ ('𐨐', '𐨓'),
+ ('𐨕', '𐨗'),
+ ('𐨙', '𐨵'),
+ ('𐩠', '𐩼'),
+ ('𐪀', '𐪜'),
+ ('𐫀', '𐫇'),
+ ('𐫉', '𐫤'),
+ ('𐬀', '𐬵'),
+ ('𐭀', '𐭕'),
+ ('𐭠', '𐭲'),
+ ('𐮀', '𐮑'),
+ ('𐰀', '𐱈'),
+ ('𐲀', '𐲲'),
+ ('𐳀', '𐳲'),
+ ('𐴀', '𐴣'),
+ ('\u{10e80}', '\u{10ea9}'),
+ ('\u{10eb0}', '\u{10eb1}'),
+ ('𐼀', '𐼜'),
+ ('𐼧', '𐼧'),
+ ('𐼰', '𐽅'),
+ ('\u{10fb0}', '\u{10fc4}'),
+ ('𐿠', '𐿶'),
+ ('𑀃', '𑀷'),
+ ('𑂃', '𑂯'),
+ ('𑃐', '𑃨'),
+ ('𑄃', '𑄦'),
+ ('𑅄', '𑅄'),
+ ('\u{11147}', '\u{11147}'),
+ ('𑅐', '𑅲'),
+ ('𑅶', '𑅶'),
+ ('𑆃', '𑆲'),
+ ('𑇁', '𑇄'),
+ ('𑇚', '𑇚'),
+ ('𑇜', '𑇜'),
+ ('𑈀', '𑈑'),
+ ('𑈓', '𑈫'),
+ ('𑊀', '𑊆'),
+ ('𑊈', '𑊈'),
+ ('𑊊', '𑊍'),
+ ('𑊏', '𑊝'),
+ ('𑊟', '𑊨'),
+ ('𑊰', '𑋞'),
+ ('𑌅', '𑌌'),
+ ('𑌏', '𑌐'),
+ ('𑌓', '𑌨'),
+ ('𑌪', '𑌰'),
+ ('𑌲', '𑌳'),
+ ('𑌵', '𑌹'),
+ ('𑌽', '𑌽'),
+ ('𑍐', '𑍐'),
+ ('𑍝', '𑍡'),
+ ('𑐀', '𑐴'),
+ ('𑑇', '𑑊'),
+ ('𑑟', '\u{11461}'),
+ ('𑒀', '𑒯'),
+ ('𑓄', '𑓅'),
+ ('𑓇', '𑓇'),
+ ('𑖀', '𑖮'),
+ ('𑗘', '𑗛'),
+ ('𑘀', '𑘯'),
+ ('𑙄', '𑙄'),
+ ('𑚀', '𑚪'),
+ ('𑚸', '𑚸'),
+ ('𑠀', '𑠫'),
+ ('𑢠', '𑣟'),
+ ('𑣿', '\u{11906}'),
+ ('\u{11909}', '\u{11909}'),
+ ('\u{1190c}', '\u{11913}'),
+ ('\u{11915}', '\u{11916}'),
+ ('\u{11918}', '\u{1192f}'),
+ ('\u{1193f}', '\u{1193f}'),
+ ('\u{11941}', '\u{11941}'),
+ ('𑦠', '𑦧'),
+ ('𑦪', '𑧐'),
+ ('𑧡', '𑧡'),
+ ('𑧣', '𑧣'),
+ ('𑨀', '𑨀'),
+ ('𑨋', '𑨲'),
+ ('𑨺', '𑨺'),
+ ('𑩐', '𑩐'),
+ ('𑩜', '𑪉'),
+ ('𑪝', '𑪝'),
+ ('𑫀', '𑫸'),
+ ('𑰀', '𑰈'),
+ ('𑰊', '𑰮'),
+ ('𑱀', '𑱀'),
+ ('𑱲', '𑲏'),
+ ('𑴀', '𑴆'),
+ ('𑴈', '𑴉'),
+ ('𑴋', '𑴰'),
+ ('𑵆', '𑵆'),
+ ('𑵠', '𑵥'),
+ ('𑵧', '𑵨'),
+ ('𑵪', '𑶉'),
+ ('𑶘', '𑶘'),
+ ('𑻠', '𑻲'),
+ ('\u{11fb0}', '\u{11fb0}'),
+ ('𒀀', '𒎙'),
+ ('𒐀', '𒑮'),
+ ('𒒀', '𒕃'),
+ ('𓀀', '𓐮'),
+ ('𔐀', '𔙆'),
+ ('𖠀', '𖨸'),
+ ('𖩀', '𖩞'),
+ ('𖫐', '𖫭'),
+ ('𖬀', '𖬯'),
+ ('𖭀', '𖭃'),
+ ('𖭣', '𖭷'),
+ ('𖭽', '𖮏'),
+ ('𖹀', '𖹿'),
+ ('𖼀', '𖽊'),
+ ('𖽐', '𖽐'),
+ ('𖾓', '𖾟'),
+ ('𖿠', '𖿡'),
+ ('𖿣', '𖿣'),
+ ('𛰀', '𛱪'),
+ ('𛱰', '𛱼'),
+ ('𛲀', '𛲈'),
+ ('𛲐', '𛲙'),
+ ('𝐀', '𝑔'),
+ ('𝑖', '𝒜'),
+ ('𝒞', '𝒟'),
+ ('𝒢', '𝒢'),
+ ('𝒥', '𝒦'),
+ ('𝒩', '𝒬'),
+ ('𝒮', '𝒹'),
+ ('𝒻', '𝒻'),
+ ('𝒽', '𝓃'),
+ ('𝓅', '𝔅'),
+ ('𝔇', '𝔊'),
+ ('𝔍', '𝔔'),
+ ('𝔖', '𝔜'),
+ ('𝔞', '𝔹'),
+ ('𝔻', '𝔾'),
+ ('𝕀', '𝕄'),
+ ('𝕆', '𝕆'),
+ ('𝕊', '𝕐'),
+ ('𝕒', '𝚥'),
+ ('𝚨', '𝛀'),
+ ('𝛂', '𝛚'),
+ ('𝛜', '𝛺'),
+ ('𝛼', '𝜔'),
+ ('𝜖', '𝜴'),
+ ('𝜶', '𝝎'),
+ ('𝝐', '𝝮'),
+ ('𝝰', '𝞈'),
+ ('𝞊', '𝞨'),
+ ('𝞪', '𝟂'),
+ ('𝟄', '𝟋'),
+ ('𞄀', '𞄬'),
+ ('𞄷', '𞄽'),
+ ('𞅎', '𞅎'),
+ ('𞋀', '𞋫'),
+ ('𞠀', '𞣄'),
+ ('𞤀', '𞥃'),
+ ('𞥋', '𞥋'),
+ ('𞸀', '𞸃'),
+ ('𞸅', '𞸟'),
+ ('𞸡', '𞸢'),
+ ('𞸤', '𞸤'),
+ ('𞸧', '𞸧'),
+ ('𞸩', '𞸲'),
+ ('𞸴', '𞸷'),
+ ('𞸹', '𞸹'),
+ ('𞸻', '𞸻'),
+ ('𞹂', '𞹂'),
+ ('𞹇', '𞹇'),
+ ('𞹉', '𞹉'),
+ ('𞹋', '𞹋'),
+ ('𞹍', '𞹏'),
+ ('𞹑', '𞹒'),
+ ('𞹔', '𞹔'),
+ ('𞹗', '𞹗'),
+ ('𞹙', '𞹙'),
+ ('𞹛', '𞹛'),
+ ('𞹝', '𞹝'),
+ ('𞹟', '𞹟'),
+ ('𞹡', '𞹢'),
+ ('𞹤', '𞹤'),
+ ('𞹧', '𞹪'),
+ ('𞹬', '𞹲'),
+ ('𞹴', '𞹷'),
+ ('𞹹', '𞹼'),
+ ('𞹾', '𞹾'),
+ ('𞺀', '𞺉'),
+ ('𞺋', '𞺛'),
+ ('𞺡', '𞺣'),
+ ('𞺥', '𞺩'),
+ ('𞺫', '𞺻'),
+ ('🄰', '🅉'),
+ ('🅐', '🅩'),
+ ('🅰', '🆉'),
+];
+
+pub const CR: &'static [(char, char)] = &[('\r', '\r')];
+
+pub const DOUBLE_QUOTE: &'static [(char, char)] = &[('\"', '\"')];
+
+pub const EXTEND: &'static [(char, char)] = &[
+ ('\u{300}', '\u{36f}'),
+ ('\u{483}', '\u{489}'),
+ ('\u{591}', '\u{5bd}'),
+ ('\u{5bf}', '\u{5bf}'),
+ ('\u{5c1}', '\u{5c2}'),
+ ('\u{5c4}', '\u{5c5}'),
+ ('\u{5c7}', '\u{5c7}'),
+ ('\u{610}', '\u{61a}'),
+ ('\u{64b}', '\u{65f}'),
+ ('\u{670}', '\u{670}'),
+ ('\u{6d6}', '\u{6dc}'),
+ ('\u{6df}', '\u{6e4}'),
+ ('\u{6e7}', '\u{6e8}'),
+ ('\u{6ea}', '\u{6ed}'),
+ ('\u{711}', '\u{711}'),
+ ('\u{730}', '\u{74a}'),
+ ('\u{7a6}', '\u{7b0}'),
+ ('\u{7eb}', '\u{7f3}'),
+ ('\u{7fd}', '\u{7fd}'),
+ ('\u{816}', '\u{819}'),
+ ('\u{81b}', '\u{823}'),
+ ('\u{825}', '\u{827}'),
+ ('\u{829}', '\u{82d}'),
+ ('\u{859}', '\u{85b}'),
+ ('\u{8d3}', '\u{8e1}'),
+ ('\u{8e3}', 'ः'),
+ ('\u{93a}', '\u{93c}'),
+ ('ा', 'ॏ'),
+ ('\u{951}', '\u{957}'),
+ ('\u{962}', '\u{963}'),
+ ('\u{981}', 'ঃ'),
+ ('\u{9bc}', '\u{9bc}'),
+ ('\u{9be}', '\u{9c4}'),
+ ('ে', 'ৈ'),
+ ('ো', '\u{9cd}'),
+ ('\u{9d7}', '\u{9d7}'),
+ ('\u{9e2}', '\u{9e3}'),
+ ('\u{9fe}', '\u{9fe}'),
+ ('\u{a01}', 'ਃ'),
+ ('\u{a3c}', '\u{a3c}'),
+ ('ਾ', '\u{a42}'),
+ ('\u{a47}', '\u{a48}'),
+ ('\u{a4b}', '\u{a4d}'),
+ ('\u{a51}', '\u{a51}'),
+ ('\u{a70}', '\u{a71}'),
+ ('\u{a75}', '\u{a75}'),
+ ('\u{a81}', 'ઃ'),
+ ('\u{abc}', '\u{abc}'),
+ ('ા', '\u{ac5}'),
+ ('\u{ac7}', 'ૉ'),
+ ('ો', '\u{acd}'),
+ ('\u{ae2}', '\u{ae3}'),
+ ('\u{afa}', '\u{aff}'),
+ ('\u{b01}', 'ଃ'),
+ ('\u{b3c}', '\u{b3c}'),
+ ('\u{b3e}', '\u{b44}'),
+ ('େ', 'ୈ'),
+ ('ୋ', '\u{b4d}'),
+ ('\u{b55}', '\u{b57}'),
+ ('\u{b62}', '\u{b63}'),
+ ('\u{b82}', '\u{b82}'),
+ ('\u{bbe}', 'ூ'),
+ ('ெ', 'ை'),
+ ('ொ', '\u{bcd}'),
+ ('\u{bd7}', '\u{bd7}'),
+ ('\u{c00}', '\u{c04}'),
+ ('\u{c3e}', 'ౄ'),
+ ('\u{c46}', '\u{c48}'),
+ ('\u{c4a}', '\u{c4d}'),
+ ('\u{c55}', '\u{c56}'),
+ ('\u{c62}', '\u{c63}'),
+ ('\u{c81}', 'ಃ'),
+ ('\u{cbc}', '\u{cbc}'),
+ ('ಾ', 'ೄ'),
+ ('\u{cc6}', 'ೈ'),
+ ('ೊ', '\u{ccd}'),
+ ('\u{cd5}', '\u{cd6}'),
+ ('\u{ce2}', '\u{ce3}'),
+ ('\u{d00}', 'ഃ'),
+ ('\u{d3b}', '\u{d3c}'),
+ ('\u{d3e}', '\u{d44}'),
+ ('െ', 'ൈ'),
+ ('ൊ', '\u{d4d}'),
+ ('\u{d57}', '\u{d57}'),
+ ('\u{d62}', '\u{d63}'),
+ ('\u{d81}', 'ඃ'),
+ ('\u{dca}', '\u{dca}'),
+ ('\u{dcf}', '\u{dd4}'),
+ ('\u{dd6}', '\u{dd6}'),
+ ('ෘ', '\u{ddf}'),
+ ('ෲ', 'ෳ'),
+ ('\u{e31}', '\u{e31}'),
+ ('\u{e34}', '\u{e3a}'),
+ ('\u{e47}', '\u{e4e}'),
+ ('\u{eb1}', '\u{eb1}'),
+ ('\u{eb4}', '\u{ebc}'),
+ ('\u{ec8}', '\u{ecd}'),
+ ('\u{f18}', '\u{f19}'),
+ ('\u{f35}', '\u{f35}'),
+ ('\u{f37}', '\u{f37}'),
+ ('\u{f39}', '\u{f39}'),
+ ('༾', '༿'),
+ ('\u{f71}', '\u{f84}'),
+ ('\u{f86}', '\u{f87}'),
+ ('\u{f8d}', '\u{f97}'),
+ ('\u{f99}', '\u{fbc}'),
+ ('\u{fc6}', '\u{fc6}'),
+ ('ါ', '\u{103e}'),
+ ('ၖ', '\u{1059}'),
+ ('\u{105e}', '\u{1060}'),
+ ('ၢ', 'ၤ'),
+ ('ၧ', 'ၭ'),
+ ('\u{1071}', '\u{1074}'),
+ ('\u{1082}', '\u{108d}'),
+ ('ႏ', 'ႏ'),
+ ('ႚ', '\u{109d}'),
+ ('\u{135d}', '\u{135f}'),
+ ('\u{1712}', '\u{1714}'),
+ ('\u{1732}', '\u{1734}'),
+ ('\u{1752}', '\u{1753}'),
+ ('\u{1772}', '\u{1773}'),
+ ('\u{17b4}', '\u{17d3}'),
+ ('\u{17dd}', '\u{17dd}'),
+ ('\u{180b}', '\u{180d}'),
+ ('\u{1885}', '\u{1886}'),
+ ('\u{18a9}', '\u{18a9}'),
+ ('\u{1920}', 'ᤫ'),
+ ('ᤰ', '\u{193b}'),
+ ('\u{1a17}', '\u{1a1b}'),
+ ('ᩕ', '\u{1a5e}'),
+ ('\u{1a60}', '\u{1a7c}'),
+ ('\u{1a7f}', '\u{1a7f}'),
+ ('\u{1ab0}', '\u{1ac0}'),
+ ('\u{1b00}', 'ᬄ'),
+ ('\u{1b34}', '᭄'),
+ ('\u{1b6b}', '\u{1b73}'),
+ ('\u{1b80}', 'ᮂ'),
+ ('ᮡ', '\u{1bad}'),
+ ('\u{1be6}', '᯳'),
+ ('ᰤ', '\u{1c37}'),
+ ('\u{1cd0}', '\u{1cd2}'),
+ ('\u{1cd4}', '\u{1ce8}'),
+ ('\u{1ced}', '\u{1ced}'),
+ ('\u{1cf4}', '\u{1cf4}'),
+ ('᳷', '\u{1cf9}'),
+ ('\u{1dc0}', '\u{1df9}'),
+ ('\u{1dfb}', '\u{1dff}'),
+ ('\u{200c}', '\u{200c}'),
+ ('\u{20d0}', '\u{20f0}'),
+ ('\u{2cef}', '\u{2cf1}'),
+ ('\u{2d7f}', '\u{2d7f}'),
+ ('\u{2de0}', '\u{2dff}'),
+ ('\u{302a}', '\u{302f}'),
+ ('\u{3099}', '\u{309a}'),
+ ('\u{a66f}', '\u{a672}'),
+ ('\u{a674}', '\u{a67d}'),
+ ('\u{a69e}', '\u{a69f}'),
+ ('\u{a6f0}', '\u{a6f1}'),
+ ('\u{a802}', '\u{a802}'),
+ ('\u{a806}', '\u{a806}'),
+ ('\u{a80b}', '\u{a80b}'),
+ ('ꠣ', 'ꠧ'),
+ ('\u{a82c}', '\u{a82c}'),
+ ('ꢀ', 'ꢁ'),
+ ('ꢴ', '\u{a8c5}'),
+ ('\u{a8e0}', '\u{a8f1}'),
+ ('\u{a8ff}', '\u{a8ff}'),
+ ('\u{a926}', '\u{a92d}'),
+ ('\u{a947}', '꥓'),
+ ('\u{a980}', 'ꦃ'),
+ ('\u{a9b3}', '꧀'),
+ ('\u{a9e5}', '\u{a9e5}'),
+ ('\u{aa29}', '\u{aa36}'),
+ ('\u{aa43}', '\u{aa43}'),
+ ('\u{aa4c}', 'ꩍ'),
+ ('ꩻ', 'ꩽ'),
+ ('\u{aab0}', '\u{aab0}'),
+ ('\u{aab2}', '\u{aab4}'),
+ ('\u{aab7}', '\u{aab8}'),
+ ('\u{aabe}', '\u{aabf}'),
+ ('\u{aac1}', '\u{aac1}'),
+ ('ꫫ', 'ꫯ'),
+ ('ꫵ', '\u{aaf6}'),
+ ('ꯣ', 'ꯪ'),
+ ('꯬', '\u{abed}'),
+ ('\u{fb1e}', '\u{fb1e}'),
+ ('\u{fe00}', '\u{fe0f}'),
+ ('\u{fe20}', '\u{fe2f}'),
+ ('\u{ff9e}', '\u{ff9f}'),
+ ('\u{101fd}', '\u{101fd}'),
+ ('\u{102e0}', '\u{102e0}'),
+ ('\u{10376}', '\u{1037a}'),
+ ('\u{10a01}', '\u{10a03}'),
+ ('\u{10a05}', '\u{10a06}'),
+ ('\u{10a0c}', '\u{10a0f}'),
+ ('\u{10a38}', '\u{10a3a}'),
+ ('\u{10a3f}', '\u{10a3f}'),
+ ('\u{10ae5}', '\u{10ae6}'),
+ ('\u{10d24}', '\u{10d27}'),
+ ('\u{10eab}', '\u{10eac}'),
+ ('\u{10f46}', '\u{10f50}'),
+ ('𑀀', '𑀂'),
+ ('\u{11038}', '\u{11046}'),
+ ('\u{1107f}', '𑂂'),
+ ('𑂰', '\u{110ba}'),
+ ('\u{11100}', '\u{11102}'),
+ ('\u{11127}', '\u{11134}'),
+ ('𑅅', '𑅆'),
+ ('\u{11173}', '\u{11173}'),
+ ('\u{11180}', '𑆂'),
+ ('𑆳', '𑇀'),
+ ('\u{111c9}', '\u{111cc}'),
+ ('\u{111ce}', '\u{111cf}'),
+ ('𑈬', '\u{11237}'),
+ ('\u{1123e}', '\u{1123e}'),
+ ('\u{112df}', '\u{112ea}'),
+ ('\u{11300}', '𑌃'),
+ ('\u{1133b}', '\u{1133c}'),
+ ('\u{1133e}', '𑍄'),
+ ('𑍇', '𑍈'),
+ ('𑍋', '𑍍'),
+ ('\u{11357}', '\u{11357}'),
+ ('𑍢', '𑍣'),
+ ('\u{11366}', '\u{1136c}'),
+ ('\u{11370}', '\u{11374}'),
+ ('𑐵', '\u{11446}'),
+ ('\u{1145e}', '\u{1145e}'),
+ ('\u{114b0}', '\u{114c3}'),
+ ('\u{115af}', '\u{115b5}'),
+ ('𑖸', '\u{115c0}'),
+ ('\u{115dc}', '\u{115dd}'),
+ ('𑘰', '\u{11640}'),
+ ('\u{116ab}', '\u{116b7}'),
+ ('\u{1171d}', '\u{1172b}'),
+ ('𑠬', '\u{1183a}'),
+ ('\u{11930}', '\u{11935}'),
+ ('\u{11937}', '\u{11938}'),
+ ('\u{1193b}', '\u{1193e}'),
+ ('\u{11940}', '\u{11940}'),
+ ('\u{11942}', '\u{11943}'),
+ ('𑧑', '\u{119d7}'),
+ ('\u{119da}', '\u{119e0}'),
+ ('𑧤', '𑧤'),
+ ('\u{11a01}', '\u{11a0a}'),
+ ('\u{11a33}', '𑨹'),
+ ('\u{11a3b}', '\u{11a3e}'),
+ ('\u{11a47}', '\u{11a47}'),
+ ('\u{11a51}', '\u{11a5b}'),
+ ('\u{11a8a}', '\u{11a99}'),
+ ('𑰯', '\u{11c36}'),
+ ('\u{11c38}', '\u{11c3f}'),
+ ('\u{11c92}', '\u{11ca7}'),
+ ('𑲩', '\u{11cb6}'),
+ ('\u{11d31}', '\u{11d36}'),
+ ('\u{11d3a}', '\u{11d3a}'),
+ ('\u{11d3c}', '\u{11d3d}'),
+ ('\u{11d3f}', '\u{11d45}'),
+ ('\u{11d47}', '\u{11d47}'),
+ ('𑶊', '𑶎'),
+ ('\u{11d90}', '\u{11d91}'),
+ ('𑶓', '\u{11d97}'),
+ ('\u{11ef3}', '𑻶'),
+ ('\u{16af0}', '\u{16af4}'),
+ ('\u{16b30}', '\u{16b36}'),
+ ('\u{16f4f}', '\u{16f4f}'),
+ ('𖽑', '𖾇'),
+ ('\u{16f8f}', '\u{16f92}'),
+ ('\u{16fe4}', '\u{16fe4}'),
+ ('\u{16ff0}', '\u{16ff1}'),
+ ('\u{1bc9d}', '\u{1bc9e}'),
+ ('\u{1d165}', '\u{1d169}'),
+ ('𝅭', '\u{1d172}'),
+ ('\u{1d17b}', '\u{1d182}'),
+ ('\u{1d185}', '\u{1d18b}'),
+ ('\u{1d1aa}', '\u{1d1ad}'),
+ ('\u{1d242}', '\u{1d244}'),
+ ('\u{1da00}', '\u{1da36}'),
+ ('\u{1da3b}', '\u{1da6c}'),
+ ('\u{1da75}', '\u{1da75}'),
+ ('\u{1da84}', '\u{1da84}'),
+ ('\u{1da9b}', '\u{1da9f}'),
+ ('\u{1daa1}', '\u{1daaf}'),
+ ('\u{1e000}', '\u{1e006}'),
+ ('\u{1e008}', '\u{1e018}'),
+ ('\u{1e01b}', '\u{1e021}'),
+ ('\u{1e023}', '\u{1e024}'),
+ ('\u{1e026}', '\u{1e02a}'),
+ ('\u{1e130}', '\u{1e136}'),
+ ('\u{1e2ec}', '\u{1e2ef}'),
+ ('\u{1e8d0}', '\u{1e8d6}'),
+ ('\u{1e944}', '\u{1e94a}'),
+ ('🏻', '🏿'),
+ ('\u{e0020}', '\u{e007f}'),
+ ('\u{e0100}', '\u{e01ef}'),
+];
+
+pub const EXTENDNUMLET: &'static [(char, char)] = &[
+ ('_', '_'),
+ ('\u{202f}', '\u{202f}'),
+ ('‿', '⁀'),
+ ('⁔', '⁔'),
+ ('︳', '︴'),
+ ('﹍', '﹏'),
+ ('_', '_'),
+];
+
+pub const FORMAT: &'static [(char, char)] = &[
+ ('\u{ad}', '\u{ad}'),
+ ('\u{600}', '\u{605}'),
+ ('\u{61c}', '\u{61c}'),
+ ('\u{6dd}', '\u{6dd}'),
+ ('\u{70f}', '\u{70f}'),
+ ('\u{8e2}', '\u{8e2}'),
+ ('\u{180e}', '\u{180e}'),
+ ('\u{200e}', '\u{200f}'),
+ ('\u{202a}', '\u{202e}'),
+ ('\u{2060}', '\u{2064}'),
+ ('\u{2066}', '\u{206f}'),
+ ('\u{feff}', '\u{feff}'),
+ ('\u{fff9}', '\u{fffb}'),
+ ('\u{110bd}', '\u{110bd}'),
+ ('\u{110cd}', '\u{110cd}'),
+ ('\u{13430}', '\u{13438}'),
+ ('\u{1bca0}', '\u{1bca3}'),
+ ('\u{1d173}', '\u{1d17a}'),
+ ('\u{e0001}', '\u{e0001}'),
+];
+
+pub const HEBREW_LETTER: &'static [(char, char)] = &[
+ ('א', 'ת'),
+ ('ׯ', 'ײ'),
+ ('יִ', 'יִ'),
+ ('ײַ', 'ﬨ'),
+ ('שׁ', 'זּ'),
+ ('טּ', 'לּ'),
+ ('מּ', 'מּ'),
+ ('נּ', 'סּ'),
+ ('ףּ', 'פּ'),
+ ('צּ', 'ﭏ'),
+];
+
+pub const KATAKANA: &'static [(char, char)] = &[
+ ('〱', '〵'),
+ ('゛', '゜'),
+ ('゠', 'ヺ'),
+ ('ー', 'ヿ'),
+ ('ㇰ', 'ㇿ'),
+ ('㋐', '㋾'),
+ ('㌀', '㍗'),
+ ('ヲ', 'ン'),
+ ('𛀀', '𛀀'),
+ ('𛅤', '𛅧'),
+];
+
+pub const LF: &'static [(char, char)] = &[('\n', '\n')];
+
+pub const MIDLETTER: &'static [(char, char)] = &[
+ (':', ':'),
+ ('·', '·'),
+ ('·', '·'),
+ ('՟', '՟'),
+ ('״', '״'),
+ ('‧', '‧'),
+ ('︓', '︓'),
+ ('﹕', '﹕'),
+ (':', ':'),
+];
+
+pub const MIDNUM: &'static [(char, char)] = &[
+ (',', ','),
+ (';', ';'),
+ (';', ';'),
+ ('։', '։'),
+ ('،', '؍'),
+ ('٬', '٬'),
+ ('߸', '߸'),
+ ('⁄', '⁄'),
+ ('︐', '︐'),
+ ('︔', '︔'),
+ ('﹐', '﹐'),
+ ('﹔', '﹔'),
+ (',', ','),
+ (';', ';'),
+];
+
+pub const MIDNUMLET: &'static [(char, char)] = &[
+ ('.', '.'),
+ ('‘', '’'),
+ ('․', '․'),
+ ('﹒', '﹒'),
+ (''', '''),
+ ('.', '.'),
+];
+
+pub const NEWLINE: &'static [(char, char)] =
+ &[('\u{b}', '\u{c}'), ('\u{85}', '\u{85}'), ('\u{2028}', '\u{2029}')];
+
+pub const NUMERIC: &'static [(char, char)] = &[
+ ('0', '9'),
+ ('٠', '٩'),
+ ('٫', '٫'),
+ ('۰', '۹'),
+ ('߀', '߉'),
+ ('०', '९'),
+ ('০', '৯'),
+ ('੦', '੯'),
+ ('૦', '૯'),
+ ('୦', '୯'),
+ ('௦', '௯'),
+ ('౦', '౯'),
+ ('೦', '೯'),
+ ('൦', '൯'),
+ ('෦', '෯'),
+ ('๐', '๙'),
+ ('໐', '໙'),
+ ('༠', '༩'),
+ ('၀', '၉'),
+ ('႐', '႙'),
+ ('០', '៩'),
+ ('᠐', '᠙'),
+ ('᥆', '᥏'),
+ ('᧐', '᧙'),
+ ('᪀', '᪉'),
+ ('᪐', '᪙'),
+ ('᭐', '᭙'),
+ ('᮰', '᮹'),
+ ('᱀', '᱉'),
+ ('᱐', '᱙'),
+ ('꘠', '꘩'),
+ ('꣐', '꣙'),
+ ('꤀', '꤉'),
+ ('꧐', '꧙'),
+ ('꧰', '꧹'),
+ ('꩐', '꩙'),
+ ('꯰', '꯹'),
+ ('0', '9'),
+ ('𐒠', '𐒩'),
+ ('𐴰', '𐴹'),
+ ('𑁦', '𑁯'),
+ ('𑃰', '𑃹'),
+ ('𑄶', '𑄿'),
+ ('𑇐', '𑇙'),
+ ('𑋰', '𑋹'),
+ ('𑑐', '𑑙'),
+ ('𑓐', '𑓙'),
+ ('𑙐', '𑙙'),
+ ('𑛀', '𑛉'),
+ ('𑜰', '𑜹'),
+ ('𑣠', '𑣩'),
+ ('\u{11950}', '\u{11959}'),
+ ('𑱐', '𑱙'),
+ ('𑵐', '𑵙'),
+ ('𑶠', '𑶩'),
+ ('𖩠', '𖩩'),
+ ('𖭐', '𖭙'),
+ ('𝟎', '𝟿'),
+ ('𞅀', '𞅉'),
+ ('𞋰', '𞋹'),
+ ('𞥐', '𞥙'),
+ ('\u{1fbf0}', '\u{1fbf9}'),
+];
+
+pub const REGIONAL_INDICATOR: &'static [(char, char)] = &[('🇦', '🇿')];
+
+pub const SINGLE_QUOTE: &'static [(char, char)] = &[('\'', '\'')];
+
+pub const WSEGSPACE: &'static [(char, char)] = &[
+ (' ', ' '),
+ ('\u{1680}', '\u{1680}'),
+ ('\u{2000}', '\u{2006}'),
+ ('\u{2008}', '\u{200a}'),
+ ('\u{205f}', '\u{205f}'),
+ ('\u{3000}', '\u{3000}'),
+];
+
+pub const ZWJ: &'static [(char, char)] = &[('\u{200d}', '\u{200d}')];