unicode.hh (4055B)
1 #ifndef unicode_hh_INCLUDED 2 #define unicode_hh_INCLUDED 3 4 #include <cwctype> 5 #include <cwchar> 6 7 #include "array_view.hh" 8 #include "units.hh" 9 10 namespace Kakoune 11 { 12 13 using Codepoint = char32_t; 14 15 inline bool is_eol(Codepoint c) noexcept 16 { 17 return c == '\n'; 18 } 19 20 inline bool is_horizontal_blank(Codepoint c) noexcept 21 { 22 // Characters considered whitespace by ECMA Regex Spec 23 // minus vertical tab 24 // <https://262.ecma-international.org/11.0/#sec-white-space> 25 return c == '\t' or 26 c == '\f' or 27 c == ' ' or 28 c == U'\u00A0' or 29 c == U'\uFEFF' or 30 c == U'\u1680' or 31 c == U'\u2000' or 32 c == U'\u2001' or 33 c == U'\u2002' or 34 c == U'\u2003' or 35 c == U'\u2004' or 36 c == U'\u2005' or 37 c == U'\u2006' or 38 c == U'\u2007' or 39 c == U'\u2008' or 40 c == U'\u2009' or 41 c == U'\u200A' or 42 c == U'\u2028' or 43 c == U'\u2029' or 44 c == U'\u202F' or 45 c == U'\u205F' or 46 c == U'\u3000' ; 47 } 48 49 inline bool is_blank(Codepoint c) noexcept 50 { 51 // Characters considered Line Terminators by ECMA Regex Spec 52 // plus vertical tab 53 // <https://262.ecma-international.org/11.0/#sec-line-terminators> 54 return c == '\n' or 55 c == '\r' or 56 c == '\v' or 57 c == U'\u2028' or 58 c == U'\u2029' or 59 is_horizontal_blank(c) ; 60 } 61 62 inline bool is_basic_alpha(Codepoint c) noexcept 63 { 64 return (c >= 'a' and c <= 'z') or (c >= 'A' and c <= 'Z'); 65 } 66 67 inline bool is_basic_digit(Codepoint c) noexcept 68 { 69 return c >= '0' and c <= '9'; 70 } 71 72 enum WordType { Word, WORD }; 73 74 template<WordType word_type = Word> 75 inline bool is_word(Codepoint c, ConstArrayView<Codepoint> extra_word_chars = {'_'}) noexcept 76 { 77 if (c < 128 ? is_basic_alpha(c) or is_basic_digit(c) : iswalnum((wchar_t)c)) 78 return true; 79 for (auto cp : extra_word_chars) 80 if (c == cp) 81 return true; 82 return false; 83 } 84 85 template<> 86 inline bool is_word<WORD>(Codepoint c, ConstArrayView<Codepoint>) noexcept 87 { 88 return not is_blank(c); 89 } 90 91 inline bool is_punctuation(Codepoint c, ConstArrayView<Codepoint> extra_word_chars = {'_'}) noexcept 92 { 93 return not (is_word(c, extra_word_chars) or is_blank(c)); 94 } 95 96 inline bool is_identifier(Codepoint c) noexcept 97 { 98 return is_basic_alpha(c) or is_basic_digit(c) or 99 c == '_' or c == '-'; 100 } 101 102 inline ColumnCount codepoint_width(Codepoint c) noexcept 103 { 104 if (c < 0x80) [[likely]] 105 return 1; 106 const auto width = wcwidth((wchar_t)c); 107 return width >= 0 ? width : 1; 108 } 109 110 enum class CharCategories 111 { 112 Blank, 113 EndOfLine, 114 Word, 115 Punctuation, 116 }; 117 118 template<WordType word_type = Word> 119 inline CharCategories categorize(Codepoint c, ConstArrayView<Codepoint> extra_word_chars) noexcept 120 { 121 if (is_eol(c)) 122 return CharCategories::EndOfLine; 123 if (is_horizontal_blank(c)) 124 return CharCategories::Blank; 125 if (word_type == WORD or is_word(c, extra_word_chars)) 126 return CharCategories::Word; 127 return CharCategories::Punctuation; 128 } 129 130 inline char to_lower(char c) noexcept { return c >= 'A' and c <= 'Z' ? c - 'A' + 'a' : c; } 131 inline char to_upper(char c) noexcept { return c >= 'a' and c <= 'z' ? c - 'a' + 'A' : c; } 132 133 inline bool is_lower(char c) noexcept { return c >= 'a' and c <= 'z'; } 134 inline bool is_upper(char c) noexcept { return c >= 'A' and c <= 'Z'; } 135 136 inline Codepoint to_lower(Codepoint cp) noexcept { return cp < 128 ? (Codepoint)to_lower((char)cp) : towlower((wchar_t)cp); } 137 inline Codepoint to_upper(Codepoint cp) noexcept { return cp < 128 ? (Codepoint)to_upper((char)cp) : towupper((wchar_t)cp); } 138 139 inline bool is_lower(Codepoint cp) noexcept { return cp < 128 ? is_lower((char)cp) : iswlower((wchar_t)cp); } 140 inline bool is_upper(Codepoint cp) noexcept { return cp < 128 ? is_upper((char)cp) : iswupper((wchar_t)cp); } 141 142 } 143 144 #endif // unicode_hh_INCLUDED