Handle non-BMP combining characters

Use a level of indirection to store combining characters. This allows
combining characters to be stored using only two bytes, even if they are
after USHORT_MAX
This commit is contained in:
Kovid Goyal
2018-01-18 16:25:42 +05:30
parent 32632264ee
commit 80301d465b
6 changed files with 36 additions and 34 deletions

View File

@@ -6,6 +6,8 @@ bool is_ignored_char(char_type ch);
bool is_word_char(char_type ch);
bool is_CZ_category(char_type);
bool is_P_category(char_type);
char_type codepoint_for_mark(combining_type m);
combining_type mark_for_codepoint(char_type c);
static inline bool
is_url_char(uint32_t ch) {