tommy/libpostal
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[utils] function to create an array of uint32_t codepoints from a UTF-8 string, a few bug fixes to string_utils

Browse Source
This commit is contained in:
Al
2017-10-19 04:48:50 -04:00
parent c61007388b
commit 245aa226e0
2 changed files with 27 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
src/string_utils.h
View File

@@ -74,7 +74,7 @@ ssize_t utf8proc_iterate_reversed(const uint8_t *str, ssize_t start, int32_t *ds
char *utf8_lower_options(const char *s, utf8proc_option_t options);
char *utf8_lower(const char *s);
char *utf8_upper_options(const char *s, utf8proc_option_t options);
char *utf8_lower(const char *s);
char *utf8_upper(const char *s);
int utf8_compare(const char *str1, const char *str2);
int utf8_compare_len(const char *str1, const char *str2, size_t len);
@@ -87,6 +87,8 @@ bool utf8_equal_ignore_separators(const char *str1, const char *str2);
ssize_t utf8_len(const char *str, size_t len);
uint32_array *unicode_codepoints(const char *str);
bool utf8_is_hyphen(int32_t ch);
bool utf8_is_letter(int cat);
bool utf8_is_number(int cat);