Detect utf8 urls

They might not be valid, but like many things they are still used
This commit is contained in:
TingPing 2014-09-07 19:39:30 -04:00
parent 3ad8776a51
commit ce04d310e1

View File

@ -415,8 +415,8 @@ regex_match (const GRegex *re, const char *word, int *start, int *end)
}
/* Miscellaneous description --- */
#define DOMAIN "[_a-z0-9][-_a-z0-9]*(\\.[-_a-z0-9]+)*"
#define TLD "\\.[a-z][-a-z0-9]*[a-z]"
#define DOMAIN "[_\\pL\\pN][-_\\pL\\pN]*(\\.[-_\\pL\\pN]+)*"
#define TLD "\\.[\\pL][-\\pL\\pN]*[\\pL]"
#define IPADDR "[0-9]{1,3}(\\.[0-9]{1,3}){3}"
#define IPV6GROUP "([0-9a-f]{0,4})"
#define IPV6ADDR "((" IPV6GROUP "(:" IPV6GROUP "){7})" \