I'm adding functions for comparing Unicode strings, while ignoring their
normalization form.
2009-03-06 Bruno Haible <br...@clisp.org>
New module 'uninorm/u32-normcmp'.
* lib/uninorm/u32-normcmp.c: New file.
* modules/uninorm/u32-normcmp: New file.
New module 'uninorm/u16-normcmp'.
* lib/uninorm/u16-normcmp.c: New file.
* modules/uninorm/u16-normcmp: New file.
New module 'uninorm/u8-normcmp'.
* lib/uninorm.h (u8_normcmp, u16_normcmp, u32_normcmp): New
declarations.
* lib/uninorm/u8-normcmp.c: New file.
* lib/uninorm/u-normcmp.h: New file.
* modules/uninorm/u8-normcmp: New file.
*** lib/uninorm.h.orig 2009-03-06 19:20:21.000000000 +0100
--- lib/uninorm.h 2009-03-06 19:01:38.000000000 +0100
***************
*** 145,150 ****
--- 145,165 ----
uint32_t *resultbuf, size_t *lengthp);
+ /* Compare S1 and S2, ignoring normalization.
+ NF must be either UNINORM_NFD or UNINORM_NFKD.
+ If successful, set *RESULT to -1 if S1 < S2, 0 if S1 = S2, 1 if S1 > S2,
and
+ return 0. Upon failure, return the error number. */
+ extern int
+ u8_normcmp (const uint8_t *s1, size_t n1, const uint8_t *s2, size_t n2,
+ uninorm_t nf, int *result);
+ extern int
+ u16_normcmp (const uint16_t *s1, size_t n1, const uint16_t *s2, size_t
n2,
+ uninorm_t nf, int *result);
+ extern int
+ u32_normcmp (const uint32_t *s1, size_t n1, const uint32_t *s2, size_t
n2,
+ uninorm_t nf, int *result);
+
+
/* Normalization of a stream of Unicode characters.
A "stream of Unicode characters" is essentially a function that accepts an
