X-Git-Url: http://mj.ucw.cz/gitweb/?a=blobdiff_plain;ds=inline;f=ucw%2Funicode.h;h=c9d6a0c1cde12477e1e88b0f0601272d179d1122;hb=da418321735b159f0ff39366d68203c772f152d4;hp=2bbbc5170635a4fa3c6c275bd42a1c2603467f76;hpb=1cf8ac51f5495ccd5187dc220ffc69e95d6e0cfc;p=libucw.git

diff --git a/ucw/unicode.h b/ucw/unicode.h
index 2bbbc517..c9d6a0c1 100644
--- a/ucw/unicode.h
+++ b/ucw/unicode.h
@@ -12,7 +12,12 @@
 #ifndef _UCW_UNICODE_H
 #define _UCW_UNICODE_H
 
-#include "ucw/unaligned.h"
+#include <ucw/unaligned.h>
+
+#ifdef CONFIG_UCW_CLEAN_ABI
+#define utf8_strlen ucw_utf8_strlen
+#define utf8_strnlen ucw_utf8_strnlen
+#endif
 
 /* Macros for handling UTF-8 */
 
@@ -168,7 +173,7 @@ get1: UTF8_GET_NEXT;
 }
 
 /**
- * Decode a value from the range `[0, 0xFFFF]` (basic multilignual plane)
+ * Decode a value from the range `[0, 0xFFFF]` (basic multilingual plane)
  * or return `UNI_REPLACEMENT` if the encoding has been corrupted.
  **/
 static inline byte *utf8_get(const byte *p, uns *uu)
@@ -213,7 +218,7 @@ static inline uns utf8_space(uns u)
 }
 
 /**
- * Compute the length of a single UTF-8 character from it's first byte. The encoding must be valid.
+ * Compute the length of a single UTF-8 character from its first byte. The encoding must be valid.
  **/
 static inline uns utf8_encoding_len(uns c)
 {
@@ -253,7 +258,7 @@ static inline void *utf16_le_put(void *p, uns u)
 }
 
 /**
- * Encode an UTF-16BE character from the range `[0, 0xD7FF]` or `[0xE000,0x11FFFF]`;
+ * Encode a UTF-16BE character from the range `[0, 0xD7FF]` or `[0xE000,0x11FFFF]`;
  * up to 4 bytes needed.
  **/
 static inline void *utf16_be_put(void *p, uns u)
@@ -274,7 +279,7 @@ static inline void *utf16_be_put(void *p, uns u)
 }
 
 /**
- * Decode an UTF-16LE character from the range `[0, 0xD7FF]` or `[0xE000,11FFFF]`
+ * Decode a UTF-16LE character from the range `[0, 0xD7FF]` or `[0xE000,11FFFF]`
  * or return @repl if the encoding has been corrupted.
  **/
 static inline void *utf16_le_get_repl(const void *p, uns *uu, uns repl)
@@ -294,7 +299,7 @@ static inline void *utf16_le_get_repl(const void *p, uns *uu, uns repl)
 }
 
 /**
- * Decode an UTF-16BE character from the range `[0, 0xD7FF]` or `[0xE000,11FFFF]`
+ * Decode a UTF-16BE character from the range `[0, 0xD7FF]` or `[0xE000,11FFFF]`
  * or return @repl if the encoding has been corrupted.
  **/
 static inline void *utf16_be_get_repl(const void *p, uns *uu, uns repl)
@@ -314,7 +319,7 @@ static inline void *utf16_be_get_repl(const void *p, uns *uu, uns repl)
 }
 
 /**
- * Decode an UTF-16LE  character from the range `[0, 0xD7FF]` or `[0xE000,11FFFF]`
+ * Decode a UTF-16LE  character from the range `[0, 0xD7FF]` or `[0xE000,11FFFF]`
  * or return `UNI_REPLACEMENT` if the encoding has been corrupted.
  **/
 static inline void *utf16_le_get(const void *p, uns *uu)
@@ -323,7 +328,7 @@ static inline void *utf16_le_get(const void *p, uns *uu)
 }
 
 /**
- * Decode an UTF-16BE  character from the range `[0, 0xD7FF]` or `[0xE000,11FFFF]`
+ * Decode a UTF-16BE  character from the range `[0, 0xD7FF]` or `[0xE000,11FFFF]`
  * or return `UNI_REPLACEMENT` if the encoding has been corrupted.
  **/
 static inline void *utf16_be_get(const void *p, uns *uu)
@@ -332,7 +337,9 @@ static inline void *utf16_be_get(const void *p, uns *uu)
 }
 
 /**
- * Check an Unicode value and if it seems to be useless (defined by Ucwlib; it may change in future) return `UNI_REPLACEMENT` instead.
+ * Basic sanity check on Unicode characters. Return `UNI_REPLACEMENT` if the input
+ * character is a surrogate, ASCII or Latin-1 control character different from the tab,
+ * or if it lies outside the basic plane. In all other cases, it acts as an identity.
  **/
 static inline uns unicode_sanitize_char(uns u)
 {
@@ -347,14 +354,14 @@ static inline uns unicode_sanitize_char(uns u)
 /* unicode-utf8.c */
 
 /**
- * Count the number of Unicode character in a zero-terminated UTF-8 string.
+ * Count the number of Unicode characters in a zero-terminated UTF-8 string.
  * Returned value for corrupted encoding is undefined, but is never greater than strlen().
  **/
-uns utf8_strlen(const byte *str);
+size_t utf8_strlen(const byte *str);
 
 /**
  * Same as @utf8_strlen(), but returns at most @n characters.
  **/
-uns utf8_strnlen(const byte *str, uns n);
+size_t utf8_strnlen(const byte *str, size_t n);
 
 #endif