(svn r9012) -Fix/Feature (UTF8): When cutting strings into multiple lines also take into consideration whitespace characters of more than 1 byte length (eg IDEOGRAPHIC SPACE, IsWhitespace() function). When trimming such strings, account for multiple-byte long sequences so use *Utf8PrevChar(v) = '\0'.

-Codechange: Add a function Utf8TrimString() that properly trims a string to an UTF8 encoding seperation instead of somewhere in the wild (and use it in the chat area)

1 files changed, 33 insertions, 0 deletions

diff --git a/src/string.h b/src/string.h
index 8b6d6792f..eba7eff63 100644
--- a/src/string.h
+++ b/src/string.h
@@ -74,6 +74,7 @@ bool IsValidChar(WChar key, CharSetFilter afilter);
 
 size_t Utf8Decode(WChar *c, const char *s);
 size_t Utf8Encode(char *buf, WChar c);
+size_t Utf8TrimString(char *s, size_t maxlen);
 
 
 static inline WChar Utf8Consume(const char **s)
@@ -100,6 +101,23 @@ static inline size_t Utf8CharLen(WChar c)
 }
 
 
+/**
+ * Return the length of an UTF-8 encoded value based on a single char. This
+ * char should be the first byte of the UTF-8 encoding. If not, or encoding
+ * is invalid, return value is 0
+ */
+static inline size_t Utf8EncodedCharLen(char c)
+{
+	if (GB(c, 3, 5) == 0x1E) return 4;
+	if (GB(c, 4, 4) == 0x0E) return 3;
+	if (GB(c, 5, 3) == 0x06) return 2;
+	if (GB(c, 7, 1) == 0x00) return 1;
+
+	/* Invalid UTF8 start encoding */
+	return 0;
+}
+
+
 /* Check if the given character is part of a UTF8 sequence */
 static inline bool IsUtf8Part(char c)
 {
@@ -129,5 +147,20 @@ static inline bool IsPrintable(WChar c)
 	return true;
 }
 
+/**
+ * Check whether UNICODE character is whitespace or not
+ * @param c UNICODE character to check
+ * @return a boolean value whether 'c' is a whitespace character or not
+ * @see http://www.fileformat.info/info/unicode/category/Zs/list.htm
+ */
+static inline bool IsWhitespace(WChar c)
+{
+	return
+	  c == 0x0020 /* SPACE */ ||
+	  c == 0x00A0 /* NO-BREAK SPACE */ ||
+	  c == 0x3000 /* IDEOGRAPHIC SPACE */
+	;
+}
+
 
 #endif /* STRING_H */