added "length" parameter to validUtf8String()

At several places, we create temporary copies of non-null-terminated
strings, just to use them in functions like validUtf8String().  We can
save this temporary allocation and avoid heap fragmentation if we
add a length parameter instead of expecting a null-terminated string.

4 files changed, 16 insertions, 10 deletions

diff --git a/src/path.c b/src/path.c
index 6aaff84c..ceb00c5d 100644
--- a/src/path.c
+++ b/src/path.c
@@ -47,7 +47,7 @@ static char *path_conv_charset(char *dest, const char *to,
 char *fs_charset_to_utf8(char *dst, const char *str)
 {
 	char *ret = path_conv_charset(dst, "UTF-8", fsCharset, str);
-	return (ret && !validUtf8String(ret)) ? NULL : ret;
+	return (ret && !validUtf8String(ret, strlen(ret))) ? NULL : ret;
 }
 
 char *utf8_to_fs_charset(char *dst, const char *str)
diff --git a/src/tag.c b/src/tag.c
index a52391b4..1d22b46d 100644
--- a/src/tag.c
+++ b/src/tag.c
@@ -352,7 +352,7 @@ static inline char *fix_utf8(char *str) {
 
 	assert(str != NULL);
 
-	if (validUtf8String(str))
+	if (validUtf8String(str, strlen(str)))
 		return str;
 
 	DEBUG("not valid utf8 in tag: %s\n",str);
diff --git a/src/utf8.c b/src/utf8.c
index e8f3dbdd..1b03f5d2 100644
--- a/src/utf8.c
+++ b/src/utf8.c
@@ -69,10 +69,12 @@ static char utf8_to_latin1_char(const char *inUtf8)
 	return (char)(c + utf8[1]);
 }
 
-static unsigned int validateUtf8Char(const char *inUtf8Char)
+static unsigned int validateUtf8Char(const char *inUtf8Char, size_t length)
 {
 	const unsigned char *utf8Char = (const unsigned char *)inUtf8Char;
 
+	assert(length > 0);
+
 	if (utf8Char[0] < 0x80)
 		return 1;
 
@@ -84,7 +86,7 @@ static unsigned int validateUtf8Char(const char *inUtf8Char)
 			t = (t >> 1);
 			count++;
 		}
-		if (count > 5)
+		if (count > 5 || (size_t)count > length)
 			return 0;
 		for (i = 1; i <= count; i++) {
 			if (utf8Char[i] < 0x80 || utf8Char[i] > 0xBF)
@@ -95,15 +97,17 @@ static unsigned int validateUtf8Char(const char *inUtf8Char)
 		return 0;
 }
 
-int validUtf8String(const char *string)
+int validUtf8String(const char *string, size_t length)
 {
 	unsigned int ret;
 
-	while (*string) {
-		ret = validateUtf8Char(string);
+	while (length > 0) {
+		ret = validateUtf8Char(string, length);
+		assert((size_t)ret <= length);
 		if (0 == ret)
 			return 0;
 		string += ret;
+		length -= ret;
 	}
 
 	return 1;
@@ -118,7 +122,7 @@ char *utf8StrToLatin1Dup(const char *utf8)
 	size_t len = 0;
 
 	while (*utf8) {
-		count = validateUtf8Char(utf8);
+		count = validateUtf8Char(utf8, INT_MAX);
 		if (!count) {
 			free(ret);
 			return NULL;
@@ -140,7 +144,7 @@ char *utf8_to_latin1(char *dest, const char *utf8)
 	size_t len = 0;
 
 	while (*utf8) {
-		count = validateUtf8Char(utf8);
+		count = validateUtf8Char(utf8, INT_MAX);
 		if (count) {
 			*(cp++) = utf8_to_latin1_char(utf8);
 			utf8 += count;
diff --git a/src/utf8.h b/src/utf8.h
index 4a498306..353977be 100644
--- a/src/utf8.h
+++ b/src/utf8.h
@@ -19,11 +19,13 @@
 #ifndef UTF_8_H
 #define UTF_8_H
 
+#include <os_compat.h>
+
 char *latin1StrToUtf8Dup(const char *latin1);
 
 char *utf8StrToLatin1Dup(const char *utf8);
 
-int validUtf8String(const char *string);
+int validUtf8String(const char *string, size_t length);
 
 char *utf8_to_latin1(char *dest, const char *utf8);