* Fixed previous version of UTF8CountBytes() and put it in again: it was

breaking out of the loop too early, and would therefore miss the last
  character.
* Replaced UTF8CountChars() with a short and safe version as well.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24173 a95241bf-73f2-0310-859d-f6bbb57e9c96
This commit is contained in:
Axel Dörfler 2008-02-28 19:17:33 +00:00
parent ccf6ad1d8b
commit d273d00822

View File

@ -1,5 +1,5 @@
/*
* Copyright 2004-2006, Haiku, Inc.
* Copyright 2004-2008, Haiku, Inc.
* Distributed under the terms of the MIT License.
*/
#ifndef _UTF8_FUNCTIONS_H
@ -64,48 +64,22 @@ UTF8PreviousCharLen(const char *text, const char *limit)
static inline uint32
UTF8CountBytes(const char *bytes, int32 numChars)
{
if (!bytes)
if (bytes == NULL)
return 0;
if (numChars < 0)
numChars = INT_MAX;
const char *base = bytes;
while (*bytes && numChars-- > 0) {
if (bytes[0] & 0x80) {
if (bytes[0] & 0x40) {
if (bytes[0] & 0x20) {
if (bytes[0] & 0x10) {
if (bytes[1] == 0 || bytes[2] == 0 || bytes[3] == 0)
return (bytes - base);
bytes += 4;
continue;
}
if (bytes[1] == 0 || bytes[2] == 0)
return (bytes - base);
bytes += 3;
continue;
}
if (bytes[1] == 0)
return (bytes - base);
bytes += 2;
continue;
}
/* Not a startbyte - skip */
bytes += 1;
continue;
while (bytes[0] != '\0') {
if ((bytes[0] & 0xc0) != 0x80) {
if (--numChars < 0)
break;
}
bytes += 1;
bytes++;
}
return (bytes - base);
return bytes - base;
}
@ -116,50 +90,19 @@ UTF8CountBytes(const char *bytes, int32 numChars)
static inline uint32
UTF8CountChars(const char *bytes, int32 numBytes)
{
if (!bytes)
if (bytes == NULL)
return 0;
uint32 length = 0;
const char *last = bytes + numBytes - 1;
const char *last;
if (numBytes < 0)
last = (const char *)UINT_MAX;
else
last = bytes + numBytes - 1;
while (*bytes && bytes <= last) {
if (bytes[0] & 0x80) {
if (bytes[0] & 0x40) {
if (bytes[0] & 0x20) {
if (bytes[0] & 0x10) {
if (bytes[1] == 0 || bytes[2] == 0 || bytes[3] == 0)
return length;
bytes += 4;
length++;
continue;
}
if (bytes[1] == 0 || bytes[2] == 0)
return length;
bytes += 3;
length++;
continue;
}
if (bytes[1] == 0)
return length;
bytes += 2;
length++;
continue;
}
/* Not a startbyte - skip */
bytes += 1;
continue;
}
bytes += 1;
length++;
while (bytes[0] && bytes <= last) {
if ((bytes++[0] & 0xc0) != 0x80)
length++;
}
return length;