amxmodx/dlls/regex/utils.cpp
2014-08-04 13:18:37 -05:00

254 lines
6.0 KiB
C++

// vim: set ts=4 sw=4 tw=99 noet:
//
// AMX Mod X, based on AMX Mod by Aleksander Naszko ("OLO").
// Copyright (C) The AMX Mod X Development Team.
//
// This software is licensed under the GNU General Public License, version 3 or higher.
// Additional exceptions apply. For full license details, see LICENSE.txt or visit:
// https://alliedmods.net/amxmodx-license
//
// Regular Expressions Module
//
#include "amxxmodule.h"
#include <string.h>
#include "utils.h"
#if defined(WIN32)
#define strcasecmp stricmp
#define strncasecmp _strnicmp
#endif
int UTIL_CheckValidChar(char *c)
{
int count;
int bytecount = 0;
for (count = 1; (*c & 0xC0) == 0x80; count++)
{
c--;
}
switch (*c & 0xF0)
{
case 0xC0:
case 0xD0:
{
bytecount = 2;
break;
}
case 0xE0:
{
bytecount = 3;
break;
}
case 0xF0:
{
bytecount = 4;
break;
}
}
if (bytecount != count)
{
return count;
}
return 0;
}
unsigned int strncopy(char *dest, const char *src, size_t count)
{
if (!count)
{
return 0;
}
char *start = dest;
while ((*src) && (--count))
{
*dest++ = *src++;
}
*dest = '\0';
return (dest - start);
}
/**
* NOTE: Do not edit this for the love of god unless you have
* read the test cases and understand the code behind each one.
* While I don't guarantee there aren't mistakes, I do guarantee
* that plugins will end up relying on tiny idiosyncrasies of this
* function, just like they did with AMX Mod X.
*
* There are explicitly more cases than the AMX Mod X version because
* we're not doing a blind copy. Each case is specifically optimized
* for what needs to be done. Even better, we don't have to error on
* bad buffer sizes. Instead, this function will smartly cut off the
* string in a way that pushes old data out.
*/
char *UTIL_ReplaceEx(char *subject, size_t maxLen, const char *search, size_t searchLen, const char *replace, size_t replaceLen, bool caseSensitive)
{
char *ptr = subject;
size_t browsed = 0;
size_t textLen = strlen(subject);
/* It's not possible to search or replace */
if (searchLen > textLen)
{
return NULL;
}
/* Handle the case of one byte replacement.
* It's only valid in one case.
*/
if (maxLen == 1)
{
/* If the search matches and the replace length is 0,
* we can just terminate the string and be done.
*/
if ((caseSensitive ? strcmp(subject, search) : strcasecmp(subject, search)) == 0 && replaceLen == 0)
{
*subject = '\0';
return subject;
}
else
{
return NULL;
}
}
/* Subtract one off the maxlength so we can include the null terminator */
maxLen--;
while (*ptr != '\0' && (browsed <= textLen - searchLen))
{
/* See if we get a comparison */
if ((caseSensitive ? strncmp(ptr, search, searchLen) : strncasecmp(ptr, search, searchLen)) == 0)
{
if (replaceLen > searchLen)
{
/* First, see if we have enough space to do this operation */
if (maxLen - textLen < replaceLen - searchLen)
{
/* First, see if the replacement length goes out of bounds. */
if (browsed + replaceLen >= maxLen)
{
/* EXAMPLE CASE:
* Subject: AABBBCCC
* Buffer : 12 bytes
* Search : BBB
* Replace: DDDDDDDDDD
* OUTPUT : AADDDDDDDDD
* POSITION: ^
*/
/* If it does, we'll just bound the length and do a strcpy. */
replaceLen = maxLen - browsed;
/* Note, we add one to the final result for the null terminator */
strncopy(ptr, replace, replaceLen + 1);
/* Don't truncate a multi-byte character */
if (*(ptr + replaceLen - 1) & 1 << 7)
{
replaceLen -= UTIL_CheckValidChar(ptr + replaceLen - 1);
*(ptr + replaceLen) = '\0';
}
}
else
{
/* EXAMPLE CASE:
* Subject: AABBBCCC
* Buffer : 12 bytes
* Search : BBB
* Replace: DDDDDDD
* OUTPUT : AADDDDDDDCC
* POSITION: ^
*/
/* We're going to have some bytes left over... */
size_t origBytesToCopy = (textLen - (browsed + searchLen)) + 1;
size_t realBytesToCopy = (maxLen - (browsed + replaceLen)) + 1;
char *moveFrom = ptr + searchLen + (origBytesToCopy - realBytesToCopy);
char *moveTo = ptr + replaceLen;
/* First, move our old data out of the way. */
memmove(moveTo, moveFrom, realBytesToCopy);
/* Now, do our replacement. */
memcpy(ptr, replace, replaceLen);
}
}
else
{
/* EXAMPLE CASE:
* Subject: AABBBCCC
* Buffer : 12 bytes
* Search : BBB
* Replace: DDDD
* OUTPUT : AADDDDCCC
* POSITION: ^
*/
/* Yes, we have enough space. Do a normal move operation. */
char *moveFrom = ptr + searchLen;
char *moveTo = ptr + replaceLen;
/* First move our old data out of the way. */
size_t bytesToCopy = (textLen - (browsed + searchLen)) + 1;
memmove(moveTo, moveFrom, bytesToCopy);
/* Now do our replacement. */
memcpy(ptr, replace, replaceLen);
}
}
else if (replaceLen < searchLen)
{
/* EXAMPLE CASE:
* Subject: AABBBCCC
* Buffer : 12 bytes
* Search : BBB
* Replace: D
* OUTPUT : AADCCC
* POSITION: ^
*/
/* If the replacement does not grow the string length, we do not
* need to do any fancy checking at all. Yay!
*/
char *moveFrom = ptr + searchLen; /* Start after the search pointer */
char *moveTo = ptr + replaceLen; /* Copy to where the replacement ends */
/* Copy our replacement in, if any */
if (replaceLen)
{
memcpy(ptr, replace, replaceLen);
}
/* Figure out how many bytes to move down, including null terminator */
size_t bytesToCopy = (textLen - (browsed + searchLen)) + 1;
/* Move the rest of the string down */
memmove(moveTo, moveFrom, bytesToCopy);
}
else
{
/* EXAMPLE CASE:
* Subject: AABBBCCC
* Buffer : 12 bytes
* Search : BBB
* Replace: DDD
* OUTPUT : AADDDCCC
* POSITION: ^
*/
/* We don't have to move anything around, just do a straight copy */
memcpy(ptr, replace, replaceLen);
}
return ptr + replaceLen;
}
ptr++;
browsed++;
}
return NULL;
}