2001-09-29 00:14:13 +04:00
|
|
|
/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
|
|
|
|
/* ***** BEGIN LICENSE BLOCK *****
|
2004-04-18 18:21:17 +04:00
|
|
|
* Version: MPL 1.1/GPL 2.0/LGPL 2.1
|
1998-04-14 00:24:54 +04:00
|
|
|
*
|
2004-04-18 18:21:17 +04:00
|
|
|
* The contents of this file are subject to the Mozilla Public License Version
|
|
|
|
* 1.1 (the "License"); you may not use this file except in compliance with
|
|
|
|
* the License. You may obtain a copy of the License at
|
|
|
|
* http://www.mozilla.org/MPL/
|
1998-04-14 00:24:54 +04:00
|
|
|
*
|
2001-09-29 00:14:13 +04:00
|
|
|
* Software distributed under the License is distributed on an "AS IS" basis,
|
|
|
|
* WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
|
|
|
|
* for the specific language governing rights and limitations under the
|
|
|
|
* License.
|
1998-04-14 00:24:54 +04:00
|
|
|
*
|
1999-11-06 06:43:54 +03:00
|
|
|
* The Original Code is mozilla.org code.
|
|
|
|
*
|
2004-04-18 18:21:17 +04:00
|
|
|
* The Initial Developer of the Original Code is
|
2001-09-29 00:14:13 +04:00
|
|
|
* Netscape Communications Corporation.
|
|
|
|
* Portions created by the Initial Developer are Copyright (C) 1998
|
|
|
|
* the Initial Developer. All Rights Reserved.
|
1999-11-06 06:43:54 +03:00
|
|
|
*
|
2001-09-29 00:14:13 +04:00
|
|
|
* Contributor(s):
|
|
|
|
*
|
|
|
|
* Alternatively, the contents of this file may be used under the terms of
|
2004-04-18 18:21:17 +04:00
|
|
|
* either of the GNU General Public License Version 2 or later (the "GPL"),
|
|
|
|
* or the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
|
2001-09-29 00:14:13 +04:00
|
|
|
* in which case the provisions of the GPL or the LGPL are applicable instead
|
|
|
|
* of those above. If you wish to allow use of your version of this file only
|
|
|
|
* under the terms of either the GPL or the LGPL, and not to allow others to
|
2004-04-18 18:21:17 +04:00
|
|
|
* use your version of this file under the terms of the MPL, indicate your
|
2001-09-29 00:14:13 +04:00
|
|
|
* decision by deleting the provisions above and replace them with the notice
|
|
|
|
* and other provisions required by the GPL or the LGPL. If you do not delete
|
|
|
|
* the provisions above, a recipient may use your version of this file under
|
2004-04-18 18:21:17 +04:00
|
|
|
* the terms of any one of the MPL, the GPL or the LGPL.
|
2001-09-29 00:14:13 +04:00
|
|
|
*
|
|
|
|
* ***** END LICENSE BLOCK ***** */
|
1998-08-04 00:46:38 +04:00
|
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
* MODULE NOTES:
|
1998-08-04 14:12:45 +04:00
|
|
|
* @update gess7/30/98
|
1998-08-04 00:46:38 +04:00
|
|
|
*
|
|
|
|
* Much as I hate to do it, we were using string compares wrong.
|
|
|
|
* Often, programmers call functions like strcmp(s1,s2), and pass
|
|
|
|
* one or more null strings. Rather than blow up on these, I've
|
|
|
|
* added quick checks to ensure that cases like this don't cause
|
|
|
|
* us to fail.
|
|
|
|
*
|
|
|
|
* In general, if you pass a null into any of these string compare
|
|
|
|
* routines, we simply return 0.
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
1998-04-14 00:24:54 +04:00
|
|
|
#include "nsCRT.h"
|
1999-02-08 21:57:21 +03:00
|
|
|
#include "nsIServiceManager.h"
|
2005-11-16 05:31:54 +03:00
|
|
|
#include "nsCharTraits.h"
|
2008-02-13 14:34:46 +03:00
|
|
|
#include "prbit.h"
|
2005-11-16 05:31:54 +03:00
|
|
|
|
|
|
|
#define ADD_TO_HASHVAL(hashval, c) \
|
2008-02-13 14:34:46 +03:00
|
|
|
hashval = PR_ROTATE_LEFT32(hashval, 4) ^ (c);
|
1998-04-14 00:24:54 +04:00
|
|
|
|
|
|
|
//----------------------------------------------------------------------
|
|
|
|
|
2000-08-23 21:27:06 +04:00
|
|
|
|
1999-02-06 07:18:15 +03:00
|
|
|
////////////////////////////////////////////////////////////////////////////////
|
|
|
|
// My lovely strtok routine
|
|
|
|
|
|
|
|
#define IS_DELIM(m, c) ((m)[(c) >> 3] & (1 << ((c) & 7)))
|
|
|
|
#define SET_DELIM(m, c) ((m)[(c) >> 3] |= (1 << ((c) & 7)))
|
|
|
|
#define DELIM_TABLE_SIZE 32
|
|
|
|
|
1999-06-05 05:15:00 +04:00
|
|
|
char* nsCRT::strtok(char* string, const char* delims, char* *newStr)
|
1998-04-14 00:24:54 +04:00
|
|
|
{
|
1999-06-05 05:15:00 +04:00
|
|
|
NS_ASSERTION(string, "Unlike regular strtok, the first argument cannot be null.");
|
1999-02-06 07:18:15 +03:00
|
|
|
|
|
|
|
char delimTable[DELIM_TABLE_SIZE];
|
|
|
|
PRUint32 i;
|
|
|
|
char* result;
|
1999-06-05 05:15:00 +04:00
|
|
|
char* str = string;
|
1999-02-06 07:18:15 +03:00
|
|
|
|
|
|
|
for (i = 0; i < DELIM_TABLE_SIZE; i++)
|
|
|
|
delimTable[i] = '\0';
|
|
|
|
|
2001-05-21 14:45:44 +04:00
|
|
|
for (i = 0; delims[i]; i++) {
|
2007-07-08 11:08:04 +04:00
|
|
|
SET_DELIM(delimTable, static_cast<PRUint8>(delims[i]));
|
1999-02-06 07:22:43 +03:00
|
|
|
}
|
|
|
|
NS_ASSERTION(delims[i] == '\0', "too many delimiters");
|
1999-02-06 07:18:15 +03:00
|
|
|
|
|
|
|
// skip to beginning
|
2007-07-08 11:08:04 +04:00
|
|
|
while (*str && IS_DELIM(delimTable, static_cast<PRUint8>(*str))) {
|
1999-02-06 07:18:15 +03:00
|
|
|
str++;
|
|
|
|
}
|
|
|
|
result = str;
|
|
|
|
|
|
|
|
// fix up the end of the token
|
|
|
|
while (*str) {
|
2007-07-08 11:08:04 +04:00
|
|
|
if (IS_DELIM(delimTable, static_cast<PRUint8>(*str))) {
|
1999-02-06 07:18:15 +03:00
|
|
|
*str++ = '\0';
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
str++;
|
|
|
|
}
|
|
|
|
*newStr = str;
|
|
|
|
|
|
|
|
return str == result ? NULL : result;
|
1999-02-06 06:56:17 +03:00
|
|
|
}
|
|
|
|
|
1999-02-06 07:18:15 +03:00
|
|
|
////////////////////////////////////////////////////////////////////////////////
|
|
|
|
|
1998-08-04 00:46:38 +04:00
|
|
|
/**
|
|
|
|
* Compare unichar string ptrs, stopping at the 1st null
|
|
|
|
* NOTE: If both are null, we return 0.
|
1999-11-14 09:22:52 +03:00
|
|
|
* NOTE: We terminate the search upon encountering a NULL
|
|
|
|
*
|
|
|
|
* @update gess 11/10/99
|
1998-08-04 00:46:38 +04:00
|
|
|
* @param s1 and s2 both point to unichar strings
|
1998-08-04 01:03:49 +04:00
|
|
|
* @return 0 if they match, -1 if s1<s2; 1 if s1>s2
|
1998-08-04 00:46:38 +04:00
|
|
|
*/
|
1999-11-14 09:22:52 +03:00
|
|
|
PRInt32 nsCRT::strcmp(const PRUnichar* s1, const PRUnichar* s2) {
|
1998-08-04 00:46:38 +04:00
|
|
|
if(s1 && s2) {
|
|
|
|
for (;;) {
|
|
|
|
PRUnichar c1 = *s1++;
|
|
|
|
PRUnichar c2 = *s2++;
|
|
|
|
if (c1 != c2) {
|
|
|
|
if (c1 < c2) return -1;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
if ((0==c1) || (0==c2)) break;
|
1998-04-14 00:24:54 +04:00
|
|
|
}
|
|
|
|
}
|
1999-12-08 01:10:31 +03:00
|
|
|
else {
|
|
|
|
if (s1) // s2 must have been null
|
|
|
|
return -1;
|
|
|
|
if (s2) // s1 must have been null
|
|
|
|
return 1;
|
|
|
|
}
|
1998-04-14 00:24:54 +04:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
1998-08-04 00:46:38 +04:00
|
|
|
/**
|
|
|
|
* Compare unichar string ptrs, stopping at the 1st null or nth char.
|
|
|
|
* NOTE: If either is null, we return 0.
|
1999-11-14 09:22:52 +03:00
|
|
|
* NOTE: We DO NOT terminate the search upon encountering NULL's before N
|
|
|
|
*
|
|
|
|
* @update gess 11/10/99
|
1998-08-04 00:46:38 +04:00
|
|
|
* @param s1 and s2 both point to unichar strings
|
1998-08-04 01:03:49 +04:00
|
|
|
* @return 0 if they match, -1 if s1<s2; 1 if s1>s2
|
1998-08-04 00:46:38 +04:00
|
|
|
*/
|
1999-11-14 09:22:52 +03:00
|
|
|
PRInt32 nsCRT::strncmp(const PRUnichar* s1, const PRUnichar* s2, PRUint32 n) {
|
1998-08-04 00:46:38 +04:00
|
|
|
if(s1 && s2) {
|
1999-02-09 01:01:08 +03:00
|
|
|
if(n != 0) {
|
|
|
|
do {
|
1998-08-04 00:46:38 +04:00
|
|
|
PRUnichar c1 = *s1++;
|
|
|
|
PRUnichar c2 = *s2++;
|
|
|
|
if (c1 != c2) {
|
|
|
|
if (c1 < c2) return -1;
|
|
|
|
return 1;
|
|
|
|
}
|
1999-02-09 01:01:08 +03:00
|
|
|
} while (--n != 0);
|
1998-04-14 00:24:54 +04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
1998-12-16 08:40:20 +03:00
|
|
|
PRUnichar* nsCRT::strdup(const PRUnichar* str)
|
|
|
|
{
|
2000-08-10 10:19:37 +04:00
|
|
|
PRUint32 len = nsCRT::strlen(str);
|
|
|
|
return strndup(str, len);
|
|
|
|
}
|
1999-07-22 22:23:02 +04:00
|
|
|
|
2000-08-10 10:19:37 +04:00
|
|
|
PRUnichar* nsCRT::strndup(const PRUnichar* str, PRUint32 len)
|
|
|
|
{
|
1999-07-22 22:23:02 +04:00
|
|
|
nsCppSharedAllocator<PRUnichar> shared_allocator;
|
2001-01-08 22:58:12 +03:00
|
|
|
PRUnichar* rslt = shared_allocator.allocate(len + 1); // add one for the null
|
|
|
|
// PRUnichar* rslt = new PRUnichar[len + 1];
|
1999-07-22 22:23:02 +04:00
|
|
|
|
1998-12-16 08:40:20 +03:00
|
|
|
if (rslt == NULL) return NULL;
|
2002-01-12 06:18:55 +03:00
|
|
|
memcpy(rslt, str, len * sizeof(PRUnichar));
|
2001-01-08 22:58:12 +03:00
|
|
|
rslt[len] = 0;
|
1998-12-16 08:40:20 +03:00
|
|
|
return rslt;
|
|
|
|
}
|
|
|
|
|
2001-03-11 05:15:36 +03:00
|
|
|
/**
|
|
|
|
* |nsCRT::HashCode| is identical to |PL_HashString|, which tests
|
|
|
|
* (http://bugzilla.mozilla.org/showattachment.cgi?attach_id=26596)
|
|
|
|
* show to be the best hash among several other choices.
|
|
|
|
*
|
|
|
|
* We re-implement it here rather than calling it for two reasons:
|
|
|
|
* (1) in this interface, we also calculate the length of the
|
|
|
|
* string being hashed; and (2) the narrow and wide and `buffer' versions here
|
|
|
|
* will hash equivalent strings to the same value, e.g., "Hello" and L"Hello".
|
|
|
|
*/
|
2000-08-21 01:29:10 +04:00
|
|
|
PRUint32 nsCRT::HashCode(const char* str, PRUint32* resultingStrLen)
|
1999-06-26 10:32:35 +04:00
|
|
|
{
|
2001-03-11 05:15:36 +03:00
|
|
|
PRUint32 h = 0;
|
|
|
|
const char* s = str;
|
|
|
|
|
2002-01-14 23:10:48 +03:00
|
|
|
if (!str) return h;
|
|
|
|
|
2001-03-11 05:15:36 +03:00
|
|
|
unsigned char c;
|
|
|
|
while ( (c = *s++) )
|
2005-11-16 05:31:54 +03:00
|
|
|
ADD_TO_HASHVAL(h, c);
|
2001-03-11 05:15:36 +03:00
|
|
|
|
|
|
|
if ( resultingStrLen )
|
|
|
|
*resultingStrLen = (s-str)-1;
|
|
|
|
return h;
|
1999-06-26 10:32:35 +04:00
|
|
|
}
|
|
|
|
|
2007-07-12 00:46:43 +04:00
|
|
|
PRUint32 nsCRT::HashCode(const char* start, PRUint32 length)
|
|
|
|
{
|
|
|
|
PRUint32 h = 0;
|
|
|
|
const char* s = start;
|
|
|
|
const char* end = start + length;
|
|
|
|
|
|
|
|
unsigned char c;
|
|
|
|
while ( s < end ) {
|
|
|
|
c = *s++;
|
|
|
|
ADD_TO_HASHVAL(h, c);
|
|
|
|
}
|
|
|
|
|
|
|
|
return h;
|
|
|
|
}
|
|
|
|
|
2000-08-21 01:29:10 +04:00
|
|
|
PRUint32 nsCRT::HashCode(const PRUnichar* str, PRUint32* resultingStrLen)
|
1998-04-14 00:24:54 +04:00
|
|
|
{
|
2001-03-11 05:15:36 +03:00
|
|
|
PRUint32 h = 0;
|
|
|
|
const PRUnichar* s = str;
|
|
|
|
|
2002-01-14 23:10:48 +03:00
|
|
|
if (!str) return h;
|
|
|
|
|
2002-02-16 22:50:32 +03:00
|
|
|
PRUnichar c;
|
|
|
|
while ( (c = *s++) )
|
2005-11-16 05:31:54 +03:00
|
|
|
ADD_TO_HASHVAL(h, c);
|
2002-02-16 22:50:32 +03:00
|
|
|
|
|
|
|
if ( resultingStrLen )
|
|
|
|
*resultingStrLen = (s-str)-1;
|
|
|
|
return h;
|
|
|
|
}
|
|
|
|
|
2007-07-12 00:46:43 +04:00
|
|
|
PRUint32 nsCRT::HashCodeAsUTF8(const PRUnichar* start, PRUint32 length)
|
2005-11-04 22:52:18 +03:00
|
|
|
{
|
|
|
|
PRUint32 h = 0;
|
2007-07-12 00:46:43 +04:00
|
|
|
const PRUnichar* s = start;
|
|
|
|
const PRUnichar* end = start + length;
|
|
|
|
|
|
|
|
PRUint16 W1 = 0; // the first UTF-16 word in a two word tuple
|
|
|
|
PRUint32 U = 0; // the current char as UCS-4
|
|
|
|
int code_length = 0; // the number of bytes in the UTF-8 sequence for the current char
|
|
|
|
|
|
|
|
PRUint16 W;
|
|
|
|
while ( s < end )
|
|
|
|
{
|
|
|
|
W = *s++;
|
|
|
|
/*
|
|
|
|
* On the fly, decoding from UTF-16 (and/or UCS-2) into UTF-8 as per
|
|
|
|
* http://www.ietf.org/rfc/rfc2781.txt
|
|
|
|
* http://www.ietf.org/rfc/rfc3629.txt
|
|
|
|
*/
|
|
|
|
|
|
|
|
if ( !W1 )
|
|
|
|
{
|
|
|
|
if ( !IS_SURROGATE(W) )
|
|
|
|
{
|
|
|
|
U = W;
|
|
|
|
if ( W <= 0x007F )
|
|
|
|
code_length = 1;
|
|
|
|
else if ( W <= 0x07FF )
|
|
|
|
code_length = 2;
|
|
|
|
else
|
|
|
|
code_length = 3;
|
|
|
|
}
|
|
|
|
else if ( NS_IS_HIGH_SURROGATE(W) && s < end)
|
|
|
|
{
|
2007-03-23 02:01:14 +03:00
|
|
|
W1 = W;
|
2007-07-12 00:46:43 +04:00
|
|
|
|
2007-07-12 00:46:43 +04:00
|
|
|
continue;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
// Treat broken characters as the Unicode replacement
|
|
|
|
// character 0xFFFD
|
|
|
|
U = 0xFFFD;
|
|
|
|
|
|
|
|
code_length = 3;
|
|
|
|
|
|
|
|
NS_WARNING("Got low surrogate but no previous high surrogate");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
// as required by the standard, this code is careful to
|
|
|
|
// throw out illegal sequences
|
|
|
|
|
|
|
|
if ( NS_IS_LOW_SURROGATE(W) )
|
|
|
|
{
|
|
|
|
U = SURROGATE_TO_UCS4(W1, W);
|
|
|
|
NS_ASSERTION(IS_VALID_CHAR(U), "How did this happen?");
|
|
|
|
code_length = 4;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
// Treat broken characters as the Unicode replacement
|
|
|
|
// character 0xFFFD
|
|
|
|
U = 0xFFFD;
|
|
|
|
|
|
|
|
code_length = 3;
|
|
|
|
|
|
|
|
NS_WARNING("High surrogate not followed by low surrogate");
|
|
|
|
}
|
|
|
|
|
|
|
|
W1 = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static const PRUint16 sBytePrefix[5] = { 0x0000, 0x0000, 0x00C0, 0x00E0, 0x00F0 };
|
|
|
|
static const PRUint16 sShift[5] = { 0, 0, 6, 12, 18 };
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Unlike the algorithm in
|
|
|
|
* http://www.ietf.org/rfc/rfc3629.txt we must calculate the
|
|
|
|
* bytes in left to right order so that our hash result
|
|
|
|
* matches what the narrow version would calculate on an
|
|
|
|
* already UTF-8 string.
|
|
|
|
*/
|
|
|
|
|
|
|
|
// hash the first (and often, only, byte)
|
|
|
|
ADD_TO_HASHVAL(h, (sBytePrefix[code_length] | (U>>sShift[code_length])));
|
|
|
|
|
|
|
|
// an unrolled loop for hashing any remaining bytes in this
|
|
|
|
// sequence
|
|
|
|
switch ( code_length )
|
|
|
|
{ // falling through in each case
|
|
|
|
case 4: ADD_TO_HASHVAL(h, (0x80 | ((U>>12) & 0x003F)));
|
|
|
|
case 3: ADD_TO_HASHVAL(h, (0x80 | ((U>>6 ) & 0x003F)));
|
|
|
|
case 2: ADD_TO_HASHVAL(h, (0x80 | ( U & 0x003F)));
|
|
|
|
default: code_length = 0;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2005-11-04 22:52:18 +03:00
|
|
|
return h;
|
|
|
|
}
|
|
|
|
|
2002-09-26 11:41:53 +04:00
|
|
|
PRUint32 nsCRT::BufferHashCode(const PRUnichar* s, PRUint32 len)
|
|
|
|
{
|
|
|
|
PRUint32 h = 0;
|
|
|
|
const PRUnichar* done = s + len;
|
|
|
|
|
|
|
|
while ( s < done )
|
2008-02-13 14:34:46 +03:00
|
|
|
h = PR_ROTATE_LEFT32(h, 4) ^ PRUint16(*s++); // cast to unsigned to prevent possible sign extension
|
2002-09-26 11:41:53 +04:00
|
|
|
return h;
|
|
|
|
}
|
|
|
|
|
2001-10-18 08:23:25 +04:00
|
|
|
// This should use NSPR but NSPR isn't exporting its PR_strtoll function
|
|
|
|
// Until then...
|
|
|
|
PRInt64 nsCRT::atoll(const char *str)
|
|
|
|
{
|
|
|
|
if (!str)
|
|
|
|
return LL_Zero();
|
|
|
|
|
|
|
|
PRInt64 ll = LL_Zero(), digitll = LL_Zero();
|
|
|
|
|
|
|
|
while (*str && *str >= '0' && *str <= '9') {
|
|
|
|
LL_MUL(ll, ll, 10);
|
|
|
|
LL_UI2L(digitll, (*str - '0'));
|
|
|
|
LL_ADD(ll, ll, digitll);
|
|
|
|
str++;
|
|
|
|
}
|
|
|
|
|
|
|
|
return ll;
|
|
|
|
}
|
|
|
|
|