CFStringScanner.c

   1 /*
   2  * Copyright (c) 2010 Apple Inc. All rights reserved.
   3  *
   4  * @APPLE_LICENSE_HEADER_START@
   5  *
   6  * This file contains Original Code and/or Modifications of Original Code
   7  * as defined in and that are subject to the Apple Public Source License
   8  * Version 2.0 (the 'License'). You may not use this file except in
   9  * compliance with the License. Please obtain a copy of the License at
  10  * http://www.opensource.apple.com/apsl/ and read it before using this
  11  * file.
  12  *
  13  * The Original Code and all software distributed under the License are
  14  * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
  15  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  16  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
  17  * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
  18  * Please see the License for the specific language governing rights and
  19  * limitations under the License.
  20  *
  21  * @APPLE_LICENSE_HEADER_END@
  22  */
  23
  24 /*      CFStringScanner.c
  25         Copyright (c) 1999-2009, Apple Inc. All rights reserved.
  26         Responsibility: Ali Ozer
  27 */
  28
  29 #include "CFInternal.h"
  30 #include <CoreFoundation/CFString.h>
  31 #include <sys/types.h>
  32 #include <limits.h>
  33 #include <stdlib.h>
  34 #include <string.h>
  35
  36 CF_INLINE Boolean __CFCharacterIsADigit(UniChar ch) {
  37     return (ch >= '0' && ch <= '9') ? true : false;
  38 }
  39
  40 /* Returns -1 on illegal value */
  41 CF_INLINE SInt32 __CFCharacterNumericOrHexValue (UniChar ch) {
  42     if (ch >= '0' && ch <= '9') {
  43         return ch - '0';
  44     } else if (ch >= 'A' && ch <= 'F') {
  45         return ch + 10 - 'A';
  46     } else if (ch >= 'a' && ch <= 'f') {
  47         return ch + 10 - 'a';
  48     } else {
  49         return -1;
  50     }
  51 }
  52
  53 /* Returns -1 on illegal value */
  54 CF_INLINE SInt32 __CFCharacterNumericValue(UniChar ch) {
  55     return (ch >= '0' && ch <= '9') ? (ch - '0') : -1;
  56 }
  57
  58 CF_INLINE UniChar __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(CFStringInlineBuffer *buf, SInt32 *indexPtr) {
  59     UniChar ch;
  60     while (__CFIsWhitespace(ch = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr))) (*indexPtr)++;
  61     return ch;
  62 }
  63
  64 /* result is int64_t or int, depending on doLonglong
  65 */
  66 __private_extern__ Boolean __CFStringScanInteger(CFStringInlineBuffer *buf, CFTypeRef locale, SInt32 *indexPtr, Boolean doLonglong, void *result) {
  67     Boolean doingLonglong = false;      /* Set to true if doLonglong, and we overflow an int... */
  68     Boolean neg = false;
  69     int intResult = 0;
  70     register int64_t longlongResult = 0;        /* ??? int64_t is slow when not in regs; I hope this does the right thing. */
  71     UniChar ch;
  72
  73     ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
  74
  75     if (ch == '-' || ch == '+') {
  76         neg = (ch == '-');
  77         (*indexPtr)++;
  78         ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
  79     }
  80
  81     if (! __CFCharacterIsADigit(ch)) return false;      /* No digits, bail out... */
  82     do {
  83         if (doingLonglong) {
  84             if ((longlongResult >= LLONG_MAX / 10) && ((longlongResult > LLONG_MAX / 10) || (__CFCharacterNumericValue(ch) - (neg ? 1 : 0) >= LLONG_MAX - longlongResult * 10))) {
  85                 /* ??? This might not handle LLONG_MIN correctly... */
  86                 longlongResult = neg ? LLONG_MIN : LLONG_MAX;
  87                 neg = false;
  88                 while (__CFCharacterIsADigit(ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr))));      /* Skip remaining digits */
  89             } else {
  90                 longlongResult = longlongResult * 10 + __CFCharacterNumericValue(ch);
  91                 ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
  92             }
  93         } else {
  94             if ((intResult >= INT_MAX / 10) && ((intResult > INT_MAX / 10) || (__CFCharacterNumericValue(ch) - (neg ? 1 : 0) >= INT_MAX - intResult * 10))) {
  95                 // Overflow, check for int64_t...
  96                 if (doLonglong) {
  97                     longlongResult = intResult;
  98                     doingLonglong = true;
  99                 } else {
 100                     /* ??? This might not handle INT_MIN correctly... */
 101                     intResult = neg ? INT_MIN : INT_MAX;
 102                     neg = false;
 103                     while (__CFCharacterIsADigit(ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr))));  /* Skip remaining digits */
 104                 }
 105             } else {
 106                 intResult = intResult * 10 + __CFCharacterNumericValue(ch);
 107                 ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
 108             }
 109         }
 110     } while (__CFCharacterIsADigit(ch));
 111
 112     if (result) {
 113         if (doLonglong) {
 114             if (!doingLonglong) longlongResult = intResult;
 115             *(int64_t *)result = neg ? -longlongResult : longlongResult;
 116         } else {
 117             *(int *)result = neg ? -intResult : intResult;
 118         }
 119     }
 120
 121     return true;
 122 }
 123
 124 __private_extern__ Boolean __CFStringScanHex(CFStringInlineBuffer *buf, SInt32 *indexPtr, unsigned *result) {
 125     UInt32 value = 0;
 126     SInt32 curDigit;
 127     UniChar ch;
 128
 129     ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
 130     /* Ignore the optional "0x" or "0X"; if it's followed by a non-hex, just parse the "0" and leave pointer at "x" */
 131     if (ch == '0') {
 132         ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
 133         if (ch == 'x' || ch == 'X') ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
 134         curDigit = __CFCharacterNumericOrHexValue(ch);
 135         if (curDigit == -1) {
 136             (*indexPtr)--;      /* Go back over the "x" or "X" */
 137             if (result) *result = 0;
 138             return true;        /* We just saw "0" */
 139         }
 140     } else {
 141         curDigit = __CFCharacterNumericOrHexValue(ch);
 142         if (curDigit == -1) return false;
 143     }
 144
 145     do {
 146         if (value > (UINT_MAX >> 4)) {
 147             value = UINT_MAX;   /* We do this over and over again, but it's an error case anyway */
 148         } else {
 149             value = (value << 4) + curDigit;
 150         }
 151         curDigit = __CFCharacterNumericOrHexValue(__CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr)));
 152     } while (curDigit != -1);
 153
 154     if (result) *result = value;
 155     return true;
 156 }
 157
 158 // Packed array of Boolean
 159 static const unsigned char __CFNumberSet[16] = {
 160     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  nul soh stx etx eot enq ack bel
 161     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  bs  ht  nl  vt  np  cr  so  si
 162     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  dle dc1 dc2 dc3 dc4 nak syn etb
 163     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  can em  sub esc fs  gs  rs  us
 164     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  sp   !   "   #   $   %   &   '
 165     0X68, // 0, 0, 0, 1, 0, 1, 1, 0, //  (   )   *   +   ,   -   .   /
 166     0xFF, // 1, 1, 1, 1, 1, 1, 1, 1, //  0   1   2   3   4   5   6   7
 167     0X03, // 1, 1, 0, 0, 0, 0, 0, 0, //  8   9   :   ;   <   =   >   ?
 168     0X20, // 0, 0, 0, 0, 0, 1, 0, 0, //  @   A   B   C   D   E   F   G
 169     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  H   I   J   K   L   M   N   O
 170     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  P   Q   R   S   T   U   V   W
 171     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  X   Y   Z   [   \   ]   ^   _
 172     0X20, // 0, 0, 0, 0, 0, 1, 0, 0, //  `   a   b   c   d   e   f   g
 173     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  h   i   j   k   l   m   n   o
 174     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  p   q   r   s   t   u   v   w
 175     0X00, // 0, 0, 0, 0, 0, 0, 0, 0  //  x   y   z   {   |   }   ~  del
 176 };
 177
 178 __private_extern__ Boolean __CFStringScanDouble(CFStringInlineBuffer *buf, CFTypeRef locale, SInt32 *indexPtr, double *resultPtr) {
 179     #define STACK_BUFFER_SIZE 256
 180     #define ALLOC_CHUNK_SIZE 256 // first and subsequent malloc size.  Should be greater than STACK_BUFFER_SIZE
 181     char localCharBuffer[STACK_BUFFER_SIZE];
 182     char *charPtr = localCharBuffer;
 183     char *endCharPtr;
 184     SInt32 numChars = 0;
 185     SInt32 capacity = STACK_BUFFER_SIZE;        // in chars
 186     double result;
 187     UniChar ch;
 188     CFAllocatorRef tmpAlloc = NULL;
 189
 190     ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
 191     // At this point indexPtr points at the first non-space char
 192 #if 0
 193 #warning need to allow, case insensitively, all of: "nan", "inf", "-inf", "+inf", "-infinity", "+infinity", "infinity";
 194 #warning -- strtod() will actually do most or all of that for us
 195 #define BITSFORDOUBLENAN        ((uint64_t)0x7ff8000000000000ULL)
 196 #define BITSFORDOUBLEPOSINF     ((uint64_t)0x7ff0000000000000ULL)
 197 #define BITSFORDOUBLENEGINF     ((uint64_t)0xfff0000000000000ULL)
 198     if ('N' == ch || 'n' == ch) {       // check for "NaN", case insensitively
 199         UniChar next1 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 1);
 200         UniChar next2 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 2);
 201         if (('a' == next1 || 'A' == next1) &&
 202             ('N' == next2 || 'n' == next2)) {
 203             *indexPtr += 3;
 204             if (resultPtr) *(uint64_t *)resultPtr = BITSFORDOUBLENAN;
 205             return true;
 206         }
 207     }
 208     if ('I' == ch || 'i' == ch) {       // check for "Inf", case insensitively
 209         UniChar next1 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 1);
 210         UniChar next2 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 2);
 211         if (('n' == next1 || 'N' == next1) &&
 212             ('f' == next2 || 'F' == next2)) {
 213             *indexPtr += 3;
 214             if (resultPtr) *(uint64_t *)resultPtr = BITSFORDOUBLEPOSINF;
 215             return true;
 216         }
 217     }
 218     if ('+' == ch || '-' == ch) {       // check for "+/-Inf", case insensitively
 219         UniChar next1 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 1);
 220         UniChar next2 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 2);
 221         UniChar next3 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 3);
 222         if (('I' == next1 || 'i' == next1) &&
 223             ('n' == next2 || 'N' == next2) &&
 224             ('f' == next3 || 'F' == next3)) {
 225             *indexPtr += 4;
 226             if (resultPtr) *(uint64_t *)resultPtr = ('-' == ch) ? BITSFORDOUBLENEGINF : BITSFORDOUBLEPOSINF;
 227             return true;
 228         }
 229     }
 230 #endif 0
 231     // Get characters until one not in __CFNumberSet[] is encountered
 232     while ((ch < 128) && (__CFNumberSet[ch >> 3] & (1 << (ch & 7)))) {
 233         if (numChars >= capacity - 1) {
 234             capacity += ALLOC_CHUNK_SIZE;
 235             if (tmpAlloc == NULL) tmpAlloc = __CFGetDefaultAllocator();
 236             if (charPtr == localCharBuffer) {
 237                 charPtr = (char *)CFAllocatorAllocate(tmpAlloc, capacity * sizeof(char), 0);
 238                 memmove(charPtr, localCharBuffer, numChars * sizeof(char));
 239             } else {
 240                 charPtr = (char *)CFAllocatorReallocate(tmpAlloc, charPtr, capacity * sizeof(char), 0);
 241             }
 242         }
 243         charPtr[numChars++] = (char)ch;
 244         ch = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + numChars);
 245     };
 246     charPtr[numChars] = 0;      // Null byte for strtod
 247
 248     result = strtod_l(charPtr, &endCharPtr, NULL);
 249
 250     if (tmpAlloc) CFAllocatorDeallocate(tmpAlloc, charPtr);
 251     if (charPtr == endCharPtr) return false;
 252     *indexPtr += (endCharPtr - charPtr);
 253     if (resultPtr) *resultPtr = result; // only store result if we succeed
 254
 255     return true;
 256 }
 257
 258
 259 #undef STACK_BUFFER_SIZE
 260 #undef ALLOC_CHUNK_SIZE
 261
 262