CFStringScanner.c

   1 /*
   2  * Copyright (c) 2008 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 /*      CFStringScanner.c
  24         Copyright 1999-2002, Apple, Inc. All rights reserved.
  25         Responsibility: Ali Ozer
  26 */
  27
  28 #include "CFInternal.h"
  29 #include <CoreFoundation/CFString.h>
  30 #include <sys/types.h>
  31 #include <limits.h>
  32 #include <stdlib.h>
  33 #include <string.h>
  34
  35 CF_INLINE Boolean __CFCharacterIsADigit(UniChar ch) {
  36     return (ch >= '0' && ch <= '9') ? true : false;
  37 }
  38
  39 /* Returns -1 on illegal value */
  40 CF_INLINE SInt32 __CFCharacterNumericOrHexValue (UniChar ch) {
  41     if (ch >= '0' && ch <= '9') {
  42         return ch - '0';
  43     } else if (ch >= 'A' && ch <= 'F') {
  44         return ch + 10 - 'A';
  45     } else if (ch >= 'a' && ch <= 'f') {
  46         return ch + 10 - 'a';
  47     } else {
  48         return -1;
  49     }
  50 }
  51
  52 /* Returns -1 on illegal value */
  53 CF_INLINE SInt32 __CFCharacterNumericValue(UniChar ch) {
  54     return (ch >= '0' && ch <= '9') ? (ch - '0') : -1;
  55 }
  56
  57 CF_INLINE UniChar __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(CFStringInlineBuffer *buf, SInt32 *indexPtr) {
  58     UniChar ch;
  59     while (__CFIsWhitespace(ch = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr))) (*indexPtr)++;
  60     return ch;
  61 }
  62
  63 /* result is int64_t or int, depending on doLonglong
  64 */
  65 __private_extern__ Boolean __CFStringScanInteger(CFStringInlineBuffer *buf, CFTypeRef locale, SInt32 *indexPtr, Boolean doLonglong, void *result) {
  66     Boolean doingLonglong = false;      /* Set to true if doLonglong, and we overflow an int... */
  67     Boolean neg = false;
  68     int intResult = 0;
  69     register int64_t longlongResult = 0;        /* ??? int64_t is slow when not in regs; I hope this does the right thing. */
  70     UniChar ch;
  71
  72     ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
  73
  74     if (ch == '-' || ch == '+') {
  75         neg = (ch == '-');
  76         (*indexPtr)++;
  77         ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
  78     }
  79
  80     if (! __CFCharacterIsADigit(ch)) return false;      /* No digits, bail out... */
  81     do {
  82         if (doingLonglong) {
  83             if ((longlongResult >= LLONG_MAX / 10) && ((longlongResult > LLONG_MAX / 10) || (__CFCharacterNumericValue(ch) - (neg ? 1 : 0) >= LLONG_MAX - longlongResult * 10))) {
  84                 /* ??? This might not handle LLONG_MIN correctly... */
  85                 longlongResult = neg ? LLONG_MIN : LLONG_MAX;
  86                 neg = false;
  87                 while (__CFCharacterIsADigit(ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr))));      /* Skip remaining digits */
  88             } else {
  89                 longlongResult = longlongResult * 10 + __CFCharacterNumericValue(ch);
  90                 ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
  91             }
  92         } else {
  93             if ((intResult >= INT_MAX / 10) && ((intResult > INT_MAX / 10) || (__CFCharacterNumericValue(ch) - (neg ? 1 : 0) >= INT_MAX - intResult * 10))) {
  94                 // Overflow, check for int64_t...
  95                 if (doLonglong) {
  96                     longlongResult = intResult;
  97                     doingLonglong = true;
  98                 } else {
  99                     /* ??? This might not handle INT_MIN correctly... */
 100                     intResult = neg ? INT_MIN : INT_MAX;
 101                     neg = false;
 102                     while (__CFCharacterIsADigit(ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr))));  /* Skip remaining digits */
 103                 }
 104             } else {
 105                 intResult = intResult * 10 + __CFCharacterNumericValue(ch);
 106                 ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
 107             }
 108         }
 109     } while (__CFCharacterIsADigit(ch));
 110
 111     if (result) {
 112         if (doLonglong) {
 113             if (!doingLonglong) longlongResult = intResult;
 114             *(int64_t *)result = neg ? -longlongResult : longlongResult;
 115         } else {
 116             *(int *)result = neg ? -intResult : intResult;
 117         }
 118     }
 119
 120     return true;
 121 }
 122
 123 __private_extern__ Boolean __CFStringScanHex(CFStringInlineBuffer *buf, SInt32 *indexPtr, unsigned *result) {
 124     UInt32 value = 0;
 125     SInt32 curDigit;
 126     UniChar ch;
 127
 128     ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
 129     /* Ignore the optional "0x" or "0X"; if it's followed by a non-hex, just parse the "0" and leave pointer at "x" */
 130     if (ch == '0') {
 131         ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
 132         if (ch == 'x' || ch == 'X') ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
 133         curDigit = __CFCharacterNumericOrHexValue(ch);
 134         if (curDigit == -1) {
 135             (*indexPtr)--;      /* Go back over the "x" or "X" */
 136             if (result) *result = 0;
 137             return true;        /* We just saw "0" */
 138         }
 139     } else {
 140         curDigit = __CFCharacterNumericOrHexValue(ch);
 141         if (curDigit == -1) return false;
 142     }
 143
 144     do {
 145         if (value > (UINT_MAX >> 4)) {
 146             value = UINT_MAX;   /* We do this over and over again, but it's an error case anyway */
 147         } else {
 148             value = (value << 4) + curDigit;
 149         }
 150         curDigit = __CFCharacterNumericOrHexValue(__CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr)));
 151     } while (curDigit != -1);
 152
 153     if (result) *result = value;
 154     return true;
 155 }
 156
 157 // Packed array of Boolean
 158 static const unsigned char __CFNumberSet[16] = {
 159     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  nul soh stx etx eot enq ack bel
 160     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  bs  ht  nl  vt  np  cr  so  si
 161     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  dle dc1 dc2 dc3 dc4 nak syn etb
 162     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  can em  sub esc fs  gs  rs  us
 163     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  sp   !   "   #   $   %   &   '
 164     0X28, // 0, 0, 0, 1, 0, 1, 0, 0, //  (   )   *   +   ,   -   .   /
 165     0xFF, // 1, 1, 1, 1, 1, 1, 1, 1, //  0   1   2   3   4   5   6   7
 166     0X03, // 1, 1, 0, 0, 0, 0, 0, 0, //  8   9   :   ;   <   =   >   ?
 167     0X20, // 0, 0, 0, 0, 0, 1, 0, 0, //  @   A   B   C   D   E   F   G
 168     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  H   I   J   K   L   M   N   O
 169     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  P   Q   R   S   T   U   V   W
 170     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  X   Y   Z   [   \   ]   ^   _
 171     0X20, // 0, 0, 0, 0, 0, 1, 0, 0, //  `   a   b   c   d   e   f   g
 172     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  h   i   j   k   l   m   n   o
 173     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  p   q   r   s   t   u   v   w
 174     0X00, // 0, 0, 0, 0, 0, 0, 0, 0  //  x   y   z   {   |   }   ~  del
 175 };
 176
 177 __private_extern__ Boolean __CFStringScanDouble(CFStringInlineBuffer *buf, CFTypeRef locale, SInt32 *indexPtr, double *resultPtr) {
 178     #define STACK_BUFFER_SIZE 256
 179     #define ALLOC_CHUNK_SIZE 256 // first and subsequent malloc size.  Should be greater than STACK_BUFFER_SIZE
 180     char localCharBuffer[STACK_BUFFER_SIZE];
 181     char *charPtr = localCharBuffer;
 182     char *endCharPtr;
 183     UniChar decimalChar = '.';
 184     SInt32 numChars = 0;
 185     SInt32 capacity = STACK_BUFFER_SIZE;        // in chars
 186     double result;
 187     UniChar ch;
 188     CFAllocatorRef tmpAlloc = NULL;
 189
 190 #if 0
 191     if (locale != NULL) {
 192         CFStringRef decimalSeparator = [locale objectForKey: @"NSDecimalSeparator"];
 193         if (decimalSeparator != nil) decimalChar = [decimalSeparator characterAtIndex:0];
 194     }
 195 #endif
 196     ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
 197     // At this point indexPtr points at the first non-space char
 198 #if 0
 199 #warning need to allow, case insensitively, all of: "nan", "inf", "-inf", "+inf", "-infinity", "+infinity", "infinity";
 200 #warning -- strtod() will actually do most or all of that for us
 201 #define BITSFORDOUBLENAN        ((uint64_t)0x7ff8000000000000ULL)
 202 #define BITSFORDOUBLEPOSINF     ((uint64_t)0x7ff0000000000000ULL)
 203 #define BITSFORDOUBLENEGINF     ((uint64_t)0xfff0000000000000ULL)
 204     if ('N' == ch || 'n' == ch) {       // check for "NaN", case insensitively
 205         UniChar next1 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 1);
 206         UniChar next2 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 2);
 207         if (('a' == next1 || 'A' == next1) &&
 208             ('N' == next2 || 'n' == next2)) {
 209             *indexPtr += 3;
 210             if (resultPtr) *(uint64_t *)resultPtr = BITSFORDOUBLENAN;
 211             return true;
 212         }
 213     }
 214     if ('I' == ch || 'i' == ch) {       // check for "Inf", case insensitively
 215         UniChar next1 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 1);
 216         UniChar next2 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 2);
 217         if (('n' == next1 || 'N' == next1) &&
 218             ('f' == next2 || 'F' == next2)) {
 219             *indexPtr += 3;
 220             if (resultPtr) *(uint64_t *)resultPtr = BITSFORDOUBLEPOSINF;
 221             return true;
 222         }
 223     }
 224     if ('+' == ch || '-' == ch) {       // check for "+/-Inf", case insensitively
 225         UniChar next1 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 1);
 226         UniChar next2 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 2);
 227         UniChar next3 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 3);
 228         if (('I' == next1 || 'i' == next1) &&
 229             ('n' == next2 || 'N' == next2) &&
 230             ('f' == next3 || 'F' == next3)) {
 231             *indexPtr += 4;
 232             if (resultPtr) *(uint64_t *)resultPtr = ('-' == ch) ? BITSFORDOUBLENEGINF : BITSFORDOUBLEPOSINF;
 233             return true;
 234         }
 235     }
 236 #endif 0
 237     do {
 238         if (ch >= 128 || (__CFNumberSet[ch >> 3] & (1 << (ch & 7))) == 0) {
 239             // Not in __CFNumberSet
 240             if (ch != decimalChar) break;
 241             ch = '.';   // Replace the decimal character with something strtod will understand
 242         }
 243         if (numChars >= capacity - 1) {
 244             capacity += ALLOC_CHUNK_SIZE;
 245             if (tmpAlloc == NULL) tmpAlloc = __CFGetDefaultAllocator();
 246             if (charPtr == localCharBuffer) {
 247                 charPtr = (char *)CFAllocatorAllocate(tmpAlloc, capacity * sizeof(char), 0);
 248                 memmove(charPtr, localCharBuffer, numChars * sizeof(char));
 249             } else {
 250                 charPtr = (char *)CFAllocatorReallocate(tmpAlloc, charPtr, capacity * sizeof(char), 0);
 251             }
 252         }
 253         charPtr[numChars++] = (char)ch;
 254         ch = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + numChars);
 255     } while (true);
 256     charPtr[numChars] = 0;      // Null byte for strtod
 257
 258     result = strtod_l(charPtr, &endCharPtr, NULL);
 259
 260     if (tmpAlloc) CFAllocatorDeallocate(tmpAlloc, charPtr);
 261     if (charPtr == endCharPtr) return false;
 262     *indexPtr += (endCharPtr - charPtr);
 263     if (resultPtr) *resultPtr = result; // only store result if we succeed
 264
 265     return true;
 266 }
 267
 268
 269 #undef STACK_BUFFER_SIZE
 270 #undef ALLOC_CHUNK_SIZE
 271
 272