String.subproj/CFStringScanner.c

   1 /*
   2  * Copyright (c) 2005 Apple Computer, 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 #if !defined(__MACOS8__)
  31 #include <sys/types.h>
  32 #endif
  33 #include <limits.h>
  34 #include <stdlib.h>
  35 #include <string.h>
  36
  37 CF_INLINE Boolean __CFCharacterIsADigit(UniChar ch) {
  38     return (ch >= '0' && ch <= '9') ? true : false;
  39 }
  40
  41 /* Returns -1 on illegal value */
  42 CF_INLINE SInt32 __CFCharacterNumericOrHexValue (UniChar ch) {
  43     if (ch >= '0' && ch <= '9') {
  44         return ch - '0';
  45     } else if (ch >= 'A' && ch <= 'F') {
  46         return ch + 10 - 'A';
  47     } else if (ch >= 'a' && ch <= 'f') {
  48         return ch + 10 - 'a';
  49     } else {
  50         return -1;
  51     }
  52 }
  53
  54 /* Returns -1 on illegal value */
  55 CF_INLINE SInt32 __CFCharacterNumericValue(UniChar ch) {
  56     return (ch >= '0' && ch <= '9') ? (ch - '0') : -1;
  57 }
  58
  59 CF_INLINE UniChar __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(CFStringInlineBuffer *buf, SInt32 *indexPtr) {
  60     UniChar ch;
  61     while (__CFIsWhitespace(ch = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr))) (*indexPtr)++;
  62     return ch;
  63 }
  64
  65 /* result is int64_t or int, depending on doLonglong
  66 */
  67 __private_extern__ Boolean __CFStringScanInteger(CFStringInlineBuffer *buf, CFDictionaryRef locale, SInt32 *indexPtr, Boolean doLonglong, void *result) {
  68     Boolean doingLonglong = false;      /* Set to true if doLonglong, and we overflow an int... */
  69     Boolean neg = false;
  70     int intResult = 0;
  71     register int64_t longlongResult = 0;        /* ??? int64_t is slow when not in regs; I hope this does the right thing. */
  72     UniChar ch;
  73
  74     ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
  75
  76     if (ch == '-' || ch == '+') {
  77         neg = (ch == '-');
  78         (*indexPtr)++;
  79         ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
  80     }
  81
  82     if (! __CFCharacterIsADigit(ch)) return false;      /* No digits, bail out... */
  83     do {
  84         if (doingLonglong) {
  85             if ((longlongResult >= LLONG_MAX / 10) && ((longlongResult > LLONG_MAX / 10) || (__CFCharacterNumericValue(ch) - (neg ? 1 : 0) >= LLONG_MAX - longlongResult * 10))) {
  86                 /* ??? This might not handle LLONG_MIN correctly... */
  87                 longlongResult = neg ? LLONG_MIN : LLONG_MAX;
  88                 neg = false;
  89                 while (__CFCharacterIsADigit(ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr))));      /* Skip remaining digits */
  90             } else {
  91                 longlongResult = longlongResult * 10 + __CFCharacterNumericValue(ch);
  92                 ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
  93             }
  94         } else {
  95             if ((intResult >= INT_MAX / 10) && ((intResult > INT_MAX / 10) || (__CFCharacterNumericValue(ch) - (neg ? 1 : 0) >= INT_MAX - intResult * 10))) {
  96                 // Overflow, check for int64_t...
  97                 if (doLonglong) {
  98                     longlongResult = intResult;
  99                     doingLonglong = true;
 100                 } else {
 101                     /* ??? This might not handle INT_MIN correctly... */
 102                     intResult = neg ? INT_MIN : INT_MAX;
 103                     neg = false;
 104                     while (__CFCharacterIsADigit(ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr))));  /* Skip remaining digits */
 105                 }
 106             } else {
 107                 intResult = intResult * 10 + __CFCharacterNumericValue(ch);
 108                 ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
 109             }
 110         }
 111     } while (__CFCharacterIsADigit(ch));
 112
 113     if (result) {
 114         if (doLonglong) {
 115             if (!doingLonglong) longlongResult = intResult;
 116             *(int64_t *)result = neg ? -longlongResult : longlongResult;
 117         } else {
 118             *(int *)result = neg ? -intResult : intResult;
 119         }
 120     }
 121
 122     return true;
 123 }
 124
 125 __private_extern__ Boolean __CFStringScanHex(CFStringInlineBuffer *buf, SInt32 *indexPtr, unsigned *result) {
 126     UInt32 value = 0;
 127     SInt32 curDigit;
 128     UniChar ch;
 129
 130     ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
 131     /* Ignore the optional "0x" or "0X"; if it's followed by a non-hex, just parse the "0" and leave pointer at "x" */
 132     if (ch == '0') {
 133         ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
 134         if (ch == 'x' || ch == 'X') ch = __CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr));
 135         curDigit = __CFCharacterNumericOrHexValue(ch);
 136         if (curDigit == -1) {
 137             (*indexPtr)--;      /* Go back over the "x" or "X" */
 138             if (result) *result = 0;
 139             return true;        /* We just saw "0" */
 140         }
 141     } else {
 142         curDigit = __CFCharacterNumericOrHexValue(ch);
 143         if (curDigit == -1) return false;
 144     }
 145
 146     do {
 147         if (value > (UINT_MAX >> 4)) {
 148             value = UINT_MAX;   /* We do this over and over again, but it's an error case anyway */
 149         } else {
 150             value = (value << 4) + curDigit;
 151         }
 152         curDigit = __CFCharacterNumericOrHexValue(__CFStringGetCharacterFromInlineBufferAux(buf, ++(*indexPtr)));
 153     } while (curDigit != -1);
 154
 155     if (result) *result = value;
 156     return true;
 157 }
 158
 159 // Packed array of Boolean
 160 static const char __CFNumberSet[16] = {
 161     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  nul soh stx etx eot enq ack bel
 162     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  bs  ht  nl  vt  np  cr  so  si
 163     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  dle dc1 dc2 dc3 dc4 nak syn etb
 164     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  can em  sub esc fs  gs  rs  us
 165     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  sp   !   "   #   $   %   &   '
 166     0X28, // 0, 0, 0, 1, 0, 1, 0, 0, //  (   )   *   +   ,   -   .   /
 167     0XFF, // 1, 1, 1, 1, 1, 1, 1, 1, //  0   1   2   3   4   5   6   7
 168     0X03, // 1, 1, 0, 0, 0, 0, 0, 0, //  8   9   :   ;   <   =   >   ?
 169     0X20, // 0, 0, 0, 0, 0, 1, 0, 0, //  @   A   B   C   D   E   F   G
 170     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  H   I   J   K   L   M   N   O
 171     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  P   Q   R   S   T   U   V   W
 172     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  X   Y   Z   [   \   ]   ^   _
 173     0X20, // 0, 0, 0, 0, 0, 1, 0, 0, //  `   a   b   c   d   e   f   g
 174     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  h   i   j   k   l   m   n   o
 175     0X00, // 0, 0, 0, 0, 0, 0, 0, 0, //  p   q   r   s   t   u   v   w
 176     0X00, // 0, 0, 0, 0, 0, 0, 0, 0  //  x   y   z   {   |   }   ~  del
 177 };
 178
 179 __private_extern__ Boolean __CFStringScanDouble(CFStringInlineBuffer *buf, CFDictionaryRef locale, SInt32 *indexPtr, double *resultPtr) {
 180     #define STACK_BUFFER_SIZE 256
 181     #define ALLOC_CHUNK_SIZE 256 // first and subsequent malloc size.  Should be greater than STACK_BUFFER_SIZE
 182     char localCharBuffer[STACK_BUFFER_SIZE];
 183     char *charPtr = localCharBuffer;
 184     char *endCharPtr;
 185     UniChar decimalChar = '.';
 186     SInt32 numChars = 0;
 187     SInt32 capacity = STACK_BUFFER_SIZE;        // in chars
 188     double result;
 189     UniChar ch;
 190     CFAllocatorRef tmpAlloc = NULL;
 191
 192 #if 0
 193     if (locale != NULL) {
 194         CFStringRef decimalSeparator = [locale objectForKey: NSDecimalSeparator];
 195         if (decimalSeparator != nil) decimalChar = [decimalSeparator characterAtIndex:0];
 196     }
 197 #endif
 198     ch = __CFStringGetFirstNonSpaceCharacterFromInlineBuffer(buf, indexPtr);
 199     // At this point indexPtr points at the first non-space char
 200 #if 0
 201 #warning need to allow, case insensitively, all of: "nan", "inf", "-inf", "+inf", "-infinity", "+infinity", "infinity";
 202 #warning -- strtod() will actually do most or all of that for us
 203 #define BITSFORDOUBLENAN        ((uint64_t)0x7ff8000000000000)
 204 #define BITSFORDOUBLEPOSINF     ((uint64_t)0x7ff0000000000000)
 205 #define BITSFORDOUBLENEGINF     ((uint64_t)0xfff0000000000000)
 206     if ('N' == ch || 'n' == ch) {       // check for "NaN", case insensitively
 207         UniChar next1 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 1);
 208         UniChar next2 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 2);
 209         if (('a' == next1 || 'A' == next1) &&
 210             ('N' == next2 || 'n' == next2)) {
 211             *indexPtr += 3;
 212             if (resultPtr) *(uint64_t *)resultPtr = BITSFORDOUBLENAN;
 213             return true;
 214         }
 215     }
 216     if ('I' == ch || 'i' == ch) {       // check for "Inf", case insensitively
 217         UniChar next1 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 1);
 218         UniChar next2 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 2);
 219         if (('n' == next1 || 'N' == next1) &&
 220             ('f' == next2 || 'F' == next2)) {
 221             *indexPtr += 3;
 222             if (resultPtr) *(uint64_t *)resultPtr = BITSFORDOUBLEPOSINF;
 223             return true;
 224         }
 225     }
 226     if ('+' == ch || '-' == ch) {       // check for "+/-Inf", case insensitively
 227         UniChar next1 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 1);
 228         UniChar next2 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 2);
 229         UniChar next3 = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + 3);
 230         if (('I' == next1 || 'i' == next1) &&
 231             ('n' == next2 || 'N' == next2) &&
 232             ('f' == next3 || 'F' == next3)) {
 233             *indexPtr += 4;
 234             if (resultPtr) *(uint64_t *)resultPtr = ('-' == ch) ? BITSFORDOUBLENEGINF : BITSFORDOUBLEPOSINF;
 235             return true;
 236         }
 237     }
 238 #endif 0
 239     do {
 240         if (ch >= 128 || (__CFNumberSet[ch >> 3] & (1 << (ch & 7))) == 0) {
 241             // Not in __CFNumberSet
 242             if (ch != decimalChar) break;
 243             ch = '.';   // Replace the decimal character with something strtod will understand
 244         }
 245         if (numChars >= capacity - 1) {
 246             capacity += ALLOC_CHUNK_SIZE;
 247             if (tmpAlloc == NULL) tmpAlloc = __CFGetDefaultAllocator();
 248             if (charPtr == localCharBuffer) {
 249                 charPtr = CFAllocatorAllocate(tmpAlloc, capacity * sizeof(char), 0);
 250                 memmove(charPtr, localCharBuffer, numChars * sizeof(char));
 251             } else {
 252                 charPtr = CFAllocatorReallocate(tmpAlloc, charPtr, capacity * sizeof(char), 0);
 253             }
 254         }
 255         charPtr[numChars++] = (char)ch;
 256         ch = __CFStringGetCharacterFromInlineBufferAux(buf, *indexPtr + numChars);
 257     } while (true);
 258     charPtr[numChars] = 0;      // Null byte for strtod
 259
 260     result = strtod_l(charPtr, &endCharPtr, NULL);
 261
 262     if (tmpAlloc) CFAllocatorDeallocate(tmpAlloc, charPtr);
 263     if (charPtr == endCharPtr) return false;
 264     *indexPtr += (endCharPtr - charPtr);
 265     if (resultPtr) *resultPtr = result; // only store result if we succeed
 266
 267     return true;
 268 }
 269
 270
 271 #undef STACK_BUFFER_SIZE
 272 #undef ALLOC_CHUNK_SIZE
 273
 274