src/base/str.cc

   1 /*
   2  * Copyright (c) 2001-2005 The Regents of The University of Michigan
   3  * All rights reserved.
   4  *
   5  * Redistribution and use in source and binary forms, with or without
   6  * modification, are permitted provided that the following conditions are
   7  * met: redistributions of source code must retain the above copyright
   8  * notice, this list of conditions and the following disclaimer;
   9  * redistributions in binary form must reproduce the above copyright
  10  * notice, this list of conditions and the following disclaimer in the
  11  * documentation and/or other materials provided with the distribution;
  12  * neither the name of the copyright holders nor the names of its
  13  * contributors may be used to endorse or promote products derived from
  14  * this software without specific prior written permission.
  15  *
  16  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
  17  * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  18  * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
  19  * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
  20  * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  21  * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
  22  * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  23  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  24  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  25  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
  26  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  27  */
  28
  29 #include <ctype.h>
  30
  31 #include <cstring>
  32 #include <iostream>
  33 #include <string>
  34 #include <vector>
  35
  36 #include "base/intmath.hh"
  37 #include "base/str.hh"
  38
  39 using namespace std;
  40
  41 bool
  42 split_first(const string &s, string &lhs, string &rhs, char c)
  43 {
  44     string::size_type offset = s.find(c);
  45     if (offset == string::npos) {
  46         lhs = s;
  47         rhs = "";
  48         return false;
  49     }
  50
  51     lhs = s.substr(0, offset);
  52     rhs = s.substr(offset + 1);
  53     return true;
  54 }
  55
  56 bool
  57 split_last(const string &s, string &lhs, string &rhs, char c)
  58 {
  59     string::size_type offset = s.rfind(c);
  60     if (offset == string::npos) {
  61         lhs = s;
  62         rhs = "";
  63         return false;
  64     }
  65
  66     lhs = s.substr(0, offset);
  67     rhs = s.substr(offset + 1);
  68     return true;
  69 }
  70
  71 void
  72 tokenize(vector<string>& v, const string &s, char token, bool ignore)
  73 {
  74     string::size_type first = 0;
  75     string::size_type last = s.find_first_of(token);
  76
  77     if (s.empty())
  78         return;
  79
  80     if (ignore && last == first) {
  81         while (last == first)
  82             last = s.find_first_of(token, ++first);
  83
  84         if (last == string::npos) {
  85             if (first != s.size())
  86                 v.push_back(s.substr(first));
  87             return;
  88         }
  89     }
  90
  91     while (last != string::npos) {
  92         v.push_back(s.substr(first, last - first));
  93
  94         if (ignore) {
  95             first = s.find_first_not_of(token, last + 1);
  96
  97             if (first == string::npos)
  98                 return;
  99         } else
 100             first = last + 1;
 101
 102         last = s.find_first_of(token, first);
 103     }
 104
 105     v.push_back(s.substr(first));
 106 }
 107
 108 /**
 109  * @todo This function will not handle the smallest negative decimal
 110  * value for a signed type
 111  */
 112
 113 template <class T>
 114 inline bool
 115 __to_number(string value, T &retval)
 116 {
 117     static const T maxnum = ((T)-1);
 118     static const bool sign = maxnum < 0;
 119     static const int bits = sizeof(T) * 8;
 120     static const T hexmax = maxnum & (((T)1 << (bits - 4 - sign)) - 1);
 121     static const T octmax = maxnum & (((T)1 << (bits - 3 - sign)) - 1);
 122     static const T signmax =
 123         (sign) ? maxnum & (((T)1 << (bits - 1)) - 1) : maxnum;
 124     static const T decmax = signmax / 10;
 125
 126 #if 0
 127     cout << "maxnum =  0x" << hex << (unsigned long long)maxnum << "\n"
 128          << "sign =    0x" << hex << (unsigned long long)sign << "\n"
 129          << "hexmax =  0x" << hex << (unsigned long long)hexmax << "\n"
 130          << "octmax =  0x" << hex << (unsigned long long)octmax << "\n"
 131          << "signmax = 0x" << hex << (unsigned long long)signmax << "\n"
 132          << "decmax =  0x" << hex << (unsigned long long)decmax << "\n";
 133 #endif
 134
 135     eat_white(value);
 136
 137     bool negative = false;
 138     bool hex = false;
 139     bool oct = false;
 140     int last = value.size() - 1;
 141     retval = 0;
 142     int i = 0;
 143
 144     char c = value[i];
 145     if (!isDec(c)) {
 146         if (c == '-' && sign)
 147             negative = true;
 148         else
 149             return false;
 150     }
 151     else {
 152         retval += c - '0';
 153         if (last == 0) return true;
 154     }
 155
 156     if (c == '0')
 157         oct = true;
 158
 159     c = value[++i];
 160     if (oct) {
 161         if (sign && negative)
 162             return false;
 163
 164         if (!isOct(c)) {
 165             if (c == 'X' || c == 'x') {
 166                 hex = true;
 167                 oct = false;
 168             } else
 169                 return false;
 170         }
 171         else
 172             retval += c - '0';
 173     } else if (!isDec(c))
 174         goto multiply;
 175     else {
 176         if (sign && negative && c == '0')
 177             return false;
 178
 179         retval *= 10;
 180         retval += c - '0';
 181         if (last == 1) {
 182             if (sign && negative) retval = -retval;
 183             return true;
 184         }
 185     }
 186
 187     if (hex) {
 188         if (last == 1)
 189             return false;
 190
 191         for (i = 2; i <= last ; i++) {
 192             c = value[i];
 193             if (!isHex(c))
 194                 return false;
 195
 196             if (retval > hexmax) return false;
 197             retval *= 16;
 198             retval += hex2Int(c);
 199         }
 200         return true;
 201     } else if (oct) {
 202         for (i = 2; i <= last ; i++) {
 203             c = value[i];
 204             if (!isOct(c))
 205                 return false;
 206
 207             if (retval > octmax) return false;
 208             retval *= 8;
 209             retval += (c - '0');
 210         }
 211         return true;
 212     }
 213
 214     for (i = 2; i < last ; i++) {
 215         c = value[i];
 216         if (!isDec(c))
 217             goto multiply;
 218
 219         if (retval > decmax) return false;
 220         bool atmax = retval == decmax;
 221         retval *= 10;
 222         retval += c - '0';
 223         if (atmax && retval < decmax) return false;
 224         if (sign && (retval & ((T)1 << (sizeof(T) * 8 - 1))))
 225             return false;
 226     }
 227
 228     c = value[last];
 229     if (isDec(c)) {
 230
 231         if (retval > decmax) return false;
 232         bool atmax = retval == decmax;
 233         retval *= 10;
 234         retval += c - '0';
 235         if (atmax && retval < decmax) return false;
 236         if (sign && negative) {
 237             if ((retval & ((T)1 << (sizeof(T) * 8 - 1))) &&
 238                 retval >= (T)-signmax)
 239                 return false;
 240             retval = -retval;
 241         }
 242         else
 243             if (sign && (retval & ((T)1 << ((sizeof(T) * 8) - 1))))
 244                 return false;
 245         return true;
 246     }
 247
 248   multiply:
 249     signed long long mult = 1;
 250     T val;
 251     switch (c) {
 252       case 'k':
 253       case 'K':
 254         if (i != last) return false;
 255         mult = 1024;
 256         val = signmax / mult;
 257         break;
 258       case 'm':
 259       case 'M':
 260         if (i != last) return false;
 261         mult = 1024 * 1024;
 262         val = signmax / mult;
 263         break;
 264       case 'g':
 265       case 'G':
 266         if (i != last) return false;
 267         mult = 1024 * 1024 * 1024;
 268         val = signmax / mult;
 269         break;
 270       case 'e':
 271       case 'E':
 272         if (i >= last) return false;
 273
 274         mult = 0;
 275         for (i++; i <= last; i++) {
 276             c = value[i];
 277             if (!isDec(c))
 278                 return false;
 279
 280             mult *= 10;
 281             mult += c - '0';
 282         }
 283
 284         for (i = 0; i < mult; i++) {
 285             if (retval > signmax / 10)
 286                 return false;
 287             retval *= 10;
 288             if (sign && (retval & ((T)1 << (sizeof(T) * 8 - 1))))
 289                 return false;
 290         }
 291         if (sign && negative) {
 292             if ((retval & ((T)1 << (sizeof(T) * 8 - 1))) &&
 293                 retval >= (T)-signmax)
 294                 return false;
 295             retval = -retval;
 296         }
 297         else
 298             if (sign && (retval & ((T)1 << ((sizeof(T) * 8) - 1))))
 299                 return false;
 300
 301         return true;
 302
 303       default:
 304         return false;
 305     }
 306
 307     if (sign && negative)
 308         return false;
 309
 310     if (mult > (unsigned long long)signmax)
 311         return false;
 312
 313     if (retval > val)
 314         return false;
 315
 316     retval *= mult;
 317
 318     return true;
 319 }
 320
 321 #define STN(type) \
 322 template<> \
 323 bool to_number<type>(const string &value, type &retval) \
 324 { return __to_number(value, retval); }
 325
 326 STN(unsigned long long);
 327 STN(signed long long);
 328 STN(unsigned long);
 329 STN(signed long);
 330 STN(unsigned int);
 331 STN(signed int);
 332 STN(unsigned short);
 333 STN(signed short);
 334 STN(unsigned char);
 335 STN(signed char);
 336
 337 template<>
 338 bool to_number<bool>(const string &value, bool &retval)
 339 {
 340     string lowered = to_lower(value);
 341
 342     if (value == "0") {
 343         retval = false;
 344         return true;
 345     }
 346
 347     if (value == "1"){
 348         retval = true;
 349         return true;
 350     }
 351
 352     if (lowered == "false") {
 353         retval = false;
 354         return true;
 355     }
 356
 357     if (lowered == "true"){
 358         retval = true;
 359         return true;
 360     }
 361
 362     if (lowered == "no") {
 363         retval = false;
 364         return true;
 365     }
 366
 367     if (lowered == "yes"){
 368         retval = true;
 369         return true;
 370     }
 371
 372     return false;
 373 }