base/str.cc

   1 /*
   2  * Copyright (c) 2001-2003 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 <iostream>
  30
  31 #include <string.h>
  32 #include <ctype.h>
  33
  34 #include <string>
  35 #include <vector>
  36
  37 #include "base/intmath.hh"
  38 #include "base/str.hh"
  39
  40 using namespace std;
  41
  42 bool
  43 split_first(const string &s, string &lhs, string &rhs, char c)
  44 {
  45     string::size_type offset = s.find(c);
  46     if (offset == string::npos) {
  47         lhs = s;
  48         rhs = "";
  49         return false;
  50     }
  51
  52     lhs = s.substr(0, offset);
  53     rhs = s.substr(offset + 1);
  54     return true;
  55 }
  56
  57 bool
  58 split_last(const string &s, string &lhs, string &rhs, char c)
  59 {
  60     string::size_type offset = s.rfind(c);
  61     if (offset == string::npos) {
  62         lhs = s;
  63         rhs = "";
  64         return false;
  65     }
  66
  67     lhs = s.substr(0, offset);
  68     rhs = s.substr(offset + 1);
  69     return true;
  70 }
  71
  72 void
  73 tokenize(vector<string>& v, const string &s, char token, bool ignore)
  74 {
  75     string::size_type first = 0;
  76     string::size_type last = s.find_first_of(token);
  77
  78     if (ignore) {
  79         if (last == first) {
  80             while (last == first)
  81                 last = s.find_first_of(token, ++first);
  82
  83             if (last == string::npos) {
  84                 v.push_back(s);
  85                 return;
  86             }
  87         }
  88     }
  89
  90     while (last != string::npos) {
  91         v.push_back(s.substr(first, last - first));
  92
  93         if (ignore) {
  94             first = s.find_first_not_of(token, last + 1);
  95
  96             if (first == string::npos)
  97                 return;
  98         } else
  99             first = last + 1;
 100
 101         last = s.find_first_of(token, first);
 102     }
 103
 104     v.push_back(s.substr(first));
 105 }
 106
 107 /**
 108  * @todo This function will not handle the smallest negative decimal
 109  * value for a signed type
 110  */
 111
 112 template <class T>
 113 inline bool
 114 __to_number(string value, T &retval)
 115 {
 116     static const T maxnum = ((T)-1);
 117     static const bool sign = maxnum < 0;
 118     static const int bits = sizeof(T) * 8;
 119     static const T hexmax = maxnum & (((T)1 << (bits - 4 - sign)) - 1);
 120     static const T octmax = maxnum & (((T)1 << (bits - 3 - sign)) - 1);
 121     static const T signmax =
 122         (sign) ? maxnum & (((T)1 << (bits - 1)) - 1) : maxnum;
 123     static const T decmax = signmax / 10;
 124
 125 #if 0
 126     cout << "maxnum =  0x" << hex << (unsigned long long)maxnum << "\n"
 127          << "sign =    0x" << hex << (unsigned long long)sign << "\n"
 128          << "hexmax =  0x" << hex << (unsigned long long)hexmax << "\n"
 129          << "octmax =  0x" << hex << (unsigned long long)octmax << "\n"
 130          << "signmax = 0x" << hex << (unsigned long long)signmax << "\n"
 131          << "decmax =  0x" << hex << (unsigned long long)decmax << "\n";
 132 #endif
 133
 134     eat_white(value);
 135
 136     bool negative = false;
 137     bool hex = false;
 138     bool oct = false;
 139     int last = value.size() - 1;
 140     retval = 0;
 141     int i = 0;
 142
 143     char c = value[i];
 144     if (!IsDec(c)) {
 145         if (c == '-' && sign)
 146             negative = true;
 147         else
 148             return false;
 149     }
 150     else {
 151         retval += c - '0';
 152         if (last == 0) return true;
 153     }
 154
 155     if (c == '0')
 156         oct = true;
 157
 158     c = value[++i];
 159     if (oct) {
 160         if (sign && negative)
 161             return false;
 162
 163         if (!IsOct(c)) {
 164             if (c == 'X' || c == 'x') {
 165                 hex = true;
 166                 oct = false;
 167             } else
 168                 return false;
 169         }
 170         else
 171             retval += c - '0';
 172     } else if (!IsDec(c))
 173         goto multiply;
 174     else {
 175         if (sign && negative && c == '0')
 176             return false;
 177
 178         retval *= 10;
 179         retval += c - '0';
 180         if (last == 1) {
 181             if (sign && negative) retval = -retval;
 182             return true;
 183         }
 184     }
 185
 186     if (hex) {
 187         if (last == 1)
 188             return false;
 189
 190         for (i = 2; i <= last ; i++) {
 191             c = value[i];
 192             if (!IsHex(c))
 193                 return false;
 194
 195             if (retval > hexmax) return false;
 196             retval *= 16;
 197             retval += Hex2Int(c);
 198         }
 199         return true;
 200     } else if (oct) {
 201         for (i = 2; i <= last ; i++) {
 202             c = value[i];
 203             if (!IsOct(c))
 204                 return false;
 205
 206             if (retval > octmax) return false;
 207             retval *= 8;
 208             retval += (c - '0');
 209         }
 210         return true;
 211     }
 212
 213     for (i = 2; i < last ; i++) {
 214         c = value[i];
 215         if (!IsDec(c))
 216             goto multiply;
 217
 218         if (retval > decmax) return false;
 219         bool atmax = retval == decmax;
 220         retval *= 10;
 221         retval += c - '0';
 222         if (atmax && retval < decmax) return false;
 223         if (sign && (retval & ((T)1 << (sizeof(T) * 8 - 1))))
 224             return false;
 225     }
 226
 227     c = value[last];
 228     if (IsDec(c)) {
 229
 230         if (retval > decmax) return false;
 231         bool atmax = retval == decmax;
 232         retval *= 10;
 233         retval += c - '0';
 234         if (atmax && retval < decmax) return false;
 235         if (sign && negative) {
 236             if ((retval & ((T)1 << (sizeof(T) * 8 - 1))) &&
 237                 retval >= (T)-signmax)
 238                 return false;
 239             retval = -retval;
 240         }
 241         else
 242             if (sign && (retval & ((T)1 << ((sizeof(T) * 8) - 1))))
 243                 return false;
 244         return true;
 245     }
 246
 247   multiply:
 248     signed long long mult = 1;
 249     T val;
 250     switch (c) {
 251       case 'k':
 252       case 'K':
 253         if (i != last) return false;
 254         mult = 1024;
 255         val = signmax / mult;
 256         break;
 257       case 'm':
 258       case 'M':
 259         if (i != last) return false;
 260         mult = 1024 * 1024;
 261         val = signmax / mult;
 262         break;
 263       case 'g':
 264       case 'G':
 265         if (i != last) return false;
 266         mult = 1024 * 1024 * 1024;
 267         val = signmax / mult;
 268         break;
 269       case 'e':
 270       case 'E':
 271         if (i >= last) return false;
 272
 273         mult = 0;
 274         for (i++; i <= last; i++) {
 275             c = value[i];
 276             if (!IsDec(c))
 277                 return false;
 278
 279             mult *= 10;
 280             mult += c - '0';
 281         }
 282
 283         for (i = 0; i < mult; i++) {
 284             if (retval > signmax / 10)
 285                 return false;
 286             retval *= 10;
 287             if (sign && (retval & ((T)1 << (sizeof(T) * 8 - 1))))
 288                 return false;
 289         }
 290         if (sign && negative) {
 291             if ((retval & ((T)1 << (sizeof(T) * 8 - 1))) &&
 292                 retval >= (T)-signmax)
 293                 return false;
 294             retval = -retval;
 295         }
 296         else
 297             if (sign && (retval & ((T)1 << ((sizeof(T) * 8) - 1))))
 298                 return false;
 299
 300         return true;
 301
 302       default:
 303         return false;
 304     }
 305
 306     if (sign && negative)
 307         return false;
 308
 309     if (mult > (unsigned long long)signmax)
 310         return false;
 311
 312     if (retval > val)
 313         return false;
 314
 315     retval *= mult;
 316
 317     return true;
 318 }
 319
 320 #define STN(type) \
 321 template<> \
 322 bool to_number<type>(const string &value, type &retval) \
 323 { return __to_number(value, retval); }
 324
 325 STN(unsigned long long);
 326 STN(signed long long);
 327 STN(unsigned long);
 328 STN(signed long);
 329 STN(unsigned int);
 330 STN(signed int);
 331 STN(unsigned short);
 332 STN(signed short);
 333 STN(unsigned char);
 334 STN(signed char);
 335
 336 template<>
 337 bool to_number<bool>(const string &value, bool &retval)
 338 {
 339     string lowered = to_lower(value);
 340
 341     if (value == "0") {
 342         retval = false;
 343         return true;
 344     }
 345
 346     if (value == "1"){
 347         retval = true;
 348         return true;
 349     }
 350
 351     if (lowered == "false") {
 352         retval = false;
 353         return true;
 354     }
 355
 356     if (lowered == "true"){
 357         retval = true;
 358         return true;
 359     }
 360
 361     if (lowered == "no") {
 362         retval = false;
 363         return true;
 364     }
 365
 366     if (lowered == "yes"){
 367         retval = true;
 368         return true;
 369     }
 370
 371     return false;
 372 }