src/gromacs/utility/stringutil.cpp

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2011,2012,2013,2014,2015, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*! \internal \file
  36  * \brief
  37  * Implements functions and classes in stringutil.h.
  38  *
  39  * \author Teemu Murtola <teemu.murtola@gmail.com>
  40  * \ingroup module_utility
  41  */
  42 #include "gmxpre.h"
  43
  44 #include "stringutil.h"
  45
  46 #include <cctype>
  47 #include <cstdarg>
  48 #include <cstdio>
  49 #include <cstring>
  50
  51 #include <algorithm>
  52 #include <string>
  53 #include <vector>
  54
  55 #include "gromacs/utility/gmxassert.h"
  56
  57 namespace gmx
  58 {
  59
  60 std::size_t
  61 countWords(const char *s)
  62 {
  63     std::size_t nWords = 0;
  64     // Use length variable to avoid N^2 complexity when executing strlen(s) every iteration
  65     std::size_t length = std::strlen(s);
  66
  67     for (std::size_t i = 0; i < length; i++)
  68     {
  69         // If we found a new word, increase counter and step through the word
  70         if (std::isalnum(s[i]))
  71         {
  72             ++nWords;
  73             // If we hit string end, '\0' is not alphanumerical
  74             while (std::isalnum(s[i]))
  75             {
  76                 // This might increment i to the string end, and then the outer
  77                 // loop will increment i one unit beyond that, but since
  78                 // we compare to the string length in the outer loop this is fine.
  79                 i++;
  80             }
  81         }
  82     }
  83     return nWords;
  84 }
  85
  86
  87 std::size_t
  88 countWords(const std::string &str)
  89 {
  90     // Under out beautiful C++ interface hides an ugly c-string implementation :-)
  91     return countWords(str.c_str());
  92 }
  93
  94 bool endsWith(const char *str, const char *suffix)
  95 {
  96     if (isNullOrEmpty(suffix))
  97     {
  98         return true;
  99     }
 100     const size_t strLength    = std::strlen(str);
 101     const size_t suffixLength = std::strlen(suffix);
 102     return (strLength >= suffixLength
 103             && std::strcmp(&str[strLength - suffixLength], suffix) == 0);
 104 }
 105
 106 std::string stripSuffixIfPresent(const std::string &str, const char *suffix)
 107 {
 108     if (suffix != NULL)
 109     {
 110         size_t suffixLength = std::strlen(suffix);
 111         if (suffixLength > 0 && endsWith(str, suffix))
 112         {
 113             return str.substr(0, str.length() - suffixLength);
 114         }
 115     }
 116     return str;
 117 }
 118
 119 std::string stripString(const std::string &str)
 120 {
 121     std::string::const_iterator start = str.begin();
 122     std::string::const_iterator end   = str.end();
 123     while (start != end && std::isspace(*start))
 124     {
 125         ++start;
 126     }
 127     while (start != end && std::isspace(*(end - 1)))
 128     {
 129         --end;
 130     }
 131     return std::string(start, end);
 132 }
 133
 134 std::string formatString(const char *fmt, ...)
 135 {
 136     va_list           ap;
 137     char              staticBuf[1024];
 138     int               length = 1024;
 139     std::vector<char> dynamicBuf;
 140     char             *buf = staticBuf;
 141
 142     // TODO: There may be a better way of doing this on Windows, Microsoft
 143     // provides their own way of doing things...
 144     while (1)
 145     {
 146         va_start(ap, fmt);
 147         int n = vsnprintf(buf, length, fmt, ap);
 148         va_end(ap);
 149         if (n > -1 && n < length)
 150         {
 151             std::string result(buf);
 152             return result;
 153         }
 154         if (n > -1)
 155         {
 156             length = n + 1;
 157         }
 158         else
 159         {
 160             length *= 2;
 161         }
 162         dynamicBuf.resize(length);
 163         buf = dynamicBuf.data();
 164     }
 165 }
 166
 167 std::vector<std::string> splitString(const std::string &str)
 168 {
 169     std::vector<std::string>          result;
 170     std::string::const_iterator       currPos = str.begin();
 171     const std::string::const_iterator end     = str.end();
 172     while (currPos != end)
 173     {
 174         while (currPos != end && std::isspace(*currPos))
 175         {
 176             ++currPos;
 177         }
 178         const std::string::const_iterator startPos = currPos;
 179         while (currPos != end && !std::isspace(*currPos))
 180         {
 181             ++currPos;
 182         }
 183         if (startPos != end)
 184         {
 185             result.push_back(std::string(startPos, currPos));
 186         }
 187     }
 188     return result;
 189 }
 190
 191 namespace
 192 {
 193
 194 /*! \brief
 195  * Helper function to identify word boundaries for replaceAllWords().
 196  *
 197  * \returns  `true` if the character is considered part of a word.
 198  *
 199  * \ingroup module_utility
 200  */
 201 bool isWordChar(char c)
 202 {
 203     return std::isalnum(c) || c == '-' || c == '_';
 204 }
 205
 206 /*! \brief
 207  * Common implementation for string replacement functions.
 208  *
 209  * \param[in] input  Input string.
 210  * \param[in] from   String to find.
 211  * \param[in] to     String to use to replace \p from.
 212  * \param[in] bWholeWords  Whether to only consider matches to whole words.
 213  * \returns   \p input with all occurrences of \p from replaced with \p to.
 214  * \throws    std::bad_alloc if out of memory.
 215  *
 216  * \ingroup module_utility
 217  */
 218 std::string
 219 replaceInternal(const std::string &input, const char *from, const char *to,
 220                 bool bWholeWords)
 221 {
 222     GMX_RELEASE_ASSERT(from != NULL && to != NULL,
 223                        "Replacement strings must not be NULL");
 224     size_t      matchLength = std::strlen(from);
 225     std::string result;
 226     size_t      inputPos = 0;
 227     size_t      matchPos = input.find(from);
 228     while (matchPos < input.length())
 229     {
 230         size_t matchEnd = matchPos + matchLength;
 231         if (bWholeWords)
 232         {
 233             if (!((matchPos == 0 || !isWordChar(input[matchPos-1]))
 234                   && (matchEnd == input.length() || !isWordChar(input[matchEnd]))))
 235             {
 236                 matchPos = input.find(from, matchPos + 1);
 237                 continue;
 238             }
 239
 240         }
 241         result.append(input, inputPos, matchPos - inputPos);
 242         result.append(to);
 243         inputPos = matchEnd;
 244         matchPos = input.find(from, inputPos);
 245     }
 246     result.append(input, inputPos, matchPos - inputPos);
 247     return result;
 248 }
 249
 250 }   // namespace
 251
 252 std::string
 253 replaceAll(const std::string &input, const char *from, const char *to)
 254 {
 255     return replaceInternal(input, from, to, false);
 256 }
 257
 258 std::string
 259 replaceAll(const std::string &input, const std::string &from,
 260            const std::string &to)
 261 {
 262     return replaceInternal(input, from.c_str(), to.c_str(), false);
 263 }
 264
 265 std::string
 266 replaceAllWords(const std::string &input, const char *from, const char *to)
 267 {
 268     return replaceInternal(input, from, to, true);
 269 }
 270
 271 std::string
 272 replaceAllWords(const std::string &input, const std::string &from,
 273                 const std::string &to)
 274 {
 275     return replaceInternal(input, from.c_str(), to.c_str(), true);
 276 }
 277
 278
 279 /********************************************************************
 280  * TextLineWrapperSettings
 281  */
 282
 283 TextLineWrapperSettings::TextLineWrapperSettings()
 284     : maxLength_(0), indent_(0), firstLineIndent_(-1),
 285       bKeepFinalSpaces_(false), continuationChar_('\0')
 286 {
 287 }
 288
 289
 290 /********************************************************************
 291  * TextLineWrapper
 292  */
 293
 294 bool TextLineWrapper::isTrivial() const
 295 {
 296     return settings_.lineLength() == 0 && settings_.indent() == 0
 297            && settings_.firstLineIndent_ <= 0;
 298 }
 299
 300 size_t
 301 TextLineWrapper::findNextLine(const char *input, size_t lineStart) const
 302 {
 303     size_t inputLength = std::strlen(input);
 304     bool   bFirstLine  = (lineStart == 0 || input[lineStart - 1] == '\n');
 305     // Ignore leading whitespace if necessary.
 306     if (!bFirstLine)
 307     {
 308         lineStart += std::strspn(input + lineStart, " ");
 309         if (lineStart >= inputLength)
 310         {
 311             return inputLength;
 312         }
 313     }
 314
 315     int    indent = (bFirstLine ? settings_.firstLineIndent() : settings_.indent());
 316     size_t lastAllowedBreakPoint
 317         = (settings_.lineLength() > 0
 318            ? std::min(lineStart + settings_.lineLength() - indent, inputLength)
 319            : inputLength);
 320     // Ignore trailing whitespace.
 321     lastAllowedBreakPoint += std::strspn(input + lastAllowedBreakPoint, " ");
 322     size_t lineEnd = lineStart;
 323     do
 324     {
 325         const char *nextBreakPtr = std::strpbrk(input + lineEnd, " \n");
 326         size_t      nextBreak
 327             = (nextBreakPtr != NULL ? nextBreakPtr - input : inputLength);
 328         if (nextBreak > lastAllowedBreakPoint && lineEnd > lineStart)
 329         {
 330             break;
 331         }
 332         lineEnd = nextBreak + 1;
 333     }
 334     while (lineEnd < lastAllowedBreakPoint && input[lineEnd - 1] != '\n');
 335     return (lineEnd < inputLength ? lineEnd : inputLength);
 336 }
 337
 338 size_t
 339 TextLineWrapper::findNextLine(const std::string &input, size_t lineStart) const
 340 {
 341     return findNextLine(input.c_str(), lineStart);
 342 }
 343
 344 std::string
 345 TextLineWrapper::formatLine(const std::string &input,
 346                             size_t lineStart, size_t lineEnd) const
 347 {
 348     size_t inputLength = input.length();
 349     bool   bFirstLine  = (lineStart == 0 || input[lineStart - 1] == '\n');
 350     // Strip leading whitespace if necessary.
 351     if (!bFirstLine)
 352     {
 353         lineStart = input.find_first_not_of(' ', lineStart);
 354         if (lineStart >= inputLength)
 355         {
 356             return std::string();
 357         }
 358     }
 359     int  indent        = (bFirstLine ? settings_.firstLineIndent() : settings_.indent());
 360     bool bContinuation = (lineEnd < inputLength && input[lineEnd - 1] != '\n');
 361     // Strip trailing whitespace.
 362     if (!settings_.bKeepFinalSpaces_ || lineEnd < inputLength || input[inputLength - 1] == '\n')
 363     {
 364         while (lineEnd > lineStart && std::isspace(input[lineEnd - 1]))
 365         {
 366             --lineEnd;
 367         }
 368     }
 369
 370     const size_t lineLength = lineEnd - lineStart;
 371     if (lineLength == 0)
 372     {
 373         return std::string();
 374     }
 375     std::string result(indent, ' ');
 376     result.append(input, lineStart, lineLength);
 377     if (bContinuation && settings_.continuationChar_ != '\0')
 378     {
 379         result.append(1, ' ');
 380         result.append(1, settings_.continuationChar_);
 381     }
 382     return result;
 383 }
 384
 385 std::string
 386 TextLineWrapper::wrapToString(const std::string &input) const
 387 {
 388     std::string result;
 389     size_t      lineStart = 0;
 390     size_t      length    = input.length();
 391     while (lineStart < length)
 392     {
 393         size_t nextLineStart = findNextLine(input, lineStart);
 394         result.append(formatLine(input, lineStart, nextLineStart));
 395         if (nextLineStart < length
 396             || (nextLineStart == length && input[length - 1] == '\n'))
 397         {
 398             result.append("\n");
 399         }
 400         lineStart = nextLineStart;
 401     }
 402     return result;
 403 }
 404
 405 std::vector<std::string>
 406 TextLineWrapper::wrapToVector(const std::string &input) const
 407 {
 408     std::vector<std::string> result;
 409     size_t                   lineStart = 0;
 410     size_t                   length    = input.length();
 411     while (lineStart < length)
 412     {
 413         size_t nextLineStart = findNextLine(input, lineStart);
 414         result.push_back(formatLine(input, lineStart, nextLineStart));
 415         lineStart = nextLineStart;
 416     }
 417     return result;
 418 }
 419
 420 } // namespace gmx