2 * This file is part of the GROMACS molecular simulation package.
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.
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.
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.
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.
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.
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.
37 * Declares common string utility and formatting routines.
39 * \author Teemu Murtola <teemu.murtola@gmail.com>
41 * \ingroup module_utility
43 #ifndef GMX_UTILITY_STRINGUTIL_H
44 #define GMX_UTILITY_STRINGUTIL_H
54 //! \addtogroup module_utility
58 * Tests whether a string is null or empty.
62 static inline bool isNullOrEmpty(const char *str)
64 return str == NULL || str[0] == '\0';
68 * Tests whether a string starts with another string.
70 * \param[in] str String to process.
71 * \param[in] prefix Prefix to find.
72 * \returns true if \p str starts with \p prefix.
74 * Returns true if \p prefix is empty.
77 static inline bool startsWith(const std::string &str, const std::string &prefix)
79 return str.compare(0, prefix.length(), prefix) == 0;
81 //! \copydoc startsWith(const std::string &, const std::string &)
82 static inline bool startsWith(const char *str, const char *prefix)
84 return std::strncmp(str, prefix, std::strlen(prefix)) == 0;
88 * Tests whether a string ends with another string.
90 * \param[in] str String to process.
91 * \param[in] suffix Suffix to find.
92 * \returns true if \p str ends with \p suffix.
94 * Returns true if \p suffix is NULL or empty.
97 bool endsWith(const char *str, const char *suffix);
98 //! \copydoc endsWith(const char *, const char *)
99 static inline bool endsWith(const std::string &str, const char *suffix)
101 return endsWith(str.c_str(), suffix);
105 * Removes a suffix from a string.
107 * \param[in] str String to process.
108 * \param[in] suffix Suffix to remove.
109 * \returns \p str with \p suffix removed, or \p str unmodified if it does
110 * not end with \p suffix.
111 * \throws std::bad_alloc if out of memory.
113 * Returns \p str if \p suffix is NULL or empty.
115 std::string stripSuffixIfPresent(const std::string &str, const char *suffix);
117 * Removes leading and trailing whitespace from a string.
119 * \param[in] str String to process.
120 * \returns \p str with leading and trailing whitespaces removed.
121 * \throws std::bad_alloc if out of memory.
123 std::string stripString(const std::string &str);
126 * Formats a string (snprintf() wrapper).
128 * \throws std::bad_alloc if out of memory.
130 * This function works like sprintf(), except that it returns an std::string
131 * instead of requiring a preallocated buffer. Arbitrary length output is
134 std::string formatString(const char *fmt, ...);
136 /*! \brief Function object that wraps a call to formatString() that
137 * expects a single conversion argument, for use with algorithms. */
138 class StringFormatter
141 /*! \brief Constructor
143 * \param[in] format The printf-style format string that will
144 * be applied to convert values of type T to
145 * string. Exactly one argument to the conversion
146 * specification(s) in `format` is supported. */
147 explicit StringFormatter(const char *format) : format_(format)
151 //! Implements the formatting functionality
152 template <typename T>
153 std::string operator()(const T &value) const
155 return formatString(format_, value);
159 //! Format string to use
163 /*! \brief Function object to implement the same interface as
164 * `StringFormatter` to use with strings that should not be formatted
166 class IdentityFormatter
169 //! Implements the formatting non-functionality
170 std::string operator()(const std::string &value) const
176 /*! \brief Formats all the range as strings, and then joins them with
177 * a separator in between.
179 * \param[in] begin Iterator the beginning of the range to join.
180 * \param[in] end Iterator the end of the range to join.
181 * \param[in] separator String to put in between the joined strings.
182 * \param[in] formatter Function object to format the objects in
183 * `container` as strings
184 * \returns All objects in the range from `begin` to `end` formatted
185 * as strings and concatenated with `separator` between each pair.
186 * \throws std::bad_alloc if out of memory.
188 template <typename InputIterator, typename FormatterType>
189 std::string formatAndJoin(InputIterator begin, InputIterator end, const char *separator, const FormatterType &formatter)
192 const char *currentSeparator = "";
193 for (InputIterator i = begin; i != end; ++i)
195 result.append(currentSeparator);
196 result.append(formatter(*i));
197 currentSeparator = separator;
202 /*! \brief Formats all elements of the container as strings, and then
203 * joins them with a separator in between.
205 * \param[in] container Objects to join.
206 * \param[in] separator String to put in between the joined strings.
207 * \param[in] formatter Function object to format the objects in
208 * `container` as strings
209 * \returns All objects from `container` formatted as strings and
210 * concatenated with `separator` between each pair.
211 * \throws std::bad_alloc if out of memory.
213 template <typename ContainerType, typename FormatterType>
214 std::string formatAndJoin(const ContainerType &container, const char *separator, const FormatterType &formatter)
216 return formatAndJoin(container.begin(), container.end(), separator, formatter);
220 * Joins strings from a range with a separator in between.
222 * \param[in] begin Iterator the beginning of the range to join.
223 * \param[in] end Iterator the end of the range to join.
224 * \param[in] separator String to put in between the joined strings.
225 * \returns All strings from (`begin`, `end`) concatenated with `separator`
227 * \throws std::bad_alloc if out of memory.
229 template <typename InputIterator>
230 std::string joinStrings(InputIterator begin, InputIterator end,
231 const char *separator)
233 return formatAndJoin(begin, end, separator, IdentityFormatter());
237 * Joins strings from a container with a separator in between.
239 * \param[in] container Strings to join.
240 * \param[in] separator String to put in between the joined strings.
241 * \returns All strings from `container` concatenated with `separator`
243 * \throws std::bad_alloc if out of memory.
245 template <typename ContainerType>
246 std::string joinStrings(const ContainerType &container, const char *separator)
248 return joinStrings(container.begin(), container.end(), separator);
252 * Joins strings from an array with a separator in between.
254 * \param[in] array Array of strings to join.
255 * \param[in] separator String to put in between the joined strings.
256 * \tparam count Deduced number of elements in \p array.
257 * \returns All strings from `aray` concatenated with `separator`
259 * \throws std::bad_alloc if out of memory.
261 template <size_t count>
262 std::string joinStrings(const char *const (&array)[count], const char *separator)
264 return joinStrings(array, array + count, separator);
268 * Splits a string to whitespace separated tokens.
270 * \param[in] str String to process.
271 * \returns \p str split into tokens at each whitespace sequence.
272 * \throws std::bad_alloc if out of memory.
274 * This function works like `split` in Python, i.e., leading and trailing
275 * whitespace is ignored, and consecutive whitespaces are treated as a single
278 std::vector<std::string> splitString(const std::string &str);
281 * Replace all occurrences of a string with another string.
283 * \param[in] input Input string.
284 * \param[in] from String to find.
285 * \param[in] to String to use to replace \p from.
286 * \returns Copy of \p input with all occurrences of \p from replaced with \p to.
287 * \throws std::bad_alloc if out of memory.
289 * The replacement is greedy and not recursive: starting from the beginning of
290 * \p input, each match of \p from is replaced with \p to, and the search for
291 * the next match begins after the end of the previous match.
293 * Compexity is O(N), where N is length of output.
295 * \see replaceAllWords()
297 std::string replaceAll(const std::string &input,
298 const char *from, const char *to);
299 //! \copydoc replaceAll(const std::string &, const char *, const char *)
300 std::string replaceAll(const std::string &input,
301 const std::string &from, const std::string &to);
303 * Replace whole words with others.
305 * \param[in] input Input string.
306 * \param[in] from String to find.
307 * \param[in] to String to use to replace \p from.
308 * \returns Copy of \p input with all \p from words replaced with \p to.
309 * \throws std::bad_alloc if out of memory.
311 * Works as replaceAll(), but a match is only considered if it is delimited by
312 * non-alphanumeric characters.
316 std::string replaceAllWords(const std::string &input,
317 const char *from, const char *to);
318 //! \copydoc replaceAllWords(const std::string &, const char *, const char *)
319 std::string replaceAllWords(const std::string &input,
320 const std::string &from, const std::string &to);
322 class TextLineWrapper;
325 * Stores settings for line wrapping.
327 * Methods in this class do not throw.
329 * \see TextLineWrapper
333 class TextLineWrapperSettings
337 * Initializes default wrapper settings.
339 * Default settings are:
340 * - No maximum line width (only explicit line breaks).
342 * - No continuation characters.
343 * - Do not keep final spaces in input strings.
345 TextLineWrapperSettings();
348 * Sets the maximum length for output lines.
350 * \param[in] length Maximum length for the lines after wrapping.
352 * If this method is not called, or is called with zero \p length, the
353 * wrapper has no maximum length (only wraps at explicit line breaks).
355 void setLineLength(int length) { maxLength_ = length; }
357 * Sets the indentation for output lines.
359 * \param[in] indent Number of spaces to add for indentation.
361 * If this method is not called, the wrapper does not add indentation.
363 void setIndent(int indent) { indent_ = indent; }
365 * Sets the indentation for first output line after a line break.
367 * \param[in] indent Number of spaces to add for indentation.
369 * If this method is not called, or called with \p indent equal to -1,
370 * the value set with setIndent() is used.
372 void setFirstLineIndent(int indent) { firstLineIndent_ = indent; }
374 * Sets whether final spaces in input should be kept.
376 * \param[in] bKeep Whether to keep spaces at the end of the input.
378 * This means that wrapping a string that ends in spaces also keeps
379 * those spaces in the output. This allows using the wrapper for
380 * partial lines where the initial part of the line may end in a space.
381 * By default, all trailing whitespace is removed. Note that this
382 * option does not affect spaces before an explicit newline: those are
385 void setKeepFinalSpaces(bool bKeep) { bKeepFinalSpaces_ = bKeep; }
387 * Sets a continuation marker for wrapped lines.
389 * \param[in] continuationChar Character to use to mark continuation
392 * If set to non-zero character code, this character is added at the
393 * end of each line where a line break is added by TextLineWrapper
394 * (but not after lines produced by explicit line breaks).
395 * The default (\c '\0') is to not add continuation markers.
397 * Note that currently, the continuation char may cause the output line
398 * length to exceed the value set with setLineLength() by at most two
401 void setContinuationChar(char continuationChar)
403 continuationChar_ = continuationChar;
406 //! Returns the maximum length set with setLineLength().
407 int lineLength() const { return maxLength_; }
408 //! Returns the indentation set with setIndent().
409 int indent() const { return indent_; }
411 * Returns the indentation set with setFirstLineIndent().
413 * If setFirstLineIndent() has not been called or has been called with
414 * -1, indent() is returned.
416 int firstLineIndent() const
418 return (firstLineIndent_ >= 0 ? firstLineIndent_ : indent_);
422 //! Maximum length of output lines, or <= 0 if no limit.
424 //! Number of spaces to indent each output line with.
427 * Number of spaces to indent the first line after a newline.
429 * If -1, \a indent_ is used.
431 int firstLineIndent_;
432 //! Whether to keep spaces at end of input.
433 bool bKeepFinalSpaces_;
434 //! If not \c '\0', mark each wrapping point with this character.
435 char continuationChar_;
437 //! Needed to access the members.
438 friend class TextLineWrapper;
442 * Wraps lines to a predefined length.
444 * This utility class wraps lines at word breaks to produce lines that are not
445 * longer than a predefined length. Explicit newlines ('\\n') are preserved.
446 * Only space is considered a word separator. If a single word exceeds the
447 * maximum line length, it is still printed on a single line.
448 * Extra whitespace is stripped from the end of produced lines.
449 * Other options on the wrapping, such as the line length or indentation,
450 * can be changed using a TextLineWrapperSettings object.
452 * Two interfaces to do the wrapping are provided:
453 * -# High-level interface using either wrapToString() (produces a single
454 * string with embedded newlines) or wrapToVector() (produces a vector of
455 * strings with each line as one element).
456 * These methods operate on std::string and wrap the entire input string.
457 * -# Low-level interface using findNextLine() and formatLine().
458 * findNextLine() operates either on a C string or an std::string, and does
459 * not do any memory allocation (so it does not throw). It finds the next
460 * line to be wrapped, considering the wrapping settings.
461 * formatLine() does whitespace operations on the line found by
462 * findNextLine() and returns an std::string.
463 * These methods allow custom wrapping implementation to either avoid
464 * exceptions or to wrap only a part of the input string.
468 gmx::TextLineWrapper wrapper;
469 wrapper.settings().setLineLength(78);
470 printf("%s\n", wrapper.wrapToString(textToWrap).c_str());
475 class TextLineWrapper
479 * Constructs a new line wrapper with default settings.
487 * Constructs a new line wrapper with given settings.
489 * \param[in] settings Wrapping settings.
493 explicit TextLineWrapper(const TextLineWrapperSettings &settings)
494 : settings_(settings)
499 * Provides access to settings of this wrapper.
501 * \returns The settings object for this wrapper.
503 * The returned object can be used to modify settings for the wrapper.
504 * All subsequent calls to wrapToString() and wrapToVector() use the
509 TextLineWrapperSettings &settings() { return settings_; }
511 //! Returns true if the wrapper would not modify the input string.
512 bool isTrivial() const;
515 * Finds the next line to be wrapped.
517 * \param[in] input String to wrap.
518 * \param[in] lineStart Index of first character of the line to find.
519 * \returns Index of first character of the next line.
521 * If this is the last line, returns the length of \p input.
522 * In determining the length of the returned line, this function
523 * considers the maximum line length, leaving space for indentation,
524 * and also whitespace stripping behavior.
525 * Thus, the line returned may be longer than the maximum line length
526 * if it has leading and/or trailing space.
527 * When wrapping a line on a space (not on an explicit line break),
528 * the returned index is always on a non-whitespace character after the
531 * To iterate over lines in a string, use the following code:
533 gmx::TextLineWrapper wrapper;
534 // <set desired wrapping settings>
535 size_t lineStart = 0;
536 size_t length = input.length();
537 while (lineStart < length)
539 size_t nextLineStart = wrapper.findNextLine(input, lineStart);
540 std::string line = wrapper.formatLine(input, lineStart, nextLineStart));
541 // <do something with the line>
542 lineStart = nextLineStart;
549 size_t findNextLine(const char *input, size_t lineStart) const;
550 //! \copydoc findNextLine(const char *, size_t)const
551 size_t findNextLine(const std::string &input, size_t lineStart) const;
553 * Formats a single line for output according to wrapping settings.
555 * \param[in] input Input string.
556 * \param[in] lineStart Index of first character of the line to format.
557 * \param[in] lineEnd Index of first character of the next line.
558 * \returns The line with leading and/or trailing whitespace removed
559 * and indentation applied.
560 * \throws std::bad_alloc if out of memory.
562 * Intended to be used on the lines found by findNextLine().
563 * When used with the lines returned from findNextLine(), the returned
564 * line conforms to the wrapper settings.
565 * Trailing whitespace is always stripped (including any newlines,
566 * i.e., the return value does not contain a newline).
568 std::string formatLine(const std::string &input,
569 size_t lineStart, size_t lineEnd) const;
572 * Formats a string, producing a single string with all the lines.
574 * \param[in] input String to wrap.
575 * \returns \p input with added newlines such that maximum line
576 * length is not exceeded.
577 * \throws std::bad_alloc if out of memory.
579 * Newlines in the input are preserved, including terminal newlines.
580 * Note that if the input does not contain a terminal newline, the
581 * output does not either.
583 std::string wrapToString(const std::string &input) const;
585 * Formats a string, producing a vector with all the lines.
587 * \param[in] input String to wrap.
588 * \returns \p input split into lines such that maximum line length
590 * \throws std::bad_alloc if out of memory.
592 * The strings in the returned vector do not contain newlines at the
594 * Note that a single terminal newline does not affect the output:
595 * "line\\n" and "line" both produce the same output (but "line\\n\\n"
596 * produces two lines, the second of which is empty).
598 std::vector<std::string> wrapToVector(const std::string &input) const;
601 TextLineWrapperSettings settings_;