3 Copyright (C) 2010-2013 celeron55, Perttu Ahola <celeron55@gmail.com>
5 This program is free software; you can redistribute it and/or modify
6 it under the terms of the GNU Lesser General Public License as published by
7 the Free Software Foundation; either version 2.1 of the License, or
8 (at your option) any later version.
10 This program is distributed in the hope that it will be useful,
11 but WITHOUT ANY WARRANTY; without even the implied warranty of
12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 GNU Lesser General Public License for more details.
15 You should have received a copy of the GNU Lesser General Public License along
16 with this program; if not, write to the Free Software Foundation, Inc.,
17 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
20 #ifndef UTIL_STRING_HEADER
21 #define UTIL_STRING_HEADER
23 #include "irrlichttypes_bloated.h"
32 #define STRINGIFY(x) #x
33 #define TOSTRING(x) STRINGIFY(x)
35 // Checks whether a value is an ASCII printable character
36 #define IS_ASCII_PRINTABLE_CHAR(x) \
37 (((unsigned int)(x) >= 0x20) && \
38 ( (unsigned int)(x) <= 0x7e))
40 // Checks whether a byte is an inner byte for an utf-8 multibyte sequence
41 #define IS_UTF8_MULTB_INNER(x) \
42 (((unsigned char)(x) >= 0x80) && \
43 ( (unsigned char)(x) <= 0xbf))
45 // Checks whether a byte is a start byte for an utf-8 multibyte sequence
46 #define IS_UTF8_MULTB_START(x) \
47 (((unsigned char)(x) >= 0xc2) && \
48 ( (unsigned char)(x) <= 0xf4))
50 // Given a start byte x for an utf-8 multibyte sequence
51 // it gives the length of the whole sequence in bytes.
52 #define UTF8_MULTB_START_LEN(x) \
53 (((unsigned char)(x) < 0xe0) ? 2 : \
54 (((unsigned char)(x) < 0xf0) ? 3 : 4))
56 typedef std::map<std::string, std::string> StringMap;
63 // try not to convert between wide/utf8 encodings; this can result in data loss
64 // try to only convert between them when you need to input/output stuff via Irrlicht
65 std::wstring utf8_to_wide(const std::string &input);
66 std::string wide_to_utf8(const std::wstring &input);
68 wchar_t *utf8_to_wide_c(const char *str);
70 // NEVER use those two functions unless you have a VERY GOOD reason to
71 // they just convert between wide and multibyte encoding
72 // multibyte encoding depends on current locale, this is no good, especially on Windows
74 // You must free the returned string!
75 // The returned string is allocated using new
76 wchar_t *narrow_to_wide_c(const char *str);
77 std::wstring narrow_to_wide(const std::string &mbs);
78 std::string wide_to_narrow(const std::wstring &wcs);
80 std::string urlencode(std::string str);
81 std::string urldecode(std::string str);
82 u32 readFlagString(std::string str, const FlagDesc *flagdesc, u32 *flagmask);
83 std::string writeFlagString(u32 flags, const FlagDesc *flagdesc, u32 flagmask);
84 size_t mystrlcpy(char *dst, const char *src, size_t size);
85 char *mystrtok_r(char *s, const char *sep, char **lasts);
86 u64 read_seed(const char *str);
87 bool parseColorString(const std::string &value, video::SColor &color, bool quiet);
91 * Returns a copy of \p str with spaces inserted at the right hand side to ensure
92 * that the string is \p len characters in length. If \p str is <= \p len then the
93 * returned string will be identical to str.
95 inline std::string padStringRight(std::string str, size_t len)
98 str.insert(str.end(), len - str.size(), ' ');
104 * Returns a version of \p str with the first occurrence of a string
105 * contained within ends[] removed from the end of the string.
108 * @param ends A NULL- or ""- terminated array of strings to remove from s in
109 * the copy produced. Note that once one of these strings is removed
110 * that no further postfixes contained within this array are removed.
112 * @return If no end could be removed then "" is returned.
114 inline std::string removeStringEnd(const std::string &str,
117 const char **p = ends;
119 for (; *p && (*p)[0] != '\0'; p++) {
120 std::string end = *p;
121 if (str.size() < end.size())
123 if (str.compare(str.size() - end.size(), end.size(), end) == 0)
124 return str.substr(0, str.size() - end.size());
132 * Check two strings for equivalence. If \p case_insensitive is true
133 * then the case of the strings is ignored (default is false).
137 * @param case_insensitive
138 * @return true if the strings match
140 template <typename T>
141 inline bool str_equal(const std::basic_string<T> &s1,
142 const std::basic_string<T> &s2,
143 bool case_insensitive = false)
145 if (!case_insensitive)
148 if (s1.size() != s2.size())
151 for (size_t i = 0; i < s1.size(); ++i)
152 if(tolower(s1[i]) != tolower(s2[i]))
160 * Check whether \p str begins with the string prefix. If \p case_insensitive
161 * is true then the check is case insensitve (default is false; i.e. case is
166 * @param case_insensitive
167 * @return true if the str begins with prefix
169 template <typename T>
170 inline bool str_starts_with(const std::basic_string<T> &str,
171 const std::basic_string<T> &prefix,
172 bool case_insensitive = false)
174 if (str.size() < prefix.size())
177 if (!case_insensitive)
178 return str.compare(0, prefix.size(), prefix) == 0;
180 for (size_t i = 0; i < prefix.size(); ++i)
181 if (tolower(str[i]) != tolower(prefix[i]))
187 * Check whether \p str begins with the string prefix. If \p case_insensitive
188 * is true then the check is case insensitve (default is false; i.e. case is
193 * @param case_insensitive
194 * @return true if the str begins with prefix
196 template <typename T>
197 inline bool str_starts_with(const std::basic_string<T> &str,
199 bool case_insensitive = false)
201 return str_starts_with(str, std::basic_string<T>(prefix),
206 * Splits a string into its component parts separated by the character
209 * @return An std::vector<std::basic_string<T> > of the component parts
211 template <typename T>
212 inline std::vector<std::basic_string<T> > str_split(
213 const std::basic_string<T> &str,
216 std::vector<std::basic_string<T> > parts;
217 std::basic_stringstream<T> sstr(str);
218 std::basic_string<T> part;
220 while (std::getline(sstr, part, delimiter))
221 parts.push_back(part);
229 * @return A copy of \p str converted to all lowercase characters.
231 inline std::string lowercase(const std::string &str)
235 s2.reserve(str.size());
237 for (size_t i = 0; i < str.size(); i++)
238 s2 += tolower(str[i]);
246 * @return A copy of \p str with leading and trailing whitespace removed.
248 inline std::string trim(const std::string &str)
252 while (std::isspace(str[front]))
255 size_t back = str.size();
256 while (back > front && std::isspace(str[back - 1]))
259 return str.substr(front, back - front);
264 * Returns whether \p str should be regarded as (bool) true. Case and leading
265 * and trailing whitespace are ignored. Values that will return
266 * true are "y", "yes", "true" and any number that is not 0.
269 inline bool is_yes(const std::string &str)
271 std::string s2 = lowercase(trim(str));
273 return s2 == "y" || s2 == "yes" || s2 == "true" || atoi(s2.c_str()) != 0;
278 * Converts the string \p str to a signed 32-bit integer. The converted value
279 * is constrained so that min <= value <= max.
281 * @see atoi(3) for limitations
284 * @param min Range minimum
285 * @param max Range maximum
286 * @return The value converted to a signed 32-bit integer and constrained
287 * within the range defined by min and max (inclusive)
289 inline s32 mystoi(const std::string &str, s32 min, s32 max)
291 s32 i = atoi(str.c_str());
302 // MSVC2010 includes it's own versions of these
303 //#if !defined(_MSC_VER) || _MSC_VER < 1600
307 * Returns a 32-bit value reprensented by the string \p str (decimal).
308 * @see atoi(3) for further limitations
310 inline s32 mystoi(const std::string &str)
312 return atoi(str.c_str());
317 * Returns s 32-bit value represented by the wide string \p str (decimal).
318 * @see atoi(3) for further limitations
320 inline s32 mystoi(const std::wstring &str)
322 return mystoi(wide_to_narrow(str));
327 * Returns a float reprensented by the string \p str (decimal).
330 inline float mystof(const std::string &str)
332 return atof(str.c_str());
340 /// Returns a value represented by the string \p val.
341 template <typename T>
342 inline T from_string(const std::string &str)
344 std::stringstream tmp(str);
350 /// Returns a 64-bit signed value represented by the string \p str (decimal).
351 inline s64 stoi64(const std::string &str) { return from_string<s64>(str); }
353 // TODO: Replace with C++11 std::to_string.
355 /// Returns a string representing the value \p val.
356 template <typename T>
357 inline std::string to_string(T val)
359 std::ostringstream oss;
364 /// Returns a string representing the decimal value of the 32-bit value \p i.
365 inline std::string itos(s32 i) { return to_string(i); }
366 /// Returns a string representing the decimal value of the 64-bit value \p i.
367 inline std::string i64tos(s64 i) { return to_string(i); }
368 /// Returns a string representing the decimal value of the float value \p f.
369 inline std::string ftos(float f) { return to_string(f); }
373 * Replace all occurrences of \p pattern in \p str with \p replacement.
375 * @param str String to replace pattern with replacement within.
376 * @param pattern The pattern to replace.
377 * @param replacement What to replace the pattern with.
379 inline void str_replace(std::string &str, const std::string &pattern,
380 const std::string &replacement)
382 std::string::size_type start = str.find(pattern, 0);
383 while (start != str.npos) {
384 str.replace(start, pattern.size(), replacement);
385 start = str.find(pattern, start + replacement.size());
390 * Replace all occurrences of the character \p from in \p str with \p to.
392 * @param str The string to (potentially) modify.
393 * @param from The character in str to replace.
394 * @param to The replacement character.
396 void str_replace(std::string &str, char from, char to);
400 * Check that a string only contains whitelisted characters. This is the
401 * opposite of string_allowed_blacklist().
403 * @param str The string to be checked.
404 * @param allowed_chars A string containing permitted characters.
405 * @return true if the string is allowed, otherwise false.
407 * @see string_allowed_blacklist()
409 inline bool string_allowed(const std::string &str, const std::string &allowed_chars)
411 return str.find_first_not_of(allowed_chars) == str.npos;
416 * Check that a string contains no blacklisted characters. This is the
417 * opposite of string_allowed().
419 * @param str The string to be checked.
420 * @param blacklisted_chars A string containing prohibited characters.
421 * @return true if the string is allowed, otherwise false.
423 * @see string_allowed()
425 inline bool string_allowed_blacklist(const std::string &str,
426 const std::string &blacklisted_chars)
428 return str.find_first_of(blacklisted_chars) == str.npos;
433 * Create a string based on \p from where a newline is forcefully inserted
434 * every \p row_len characters.
436 * @note This function does not honour word wraps and blindy inserts a newline
437 * every \p row_len characters whether it breaks a word or not. It is
438 * intended to be used for, for example, showing paths in the GUI.
440 * @note This function doesn't wrap inside utf-8 multibyte sequences and also
441 * counts multibyte sequences correcly as single characters.
443 * @param from The (utf-8) string to be wrapped into rows.
444 * @param row_len The row length (in characters).
445 * @return A new string with the wrapping applied.
447 inline std::string wrap_rows(const std::string &from,
452 size_t character_idx = 0;
453 for (size_t i = 0; i < from.size(); i++) {
454 if (!IS_UTF8_MULTB_INNER(from[i])) {
455 // Wrap string after last inner byte of char
456 if (character_idx > 0 && character_idx % row_len == 0)
468 * Removes backslashes from an escaped string (FormSpec strings)
470 template <typename T>
471 inline std::basic_string<T> unescape_string(const std::basic_string<T> &s)
473 std::basic_string<T> res;
475 for (size_t i = 0; i < s.length(); i++) {
488 * Remove all escape sequences in \p s.
490 * @param s The string in which to remove escape sequences.
491 * @return \p s, with escape sequences removed.
493 template <typename T>
494 std::basic_string<T> unescape_enriched(const std::basic_string<T> &s)
496 std::basic_string<T> output;
498 while (i < s.length()) {
499 if (s[i] == '\x1b') {
501 if (i == s.length()) continue;
504 while (i < s.length() && s[i] != ')') {
522 template <typename T>
523 std::vector<std::basic_string<T> > split(const std::basic_string<T> &s, T delim)
525 std::vector<std::basic_string<T> > tokens;
527 std::basic_string<T> current;
528 bool last_was_escape = false;
529 for (size_t i = 0; i < s.length(); i++) {
531 if (last_was_escape) {
534 last_was_escape = false;
537 tokens.push_back(current);
538 current = std::basic_string<T>();
539 last_was_escape = false;
540 } else if (si == '\\') {
541 last_was_escape = true;
544 last_was_escape = false;
549 tokens.push_back(current);
555 * Checks that all characters in \p to_check are a decimal digits.
558 * @return true if to_check is not empty and all characters in to_check are
559 * decimal digits, otherwise false
561 inline bool is_number(const std::string &to_check)
563 for (size_t i = 0; i < to_check.size(); i++)
564 if (!std::isdigit(to_check[i]))
567 return !to_check.empty();
572 * Returns a C-string, either "true" or "false", corresponding to \p val.
574 * @return If \p val is true, then "true" is returned, otherwise "false".
576 inline const char *bool_to_cstr(bool val)
578 return val ? "true" : "false";