OpenTTD
string.cpp
Go to the documentation of this file.
1 /* $Id: string.cpp 27380 2015-08-10 20:21:29Z michi_cc $ */
2 
3 /*
4  * This file is part of OpenTTD.
5  * OpenTTD is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, version 2.
6  * OpenTTD is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
7  * See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with OpenTTD. If not, see <http://www.gnu.org/licenses/>.
8  */
9 
12 #include "stdafx.h"
13 #include "debug.h"
14 #include "core/alloc_func.hpp"
15 #include "core/math_func.hpp"
16 #include "string_func.h"
17 #include "string_base.h"
18 
19 #include "table/control_codes.h"
20 
21 #include <stdarg.h>
22 #include <ctype.h> /* required for tolower() */
23 
24 #ifdef _MSC_VER
25 #include <errno.h> // required by vsnprintf implementation for MSVC
26 #endif
27 
28 #ifdef WITH_ICU_SORT
29 /* Required by strnatcmp. */
30 #include <unicode/ustring.h>
31 #include "language.h"
32 #include "gfx_func.h"
33 #endif /* WITH_ICU_SORT */
34 
35 /* The function vsnprintf is used internally to perform the required formatting
36  * tasks. As such this one must be allowed, and makes sure it's terminated. */
37 #include "safeguards.h"
38 #undef vsnprintf
39 
50 int CDECL vseprintf(char *str, const char *last, const char *format, va_list ap)
51 {
52  ptrdiff_t diff = last - str;
53  if (diff < 0) return 0;
54  return min((int)diff, vsnprintf(str, diff + 1, format, ap));
55 }
56 
73 char *strecat(char *dst, const char *src, const char *last)
74 {
75  assert(dst <= last);
76  while (*dst != '\0') {
77  if (dst == last) return dst;
78  dst++;
79  }
80 
81  return strecpy(dst, src, last);
82 }
83 
84 
101 char *strecpy(char *dst, const char *src, const char *last)
102 {
103  assert(dst <= last);
104  while (dst != last && *src != '\0') {
105  *dst++ = *src++;
106  }
107  *dst = '\0';
108 
109  if (dst == last && *src != '\0') {
110 #if defined(STRGEN) || defined(SETTINGSGEN)
111  error("String too long for destination buffer");
112 #else /* STRGEN || SETTINGSGEN */
113  DEBUG(misc, 0, "String too long for destination buffer");
114 #endif /* STRGEN || SETTINGSGEN */
115  }
116  return dst;
117 }
118 
126 char *stredup(const char *s, const char *last)
127 {
128  size_t len = last == NULL ? strlen(s) : ttd_strnlen(s, last - s + 1);
129  char *tmp = CallocT<char>(len + 1);
130  memcpy(tmp, s, len);
131  return tmp;
132 }
133 
139 char *CDECL str_fmt(const char *str, ...)
140 {
141  char buf[4096];
142  va_list va;
143 
144  va_start(va, str);
145  int len = vseprintf(buf, lastof(buf), str, va);
146  va_end(va);
147  char *p = MallocT<char>(len + 1);
148  memcpy(p, buf, len + 1);
149  return p;
150 }
151 
158 void str_fix_scc_encoded(char *str, const char *last)
159 {
160  while (str <= last && *str != '\0') {
161  size_t len = Utf8EncodedCharLen(*str);
162  if ((len == 0 && str + 4 > last) || str + len > last) break;
163 
164  WChar c;
165  len = Utf8Decode(&c, str);
166  if (c == '\0') break;
167 
168  if (c == 0xE028 || c == 0xE02A) {
169  c = SCC_ENCODED;
170  }
171  str += Utf8Encode(str, c);
172  }
173  *str = '\0';
174 }
175 
176 
184 void str_validate(char *str, const char *last, StringValidationSettings settings)
185 {
186  /* Assume the ABSOLUTE WORST to be in str as it comes from the outside. */
187 
188  char *dst = str;
189  while (str <= last && *str != '\0') {
190  size_t len = Utf8EncodedCharLen(*str);
191  /* If the character is unknown, i.e. encoded length is 0
192  * we assume worst case for the length check.
193  * The length check is needed to prevent Utf8Decode to read
194  * over the terminating '\0' if that happens to be placed
195  * within the encoding of an UTF8 character. */
196  if ((len == 0 && str + 4 > last) || str + len > last) break;
197 
198  WChar c;
199  len = Utf8Decode(&c, str);
200  /* It's possible to encode the string termination character
201  * into a multiple bytes. This prevents those termination
202  * characters to be skipped */
203  if (c == '\0') break;
204 
205  if ((IsPrintable(c) && (c < SCC_SPRITE_START || c > SCC_SPRITE_END)) || ((settings & SVS_ALLOW_CONTROL_CODE) != 0 && c == SCC_ENCODED)) {
206  /* Copy the character back. Even if dst is current the same as str
207  * (i.e. no characters have been changed) this is quicker than
208  * moving the pointers ahead by len */
209  do {
210  *dst++ = *str++;
211  } while (--len != 0);
212  } else if ((settings & SVS_ALLOW_NEWLINE) != 0 && c == '\n') {
213  *dst++ = *str++;
214  } else {
215  if ((settings & SVS_ALLOW_NEWLINE) != 0 && c == '\r' && str[1] == '\n') {
216  str += len;
217  continue;
218  }
219  /* Replace the undesirable character with a question mark */
220  str += len;
221  if ((settings & SVS_REPLACE_WITH_QUESTION_MARK) != 0) *dst++ = '?';
222  }
223  }
224 
225  *dst = '\0';
226 }
227 
233 void ValidateString(const char *str)
234 {
235  /* We know it is '\0' terminated. */
236  str_validate(const_cast<char *>(str), str + strlen(str) + 1);
237 }
238 
239 
247 bool StrValid(const char *str, const char *last)
248 {
249  /* Assume the ABSOLUTE WORST to be in str as it comes from the outside. */
250 
251  while (str <= last && *str != '\0') {
252  size_t len = Utf8EncodedCharLen(*str);
253  /* Encoded length is 0 if the character isn't known.
254  * The length check is needed to prevent Utf8Decode to read
255  * over the terminating '\0' if that happens to be placed
256  * within the encoding of an UTF8 character. */
257  if (len == 0 || str + len > last) return false;
258 
259  WChar c;
260  len = Utf8Decode(&c, str);
261  if (!IsPrintable(c) || (c >= SCC_SPRITE_START && c <= SCC_SPRITE_END)) {
262  return false;
263  }
264 
265  str += len;
266  }
267 
268  return *str == '\0';
269 }
270 
272 void str_strip_colours(char *str)
273 {
274  char *dst = str;
275  WChar c;
276  size_t len;
277 
278  for (len = Utf8Decode(&c, str); c != '\0'; len = Utf8Decode(&c, str)) {
279  if (c < SCC_BLUE || c > SCC_BLACK) {
280  /* Copy the character back. Even if dst is current the same as str
281  * (i.e. no characters have been changed) this is quicker than
282  * moving the pointers ahead by len */
283  do {
284  *dst++ = *str++;
285  } while (--len != 0);
286  } else {
287  /* Just skip (strip) the colour codes */
288  str += len;
289  }
290  }
291  *dst = '\0';
292 }
293 
300 size_t Utf8StringLength(const char *s)
301 {
302  size_t len = 0;
303  const char *t = s;
304  while (Utf8Consume(&t) != 0) len++;
305  return len;
306 }
307 
308 
320 bool strtolower(char *str)
321 {
322  bool changed = false;
323  for (; *str != '\0'; str++) {
324  char new_str = tolower(*str);
325  changed |= new_str != *str;
326  *str = new_str;
327  }
328  return changed;
329 }
330 
338 bool IsValidChar(WChar key, CharSetFilter afilter)
339 {
340  switch (afilter) {
341  case CS_ALPHANUMERAL: return IsPrintable(key);
342  case CS_NUMERAL: return (key >= '0' && key <= '9');
343  case CS_NUMERAL_SPACE: return (key >= '0' && key <= '9') || key == ' ';
344  case CS_ALPHA: return IsPrintable(key) && !(key >= '0' && key <= '9');
345  case CS_HEXADECIMAL: return (key >= '0' && key <= '9') || (key >= 'a' && key <= 'f') || (key >= 'A' && key <= 'F');
346  }
347 
348  return false;
349 }
350 
351 #ifdef WIN32
352 #if defined(_MSC_VER) && _MSC_VER < 1900
353 
360 int CDECL vsnprintf(char *str, size_t size, const char *format, va_list ap)
361 {
362  if (size == 0) return 0;
363 
364  errno = 0;
365  int ret = _vsnprintf(str, size, format, ap);
366 
367  if (ret < 0) {
368  if (errno != ERANGE) {
369  /* There's a formatting error, better get that looked
370  * at properly instead of ignoring it. */
371  NOT_REACHED();
372  }
373  } else if ((size_t)ret < size) {
374  /* The buffer is big enough for the number of
375  * characters stored (excluding null), i.e.
376  * the string has been null-terminated. */
377  return ret;
378  }
379 
380  /* The buffer is too small for _vsnprintf to write the
381  * null-terminator at its end and return size. */
382  str[size - 1] = '\0';
383  return (int)size;
384 }
385 #endif /* _MSC_VER */
386 
387 #endif /* WIN32 */
388 
398 int CDECL seprintf(char *str, const char *last, const char *format, ...)
399 {
400  va_list ap;
401 
402  va_start(ap, format);
403  int ret = vseprintf(str, last, format, ap);
404  va_end(ap);
405  return ret;
406 }
407 
408 
416 char *md5sumToString(char *buf, const char *last, const uint8 md5sum[16])
417 {
418  char *p = buf;
419 
420  for (uint i = 0; i < 16; i++) {
421  p += seprintf(p, last, "%02X", md5sum[i]);
422  }
423 
424  return p;
425 }
426 
427 
428 /* UTF-8 handling routines */
429 
430 
437 size_t Utf8Decode(WChar *c, const char *s)
438 {
439  assert(c != NULL);
440 
441  if (!HasBit(s[0], 7)) {
442  /* Single byte character: 0xxxxxxx */
443  *c = s[0];
444  return 1;
445  } else if (GB(s[0], 5, 3) == 6) {
446  if (IsUtf8Part(s[1])) {
447  /* Double byte character: 110xxxxx 10xxxxxx */
448  *c = GB(s[0], 0, 5) << 6 | GB(s[1], 0, 6);
449  if (*c >= 0x80) return 2;
450  }
451  } else if (GB(s[0], 4, 4) == 14) {
452  if (IsUtf8Part(s[1]) && IsUtf8Part(s[2])) {
453  /* Triple byte character: 1110xxxx 10xxxxxx 10xxxxxx */
454  *c = GB(s[0], 0, 4) << 12 | GB(s[1], 0, 6) << 6 | GB(s[2], 0, 6);
455  if (*c >= 0x800) return 3;
456  }
457  } else if (GB(s[0], 3, 5) == 30) {
458  if (IsUtf8Part(s[1]) && IsUtf8Part(s[2]) && IsUtf8Part(s[3])) {
459  /* 4 byte character: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
460  *c = GB(s[0], 0, 3) << 18 | GB(s[1], 0, 6) << 12 | GB(s[2], 0, 6) << 6 | GB(s[3], 0, 6);
461  if (*c >= 0x10000 && *c <= 0x10FFFF) return 4;
462  }
463  }
464 
465  /* DEBUG(misc, 1, "[utf8] invalid UTF-8 sequence"); */
466  *c = '?';
467  return 1;
468 }
469 
470 
477 size_t Utf8Encode(char *buf, WChar c)
478 {
479  if (c < 0x80) {
480  *buf = c;
481  return 1;
482  } else if (c < 0x800) {
483  *buf++ = 0xC0 + GB(c, 6, 5);
484  *buf = 0x80 + GB(c, 0, 6);
485  return 2;
486  } else if (c < 0x10000) {
487  *buf++ = 0xE0 + GB(c, 12, 4);
488  *buf++ = 0x80 + GB(c, 6, 6);
489  *buf = 0x80 + GB(c, 0, 6);
490  return 3;
491  } else if (c < 0x110000) {
492  *buf++ = 0xF0 + GB(c, 18, 3);
493  *buf++ = 0x80 + GB(c, 12, 6);
494  *buf++ = 0x80 + GB(c, 6, 6);
495  *buf = 0x80 + GB(c, 0, 6);
496  return 4;
497  }
498 
499  /* DEBUG(misc, 1, "[utf8] can't UTF-8 encode value 0x%X", c); */
500  *buf = '?';
501  return 1;
502 }
503 
511 size_t Utf8TrimString(char *s, size_t maxlen)
512 {
513  size_t length = 0;
514 
515  for (const char *ptr = strchr(s, '\0'); *s != '\0';) {
516  size_t len = Utf8EncodedCharLen(*s);
517  /* Silently ignore invalid UTF8 sequences, our only concern trimming */
518  if (len == 0) len = 1;
519 
520  /* Take care when a hard cutoff was made for the string and
521  * the last UTF8 sequence is invalid */
522  if (length + len >= maxlen || (s + len > ptr)) break;
523  s += len;
524  length += len;
525  }
526 
527  *s = '\0';
528  return length;
529 }
530 
531 #ifdef DEFINE_STRCASESTR
532 char *strcasestr(const char *haystack, const char *needle)
533 {
534  size_t hay_len = strlen(haystack);
535  size_t needle_len = strlen(needle);
536  while (hay_len >= needle_len) {
537  if (strncasecmp(haystack, needle, needle_len) == 0) return const_cast<char *>(haystack);
538 
539  haystack++;
540  hay_len--;
541  }
542 
543  return NULL;
544 }
545 #endif /* DEFINE_STRCASESTR */
546 
555 static const char *SkipGarbage(const char *str)
556 {
557  while (*str != '\0' && (*str < '0' || IsInsideMM(*str, ';', '@' + 1) || IsInsideMM(*str, '[', '`' + 1) || IsInsideMM(*str, '{', '~' + 1))) str++;
558  return str;
559 }
560 
569 int strnatcmp(const char *s1, const char *s2, bool ignore_garbage_at_front)
570 {
571  if (ignore_garbage_at_front) {
572  s1 = SkipGarbage(s1);
573  s2 = SkipGarbage(s2);
574  }
575 #ifdef WITH_ICU_SORT
576  if (_current_collator != NULL) {
577  UErrorCode status = U_ZERO_ERROR;
578  int result = _current_collator->compareUTF8(s1, s2, status);
579  if (U_SUCCESS(status)) return result;
580  }
581 
582 #endif /* WITH_ICU_SORT */
583 
584  /* Do a normal comparison if ICU is missing or if we cannot create a collator. */
585  return strcasecmp(s1, s2);
586 }
587 
588 #ifdef WITH_ICU_SORT
589 
590 #include <unicode/utext.h>
591 #include <unicode/brkiter.h>
592 
595 {
596  icu::BreakIterator *char_itr;
597  icu::BreakIterator *word_itr;
598 
601 
602 public:
603  IcuStringIterator() : char_itr(NULL), word_itr(NULL)
604  {
605  UErrorCode status = U_ZERO_ERROR;
606  this->char_itr = icu::BreakIterator::createCharacterInstance(icu::Locale(_current_language != NULL ? _current_language->isocode : "en"), status);
607  this->word_itr = icu::BreakIterator::createWordInstance(icu::Locale(_current_language != NULL ? _current_language->isocode : "en"), status);
608 
609  *this->utf16_str.Append() = '\0';
610  *this->utf16_to_utf8.Append() = 0;
611  }
612 
613  virtual ~IcuStringIterator()
614  {
615  delete this->char_itr;
616  delete this->word_itr;
617  }
618 
619  virtual void SetString(const char *s)
620  {
621  const char *string_base = s;
622 
623  /* Unfortunately current ICU versions only provide rudimentary support
624  * for word break iterators (especially for CJK languages) in combination
625  * with UTF-8 input. As a work around we have to convert the input to
626  * UTF-16 and create a mapping back to UTF-8 character indices. */
627  this->utf16_str.Clear();
628  this->utf16_to_utf8.Clear();
629 
630  while (*s != '\0') {
631  size_t idx = s - string_base;
632 
633  WChar c = Utf8Consume(&s);
634  if (c < 0x10000) {
635  *this->utf16_str.Append() = (UChar)c;
636  } else {
637  /* Make a surrogate pair. */
638  *this->utf16_str.Append() = (UChar)(0xD800 + ((c - 0x10000) >> 10));
639  *this->utf16_str.Append() = (UChar)(0xDC00 + ((c - 0x10000) & 0x3FF));
640  *this->utf16_to_utf8.Append() = idx;
641  }
642  *this->utf16_to_utf8.Append() = idx;
643  }
644  *this->utf16_str.Append() = '\0';
645  *this->utf16_to_utf8.Append() = s - string_base;
646 
647  UText text = UTEXT_INITIALIZER;
648  UErrorCode status = U_ZERO_ERROR;
649  utext_openUChars(&text, this->utf16_str.Begin(), this->utf16_str.Length() - 1, &status);
650  this->char_itr->setText(&text, status);
651  this->word_itr->setText(&text, status);
652  this->char_itr->first();
653  this->word_itr->first();
654  }
655 
656  virtual size_t SetCurPosition(size_t pos)
657  {
658  /* Convert incoming position to an UTF-16 string index. */
659  uint utf16_pos = 0;
660  for (uint i = 0; i < this->utf16_to_utf8.Length(); i++) {
661  if (this->utf16_to_utf8[i] == pos) {
662  utf16_pos = i;
663  break;
664  }
665  }
666 
667  /* isBoundary has the documented side-effect of setting the current
668  * position to the first valid boundary equal to or greater than
669  * the passed value. */
670  this->char_itr->isBoundary(utf16_pos);
671  return this->utf16_to_utf8[this->char_itr->current()];
672  }
673 
674  virtual size_t Next(IterType what)
675  {
676  int32_t pos;
677  switch (what) {
678  case ITER_CHARACTER:
679  pos = this->char_itr->next();
680  break;
681 
682  case ITER_WORD:
683  pos = this->word_itr->following(this->char_itr->current());
684  /* The ICU word iterator considers both the start and the end of a word a valid
685  * break point, but we only want word starts. Move to the next location in
686  * case the new position points to whitespace. */
687  while (pos != icu::BreakIterator::DONE &&
688  IsWhitespace(Utf16DecodeChar((const uint16 *)&this->utf16_str[pos]))) {
689  int32_t new_pos = this->word_itr->next();
690  /* Don't set it to DONE if it was valid before. Otherwise we'll return END
691  * even though the iterator wasn't at the end of the string before. */
692  if (new_pos == icu::BreakIterator::DONE) break;
693  pos = new_pos;
694  }
695 
696  this->char_itr->isBoundary(pos);
697  break;
698 
699  default:
700  NOT_REACHED();
701  }
702 
703  return pos == icu::BreakIterator::DONE ? END : this->utf16_to_utf8[pos];
704  }
705 
706  virtual size_t Prev(IterType what)
707  {
708  int32_t pos;
709  switch (what) {
710  case ITER_CHARACTER:
711  pos = this->char_itr->previous();
712  break;
713 
714  case ITER_WORD:
715  pos = this->word_itr->preceding(this->char_itr->current());
716  /* The ICU word iterator considers both the start and the end of a word a valid
717  * break point, but we only want word starts. Move to the previous location in
718  * case the new position points to whitespace. */
719  while (pos != icu::BreakIterator::DONE &&
720  IsWhitespace(Utf16DecodeChar((const uint16 *)&this->utf16_str[pos]))) {
721  int32_t new_pos = this->word_itr->previous();
722  /* Don't set it to DONE if it was valid before. Otherwise we'll return END
723  * even though the iterator wasn't at the start of the string before. */
724  if (new_pos == icu::BreakIterator::DONE) break;
725  pos = new_pos;
726  }
727 
728  this->char_itr->isBoundary(pos);
729  break;
730 
731  default:
732  NOT_REACHED();
733  }
734 
735  return pos == icu::BreakIterator::DONE ? END : this->utf16_to_utf8[pos];
736  }
737 };
738 
740 {
741  return new IcuStringIterator();
742 }
743 
744 #else
745 
747 class DefaultStringIterator : public StringIterator
748 {
749  const char *string;
750  size_t len;
751  size_t cur_pos;
752 
753 public:
754  DefaultStringIterator() : string(NULL), len(0), cur_pos(0)
755  {
756  }
757 
758  virtual void SetString(const char *s)
759  {
760  this->string = s;
761  this->len = strlen(s);
762  this->cur_pos = 0;
763  }
764 
765  virtual size_t SetCurPosition(size_t pos)
766  {
767  assert(this->string != NULL && pos <= this->len);
768  /* Sanitize in case we get a position inside an UTF-8 sequence. */
769  while (pos > 0 && IsUtf8Part(this->string[pos])) pos--;
770  return this->cur_pos = pos;
771  }
772 
773  virtual size_t Next(IterType what)
774  {
775  assert(this->string != NULL);
776 
777  /* Already at the end? */
778  if (this->cur_pos >= this->len) return END;
779 
780  switch (what) {
781  case ITER_CHARACTER: {
782  WChar c;
783  this->cur_pos += Utf8Decode(&c, this->string + this->cur_pos);
784  return this->cur_pos;
785  }
786 
787  case ITER_WORD: {
788  WChar c;
789  /* Consume current word. */
790  size_t offs = Utf8Decode(&c, this->string + this->cur_pos);
791  while (this->cur_pos < this->len && !IsWhitespace(c)) {
792  this->cur_pos += offs;
793  offs = Utf8Decode(&c, this->string + this->cur_pos);
794  }
795  /* Consume whitespace to the next word. */
796  while (this->cur_pos < this->len && IsWhitespace(c)) {
797  this->cur_pos += offs;
798  offs = Utf8Decode(&c, this->string + this->cur_pos);
799  }
800 
801  return this->cur_pos;
802  }
803 
804  default:
805  NOT_REACHED();
806  }
807 
808  return END;
809  }
810 
811  virtual size_t Prev(IterType what)
812  {
813  assert(this->string != NULL);
814 
815  /* Already at the beginning? */
816  if (this->cur_pos == 0) return END;
817 
818  switch (what) {
819  case ITER_CHARACTER:
820  return this->cur_pos = Utf8PrevChar(this->string + this->cur_pos) - this->string;
821 
822  case ITER_WORD: {
823  const char *s = this->string + this->cur_pos;
824  WChar c;
825  /* Consume preceding whitespace. */
826  do {
827  s = Utf8PrevChar(s);
828  Utf8Decode(&c, s);
829  } while (s > this->string && IsWhitespace(c));
830  /* Consume preceding word. */
831  while (s > this->string && !IsWhitespace(c)) {
832  s = Utf8PrevChar(s);
833  Utf8Decode(&c, s);
834  }
835  /* Move caret back to the beginning of the word. */
836  if (IsWhitespace(c)) Utf8Consume(&s);
837 
838  return this->cur_pos = s - this->string;
839  }
840 
841  default:
842  NOT_REACHED();
843  }
844 
845  return END;
846  }
847 };
848 
850 {
851  return new DefaultStringIterator();
852 }
853 
854 #endif
char *CDECL str_fmt(const char *str,...)
Format, "printf", into a newly allocated string.
Definition: string.cpp:139
Control codes that are embedded in the translation strings.
virtual void SetString(const char *s)
Set a new iteration string.
Definition: string.cpp:619
static bool IsInsideMM(const T x, const uint min, const uint max)
Checks if a value is in an interval.
Definition: math_func.hpp:266
int CDECL seprintf(char *str, const char *last, const char *format,...)
Safer implementation of snprintf; same as snprintf except:
Definition: string.cpp:398
Only hexadecimal characters.
Definition: string_type.h:31
static const size_t END
Sentinel to indicate end-of-iteration.
Definition: string_base.h:25
Functions related to debugging.
static StringIterator * Create()
Create a new iterator instance.
Definition: string.cpp:739
SmallVector< size_t, 32 > utf16_to_utf8
Mapping from UTF-16 code point position to index in the UTF-8 source string.
Definition: string.cpp:600
icu::BreakIterator * word_itr
ICU iterator for words.
Definition: string.cpp:597
int CDECL vseprintf(char *str, const char *last, const char *format, va_list ap)
Safer implementation of vsnprintf; same as vsnprintf except:
Definition: string.cpp:50
char * strecpy(char *dst, const char *src, const char *last)
Copies characters from one buffer to another.
Definition: string.cpp:101
const LanguageMetadata * _current_language
The currently loaded language.
Definition: strings.cpp:50
static bool IsWhitespace(WChar c)
Check whether UNICODE character is whitespace or not, i.e.
Definition: string_func.h:242
char * strecat(char *dst, const char *src, const char *last)
Appends characters from one string to another.
Definition: string.cpp:73
char * md5sumToString(char *buf, const char *last, const uint8 md5sum[16])
Convert the md5sum to a hexadecimal string representation.
Definition: string.cpp:416
static const char * SkipGarbage(const char *str)
Skip some of the &#39;garbage&#39; in the string that we don&#39;t want to use to sort on.
Definition: string.cpp:555
void Clear()
Remove all items from the list.
const T * Begin() const
Get the pointer to the first item (const)
icu::BreakIterator * char_itr
ICU iterator for characters.
Definition: string.cpp:596
size_t Utf8Decode(WChar *c, const char *s)
Decode and consume the next UTF-8 encoded character.
Definition: string.cpp:437
#define lastof(x)
Get the last element of an fixed size array.
Definition: depend.cpp:50
bool strtolower(char *str)
Convert a given ASCII string to lowercase.
Definition: string.cpp:320
virtual size_t SetCurPosition(size_t pos)
Change the current string cursor.
Definition: string.cpp:656
char isocode[16]
the ISO code for the language (not country code)
Definition: language.h:33
T * Append(uint to_add=1)
Append an item and return it.
StringValidationSettings
Settings for the string validation.
Definition: string_type.h:48
Iterate over characters (or more exactly grapheme clusters).
Definition: string_base.h:20
static int8 Utf8EncodedCharLen(char c)
Return the length of an UTF-8 encoded value based on a single char.
Definition: string_func.h:118
Functions related to low-level strings.
bool IsValidChar(WChar key, CharSetFilter afilter)
Only allow certain keys.
Definition: string.cpp:338
Only numeric ones.
Definition: string_type.h:28
SmallVector< UChar, 32 > utf16_str
UTF-16 copy of the string.
Definition: string.cpp:599
uint Length() const
Get the number of items in the list.
void str_validate(char *str, const char *last, StringValidationSettings settings)
Scans the string for valid characters and if it finds invalid ones, replaces them with a question mar...
Definition: string.cpp:184
virtual size_t Next(IterType what)
Advance the cursor by one iteration unit.
Definition: string.cpp:674
static WChar Utf16DecodeChar(const uint16 *c)
Decode an UTF-16 character.
Definition: string_func.h:195
Functions related to the allocation of memory.
size_t Utf8TrimString(char *s, size_t maxlen)
Properly terminate an UTF8 string to some maximum length.
Definition: string.cpp:511
Functions related to the gfx engine.
Definition of base types and functions in a cross-platform compatible way.
Allow newlines.
Definition: string_type.h:51
A number of safeguards to prevent using unsafe methods.
IterType
Type of the iterator.
Definition: string_base.h:19
Information about languages and their files.
Only numbers and spaces.
Definition: string_type.h:29
char * stredup(const char *s, const char *last)
Create a duplicate of the given string.
Definition: string.cpp:126
static T min(const T a, const T b)
Returns the minimum of two values.
Definition: math_func.hpp:42
Iterate over words.
Definition: string_base.h:21
void str_strip_colours(char *str)
Scans the string for colour codes and strips them.
Definition: string.cpp:272
CharSetFilter
Valid filter types for IsValidChar.
Definition: string_type.h:26
Integer math functions.
#define DEBUG(name, level,...)
Output a line of debugging information.
Definition: debug.h:39
Both numeric and alphabetic and spaces and stuff.
Definition: string_type.h:27
int strnatcmp(const char *s1, const char *s2, bool ignore_garbage_at_front)
Compares two strings using case insensitive natural sort.
Definition: string.cpp:569
static size_t ttd_strnlen(const char *str, size_t maxlen)
Get the length of a string, within a limited buffer.
Definition: string_func.h:71
size_t Utf8StringLength(const char *s)
Get the length of an UTF-8 encoded string in number of characters and thus not the number of bytes th...
Definition: string.cpp:300
Replace the unknown/bad bits with question marks.
Definition: string_type.h:50
void CDECL error(const char *s,...)
Error handling for fatal non-user errors.
Definition: openttd.cpp:110
static uint GB(const T x, const uint8 s, const uint8 n)
Fetch n bits from x, started at bit s.
static char * Utf8PrevChar(char *s)
Retrieve the previous UNICODE character in an UTF-8 encoded string.
Definition: string_func.h:143
Collator * _current_collator
Collator for the language currently in use.
Definition: strings.cpp:55
Class for iterating over different kind of parts of a string.
Definition: string_base.h:16
void str_fix_scc_encoded(char *str, const char *last)
Scan the string for old values of SCC_ENCODED and fix it to it&#39;s new, static value.
Definition: string.cpp:158
Allow the special control codes.
Definition: string_type.h:52
size_t Utf8Encode(char *buf, WChar c)
Encode a unicode character and place it in the buffer.
Definition: string.cpp:477
static bool HasBit(const T x, const uint8 y)
Checks if a bit in a value is set.
void ValidateString(const char *str)
Scans the string for valid characters and if it finds invalid ones, replaces them with a question mar...
Definition: string.cpp:233
String iterator using ICU as a backend.
Definition: string.cpp:594
Only alphabetic values.
Definition: string_type.h:30
virtual size_t Prev(IterType what)
Move the cursor back by one iteration unit.
Definition: string.cpp:706
uint32 WChar
Type for wide characters, i.e.
Definition: string_type.h:35
bool StrValid(const char *str, const char *last)
Checks whether the given string is valid, i.e.
Definition: string.cpp:247