src/string_func.h

   1 /* $Id: string_func.h 25653 2013-08-05 20:35:31Z 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
  10 /**
  11  * @file string_func.h Functions related to low-level strings.
  12  *
  13  * @note Be aware of "dangerous" string functions; string functions that
  14  * have behaviour that could easily cause buffer overruns and such:
  15  * - strncpy: does not '\0' terminate when input string is longer than
  16  *   the size of the output string. Use strecpy instead.
  17  * - [v]snprintf: returns the length of the string as it would be written
  18  *   when the output is large enough, so it can be more than the size of
  19  *   the buffer and than can underflow size_t (uint-ish) which makes all
  20  *   subsequent snprintf alikes write outside of the buffer. Use
  21  *   [v]seprintf instead; it will return the number of bytes actually
  22  *   added so no [v]seprintf will cause outside of bounds writes.
  23  * - [v]sprintf: does not bounds checking: use [v]seprintf instead.
  24  */
  25
  26 #ifndef STRING_FUNC_H
  27 #define STRING_FUNC_H
  28
  29 #include <stdarg.h>
  30
  31 #include "core/bitmath_func.hpp"
  32 #include "string_type.h"
  33
  34 char *strecat(char *dst, const char *src, const char *last);
  35 char *strecpy(char *dst, const char *src, const char *last);
  36 char *stredup(const char *src, const char *last = NULL);
  37
  38 int CDECL seprintf(char *str, const char *last, const char *format, ...) WARN_FORMAT(3, 4);
  39 int CDECL vseprintf(char *str, const char *last, const char *format, va_list ap);
  40
  41 char *CDECL str_fmt(const char *str, ...) WARN_FORMAT(1, 2);
  42
  43 char *str_validate_intl(char *str, const char *last, StringValidationSettings settings);
  44
  45 /**
  46  * Scans the string for valid characters and if it finds invalid ones,
  47  * replaces them with a question mark '?' (if not ignored)
  48  * @param str the string to validate
  49  * @param last the last valid character of str
  50  * @param settings the settings for the string validation.
  51  */
  52 static inline void str_validate(char *str, const char *last, StringValidationSettings settings = SVS_REPLACE_WITH_QUESTION_MARK)
  53 {
  54         *str_validate_intl(str, last, settings) = '\0';
  55 }
  56
  57 void ValidateString(const char *str);
  58
  59 void str_fix_scc_encoded(char *str, const char *last);
  60 void str_strip_colours(char *str);
  61 bool strtolower(char *str);
  62
  63 bool StrValid(const char *str, const char *last);
  64
  65 /**
  66  * Check if a string buffer is empty.
  67  *
  68  * @param s The pointer to the first element of the buffer
  69  * @return true if the buffer starts with the terminating null-character or
  70  *         if the given pointer points to NULL else return false
  71  */
  72 static inline bool StrEmpty(const char *s)
  73 {
  74         return s == NULL || s[0] == '\0';
  75 }
  76
  77 /**
  78  * Get the length of a string, within a limited buffer.
  79  *
  80  * @param str The pointer to the first element of the buffer
  81  * @param maxlen The maximum size of the buffer
  82  * @return The length of the string
  83  */
  84 static inline size_t ttd_strnlen(const char *str, size_t maxlen)
  85 {
  86         const char *t;
  87         for (t = str; (size_t)(t - str) < maxlen && *t != '\0'; t++) {}
  88         return t - str;
  89 }
  90
  91 char *md5sumToString(char *buf, const char *last, const uint8 md5sum[16]);
  92
  93 bool IsValidChar(WChar key, CharSetFilter afilter);
  94
  95 size_t Utf8Decode(WChar *c, const char *s);
  96 size_t Utf8Encode(char *buf, WChar c);
  97 size_t Utf8TrimString(char *s, size_t maxlen);
  98
  99
 100 static inline WChar Utf8Consume(const char **s)
 101 {
 102         WChar c;
 103         *s += Utf8Decode(&c, *s);
 104         return c;
 105 }
 106
 107 /**
 108  * Return the length of a UTF-8 encoded character.
 109  * @param c Unicode character.
 110  * @return Length of UTF-8 encoding for character.
 111  */
 112 static inline int8 Utf8CharLen(WChar c)
 113 {
 114         if (c < 0x80)       return 1;
 115         if (c < 0x800)      return 2;
 116         if (c < 0x10000)    return 3;
 117         if (c < 0x110000)   return 4;
 118
 119         /* Invalid valid, we encode as a '?' */
 120         return 1;
 121 }
 122
 123
 124 /**
 125  * Return the length of an UTF-8 encoded value based on a single char. This
 126  * char should be the first byte of the UTF-8 encoding. If not, or encoding
 127  * is invalid, return value is 0
 128  * @param c char to query length of
 129  * @return requested size
 130  */
 131 static inline int8 Utf8EncodedCharLen(char c)
 132 {
 133         if (GB(c, 3, 5) == 0x1E) return 4;
 134         if (GB(c, 4, 4) == 0x0E) return 3;
 135         if (GB(c, 5, 3) == 0x06) return 2;
 136         if (GB(c, 7, 1) == 0x00) return 1;
 137
 138         /* Invalid UTF8 start encoding */
 139         return 0;
 140 }
 141
 142
 143 /* Check if the given character is part of a UTF8 sequence */
 144 static inline bool IsUtf8Part(char c)
 145 {
 146         return GB(c, 6, 2) == 2;
 147 }
 148
 149 /**
 150  * Retrieve the previous UNICODE character in an UTF-8 encoded string.
 151  * @param s char pointer pointing to (the first char of) the next character
 152  * @return a pointer in 's' to the previous UNICODE character's first byte
 153  * @note The function should not be used to determine the length of the previous
 154  * encoded char because it might be an invalid/corrupt start-sequence
 155  */
 156 static inline char *Utf8PrevChar(char *s)
 157 {
 158         char *ret = s;
 159         while (IsUtf8Part(*--ret)) {}
 160         return ret;
 161 }
 162
 163 static inline const char *Utf8PrevChar(const char *s)
 164 {
 165         const char *ret = s;
 166         while (IsUtf8Part(*--ret)) {}
 167         return ret;
 168 }
 169
 170 size_t Utf8StringLength(const char *s);
 171
 172 /**
 173  * Is the given character a lead surrogate code point?
 174  * @param c The character to test.
 175  * @return True if the character is a lead surrogate code point.
 176  */
 177 static inline bool Utf16IsLeadSurrogate(uint c)
 178 {
 179         return c >= 0xD800 && c <= 0xDBFF;
 180 }
 181
 182 /**
 183  * Is the given character a lead surrogate code point?
 184  * @param c The character to test.
 185  * @return True if the character is a lead surrogate code point.
 186  */
 187 static inline bool Utf16IsTrailSurrogate(uint c)
 188 {
 189         return c >= 0xDC00 && c <= 0xDFFF;
 190 }
 191
 192 /**
 193  * Convert an UTF-16 surrogate pair to the corresponding Unicode character.
 194  * @param lead Lead surrogate code point.
 195  * @param trail Trail surrogate code point.
 196  * @return Decoded Unicode character.
 197  */
 198 static inline WChar Utf16DecodeSurrogate(uint lead, uint trail)
 199 {
 200         return 0x10000 + (((lead - 0xD800) << 10) | (trail - 0xDC00));
 201 }
 202
 203 /**
 204  * Decode an UTF-16 character.
 205  * @param c Pointer to one or two UTF-16 code points.
 206  * @return Decoded Unicode character.
 207  */
 208 static inline WChar Utf16DecodeChar(const uint16 *c)
 209 {
 210         if (Utf16IsLeadSurrogate(c[0])) {
 211                 return Utf16DecodeSurrogate(c[0], c[1]);
 212         } else {
 213                 return *c;
 214         }
 215 }
 216
 217 /**
 218  * Is the given character a text direction character.
 219  * @param c The character to test.
 220  * @return true iff the character is used to influence
 221  *         the text direction.
 222  */
 223 static inline bool IsTextDirectionChar(WChar c)
 224 {
 225         switch (c) {
 226                 case CHAR_TD_LRM:
 227                 case CHAR_TD_RLM:
 228                 case CHAR_TD_LRE:
 229                 case CHAR_TD_RLE:
 230                 case CHAR_TD_LRO:
 231                 case CHAR_TD_RLO:
 232                 case CHAR_TD_PDF:
 233                         return true;
 234
 235                 default:
 236                         return false;
 237         }
 238 }
 239
 240 static inline bool IsPrintable(WChar c)
 241 {
 242         if (c < 0x20)   return false;
 243         if (c < 0xE000) return true;
 244         if (c < 0xE200) return false;
 245         return true;
 246 }
 247
 248 /**
 249  * Check whether UNICODE character is whitespace or not, i.e. whether
 250  * this is a potential line-break character.
 251  * @param c UNICODE character to check
 252  * @return a boolean value whether 'c' is a whitespace character or not
 253  * @see http://www.fileformat.info/info/unicode/category/Zs/list.htm
 254  */
 255 static inline bool IsWhitespace(WChar c)
 256 {
 257         return c == 0x0020 /* SPACE */ || c == 0x3000; /* IDEOGRAPHIC SPACE */
 258 }
 259
 260 /* Needed for NetBSD version (so feature) testing */
 261 #if defined(__NetBSD__) || defined(__FreeBSD__)
 262 #include <sys/param.h>
 263 #endif
 264
 265 /* strcasestr is available for _GNU_SOURCE, BSD and some Apple */
 266 #if defined(_GNU_SOURCE) || (defined(__BSD_VISIBLE) && __BSD_VISIBLE) || (defined(__APPLE__) && (!defined(_POSIX_C_SOURCE) || defined(_DARWIN_C_SOURCE))) || defined(_NETBSD_SOURCE)
 267 #       undef DEFINE_STRCASESTR
 268 #else
 269 #       define DEFINE_STRCASESTR
 270 char *strcasestr(const char *haystack, const char *needle);
 271 #endif /* strcasestr is available */
 272
 273 int strnatcmp(const char *s1, const char *s2, bool ignore_garbage_at_front = false);
 274
 275 #endif /* STRING_FUNC_H */