src/string_func.h

   1 /* $Id$ */
   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 void str_validate(char *str, const char *last, StringValidationSettings settings = SVS_REPLACE_WITH_QUESTION_MARK);
  44 void ValidateString(const char *str);
  45
  46 void str_fix_scc_encoded(char *str, const char *last);
  47 void str_strip_colours(char *str);
  48 bool strtolower(char *str);
  49
  50 bool StrValid(const char *str, const char *last);
  51
  52 /**
  53  * Check if a string buffer is empty.
  54  *
  55  * @param s The pointer to the first element of the buffer
  56  * @return true if the buffer starts with the terminating null-character or
  57  *         if the given pointer points to NULL else return false
  58  */
  59 static inline bool StrEmpty(const char *s)
  60 {
  61         return s == NULL || s[0] == '\0';
  62 }
  63
  64 /**
  65  * Get the length of a string, within a limited buffer.
  66  *
  67  * @param str The pointer to the first element of the buffer
  68  * @param maxlen The maximum size of the buffer
  69  * @return The length of the string
  70  */
  71 static inline size_t ttd_strnlen(const char *str, size_t maxlen)
  72 {
  73         const char *t;
  74         for (t = str; (size_t)(t - str) < maxlen && *t != '\0'; t++) {}
  75         return t - str;
  76 }
  77
  78 char *md5sumToString(char *buf, const char *last, const uint8 md5sum[16]);
  79
  80 bool IsValidChar(WChar key, CharSetFilter afilter);
  81
  82 size_t Utf8Decode(WChar *c, const char *s);
  83 size_t Utf8Encode(char *buf, WChar c);
  84 size_t Utf8TrimString(char *s, size_t maxlen);
  85
  86
  87 static inline WChar Utf8Consume(const char **s)
  88 {
  89         WChar c;
  90         *s += Utf8Decode(&c, *s);
  91         return c;
  92 }
  93
  94 /**
  95  * Return the length of a UTF-8 encoded character.
  96  * @param c Unicode character.
  97  * @return Length of UTF-8 encoding for character.
  98  */
  99 static inline int8 Utf8CharLen(WChar c)
 100 {
 101         if (c < 0x80)       return 1;
 102         if (c < 0x800)      return 2;
 103         if (c < 0x10000)    return 3;
 104         if (c < 0x110000)   return 4;
 105
 106         /* Invalid valid, we encode as a '?' */
 107         return 1;
 108 }
 109
 110
 111 /**
 112  * Return the length of an UTF-8 encoded value based on a single char. This
 113  * char should be the first byte of the UTF-8 encoding. If not, or encoding
 114  * is invalid, return value is 0
 115  * @param c char to query length of
 116  * @return requested size
 117  */
 118 static inline int8 Utf8EncodedCharLen(char c)
 119 {
 120         if (GB(c, 3, 5) == 0x1E) return 4;
 121         if (GB(c, 4, 4) == 0x0E) return 3;
 122         if (GB(c, 5, 3) == 0x06) return 2;
 123         if (GB(c, 7, 1) == 0x00) return 1;
 124
 125         /* Invalid UTF8 start encoding */
 126         return 0;
 127 }
 128
 129
 130 /* Check if the given character is part of a UTF8 sequence */
 131 static inline bool IsUtf8Part(char c)
 132 {
 133         return GB(c, 6, 2) == 2;
 134 }
 135
 136 /**
 137  * Retrieve the previous UNICODE character in an UTF-8 encoded string.
 138  * @param s char pointer pointing to (the first char of) the next character
 139  * @return a pointer in 's' to the previous UNICODE character's first byte
 140  * @note The function should not be used to determine the length of the previous
 141  * encoded char because it might be an invalid/corrupt start-sequence
 142  */
 143 static inline char *Utf8PrevChar(char *s)
 144 {
 145         char *ret = s;
 146         while (IsUtf8Part(*--ret)) {}
 147         return ret;
 148 }
 149
 150 static inline const char *Utf8PrevChar(const char *s)
 151 {
 152         const char *ret = s;
 153         while (IsUtf8Part(*--ret)) {}
 154         return ret;
 155 }
 156
 157 size_t Utf8StringLength(const char *s);
 158
 159 /**
 160  * Is the given character a lead surrogate code point?
 161  * @param c The character to test.
 162  * @return True if the character is a lead surrogate code point.
 163  */
 164 static inline bool Utf16IsLeadSurrogate(uint c)
 165 {
 166         return c >= 0xD800 && c <= 0xDBFF;
 167 }
 168
 169 /**
 170  * Is the given character a lead surrogate code point?
 171  * @param c The character to test.
 172  * @return True if the character is a lead surrogate code point.
 173  */
 174 static inline bool Utf16IsTrailSurrogate(uint c)
 175 {
 176         return c >= 0xDC00 && c <= 0xDFFF;
 177 }
 178
 179 /**
 180  * Convert an UTF-16 surrogate pair to the corresponding Unicode character.
 181  * @param lead Lead surrogate code point.
 182  * @param trail Trail surrogate code point.
 183  * @return Decoded Unicode character.
 184  */
 185 static inline WChar Utf16DecodeSurrogate(uint lead, uint trail)
 186 {
 187         return 0x10000 + (((lead - 0xD800) << 10) | (trail - 0xDC00));
 188 }
 189
 190 /**
 191  * Decode an UTF-16 character.
 192  * @param c Pointer to one or two UTF-16 code points.
 193  * @return Decoded Unicode character.
 194  */
 195 static inline WChar Utf16DecodeChar(const uint16 *c)
 196 {
 197         if (Utf16IsLeadSurrogate(c[0])) {
 198                 return Utf16DecodeSurrogate(c[0], c[1]);
 199         } else {
 200                 return *c;
 201         }
 202 }
 203
 204 /**
 205  * Is the given character a text direction character.
 206  * @param c The character to test.
 207  * @return true iff the character is used to influence
 208  *         the text direction.
 209  */
 210 static inline bool IsTextDirectionChar(WChar c)
 211 {
 212         switch (c) {
 213                 case CHAR_TD_LRM:
 214                 case CHAR_TD_RLM:
 215                 case CHAR_TD_LRE:
 216                 case CHAR_TD_RLE:
 217                 case CHAR_TD_LRO:
 218                 case CHAR_TD_RLO:
 219                 case CHAR_TD_PDF:
 220                         return true;
 221
 222                 default:
 223                         return false;
 224         }
 225 }
 226
 227 static inline bool IsPrintable(WChar c)
 228 {
 229         if (c < 0x20)   return false;
 230         if (c < 0xE000) return true;
 231         if (c < 0xE200) return false;
 232         return true;
 233 }
 234
 235 /**
 236  * Check whether UNICODE character is whitespace or not, i.e. whether
 237  * this is a potential line-break character.
 238  * @param c UNICODE character to check
 239  * @return a boolean value whether 'c' is a whitespace character or not
 240  * @see http://www.fileformat.info/info/unicode/category/Zs/list.htm
 241  */
 242 static inline bool IsWhitespace(WChar c)
 243 {
 244         return c == 0x0020 /* SPACE */ || c == 0x3000; /* IDEOGRAPHIC SPACE */
 245 }
 246
 247 /* Needed for NetBSD version (so feature) testing */
 248 #if defined(__NetBSD__) || defined(__FreeBSD__)
 249 #include <sys/param.h>
 250 #endif
 251
 252 /* strcasestr is available for _GNU_SOURCE, BSD and some Apple */
 253 #if defined(_GNU_SOURCE) || (defined(__BSD_VISIBLE) && __BSD_VISIBLE) || (defined(__APPLE__) && (!defined(_POSIX_C_SOURCE) || defined(_DARWIN_C_SOURCE))) || defined(_NETBSD_SOURCE)
 254 #       undef DEFINE_STRCASESTR
 255 #else
 256 #       define DEFINE_STRCASESTR
 257 char *strcasestr(const char *haystack, const char *needle);
 258 #endif /* strcasestr is available */
 259
 260 int strnatcmp(const char *s1, const char *s2, bool ignore_garbage_at_front = false);
 261
 262 #endif /* STRING_FUNC_H */