src/string_func.h

   1 /*
   2  * This file is part of OpenTTD.
   3  * 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.
   4  * 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.
   5  * 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/>.
   6  */
   7
   8 /**
   9  * @file string_func.h Functions related to low-level strings.
  10  *
  11  * @note Be aware of "dangerous" string functions; string functions that
  12  * have behaviour that could easily cause buffer overruns and such:
  13  * - strncpy: does not '\0' terminate when input string is longer than
  14  *   the size of the output string. Use strecpy instead.
  15  * - [v]snprintf: returns the length of the string as it would be written
  16  *   when the output is large enough, so it can be more than the size of
  17  *   the buffer and than can underflow size_t (uint-ish) which makes all
  18  *   subsequent snprintf alikes write outside of the buffer. Use
  19  *   [v]seprintf instead; it will return the number of bytes actually
  20  *   added so no [v]seprintf will cause outside of bounds writes.
  21  * - [v]sprintf: does not bounds checking: use [v]seprintf instead.
  22  */
  23
  24 #ifndef STRING_FUNC_H
  25 #define STRING_FUNC_H
  26
  27 #include <stdarg.h>
  28 #include <iosfwd>
  29
  30 #include "core/bitmath_func.hpp"
  31 #include "core/span_type.hpp"
  32 #include "string_type.h"
  33
  34 char *strecat(char *dst, const char *src, const char *last) NOACCESS(3);
  35 char *strecpy(char *dst, const char *src, const char *last) NOACCESS(3);
  36 char *stredup(const char *src, const char *last = nullptr) NOACCESS(2);
  37
  38 int CDECL seprintf(char *str, const char *last, const char *format, ...) WARN_FORMAT(3, 4) NOACCESS(2);
  39 int CDECL vseprintf(char *str, const char *last, const char *format, va_list ap) WARN_FORMAT(3, 0) NOACCESS(2);
  40
  41 char *CDECL str_fmt(const char *str, ...) WARN_FORMAT(1, 2);
  42
  43 std::string FormatArrayAsHex(span<const byte> data);
  44
  45 void StrMakeValidInPlace(char *str, const char *last, StringValidationSettings settings = SVS_REPLACE_WITH_QUESTION_MARK) NOACCESS(2);
  46 [[nodiscard]] std::string StrMakeValid(const std::string &str, StringValidationSettings settings = SVS_REPLACE_WITH_QUESTION_MARK);
  47 void StrMakeValidInPlace(char *str, StringValidationSettings settings = SVS_REPLACE_WITH_QUESTION_MARK);
  48
  49 void str_fix_scc_encoded(char *str, const char *last) NOACCESS(2);
  50 void str_strip_colours(char *str);
  51 bool strtolower(char *str);
  52 bool strtolower(std::string &str, std::string::size_type offs = 0);
  53
  54 bool StrValid(const char *str, const char *last) NOACCESS(2);
  55 void StrTrimInPlace(std::string &str);
  56
  57 bool StrStartsWith(const std::string_view str, const std::string_view prefix);
  58 bool StrEndsWith(const std::string_view str, const std::string_view suffix);
  59
  60 /**
  61  * Check if a string buffer is empty.
  62  *
  63  * @param s The pointer to the first element of the buffer
  64  * @return true if the buffer starts with the terminating null-character or
  65  *         if the given pointer points to nullptr else return false
  66  */
  67 static inline bool StrEmpty(const char *s)
  68 {
  69         return s == nullptr || s[0] == '\0';
  70 }
  71
  72 /**
  73  * Get the length of a string, within a limited buffer.
  74  *
  75  * @param str The pointer to the first element of the buffer
  76  * @param maxlen The maximum size of the buffer
  77  * @return The length of the string
  78  */
  79 static inline size_t ttd_strnlen(const char *str, size_t maxlen)
  80 {
  81         const char *t;
  82         for (t = str; (size_t)(t - str) < maxlen && *t != '\0'; t++) {}
  83         return t - str;
  84 }
  85
  86 char *md5sumToString(char *buf, const char *last, const uint8 md5sum[16]);
  87
  88 bool IsValidChar(WChar key, CharSetFilter afilter);
  89
  90 size_t Utf8Decode(WChar *c, const char *s);
  91 size_t Utf8Encode(char *buf, WChar c);
  92 size_t Utf8Encode(std::ostreambuf_iterator<char> &buf, WChar c);
  93 size_t Utf8TrimString(char *s, size_t maxlen);
  94
  95
  96 static inline WChar Utf8Consume(const char **s)
  97 {
  98         WChar c;
  99         *s += Utf8Decode(&c, *s);
 100         return c;
 101 }
 102
 103 template <class Titr>
 104 static inline WChar Utf8Consume(Titr &s)
 105 {
 106         WChar c;
 107         s += Utf8Decode(&c, &*s);
 108         return c;
 109 }
 110
 111 /**
 112  * Return the length of a UTF-8 encoded character.
 113  * @param c Unicode character.
 114  * @return Length of UTF-8 encoding for character.
 115  */
 116 static inline int8 Utf8CharLen(WChar c)
 117 {
 118         if (c < 0x80)       return 1;
 119         if (c < 0x800)      return 2;
 120         if (c < 0x10000)    return 3;
 121         if (c < 0x110000)   return 4;
 122
 123         /* Invalid valid, we encode as a '?' */
 124         return 1;
 125 }
 126
 127
 128 /**
 129  * Return the length of an UTF-8 encoded value based on a single char. This
 130  * char should be the first byte of the UTF-8 encoding. If not, or encoding
 131  * is invalid, return value is 0
 132  * @param c char to query length of
 133  * @return requested size
 134  */
 135 static inline int8 Utf8EncodedCharLen(char c)
 136 {
 137         if (GB(c, 3, 5) == 0x1E) return 4;
 138         if (GB(c, 4, 4) == 0x0E) return 3;
 139         if (GB(c, 5, 3) == 0x06) return 2;
 140         if (GB(c, 7, 1) == 0x00) return 1;
 141
 142         /* Invalid UTF8 start encoding */
 143         return 0;
 144 }
 145
 146
 147 /* Check if the given character is part of a UTF8 sequence */
 148 static inline bool IsUtf8Part(char c)
 149 {
 150         return GB(c, 6, 2) == 2;
 151 }
 152
 153 /**
 154  * Retrieve the previous UNICODE character in an UTF-8 encoded string.
 155  * @param s char pointer pointing to (the first char of) the next character
 156  * @return a pointer in 's' to the previous UNICODE character's first byte
 157  * @note The function should not be used to determine the length of the previous
 158  * encoded char because it might be an invalid/corrupt start-sequence
 159  */
 160 static inline char *Utf8PrevChar(char *s)
 161 {
 162         char *ret = s;
 163         while (IsUtf8Part(*--ret)) {}
 164         return ret;
 165 }
 166
 167 static inline const char *Utf8PrevChar(const char *s)
 168 {
 169         const char *ret = s;
 170         while (IsUtf8Part(*--ret)) {}
 171         return ret;
 172 }
 173
 174 size_t Utf8StringLength(const char *s);
 175 size_t Utf8StringLength(const std::string &str);
 176
 177 /**
 178  * Is the given character a lead surrogate code point?
 179  * @param c The character to test.
 180  * @return True if the character is a lead surrogate code point.
 181  */
 182 static inline bool Utf16IsLeadSurrogate(uint c)
 183 {
 184         return c >= 0xD800 && c <= 0xDBFF;
 185 }
 186
 187 /**
 188  * Is the given character a lead surrogate code point?
 189  * @param c The character to test.
 190  * @return True if the character is a lead surrogate code point.
 191  */
 192 static inline bool Utf16IsTrailSurrogate(uint c)
 193 {
 194         return c >= 0xDC00 && c <= 0xDFFF;
 195 }
 196
 197 /**
 198  * Convert an UTF-16 surrogate pair to the corresponding Unicode character.
 199  * @param lead Lead surrogate code point.
 200  * @param trail Trail surrogate code point.
 201  * @return Decoded Unicode character.
 202  */
 203 static inline WChar Utf16DecodeSurrogate(uint lead, uint trail)
 204 {
 205         return 0x10000 + (((lead - 0xD800) << 10) | (trail - 0xDC00));
 206 }
 207
 208 /**
 209  * Decode an UTF-16 character.
 210  * @param c Pointer to one or two UTF-16 code points.
 211  * @return Decoded Unicode character.
 212  */
 213 static inline WChar Utf16DecodeChar(const uint16 *c)
 214 {
 215         if (Utf16IsLeadSurrogate(c[0])) {
 216                 return Utf16DecodeSurrogate(c[0], c[1]);
 217         } else {
 218                 return *c;
 219         }
 220 }
 221
 222 /**
 223  * Is the given character a text direction character.
 224  * @param c The character to test.
 225  * @return true iff the character is used to influence
 226  *         the text direction.
 227  */
 228 static inline bool IsTextDirectionChar(WChar c)
 229 {
 230         switch (c) {
 231                 case CHAR_TD_LRM:
 232                 case CHAR_TD_RLM:
 233                 case CHAR_TD_LRE:
 234                 case CHAR_TD_RLE:
 235                 case CHAR_TD_LRO:
 236                 case CHAR_TD_RLO:
 237                 case CHAR_TD_PDF:
 238                         return true;
 239
 240                 default:
 241                         return false;
 242         }
 243 }
 244
 245 static inline bool IsPrintable(WChar c)
 246 {
 247         if (c < 0x20)   return false;
 248         if (c < 0xE000) return true;
 249         if (c < 0xE200) return false;
 250         return true;
 251 }
 252
 253 /**
 254  * Check whether UNICODE character is whitespace or not, i.e. whether
 255  * this is a potential line-break character.
 256  * @param c UNICODE character to check
 257  * @return a boolean value whether 'c' is a whitespace character or not
 258  * @see http://www.fileformat.info/info/unicode/category/Zs/list.htm
 259  */
 260 static inline bool IsWhitespace(WChar c)
 261 {
 262         return c == 0x0020 /* SPACE */ || c == 0x3000; /* IDEOGRAPHIC SPACE */
 263 }
 264
 265 /* Needed for NetBSD version (so feature) testing */
 266 #if defined(__NetBSD__) || defined(__FreeBSD__)
 267 #include <sys/param.h>
 268 #endif
 269
 270 /* strcasestr is available for _GNU_SOURCE, BSD and some Apple */
 271 #if defined(_GNU_SOURCE) || (defined(__BSD_VISIBLE) && __BSD_VISIBLE) || (defined(__APPLE__) && (!defined(_POSIX_C_SOURCE) || defined(_DARWIN_C_SOURCE))) || defined(_NETBSD_SOURCE)
 272 #       undef DEFINE_STRCASESTR
 273 #else
 274 #       define DEFINE_STRCASESTR
 275 char *strcasestr(const char *haystack, const char *needle);
 276 #endif /* strcasestr is available */
 277
 278 int strnatcmp(const char *s1, const char *s2, bool ignore_garbage_at_front = false);
 279
 280 #endif /* STRING_FUNC_H */