string.cpp

Go to the documentation of this file.
00001 /* $Id: string.cpp 17920 2009-10-31 14:06:16Z rubidium $ */
00002 
00003 /*
00004  * This file is part of OpenTTD.
00005  * 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.
00006  * 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.
00007  * 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/>.
00008  */
00009 
00012 #include "stdafx.h"
00013 #include "debug.h"
00014 #include "core/alloc_func.hpp"
00015 #include "core/math_func.hpp"
00016 #include "string_func.h"
00017 
00018 #include "table/control_codes.h"
00019 
00020 #include <stdarg.h>
00021 #include <ctype.h> /* required for tolower() */
00022 
00033 static int CDECL vseprintf(char *str, const char *last, const char *format, va_list ap)
00034 {
00035   if (str >= last) return 0;
00036   size_t size = last - str + 1;
00037   return min((int)size, vsnprintf(str, size, format, ap));
00038 }
00039 
00040 void ttd_strlcat(char *dst, const char *src, size_t size)
00041 {
00042   assert(size > 0);
00043   while (size > 0 && *dst != '\0') {
00044     size--;
00045     dst++;
00046   }
00047 
00048   ttd_strlcpy(dst, src, size);
00049 }
00050 
00051 
00052 void ttd_strlcpy(char *dst, const char *src, size_t size)
00053 {
00054   assert(size > 0);
00055   while (--size > 0 && *src != '\0') {
00056     *dst++ = *src++;
00057   }
00058   *dst = '\0';
00059 }
00060 
00061 
00062 char *strecat(char *dst, const char *src, const char *last)
00063 {
00064   assert(dst <= last);
00065   while (*dst != '\0') {
00066     if (dst == last) return dst;
00067     dst++;
00068   }
00069 
00070   return strecpy(dst, src, last);
00071 }
00072 
00073 
00074 char *strecpy(char *dst, const char *src, const char *last)
00075 {
00076   assert(dst <= last);
00077   while (dst != last && *src != '\0') {
00078     *dst++ = *src++;
00079   }
00080   *dst = '\0';
00081 
00082   if (dst == last && *src != '\0') {
00083 #ifdef STRGEN
00084     error("String too long for destination buffer");
00085 #else /* STRGEN */
00086     DEBUG(misc, 0, "String too long for destination buffer");
00087 #endif /* STRGEN */
00088   }
00089   return dst;
00090 }
00091 
00092 
00093 char *CDECL str_fmt(const char *str, ...)
00094 {
00095   char buf[4096];
00096   va_list va;
00097 
00098   va_start(va, str);
00099   int len = vseprintf(buf, lastof(buf), str, va);
00100   va_end(va);
00101   char *p = MallocT<char>(len + 1);
00102   memcpy(p, buf, len + 1);
00103   return p;
00104 }
00105 
00106 
00107 void str_validate(char *str, const char *last, bool allow_newlines, bool ignore)
00108 {
00109   /* Assume the ABSOLUTE WORST to be in str as it comes from the outside. */
00110 
00111   char *dst = str;
00112   while (*str != '\0') {
00113     size_t len = Utf8EncodedCharLen(*str);
00114     /* If the character is unknown, i.e. encoded length is 0
00115      * we assume worst case for the length check.
00116      * The length check is needed to prevent Utf8Decode to read
00117      * over the terminating '\0' if that happens to be placed
00118      * within the encoding of an UTF8 character. */
00119     if ((len == 0 && str + 4 > last) || str + len > last) break;
00120 
00121     WChar c;
00122     len = Utf8Decode(&c, str);
00123     /* It's possible to encode the string termination character
00124      * into a multiple bytes. This prevents those termination
00125      * characters to be skipped */
00126     if (c == '\0') break;
00127 
00128     if (IsPrintable(c) && (c < SCC_SPRITE_START || c > SCC_SPRITE_END)) {
00129       /* Copy the character back. Even if dst is current the same as str
00130        * (i.e. no characters have been changed) this is quicker than
00131        * moving the pointers ahead by len */
00132       do {
00133         *dst++ = *str++;
00134       } while (--len != 0);
00135     } else if (allow_newlines && c == '\n') {
00136       *dst++ = *str++;
00137     } else {
00138       if (allow_newlines && c == '\r' && str[1] == '\n') {
00139         str += len;
00140         continue;
00141       }
00142       /* Replace the undesirable character with a question mark */
00143       str += len;
00144       if (!ignore) *dst++ = '?';
00145     }
00146   }
00147 
00148   *dst = '\0';
00149 }
00150 
00151 
00152 void str_strip_colours(char *str)
00153 {
00154   char *dst = str;
00155   WChar c;
00156   size_t len;
00157 
00158   for (len = Utf8Decode(&c, str); c != '\0'; len = Utf8Decode(&c, str)) {
00159     if (c < SCC_BLUE || c > SCC_BLACK) {
00160       /* Copy the character back. Even if dst is current the same as str
00161        * (i.e. no characters have been changed) this is quicker than
00162        * moving the pointers ahead by len */
00163       do {
00164         *dst++ = *str++;
00165       } while (--len != 0);
00166     } else {
00167       /* Just skip (strip) the colour codes */
00168       str += len;
00169     }
00170   }
00171   *dst = '\0';
00172 }
00173 
00182 void strtolower(char *str)
00183 {
00184   for (; *str != '\0'; str++) *str = tolower(*str);
00185 }
00186 
00194 bool IsValidChar(WChar key, CharSetFilter afilter)
00195 {
00196   switch (afilter) {
00197     case CS_ALPHANUMERAL:  return IsPrintable(key);
00198     case CS_NUMERAL:       return (key >= '0' && key <= '9');
00199     case CS_NUMERAL_SPACE: return (key >= '0' && key <= '9') || key == ' ';
00200     case CS_ALPHA:         return IsPrintable(key) && !(key >= '0' && key <= '9');
00201   }
00202 
00203   return false;
00204 }
00205 
00206 #ifdef WIN32
00207 /* Since version 3.14, MinGW Runtime has snprintf() and vsnprintf() conform to C99 but it's not the case for older versions */
00208 #if (__MINGW32_MAJOR_VERSION < 3) || ((__MINGW32_MAJOR_VERSION == 3) && (__MINGW32_MINOR_VERSION < 14))
00209 int CDECL snprintf(char *str, size_t size, const char *format, ...)
00210 {
00211   va_list ap;
00212   int ret;
00213 
00214   va_start(ap, format);
00215   ret = vsnprintf(str, size, format, ap);
00216   va_end(ap);
00217   return ret;
00218 }
00219 #endif /* MinGW Runtime < 3.14 */
00220 
00221 #ifdef _MSC_VER
00222 /* *nprintf broken, not POSIX compliant, MSDN description
00223  * - If len < count, then len characters are stored in buffer, a null-terminator is appended, and len is returned.
00224  * - If len = count, then len characters are stored in buffer, no null-terminator is appended, and len is returned.
00225  * - If len > count, then count characters are stored in buffer, no null-terminator is appended, and a negative value is returned
00226  */
00227 int CDECL vsnprintf(char *str, size_t size, const char *format, va_list ap)
00228 {
00229   int ret;
00230   ret = _vsnprintf(str, size, format, ap);
00231   if (ret < 0 || ret == size) str[size - 1] = '\0';
00232   return ret;
00233 }
00234 #endif /* _MSC_VER */
00235 
00236 #endif /* WIN32 */
00237 
00247 int CDECL seprintf(char *str, const char *last, const char *format, ...)
00248 {
00249   va_list ap;
00250 
00251   va_start(ap, format);
00252   int ret = vseprintf(str, last, format, ap);
00253   va_end(ap);
00254   return ret;
00255 }
00256 
00257 
00263 char *md5sumToString(char *buf, const char *last, const uint8 md5sum[16])
00264 {
00265   char *p = buf;
00266 
00267   for (uint i = 0; i < 16; i++) {
00268     p += seprintf(p, last, "%02X", md5sum[i]);
00269   }
00270 
00271   return p;
00272 }
00273 
00274 
00275 /* UTF-8 handling routines */
00276 
00277 
00278 /* Decode and consume the next UTF-8 encoded character
00279  * @param c Buffer to place decoded character.
00280  * @param s Character stream to retrieve character from.
00281  * @return Number of characters in the sequence.
00282  */
00283 size_t Utf8Decode(WChar *c, const char *s)
00284 {
00285   assert(c != NULL);
00286 
00287   if (!HasBit(s[0], 7)) {
00288     /* Single byte character: 0xxxxxxx */
00289     *c = s[0];
00290     return 1;
00291   } else if (GB(s[0], 5, 3) == 6) {
00292     if (IsUtf8Part(s[1])) {
00293       /* Double byte character: 110xxxxx 10xxxxxx */
00294       *c = GB(s[0], 0, 5) << 6 | GB(s[1], 0, 6);
00295       if (*c >= 0x80) return 2;
00296     }
00297   } else if (GB(s[0], 4, 4) == 14) {
00298     if (IsUtf8Part(s[1]) && IsUtf8Part(s[2])) {
00299       /* Triple byte character: 1110xxxx 10xxxxxx 10xxxxxx */
00300       *c = GB(s[0], 0, 4) << 12 | GB(s[1], 0, 6) << 6 | GB(s[2], 0, 6);
00301       if (*c >= 0x800) return 3;
00302     }
00303   } else if (GB(s[0], 3, 5) == 30) {
00304     if (IsUtf8Part(s[1]) && IsUtf8Part(s[2]) && IsUtf8Part(s[3])) {
00305       /* 4 byte character: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
00306       *c = GB(s[0], 0, 3) << 18 | GB(s[1], 0, 6) << 12 | GB(s[2], 0, 6) << 6 | GB(s[3], 0, 6);
00307       if (*c >= 0x10000 && *c <= 0x10FFFF) return 4;
00308     }
00309   }
00310 
00311   /* DEBUG(misc, 1, "[utf8] invalid UTF-8 sequence"); */
00312   *c = '?';
00313   return 1;
00314 }
00315 
00316 
00317 /* Encode a unicode character and place it in the buffer
00318  * @param buf Buffer to place character.
00319  * @param c   Unicode character to encode.
00320  * @return Number of characters in the encoded sequence.
00321  */
00322 size_t Utf8Encode(char *buf, WChar c)
00323 {
00324   if (c < 0x80) {
00325     *buf = c;
00326     return 1;
00327   } else if (c < 0x800) {
00328     *buf++ = 0xC0 + GB(c,  6, 5);
00329     *buf   = 0x80 + GB(c,  0, 6);
00330     return 2;
00331   } else if (c < 0x10000) {
00332     *buf++ = 0xE0 + GB(c, 12, 4);
00333     *buf++ = 0x80 + GB(c,  6, 6);
00334     *buf   = 0x80 + GB(c,  0, 6);
00335     return 3;
00336   } else if (c < 0x110000) {
00337     *buf++ = 0xF0 + GB(c, 18, 3);
00338     *buf++ = 0x80 + GB(c, 12, 6);
00339     *buf++ = 0x80 + GB(c,  6, 6);
00340     *buf   = 0x80 + GB(c,  0, 6);
00341     return 4;
00342   }
00343 
00344   /* DEBUG(misc, 1, "[utf8] can't UTF-8 encode value 0x%X", c); */
00345   *buf = '?';
00346   return 1;
00347 }
00348 
00356 size_t Utf8TrimString(char *s, size_t maxlen)
00357 {
00358   size_t length = 0;
00359 
00360   for (const char *ptr = strchr(s, '\0'); *s != '\0';) {
00361     size_t len = Utf8EncodedCharLen(*s);
00362     /* Silently ignore invalid UTF8 sequences, our only concern trimming */
00363     if (len == 0) len = 1;
00364 
00365     /* Take care when a hard cutoff was made for the string and
00366      * the last UTF8 sequence is invalid */
00367     if (length + len >= maxlen || (s + len > ptr)) break;
00368     s += len;
00369     length += len;
00370   }
00371 
00372   *s = '\0';
00373   return length;
00374 }
00375 
00376 #ifndef _GNU_SOURCE
00377 #include "core/math_func.hpp"
00378 char *strndup(const char *s, size_t len)
00379 {
00380   len = min(strlen(s), len);
00381   char *tmp = CallocT<char>(len + 1);
00382   memcpy(tmp, s, len);
00383   return tmp;
00384 }
00385 #endif /* !_GNU_SOURCE */
00386 
00387 #ifdef DEFINE_STRCASESTR
00388 const char *strcasestr(const char *haystack, const char *needle)
00389 {
00390   size_t hay_len = strlen(haystack);
00391   size_t needle_len = strlen(needle);
00392   while (hay_len >= needle_len) {
00393     if (strncasecmp(haystack, needle, needle_len) == 0) return haystack;
00394 
00395     haystack++;
00396     hay_len--;
00397   }
00398 
00399   return NULL;
00400 }
00401 #endif /* DEFINE_STRCASESTR */

Generated on Wed Jan 20 23:38:40 2010 for OpenTTD by  doxygen 1.5.6