string.cpp

Go to the documentation of this file.
00001 /* $Id: string.cpp 11828 2008-01-13 01:21:35Z rubidium $ */
00002 
00005 #include "stdafx.h"
00006 #include "openttd.h"
00007 #include "debug.h"
00008 #include "core/alloc_func.hpp"
00009 #include "string_func.h"
00010 
00011 #include "table/control_codes.h"
00012 
00013 #include <stdarg.h>
00014 #include <ctype.h> // required for tolower()
00015 
00016 void ttd_strlcat(char *dst, const char *src, size_t size)
00017 {
00018   assert(size > 0);
00019   for (; size > 0 && *dst != '\0'; --size, ++dst) {}
00020   assert(size > 0);
00021   while (--size > 0 && *src != '\0') *dst++ = *src++;
00022   *dst = '\0';
00023 }
00024 
00025 
00026 void ttd_strlcpy(char *dst, const char *src, size_t size)
00027 {
00028   assert(size > 0);
00029   while (--size > 0 && *src != '\0') *dst++ = *src++;
00030   *dst = '\0';
00031 }
00032 
00033 
00034 char* strecat(char* dst, const char* src, const char* last)
00035 {
00036   assert(dst <= last);
00037   for (; *dst != '\0'; ++dst)
00038     if (dst == last) return dst;
00039   for (; *src != '\0' && dst != last; ++dst, ++src) *dst = *src;
00040   *dst = '\0';
00041   return strecpy(dst, src, last);
00042 }
00043 
00044 
00045 char* strecpy(char* dst, const char* src, const char* last)
00046 {
00047   assert(dst <= last);
00048   for (; *src != '\0' && dst != last; ++dst, ++src) *dst = *src;
00049   *dst = '\0';
00050   if (dst == last && *src != '\0') {
00051 #ifdef STRGEN
00052     error("String too long for destination buffer");
00053 #else /* STRGEN */
00054     DEBUG(misc, 0, "String too long for destination buffer");
00055     *dst = '\0';
00056 #endif /* STRGEN */
00057   }
00058   return dst;
00059 }
00060 
00061 
00062 char* CDECL str_fmt(const char* str, ...)
00063 {
00064   char buf[4096];
00065   va_list va;
00066   int len;
00067 
00068   va_start(va, str);
00069   len = vsnprintf(buf, lengthof(buf), str, va);
00070   va_end(va);
00071   char* p = MallocT<char>(len + 1);
00072   if (p != NULL) memcpy(p, buf, len + 1);
00073   return p;
00074 }
00075 
00076 
00077 void str_validate(char *str)
00078 {
00079   char *dst = str;
00080   WChar c;
00081   size_t len;
00082 
00083   for (len = Utf8Decode(&c, str); c != '\0'; len = Utf8Decode(&c, str)) {
00084     if (IsPrintable(c) && (c < SCC_SPRITE_START || c > SCC_SPRITE_END ||
00085       IsValidChar(c - SCC_SPRITE_START, CS_ALPHANUMERAL))) {
00086       /* Copy the character back. Even if dst is current the same as str
00087        * (i.e. no characters have been changed) this is quicker than
00088        * moving the pointers ahead by len */
00089       do {
00090         *dst++ = *str++;
00091       } while (--len != 0);
00092     } else {
00093       /* Replace the undesirable character with a question mark */
00094       str += len;
00095       *dst++ = '?';
00096     }
00097   }
00098 
00099   *dst = '\0';
00100 }
00101 
00102 
00103 void str_strip_colours(char *str)
00104 {
00105   char *dst = str;
00106   WChar c;
00107   size_t len;
00108 
00109   for (len = Utf8Decode(&c, str); c != '\0'; len = Utf8Decode(&c, str)) {
00110     if (c < SCC_BLUE || c > SCC_BLACK) {
00111       /* Copy the character back. Even if dst is current the same as str
00112        * (i.e. no characters have been changed) this is quicker than
00113        * moving the pointers ahead by len */
00114       do {
00115         *dst++ = *str++;
00116       } while (--len != 0);
00117     } else {
00118       /* Just skip (strip) the colour codes */
00119       str += len;
00120     }
00121   }
00122   *dst = '\0';
00123 }
00124 
00133 void strtolower(char *str)
00134 {
00135   for (; *str != '\0'; str++) *str = tolower(*str);
00136 }
00137 
00145 bool IsValidChar(WChar key, CharSetFilter afilter)
00146 {
00147   switch (afilter) {
00148     case CS_ALPHANUMERAL: return IsPrintable(key);
00149     case CS_NUMERAL:      return (key >= '0' && key <= '9');
00150     case CS_ALPHA:        return IsPrintable(key) && !(key >= '0' && key <= '9');
00151   }
00152 
00153   return false;
00154 }
00155 
00156 #ifdef WIN32
00157 /* Since version 3.14, MinGW Runtime has snprintf() and vsnprintf() conform to C99 but it's not the case for older versions */
00158 #if (__MINGW32_MAJOR_VERSION < 3) || ((__MINGW32_MAJOR_VERSION == 3) && (__MINGW32_MINOR_VERSION < 14))
00159 int CDECL snprintf(char *str, size_t size, const char *format, ...)
00160 {
00161   va_list ap;
00162   int ret;
00163 
00164   va_start(ap, format);
00165   ret = vsnprintf(str, size, format, ap);
00166   va_end(ap);
00167   return ret;
00168 }
00169 #endif /* MinGW Runtime < 3.14 */
00170 
00171 #ifdef _MSC_VER
00172 /* *nprintf broken, not POSIX compliant, MSDN description
00173  * - If len < count, then len characters are stored in buffer, a null-terminator is appended, and len is returned.
00174  * - If len = count, then len characters are stored in buffer, no null-terminator is appended, and len is returned.
00175  * - If len > count, then count characters are stored in buffer, no null-terminator is appended, and a negative value is returned
00176  */
00177 int CDECL vsnprintf(char *str, size_t size, const char *format, va_list ap)
00178 {
00179   int ret;
00180   ret = _vsnprintf(str, size, format, ap);
00181   if (ret < 0 || ret == size) str[size - 1] = '\0';
00182   return ret;
00183 }
00184 #endif /* _MSC_VER */
00185 
00186 #endif /* WIN32 */
00187 
00188 
00194 char *md5sumToString(char *buf, const char *last, const uint8 md5sum[16])
00195 {
00196   char *p = buf;
00197 
00198   for (uint i = 0; i < 16; i++) {
00199     p += snprintf(p, last + 1 - p, "%02X", md5sum[i]);
00200     if (p >= last) break;
00201   }
00202 
00203   return p;
00204 }
00205 
00206 
00207 /* UTF-8 handling routines */
00208 
00209 
00210 /* Decode and consume the next UTF-8 encoded character
00211  * @param c Buffer to place decoded character.
00212  * @param s Character stream to retrieve character from.
00213  * @return Number of characters in the sequence.
00214  */
00215 size_t Utf8Decode(WChar *c, const char *s)
00216 {
00217   assert(c != NULL);
00218 
00219   if (!HasBit(s[0], 7)) {
00220     /* Single byte character: 0xxxxxxx */
00221     *c = s[0];
00222     return 1;
00223   } else if (GB(s[0], 5, 3) == 6) {
00224     if (IsUtf8Part(s[1])) {
00225       /* Double byte character: 110xxxxx 10xxxxxx */
00226       *c = GB(s[0], 0, 5) << 6 | GB(s[1], 0, 6);
00227       if (*c >= 0x80) return 2;
00228     }
00229   } else if (GB(s[0], 4, 4) == 14) {
00230     if (IsUtf8Part(s[1]) && IsUtf8Part(s[2])) {
00231       /* Triple byte character: 1110xxxx 10xxxxxx 10xxxxxx */
00232       *c = GB(s[0], 0, 4) << 12 | GB(s[1], 0, 6) << 6 | GB(s[2], 0, 6);
00233       if (*c >= 0x800) return 3;
00234     }
00235   } else if (GB(s[0], 3, 5) == 30) {
00236     if (IsUtf8Part(s[1]) && IsUtf8Part(s[2]) && IsUtf8Part(s[3])) {
00237       /* 4 byte character: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
00238       *c = GB(s[0], 0, 3) << 18 | GB(s[1], 0, 6) << 12 | GB(s[2], 0, 6) << 6 | GB(s[3], 0, 6);
00239       if (*c >= 0x10000 && *c <= 0x10FFFF) return 4;
00240     }
00241   }
00242 
00243   //DEBUG(misc, 1, "[utf8] invalid UTF-8 sequence");
00244   *c = '?';
00245   return 1;
00246 }
00247 
00248 
00249 /* Encode a unicode character and place it in the buffer
00250  * @param buf Buffer to place character.
00251  * @param c   Unicode character to encode.
00252  * @return Number of characters in the encoded sequence.
00253  */
00254 size_t Utf8Encode(char *buf, WChar c)
00255 {
00256   if (c < 0x80) {
00257     *buf = c;
00258     return 1;
00259   } else if (c < 0x800) {
00260     *buf++ = 0xC0 + GB(c,  6, 5);
00261     *buf   = 0x80 + GB(c,  0, 6);
00262     return 2;
00263   } else if (c < 0x10000) {
00264     *buf++ = 0xE0 + GB(c, 12, 4);
00265     *buf++ = 0x80 + GB(c,  6, 6);
00266     *buf   = 0x80 + GB(c,  0, 6);
00267     return 3;
00268   } else if (c < 0x110000) {
00269     *buf++ = 0xF0 + GB(c, 18, 3);
00270     *buf++ = 0x80 + GB(c, 12, 6);
00271     *buf++ = 0x80 + GB(c,  6, 6);
00272     *buf   = 0x80 + GB(c,  0, 6);
00273     return 4;
00274   }
00275 
00276   //DEBUG(misc, 1, "[utf8] can't UTF-8 encode value 0x%X", c);
00277   *buf = '?';
00278   return 1;
00279 }
00280 
00288 size_t Utf8TrimString(char *s, size_t maxlen)
00289 {
00290   size_t length = 0;
00291 
00292   for (const char *ptr = strchr(s, '\0'); *s != '\0';) {
00293     size_t len = Utf8EncodedCharLen(*s);
00294     /* Silently ignore invalid UTF8 sequences, our only concern trimming */
00295     if (len == 0) len = 1;
00296 
00297     /* Take care when a hard cutoff was made for the string and
00298      * the last UTF8 sequence is invalid */
00299     if (length + len >= maxlen || (s + len > ptr)) break;
00300     s += len;
00301     length += len;
00302   }
00303 
00304   *s = '\0';
00305   return length;
00306 }

Generated on Wed Oct 1 17:03:24 2008 for openttd by  doxygen 1.5.6