rubidium@9723: /* $Id$ */ rubidium@9723: rubidium@9724: /** @file string_func.h Functions related to low-level strings. */ rubidium@9723: rubidium@9723: #ifndef STRING_FUNC_H rubidium@9723: #define STRING_FUNC_H rubidium@9723: rubidium@9723: #include "core/bitmath_func.hpp" rubidium@9723: #include "string_type.h" rubidium@9723: rubidium@9723: /** rubidium@9723: * usage ttd_strlcpy(dst, src, lengthof(dst)); rubidium@9723: * @param dst destination buffer rubidium@9723: * @param src string to copy/concatenate rubidium@9723: * @param size size of the destination buffer rubidium@9723: */ rubidium@9723: void ttd_strlcat(char *dst, const char *src, size_t size); rubidium@9723: void ttd_strlcpy(char *dst, const char *src, size_t size); rubidium@9723: rubidium@9723: /** rubidium@9723: * usage: strecpy(dst, src, lastof(dst)); rubidium@9723: * @param dst destination buffer rubidium@9723: * @param src string to copy rubidium@9723: * @param last pointer to the last element in the dst array rubidium@9723: * if NULL no boundary check is performed rubidium@9723: * @return a pointer to the terminating \0 in the destination buffer rubidium@9723: */ rubidium@9723: char *strecat(char *dst, const char *src, const char *last); rubidium@9723: char *strecpy(char *dst, const char *src, const char *last); rubidium@9723: rubidium@9723: char *CDECL str_fmt(const char *str, ...); rubidium@9723: rubidium@9723: /** Scans the string for valid characters and if it finds invalid ones, rubidium@9723: * replaces them with a question mark '?' */ rubidium@9723: void str_validate(char *str); rubidium@9723: rubidium@9723: /** Scans the string for colour codes and strips them */ rubidium@9723: void str_strip_colours(char *str); rubidium@9723: rubidium@9723: /** Convert the given string to lowercase, only works with ASCII! */ rubidium@9723: void strtolower(char *str); rubidium@9723: rubidium@9723: rubidium@9723: static inline bool StrEmpty(const char *s) { return s == NULL || s[0] == '\0'; } rubidium@9723: rubidium@9723: rubidium@9723: /** Get the length of a string, within a limited buffer */ rubidium@10513: static inline size_t ttd_strnlen(const char *str, size_t maxlen) rubidium@9723: { rubidium@9723: const char *t; rubidium@10513: for (t = str; *t != '\0' && (size_t)(t - str) < maxlen; t++) {} rubidium@9723: return t - str; rubidium@9723: } rubidium@9723: rubidium@9723: /** Convert the md5sum number to a 'hexadecimal' string, return next pos in buffer */ rubidium@9723: char *md5sumToString(char *buf, const char *last, const uint8 md5sum[16]); rubidium@9723: rubidium@9723: /** rubidium@9723: * Only allow certain keys. You can define the filter to be used. This makes rubidium@9723: * sure no invalid keys can get into an editbox, like BELL. rubidium@9723: * @param key character to be checked rubidium@9723: * @param afilter the filter to use rubidium@9723: * @return true or false depending if the character is printable/valid or not rubidium@9723: */ rubidium@9723: bool IsValidChar(WChar key, CharSetFilter afilter); rubidium@9723: rubidium@9723: size_t Utf8Decode(WChar *c, const char *s); rubidium@9723: size_t Utf8Encode(char *buf, WChar c); rubidium@9723: size_t Utf8TrimString(char *s, size_t maxlen); rubidium@9723: rubidium@9723: rubidium@9723: static inline WChar Utf8Consume(const char **s) rubidium@9723: { rubidium@9723: WChar c; rubidium@9723: *s += Utf8Decode(&c, *s); rubidium@9723: return c; rubidium@9723: } rubidium@9723: rubidium@9723: rubidium@9723: /** Return the length of a UTF-8 encoded character. rubidium@9723: * @param c Unicode character. rubidium@9723: * @return Length of UTF-8 encoding for character. rubidium@9723: */ glx@10776: static inline int8 Utf8CharLen(WChar c) rubidium@9723: { rubidium@9723: if (c < 0x80) return 1; rubidium@9723: if (c < 0x800) return 2; rubidium@9723: if (c < 0x10000) return 3; rubidium@9723: if (c < 0x110000) return 4; rubidium@9723: rubidium@9723: /* Invalid valid, we encode as a '?' */ rubidium@9723: return 1; rubidium@9723: } rubidium@9723: rubidium@9723: rubidium@9723: /** rubidium@9723: * Return the length of an UTF-8 encoded value based on a single char. This rubidium@9723: * char should be the first byte of the UTF-8 encoding. If not, or encoding rubidium@9723: * is invalid, return value is 0 rubidium@9723: * @param c char to query length of rubidium@9723: * @return requested size rubidium@9723: */ glx@10776: static inline int8 Utf8EncodedCharLen(char c) rubidium@9723: { rubidium@9723: if (GB(c, 3, 5) == 0x1E) return 4; rubidium@9723: if (GB(c, 4, 4) == 0x0E) return 3; rubidium@9723: if (GB(c, 5, 3) == 0x06) return 2; rubidium@9723: if (GB(c, 7, 1) == 0x00) return 1; rubidium@9723: rubidium@9723: /* Invalid UTF8 start encoding */ rubidium@9723: return 0; rubidium@9723: } rubidium@9723: rubidium@9723: rubidium@9723: /* Check if the given character is part of a UTF8 sequence */ rubidium@9723: static inline bool IsUtf8Part(char c) rubidium@9723: { rubidium@9723: return GB(c, 6, 2) == 2; rubidium@9723: } rubidium@9723: rubidium@9723: /** rubidium@9723: * Retrieve the previous UNICODE character in an UTF-8 encoded string. rubidium@9723: * @param s char pointer pointing to (the first char of) the next character rubidium@9723: * @return a pointer in 's' to the previous UNICODE character's first byte rubidium@9723: * @note The function should not be used to determine the length of the previous rubidium@9723: * encoded char because it might be an invalid/corrupt start-sequence rubidium@9723: */ rubidium@9723: static inline char *Utf8PrevChar(const char *s) rubidium@9723: { rubidium@9723: const char *ret = s; rubidium@9826: while (IsUtf8Part(*--ret)) {} rubidium@9723: return (char*)ret; rubidium@9723: } rubidium@9723: rubidium@9723: rubidium@9723: static inline bool IsPrintable(WChar c) rubidium@9723: { rubidium@9723: if (c < 0x20) return false; rubidium@9723: if (c < 0xE000) return true; rubidium@9723: if (c < 0xE200) return false; rubidium@9723: return true; rubidium@9723: } rubidium@9723: rubidium@9723: /** rubidium@9826: * Check whether UNICODE character is whitespace or not, i.e. whether rubidium@9826: * this is a potential line-break character. rubidium@9723: * @param c UNICODE character to check rubidium@9723: * @return a boolean value whether 'c' is a whitespace character or not rubidium@9723: * @see http://www.fileformat.info/info/unicode/category/Zs/list.htm rubidium@9723: */ rubidium@9723: static inline bool IsWhitespace(WChar c) rubidium@9723: { rubidium@9723: return rubidium@9723: c == 0x0020 /* SPACE */ || rubidium@9723: c == 0x3000 /* IDEOGRAPHIC SPACE */ rubidium@9723: ; rubidium@9723: } rubidium@9723: rubidium@9723: #endif /* STRING_FUNC_H */