1/*
2 * Copyright (c) 2007 Mans Rullgard
3 *
4 * This file is part of FFmpeg.
5 *
6 * FFmpeg is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2.1 of the License, or (at your option) any later version.
10 *
11 * FFmpeg is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
14 * Lesser General Public License for more details.
15 *
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with FFmpeg; if not, write to the Free Software
18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19 */
20
21#ifndef AVUTIL_AVSTRING_H
22#define AVUTIL_AVSTRING_H
23
24#include <stddef.h>
25#include <stdint.h>
26#include "attributes.h"
27
28/**
29 * @addtogroup lavu_string
30 * @{
31 */
32
33/**
34 * Return non-zero if pfx is a prefix of str. If it is, *ptr is set to
35 * the address of the first character in str after the prefix.
36 *
37 * @param str input string
38 * @param pfx prefix to test
39 * @param ptr updated if the prefix is matched inside str
40 * @return non-zero if the prefix matches, zero otherwise
41 */
42int av_strstart(const char *str, const char *pfx, const char **ptr);
43
44/**
45 * Return non-zero if pfx is a prefix of str independent of case. If
46 * it is, *ptr is set to the address of the first character in str
47 * after the prefix.
48 *
49 * @param str input string
50 * @param pfx prefix to test
51 * @param ptr updated if the prefix is matched inside str
52 * @return non-zero if the prefix matches, zero otherwise
53 */
54int av_stristart(const char *str, const char *pfx, const char **ptr);
55
56/**
57 * Locate the first case-independent occurrence in the string haystack
58 * of the string needle.  A zero-length string needle is considered to
59 * match at the start of haystack.
60 *
61 * This function is a case-insensitive version of the standard strstr().
62 *
63 * @param haystack string to search in
64 * @param needle   string to search for
65 * @return         pointer to the located match within haystack
66 *                 or a null pointer if no match
67 */
68char *av_stristr(const char *haystack, const char *needle);
69
70/**
71 * Locate the first occurrence of the string needle in the string haystack
72 * where not more than hay_length characters are searched. A zero-length
73 * string needle is considered to match at the start of haystack.
74 *
75 * This function is a length-limited version of the standard strstr().
76 *
77 * @param haystack   string to search in
78 * @param needle     string to search for
79 * @param hay_length length of string to search in
80 * @return           pointer to the located match within haystack
81 *                   or a null pointer if no match
82 */
83char *av_strnstr(const char *haystack, const char *needle, size_t hay_length);
84
85/**
86 * Copy the string src to dst, but no more than size - 1 bytes, and
87 * null-terminate dst.
88 *
89 * This function is the same as BSD strlcpy().
90 *
91 * @param dst destination buffer
92 * @param src source string
93 * @param size size of destination buffer
94 * @return the length of src
95 *
96 * @warning since the return value is the length of src, src absolutely
97 * _must_ be a properly 0-terminated string, otherwise this will read beyond
98 * the end of the buffer and possibly crash.
99 */
100size_t av_strlcpy(char *dst, const char *src, size_t size);
101
102/**
103 * Append the string src to the string dst, but to a total length of
104 * no more than size - 1 bytes, and null-terminate dst.
105 *
106 * This function is similar to BSD strlcat(), but differs when
107 * size <= strlen(dst).
108 *
109 * @param dst destination buffer
110 * @param src source string
111 * @param size size of destination buffer
112 * @return the total length of src and dst
113 *
114 * @warning since the return value use the length of src and dst, these
115 * absolutely _must_ be a properly 0-terminated strings, otherwise this
116 * will read beyond the end of the buffer and possibly crash.
117 */
118size_t av_strlcat(char *dst, const char *src, size_t size);
119
120/**
121 * Append output to a string, according to a format. Never write out of
122 * the destination buffer, and always put a terminating 0 within
123 * the buffer.
124 * @param dst destination buffer (string to which the output is
125 *  appended)
126 * @param size total size of the destination buffer
127 * @param fmt printf-compatible format string, specifying how the
128 *  following parameters are used
129 * @return the length of the string that would have been generated
130 *  if enough space had been available
131 */
132size_t av_strlcatf(char *dst, size_t size, const char *fmt, ...) av_printf_format(3, 4);
133
134/**
135 * Get the count of continuous non zero chars starting from the beginning.
136 *
137 * @param len maximum number of characters to check in the string, that
138 *            is the maximum value which is returned by the function
139 */
140static inline size_t av_strnlen(const char *s, size_t len)
141{
142    size_t i;
143    for (i = 0; i < len && s[i]; i++)
144        ;
145    return i;
146}
147
148/**
149 * Print arguments following specified format into a large enough auto
150 * allocated buffer. It is similar to GNU asprintf().
151 * @param fmt printf-compatible format string, specifying how the
152 *            following parameters are used.
153 * @return the allocated string
154 * @note You have to free the string yourself with av_free().
155 */
156char *av_asprintf(const char *fmt, ...) av_printf_format(1, 2);
157
158/**
159 * Convert a number to a av_malloced string.
160 */
161char *av_d2str(double d);
162
163/**
164 * Unescape the given string until a non escaped terminating char,
165 * and return the token corresponding to the unescaped string.
166 *
167 * The normal \ and ' escaping is supported. Leading and trailing
168 * whitespaces are removed, unless they are escaped with '\' or are
169 * enclosed between ''.
170 *
171 * @param buf the buffer to parse, buf will be updated to point to the
172 * terminating char
173 * @param term a 0-terminated list of terminating chars
174 * @return the malloced unescaped string, which must be av_freed by
175 * the user, NULL in case of allocation failure
176 */
177char *av_get_token(const char **buf, const char *term);
178
179/**
180 * Split the string into several tokens which can be accessed by
181 * successive calls to av_strtok().
182 *
183 * A token is defined as a sequence of characters not belonging to the
184 * set specified in delim.
185 *
186 * On the first call to av_strtok(), s should point to the string to
187 * parse, and the value of saveptr is ignored. In subsequent calls, s
188 * should be NULL, and saveptr should be unchanged since the previous
189 * call.
190 *
191 * This function is similar to strtok_r() defined in POSIX.1.
192 *
193 * @param s the string to parse, may be NULL
194 * @param delim 0-terminated list of token delimiters, must be non-NULL
195 * @param saveptr user-provided pointer which points to stored
196 * information necessary for av_strtok() to continue scanning the same
197 * string. saveptr is updated to point to the next character after the
198 * first delimiter found, or to NULL if the string was terminated
199 * @return the found token, or NULL when no token is found
200 */
201char *av_strtok(char *s, const char *delim, char **saveptr);
202
203/**
204 * Locale-independent conversion of ASCII isdigit.
205 */
206int av_isdigit(int c);
207
208/**
209 * Locale-independent conversion of ASCII isgraph.
210 */
211int av_isgraph(int c);
212
213/**
214 * Locale-independent conversion of ASCII isspace.
215 */
216int av_isspace(int c);
217
218/**
219 * Locale-independent conversion of ASCII characters to uppercase.
220 */
221static inline int av_toupper(int c)
222{
223    if (c >= 'a' && c <= 'z')
224        c ^= 0x20;
225    return c;
226}
227
228/**
229 * Locale-independent conversion of ASCII characters to lowercase.
230 */
231static inline int av_tolower(int c)
232{
233    if (c >= 'A' && c <= 'Z')
234        c ^= 0x20;
235    return c;
236}
237
238/**
239 * Locale-independent conversion of ASCII isxdigit.
240 */
241int av_isxdigit(int c);
242
243/**
244 * Locale-independent case-insensitive compare.
245 * @note This means only ASCII-range characters are case-insensitive
246 */
247int av_strcasecmp(const char *a, const char *b);
248
249/**
250 * Locale-independent case-insensitive compare.
251 * @note This means only ASCII-range characters are case-insensitive
252 */
253int av_strncasecmp(const char *a, const char *b, size_t n);
254
255
256/**
257 * Thread safe basename.
258 * @param path the path, on DOS both \ and / are considered separators.
259 * @return pointer to the basename substring.
260 */
261const char *av_basename(const char *path);
262
263/**
264 * Thread safe dirname.
265 * @param path the path, on DOS both \ and / are considered separators.
266 * @return the path with the separator replaced by the string terminator or ".".
267 * @note the function may change the input string.
268 */
269const char *av_dirname(char *path);
270
271enum AVEscapeMode {
272    AV_ESCAPE_MODE_AUTO,      ///< Use auto-selected escaping mode.
273    AV_ESCAPE_MODE_BACKSLASH, ///< Use backslash escaping.
274    AV_ESCAPE_MODE_QUOTE,     ///< Use single-quote escaping.
275};
276
277/**
278 * Consider spaces special and escape them even in the middle of the
279 * string.
280 *
281 * This is equivalent to adding the whitespace characters to the special
282 * characters lists, except it is guaranteed to use the exact same list
283 * of whitespace characters as the rest of libavutil.
284 */
285#define AV_ESCAPE_FLAG_WHITESPACE 0x01
286
287/**
288 * Escape only specified special characters.
289 * Without this flag, escape also any characters that may be considered
290 * special by av_get_token(), such as the single quote.
291 */
292#define AV_ESCAPE_FLAG_STRICT 0x02
293
294/**
295 * Escape string in src, and put the escaped string in an allocated
296 * string in *dst, which must be freed with av_free().
297 *
298 * @param dst           pointer where an allocated string is put
299 * @param src           string to escape, must be non-NULL
300 * @param special_chars string containing the special characters which
301 *                      need to be escaped, can be NULL
302 * @param mode          escape mode to employ, see AV_ESCAPE_MODE_* macros.
303 *                      Any unknown value for mode will be considered equivalent to
304 *                      AV_ESCAPE_MODE_BACKSLASH, but this behaviour can change without
305 *                      notice.
306 * @param flags         flags which control how to escape, see AV_ESCAPE_FLAG_ macros
307 * @return the length of the allocated string, or a negative error code in case of error
308 * @see av_bprint_escape()
309 */
310int av_escape(char **dst, const char *src, const char *special_chars,
311              enum AVEscapeMode mode, int flags);
312
313#define AV_UTF8_FLAG_ACCEPT_INVALID_BIG_CODES          1 ///< accept codepoints over 0x10FFFF
314#define AV_UTF8_FLAG_ACCEPT_NON_CHARACTERS             2 ///< accept non-characters - 0xFFFE and 0xFFFF
315#define AV_UTF8_FLAG_ACCEPT_SURROGATES                 4 ///< accept UTF-16 surrogates codes
316#define AV_UTF8_FLAG_EXCLUDE_XML_INVALID_CONTROL_CODES 8 ///< exclude control codes not accepted by XML
317
318#define AV_UTF8_FLAG_ACCEPT_ALL \
319    AV_UTF8_FLAG_ACCEPT_INVALID_BIG_CODES|AV_UTF8_FLAG_ACCEPT_NON_CHARACTERS|AV_UTF8_FLAG_ACCEPT_SURROGATES
320
321/**
322 * Read and decode a single UTF-8 code point (character) from the
323 * buffer in *buf, and update *buf to point to the next byte to
324 * decode.
325 *
326 * In case of an invalid byte sequence, the pointer will be updated to
327 * the next byte after the invalid sequence and the function will
328 * return an error code.
329 *
330 * Depending on the specified flags, the function will also fail in
331 * case the decoded code point does not belong to a valid range.
332 *
333 * @note For speed-relevant code a carefully implemented use of
334 * GET_UTF8() may be preferred.
335 *
336 * @param codep   pointer used to return the parsed code in case of success.
337 *                The value in *codep is set even in case the range check fails.
338 * @param bufp    pointer to the address the first byte of the sequence
339 *                to decode, updated by the function to point to the
340 *                byte next after the decoded sequence
341 * @param buf_end pointer to the end of the buffer, points to the next
342 *                byte past the last in the buffer. This is used to
343 *                avoid buffer overreads (in case of an unfinished
344 *                UTF-8 sequence towards the end of the buffer).
345 * @param flags   a collection of AV_UTF8_FLAG_* flags
346 * @return >= 0 in case a sequence was successfully read, a negative
347 * value in case of invalid sequence
348 */
349int av_utf8_decode(int32_t *codep, const uint8_t **bufp, const uint8_t *buf_end,
350                   unsigned int flags);
351
352/**
353 * @}
354 */
355
356#endif /* AVUTIL_AVSTRING_H */
357