Merge tag 'v2.12.4' into maint
[git/git.git] / strbuf.h
1 #ifndef STRBUF_H
2 #define STRBUF_H
3
4 /**
5 * strbuf's are meant to be used with all the usual C string and memory
6 * APIs. Given that the length of the buffer is known, it's often better to
7 * use the mem* functions than a str* one (memchr vs. strchr e.g.).
8 * Though, one has to be careful about the fact that str* functions often
9 * stop on NULs and that strbufs may have embedded NULs.
10 *
11 * A strbuf is NUL terminated for convenience, but no function in the
12 * strbuf API actually relies on the string being free of NULs.
13 *
14 * strbufs have some invariants that are very important to keep in mind:
15 *
16 * - The `buf` member is never NULL, so it can be used in any usual C
17 * string operations safely. strbuf's _have_ to be initialized either by
18 * `strbuf_init()` or by `= STRBUF_INIT` before the invariants, though.
19 *
20 * Do *not* assume anything on what `buf` really is (e.g. if it is
21 * allocated memory or not), use `strbuf_detach()` to unwrap a memory
22 * buffer from its strbuf shell in a safe way. That is the sole supported
23 * way. This will give you a malloced buffer that you can later `free()`.
24 *
25 * However, it is totally safe to modify anything in the string pointed by
26 * the `buf` member, between the indices `0` and `len-1` (inclusive).
27 *
28 * - The `buf` member is a byte array that has at least `len + 1` bytes
29 * allocated. The extra byte is used to store a `'\0'`, allowing the
30 * `buf` member to be a valid C-string. Every strbuf function ensure this
31 * invariant is preserved.
32 *
33 * NOTE: It is OK to "play" with the buffer directly if you work it this
34 * way:
35 *
36 * strbuf_grow(sb, SOME_SIZE); <1>
37 * strbuf_setlen(sb, sb->len + SOME_OTHER_SIZE);
38 *
39 * <1> Here, the memory array starting at `sb->buf`, and of length
40 * `strbuf_avail(sb)` is all yours, and you can be sure that
41 * `strbuf_avail(sb)` is at least `SOME_SIZE`.
42 *
43 * NOTE: `SOME_OTHER_SIZE` must be smaller or equal to `strbuf_avail(sb)`.
44 *
45 * Doing so is safe, though if it has to be done in many places, adding the
46 * missing API to the strbuf module is the way to go.
47 *
48 * WARNING: Do _not_ assume that the area that is yours is of size `alloc
49 * - 1` even if it's true in the current implementation. Alloc is somehow a
50 * "private" member that should not be messed with. Use `strbuf_avail()`
51 * instead.
52 */
53
54 /**
55 * Data Structures
56 * ---------------
57 */
58
59 /**
60 * This is the string buffer structure. The `len` member can be used to
61 * determine the current length of the string, and `buf` member provides
62 * access to the string itself.
63 */
64 struct strbuf {
65 size_t alloc;
66 size_t len;
67 char *buf;
68 };
69
70 extern char strbuf_slopbuf[];
71 #define STRBUF_INIT { 0, 0, strbuf_slopbuf }
72
73 /**
74 * Life Cycle Functions
75 * --------------------
76 */
77
78 /**
79 * Initialize the structure. The second parameter can be zero or a bigger
80 * number to allocate memory, in case you want to prevent further reallocs.
81 */
82 extern void strbuf_init(struct strbuf *, size_t);
83
84 /**
85 * Release a string buffer and the memory it used. You should not use the
86 * string buffer after using this function, unless you initialize it again.
87 */
88 extern void strbuf_release(struct strbuf *);
89
90 /**
91 * Detach the string from the strbuf and returns it; you now own the
92 * storage the string occupies and it is your responsibility from then on
93 * to release it with `free(3)` when you are done with it.
94 */
95 extern char *strbuf_detach(struct strbuf *, size_t *);
96
97 /**
98 * Attach a string to a buffer. You should specify the string to attach,
99 * the current length of the string and the amount of allocated memory.
100 * The amount must be larger than the string length, because the string you
101 * pass is supposed to be a NUL-terminated string. This string _must_ be
102 * malloc()ed, and after attaching, the pointer cannot be relied upon
103 * anymore, and neither be free()d directly.
104 */
105 extern void strbuf_attach(struct strbuf *, void *, size_t, size_t);
106
107 /**
108 * Swap the contents of two string buffers.
109 */
110 static inline void strbuf_swap(struct strbuf *a, struct strbuf *b)
111 {
112 SWAP(*a, *b);
113 }
114
115
116 /**
117 * Functions related to the size of the buffer
118 * -------------------------------------------
119 */
120
121 /**
122 * Determine the amount of allocated but unused memory.
123 */
124 static inline size_t strbuf_avail(const struct strbuf *sb)
125 {
126 return sb->alloc ? sb->alloc - sb->len - 1 : 0;
127 }
128
129 /**
130 * Ensure that at least this amount of unused memory is available after
131 * `len`. This is used when you know a typical size for what you will add
132 * and want to avoid repetitive automatic resizing of the underlying buffer.
133 * This is never a needed operation, but can be critical for performance in
134 * some cases.
135 */
136 extern void strbuf_grow(struct strbuf *, size_t);
137
138 /**
139 * Set the length of the buffer to a given value. This function does *not*
140 * allocate new memory, so you should not perform a `strbuf_setlen()` to a
141 * length that is larger than `len + strbuf_avail()`. `strbuf_setlen()` is
142 * just meant as a 'please fix invariants from this strbuf I just messed
143 * with'.
144 */
145 static inline void strbuf_setlen(struct strbuf *sb, size_t len)
146 {
147 if (len > (sb->alloc ? sb->alloc - 1 : 0))
148 die("BUG: strbuf_setlen() beyond buffer");
149 sb->len = len;
150 sb->buf[len] = '\0';
151 }
152
153 /**
154 * Empty the buffer by setting the size of it to zero.
155 */
156 #define strbuf_reset(sb) strbuf_setlen(sb, 0)
157
158
159 /**
160 * Functions related to the contents of the buffer
161 * -----------------------------------------------
162 */
163
164 /**
165 * Strip whitespace from the beginning (`ltrim`), end (`rtrim`), or both side
166 * (`trim`) of a string.
167 */
168 extern void strbuf_trim(struct strbuf *);
169 extern void strbuf_rtrim(struct strbuf *);
170 extern void strbuf_ltrim(struct strbuf *);
171
172 /**
173 * Replace the contents of the strbuf with a reencoded form. Returns -1
174 * on error, 0 on success.
175 */
176 extern int strbuf_reencode(struct strbuf *sb, const char *from, const char *to);
177
178 /**
179 * Lowercase each character in the buffer using `tolower`.
180 */
181 extern void strbuf_tolower(struct strbuf *sb);
182
183 /**
184 * Compare two buffers. Returns an integer less than, equal to, or greater
185 * than zero if the first buffer is found, respectively, to be less than,
186 * to match, or be greater than the second buffer.
187 */
188 extern int strbuf_cmp(const struct strbuf *, const struct strbuf *);
189
190
191 /**
192 * Adding data to the buffer
193 * -------------------------
194 *
195 * NOTE: All of the functions in this section will grow the buffer as
196 * necessary. If they fail for some reason other than memory shortage and the
197 * buffer hadn't been allocated before (i.e. the `struct strbuf` was set to
198 * `STRBUF_INIT`), then they will free() it.
199 */
200
201 /**
202 * Add a single character to the buffer.
203 */
204 static inline void strbuf_addch(struct strbuf *sb, int c)
205 {
206 if (!strbuf_avail(sb))
207 strbuf_grow(sb, 1);
208 sb->buf[sb->len++] = c;
209 sb->buf[sb->len] = '\0';
210 }
211
212 /**
213 * Add a character the specified number of times to the buffer.
214 */
215 extern void strbuf_addchars(struct strbuf *sb, int c, size_t n);
216
217 /**
218 * Insert data to the given position of the buffer. The remaining contents
219 * will be shifted, not overwritten.
220 */
221 extern void strbuf_insert(struct strbuf *, size_t pos, const void *, size_t);
222
223 /**
224 * Remove given amount of data from a given position of the buffer.
225 */
226 extern void strbuf_remove(struct strbuf *, size_t pos, size_t len);
227
228 /**
229 * Remove the bytes between `pos..pos+len` and replace it with the given
230 * data.
231 */
232 extern void strbuf_splice(struct strbuf *, size_t pos, size_t len,
233 const void *, size_t);
234
235 /**
236 * Add a NUL-terminated string to the buffer. Each line will be prepended
237 * by a comment character and a blank.
238 */
239 extern void strbuf_add_commented_lines(struct strbuf *out, const char *buf, size_t size);
240
241
242 /**
243 * Add data of given length to the buffer.
244 */
245 extern void strbuf_add(struct strbuf *, const void *, size_t);
246
247 /**
248 * Add a NUL-terminated string to the buffer.
249 *
250 * NOTE: This function will *always* be implemented as an inline or a macro
251 * using strlen, meaning that this is efficient to write things like:
252 *
253 * strbuf_addstr(sb, "immediate string");
254 *
255 */
256 static inline void strbuf_addstr(struct strbuf *sb, const char *s)
257 {
258 strbuf_add(sb, s, strlen(s));
259 }
260
261 /**
262 * Copy the contents of another buffer at the end of the current one.
263 */
264 extern void strbuf_addbuf(struct strbuf *sb, const struct strbuf *sb2);
265
266 /**
267 * This function can be used to expand a format string containing
268 * placeholders. To that end, it parses the string and calls the specified
269 * function for every percent sign found.
270 *
271 * The callback function is given a pointer to the character after the `%`
272 * and a pointer to the struct strbuf. It is expected to add the expanded
273 * version of the placeholder to the strbuf, e.g. to add a newline
274 * character if the letter `n` appears after a `%`. The function returns
275 * the length of the placeholder recognized and `strbuf_expand()` skips
276 * over it.
277 *
278 * The format `%%` is automatically expanded to a single `%` as a quoting
279 * mechanism; callers do not need to handle the `%` placeholder themselves,
280 * and the callback function will not be invoked for this placeholder.
281 *
282 * All other characters (non-percent and not skipped ones) are copied
283 * verbatim to the strbuf. If the callback returned zero, meaning that the
284 * placeholder is unknown, then the percent sign is copied, too.
285 *
286 * In order to facilitate caching and to make it possible to give
287 * parameters to the callback, `strbuf_expand()` passes a context pointer,
288 * which can be used by the programmer of the callback as she sees fit.
289 */
290 typedef size_t (*expand_fn_t) (struct strbuf *sb, const char *placeholder, void *context);
291 extern void strbuf_expand(struct strbuf *sb, const char *format, expand_fn_t fn, void *context);
292
293 /**
294 * Used as callback for `strbuf_expand()`, expects an array of
295 * struct strbuf_expand_dict_entry as context, i.e. pairs of
296 * placeholder and replacement string. The array needs to be
297 * terminated by an entry with placeholder set to NULL.
298 */
299 struct strbuf_expand_dict_entry {
300 const char *placeholder;
301 const char *value;
302 };
303 extern size_t strbuf_expand_dict_cb(struct strbuf *sb, const char *placeholder, void *context);
304
305 /**
306 * Append the contents of one strbuf to another, quoting any
307 * percent signs ("%") into double-percents ("%%") in the
308 * destination. This is useful for literal data to be fed to either
309 * strbuf_expand or to the *printf family of functions.
310 */
311 extern void strbuf_addbuf_percentquote(struct strbuf *dst, const struct strbuf *src);
312
313 /**
314 * Append the given byte size as a human-readable string (i.e. 12.23 KiB,
315 * 3.50 MiB).
316 */
317 extern void strbuf_humanise_bytes(struct strbuf *buf, off_t bytes);
318
319 /**
320 * Add a formatted string to the buffer.
321 */
322 __attribute__((format (printf,2,3)))
323 extern void strbuf_addf(struct strbuf *sb, const char *fmt, ...);
324
325 /**
326 * Add a formatted string prepended by a comment character and a
327 * blank to the buffer.
328 */
329 __attribute__((format (printf, 2, 3)))
330 extern void strbuf_commented_addf(struct strbuf *sb, const char *fmt, ...);
331
332 __attribute__((format (printf,2,0)))
333 extern void strbuf_vaddf(struct strbuf *sb, const char *fmt, va_list ap);
334
335 /**
336 * Add the time specified by `tm`, as formatted by `strftime`.
337 */
338 extern void strbuf_addftime(struct strbuf *sb, const char *fmt, const struct tm *tm);
339
340 /**
341 * Read a given size of data from a FILE* pointer to the buffer.
342 *
343 * NOTE: The buffer is rewound if the read fails. If -1 is returned,
344 * `errno` must be consulted, like you would do for `read(3)`.
345 * `strbuf_read()`, `strbuf_read_file()` and `strbuf_getline_*()`
346 * family of functions have the same behaviour as well.
347 */
348 extern size_t strbuf_fread(struct strbuf *, size_t, FILE *);
349
350 /**
351 * Read the contents of a given file descriptor. The third argument can be
352 * used to give a hint about the file size, to avoid reallocs. If read fails,
353 * any partial read is undone.
354 */
355 extern ssize_t strbuf_read(struct strbuf *, int fd, size_t hint);
356
357 /**
358 * Read the contents of a given file descriptor partially by using only one
359 * attempt of xread. The third argument can be used to give a hint about the
360 * file size, to avoid reallocs. Returns the number of new bytes appended to
361 * the sb.
362 */
363 extern ssize_t strbuf_read_once(struct strbuf *, int fd, size_t hint);
364
365 /**
366 * Read the contents of a file, specified by its path. The third argument
367 * can be used to give a hint about the file size, to avoid reallocs.
368 * Return the number of bytes read or a negative value if some error
369 * occurred while opening or reading the file.
370 */
371 extern ssize_t strbuf_read_file(struct strbuf *sb, const char *path, size_t hint);
372
373 /**
374 * Read the target of a symbolic link, specified by its path. The third
375 * argument can be used to give a hint about the size, to avoid reallocs.
376 */
377 extern int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint);
378
379 /**
380 * Write the whole content of the strbuf to the stream not stopping at
381 * NUL bytes.
382 */
383 extern ssize_t strbuf_write(struct strbuf *sb, FILE *stream);
384
385 /**
386 * Read a line from a FILE *, overwriting the existing contents of
387 * the strbuf. The strbuf_getline*() family of functions share
388 * this signature, but have different line termination conventions.
389 *
390 * Reading stops after the terminator or at EOF. The terminator
391 * is removed from the buffer before returning. Returns 0 unless
392 * there was nothing left before EOF, in which case it returns `EOF`.
393 */
394 typedef int (*strbuf_getline_fn)(struct strbuf *, FILE *);
395
396 /* Uses LF as the line terminator */
397 extern int strbuf_getline_lf(struct strbuf *sb, FILE *fp);
398
399 /* Uses NUL as the line terminator */
400 extern int strbuf_getline_nul(struct strbuf *sb, FILE *fp);
401
402 /*
403 * Similar to strbuf_getline_lf(), but additionally treats a CR that
404 * comes immediately before the LF as part of the terminator.
405 * This is the most friendly version to be used to read "text" files
406 * that can come from platforms whose native text format is CRLF
407 * terminated.
408 */
409 extern int strbuf_getline(struct strbuf *, FILE *);
410
411
412 /**
413 * Like `strbuf_getline`, but keeps the trailing terminator (if
414 * any) in the buffer.
415 */
416 extern int strbuf_getwholeline(struct strbuf *, FILE *, int);
417
418 /**
419 * Like `strbuf_getwholeline`, but operates on a file descriptor.
420 * It reads one character at a time, so it is very slow. Do not
421 * use it unless you need the correct position in the file
422 * descriptor.
423 */
424 extern int strbuf_getwholeline_fd(struct strbuf *, int, int);
425
426 /**
427 * Set the buffer to the path of the current working directory.
428 */
429 extern int strbuf_getcwd(struct strbuf *sb);
430
431 /**
432 * Add a path to a buffer, converting a relative path to an
433 * absolute one in the process. Symbolic links are not
434 * resolved.
435 */
436 extern void strbuf_add_absolute_path(struct strbuf *sb, const char *path);
437
438 /**
439 * Canonize `path` (make it absolute, resolve symlinks, remove extra
440 * slashes) and append it to `sb`. Die with an informative error
441 * message if there is a problem.
442 *
443 * The directory part of `path` (i.e., everything up to the last
444 * dir_sep) must denote a valid, existing directory, but the last
445 * component need not exist.
446 *
447 * Callers that don't mind links should use the more lightweight
448 * strbuf_add_absolute_path() instead.
449 */
450 extern void strbuf_add_real_path(struct strbuf *sb, const char *path);
451
452
453 /**
454 * Normalize in-place the path contained in the strbuf. See
455 * normalize_path_copy() for details. If an error occurs, the contents of "sb"
456 * are left untouched, and -1 is returned.
457 */
458 extern int strbuf_normalize_path(struct strbuf *sb);
459
460 /**
461 * Strip whitespace from a buffer. The second parameter controls if
462 * comments are considered contents to be removed or not.
463 */
464 extern void strbuf_stripspace(struct strbuf *buf, int skip_comments);
465
466 /**
467 * Temporary alias until all topic branches have switched to use
468 * strbuf_stripspace directly.
469 */
470 static inline void stripspace(struct strbuf *buf, int skip_comments)
471 {
472 strbuf_stripspace(buf, skip_comments);
473 }
474
475 static inline int strbuf_strip_suffix(struct strbuf *sb, const char *suffix)
476 {
477 if (strip_suffix_mem(sb->buf, &sb->len, suffix)) {
478 strbuf_setlen(sb, sb->len);
479 return 1;
480 } else
481 return 0;
482 }
483
484 /**
485 * Split str (of length slen) at the specified terminator character.
486 * Return a null-terminated array of pointers to strbuf objects
487 * holding the substrings. The substrings include the terminator,
488 * except for the last substring, which might be unterminated if the
489 * original string did not end with a terminator. If max is positive,
490 * then split the string into at most max substrings (with the last
491 * substring containing everything following the (max-1)th terminator
492 * character).
493 *
494 * The most generic form is `strbuf_split_buf`, which takes an arbitrary
495 * pointer/len buffer. The `_str` variant takes a NUL-terminated string,
496 * the `_max` variant takes a strbuf, and just `strbuf_split` is a convenience
497 * wrapper to drop the `max` parameter.
498 *
499 * For lighter-weight alternatives, see string_list_split() and
500 * string_list_split_in_place().
501 */
502 extern struct strbuf **strbuf_split_buf(const char *, size_t,
503 int terminator, int max);
504
505 static inline struct strbuf **strbuf_split_str(const char *str,
506 int terminator, int max)
507 {
508 return strbuf_split_buf(str, strlen(str), terminator, max);
509 }
510
511 static inline struct strbuf **strbuf_split_max(const struct strbuf *sb,
512 int terminator, int max)
513 {
514 return strbuf_split_buf(sb->buf, sb->len, terminator, max);
515 }
516
517 static inline struct strbuf **strbuf_split(const struct strbuf *sb,
518 int terminator)
519 {
520 return strbuf_split_max(sb, terminator, 0);
521 }
522
523 /**
524 * Free a NULL-terminated list of strbufs (for example, the return
525 * values of the strbuf_split*() functions).
526 */
527 extern void strbuf_list_free(struct strbuf **);
528
529 /**
530 * Add the abbreviation, as generated by find_unique_abbrev, of `sha1` to
531 * the strbuf `sb`.
532 */
533 extern void strbuf_add_unique_abbrev(struct strbuf *sb,
534 const unsigned char *sha1,
535 int abbrev_len);
536
537 /**
538 * Launch the user preferred editor to edit a file and fill the buffer
539 * with the file's contents upon the user completing their editing. The
540 * third argument can be used to set the environment which the editor is
541 * run in. If the buffer is NULL the editor is launched as usual but the
542 * file's contents are not read into the buffer upon completion.
543 */
544 extern int launch_editor(const char *path, struct strbuf *buffer, const char *const *env);
545
546 extern void strbuf_add_lines(struct strbuf *sb, const char *prefix, const char *buf, size_t size);
547
548 /**
549 * Append s to sb, with the characters '<', '>', '&' and '"' converted
550 * into XML entities.
551 */
552 extern void strbuf_addstr_xml_quoted(struct strbuf *sb, const char *s);
553
554 /**
555 * "Complete" the contents of `sb` by ensuring that either it ends with the
556 * character `term`, or it is empty. This can be used, for example,
557 * to ensure that text ends with a newline, but without creating an empty
558 * blank line if there is no content in the first place.
559 */
560 static inline void strbuf_complete(struct strbuf *sb, char term)
561 {
562 if (sb->len && sb->buf[sb->len - 1] != term)
563 strbuf_addch(sb, term);
564 }
565
566 static inline void strbuf_complete_line(struct strbuf *sb)
567 {
568 strbuf_complete(sb, '\n');
569 }
570
571 /*
572 * Copy "name" to "sb", expanding any special @-marks as handled by
573 * interpret_branch_name(). The result is a non-qualified branch name
574 * (so "foo" or "origin/master" instead of "refs/heads/foo" or
575 * "refs/remotes/origin/master").
576 *
577 * Note that the resulting name may not be a syntactically valid refname.
578 *
579 * If "allowed" is non-zero, restrict the set of allowed expansions. See
580 * interpret_branch_name() for details.
581 */
582 extern void strbuf_branchname(struct strbuf *sb, const char *name,
583 unsigned allowed);
584
585 /*
586 * Like strbuf_branchname() above, but confirm that the result is
587 * syntactically valid to be used as a local branch name in refs/heads/.
588 *
589 * The return value is "0" if the result is valid, and "-1" otherwise.
590 */
591 extern int strbuf_check_branch_ref(struct strbuf *sb, const char *name);
592
593 extern void strbuf_addstr_urlencode(struct strbuf *, const char *,
594 int reserved);
595
596 __attribute__((format (printf,1,2)))
597 extern int printf_ln(const char *fmt, ...);
598 __attribute__((format (printf,2,3)))
599 extern int fprintf_ln(FILE *fp, const char *fmt, ...);
600
601 char *xstrdup_tolower(const char *);
602
603 /**
604 * Create a newly allocated string using printf format. You can do this easily
605 * with a strbuf, but this provides a shortcut to save a few lines.
606 */
607 __attribute__((format (printf, 1, 0)))
608 char *xstrvfmt(const char *fmt, va_list ap);
609 __attribute__((format (printf, 1, 2)))
610 char *xstrfmt(const char *fmt, ...);
611
612 #endif /* STRBUF_H */