strbuf_setlen: don't write to strbuf_slopbuf
[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 if (sb->buf != strbuf_slopbuf)
151 sb->buf[len] = '\0';
152 else
153 assert(!strbuf_slopbuf[0]);
154 }
155
156 /**
157 * Empty the buffer by setting the size of it to zero.
158 */
159 #define strbuf_reset(sb) strbuf_setlen(sb, 0)
160
161
162 /**
163 * Functions related to the contents of the buffer
164 * -----------------------------------------------
165 */
166
167 /**
168 * Strip whitespace from the beginning (`ltrim`), end (`rtrim`), or both side
169 * (`trim`) of a string.
170 */
171 extern void strbuf_trim(struct strbuf *);
172 extern void strbuf_rtrim(struct strbuf *);
173 extern void strbuf_ltrim(struct strbuf *);
174
175 /**
176 * Replace the contents of the strbuf with a reencoded form. Returns -1
177 * on error, 0 on success.
178 */
179 extern int strbuf_reencode(struct strbuf *sb, const char *from, const char *to);
180
181 /**
182 * Lowercase each character in the buffer using `tolower`.
183 */
184 extern void strbuf_tolower(struct strbuf *sb);
185
186 /**
187 * Compare two buffers. Returns an integer less than, equal to, or greater
188 * than zero if the first buffer is found, respectively, to be less than,
189 * to match, or be greater than the second buffer.
190 */
191 extern int strbuf_cmp(const struct strbuf *, const struct strbuf *);
192
193
194 /**
195 * Adding data to the buffer
196 * -------------------------
197 *
198 * NOTE: All of the functions in this section will grow the buffer as
199 * necessary. If they fail for some reason other than memory shortage and the
200 * buffer hadn't been allocated before (i.e. the `struct strbuf` was set to
201 * `STRBUF_INIT`), then they will free() it.
202 */
203
204 /**
205 * Add a single character to the buffer.
206 */
207 static inline void strbuf_addch(struct strbuf *sb, int c)
208 {
209 if (!strbuf_avail(sb))
210 strbuf_grow(sb, 1);
211 sb->buf[sb->len++] = c;
212 sb->buf[sb->len] = '\0';
213 }
214
215 /**
216 * Add a character the specified number of times to the buffer.
217 */
218 extern void strbuf_addchars(struct strbuf *sb, int c, size_t n);
219
220 /**
221 * Insert data to the given position of the buffer. The remaining contents
222 * will be shifted, not overwritten.
223 */
224 extern void strbuf_insert(struct strbuf *, size_t pos, const void *, size_t);
225
226 /**
227 * Remove given amount of data from a given position of the buffer.
228 */
229 extern void strbuf_remove(struct strbuf *, size_t pos, size_t len);
230
231 /**
232 * Remove the bytes between `pos..pos+len` and replace it with the given
233 * data.
234 */
235 extern void strbuf_splice(struct strbuf *, size_t pos, size_t len,
236 const void *, size_t);
237
238 /**
239 * Add a NUL-terminated string to the buffer. Each line will be prepended
240 * by a comment character and a blank.
241 */
242 extern void strbuf_add_commented_lines(struct strbuf *out, const char *buf, size_t size);
243
244
245 /**
246 * Add data of given length to the buffer.
247 */
248 extern void strbuf_add(struct strbuf *, const void *, size_t);
249
250 /**
251 * Add a NUL-terminated string to the buffer.
252 *
253 * NOTE: This function will *always* be implemented as an inline or a macro
254 * using strlen, meaning that this is efficient to write things like:
255 *
256 * strbuf_addstr(sb, "immediate string");
257 *
258 */
259 static inline void strbuf_addstr(struct strbuf *sb, const char *s)
260 {
261 strbuf_add(sb, s, strlen(s));
262 }
263
264 /**
265 * Copy the contents of another buffer at the end of the current one.
266 */
267 extern void strbuf_addbuf(struct strbuf *sb, const struct strbuf *sb2);
268
269 /**
270 * This function can be used to expand a format string containing
271 * placeholders. To that end, it parses the string and calls the specified
272 * function for every percent sign found.
273 *
274 * The callback function is given a pointer to the character after the `%`
275 * and a pointer to the struct strbuf. It is expected to add the expanded
276 * version of the placeholder to the strbuf, e.g. to add a newline
277 * character if the letter `n` appears after a `%`. The function returns
278 * the length of the placeholder recognized and `strbuf_expand()` skips
279 * over it.
280 *
281 * The format `%%` is automatically expanded to a single `%` as a quoting
282 * mechanism; callers do not need to handle the `%` placeholder themselves,
283 * and the callback function will not be invoked for this placeholder.
284 *
285 * All other characters (non-percent and not skipped ones) are copied
286 * verbatim to the strbuf. If the callback returned zero, meaning that the
287 * placeholder is unknown, then the percent sign is copied, too.
288 *
289 * In order to facilitate caching and to make it possible to give
290 * parameters to the callback, `strbuf_expand()` passes a context pointer,
291 * which can be used by the programmer of the callback as she sees fit.
292 */
293 typedef size_t (*expand_fn_t) (struct strbuf *sb, const char *placeholder, void *context);
294 extern void strbuf_expand(struct strbuf *sb, const char *format, expand_fn_t fn, void *context);
295
296 /**
297 * Used as callback for `strbuf_expand()`, expects an array of
298 * struct strbuf_expand_dict_entry as context, i.e. pairs of
299 * placeholder and replacement string. The array needs to be
300 * terminated by an entry with placeholder set to NULL.
301 */
302 struct strbuf_expand_dict_entry {
303 const char *placeholder;
304 const char *value;
305 };
306 extern size_t strbuf_expand_dict_cb(struct strbuf *sb, const char *placeholder, void *context);
307
308 /**
309 * Append the contents of one strbuf to another, quoting any
310 * percent signs ("%") into double-percents ("%%") in the
311 * destination. This is useful for literal data to be fed to either
312 * strbuf_expand or to the *printf family of functions.
313 */
314 extern void strbuf_addbuf_percentquote(struct strbuf *dst, const struct strbuf *src);
315
316 /**
317 * Append the given byte size as a human-readable string (i.e. 12.23 KiB,
318 * 3.50 MiB).
319 */
320 extern void strbuf_humanise_bytes(struct strbuf *buf, off_t bytes);
321
322 /**
323 * Add a formatted string to the buffer.
324 */
325 __attribute__((format (printf,2,3)))
326 extern void strbuf_addf(struct strbuf *sb, const char *fmt, ...);
327
328 /**
329 * Add a formatted string prepended by a comment character and a
330 * blank to the buffer.
331 */
332 __attribute__((format (printf, 2, 3)))
333 extern void strbuf_commented_addf(struct strbuf *sb, const char *fmt, ...);
334
335 __attribute__((format (printf,2,0)))
336 extern void strbuf_vaddf(struct strbuf *sb, const char *fmt, va_list ap);
337
338 /**
339 * Add the time specified by `tm`, as formatted by `strftime`.
340 * `tz_offset` is in decimal hhmm format, e.g. -600 means six hours west
341 * of Greenwich, and it's used to expand %z internally. However, tokens
342 * with modifiers (e.g. %Ez) are passed to `strftime`.
343 * `suppress_tz_name`, when set, expands %Z internally to the empty
344 * string rather than passing it to `strftime`.
345 */
346 extern void strbuf_addftime(struct strbuf *sb, const char *fmt,
347 const struct tm *tm, int tz_offset,
348 int suppress_tz_name);
349
350 /**
351 * Read a given size of data from a FILE* pointer to the buffer.
352 *
353 * NOTE: The buffer is rewound if the read fails. If -1 is returned,
354 * `errno` must be consulted, like you would do for `read(3)`.
355 * `strbuf_read()`, `strbuf_read_file()` and `strbuf_getline_*()`
356 * family of functions have the same behaviour as well.
357 */
358 extern size_t strbuf_fread(struct strbuf *, size_t, FILE *);
359
360 /**
361 * Read the contents of a given file descriptor. The third argument can be
362 * used to give a hint about the file size, to avoid reallocs. If read fails,
363 * any partial read is undone.
364 */
365 extern ssize_t strbuf_read(struct strbuf *, int fd, size_t hint);
366
367 /**
368 * Read the contents of a given file descriptor partially by using only one
369 * attempt of xread. The third argument can be used to give a hint about the
370 * file size, to avoid reallocs. Returns the number of new bytes appended to
371 * the sb.
372 */
373 extern ssize_t strbuf_read_once(struct strbuf *, int fd, size_t hint);
374
375 /**
376 * Read the contents of a file, specified by its path. The third argument
377 * can be used to give a hint about the file size, to avoid reallocs.
378 * Return the number of bytes read or a negative value if some error
379 * occurred while opening or reading the file.
380 */
381 extern ssize_t strbuf_read_file(struct strbuf *sb, const char *path, size_t hint);
382
383 /**
384 * Read the target of a symbolic link, specified by its path. The third
385 * argument can be used to give a hint about the size, to avoid reallocs.
386 */
387 extern int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint);
388
389 /**
390 * Write the whole content of the strbuf to the stream not stopping at
391 * NUL bytes.
392 */
393 extern ssize_t strbuf_write(struct strbuf *sb, FILE *stream);
394
395 /**
396 * Read a line from a FILE *, overwriting the existing contents of
397 * the strbuf. The strbuf_getline*() family of functions share
398 * this signature, but have different line termination conventions.
399 *
400 * Reading stops after the terminator or at EOF. The terminator
401 * is removed from the buffer before returning. Returns 0 unless
402 * there was nothing left before EOF, in which case it returns `EOF`.
403 */
404 typedef int (*strbuf_getline_fn)(struct strbuf *, FILE *);
405
406 /* Uses LF as the line terminator */
407 extern int strbuf_getline_lf(struct strbuf *sb, FILE *fp);
408
409 /* Uses NUL as the line terminator */
410 extern int strbuf_getline_nul(struct strbuf *sb, FILE *fp);
411
412 /*
413 * Similar to strbuf_getline_lf(), but additionally treats a CR that
414 * comes immediately before the LF as part of the terminator.
415 * This is the most friendly version to be used to read "text" files
416 * that can come from platforms whose native text format is CRLF
417 * terminated.
418 */
419 extern int strbuf_getline(struct strbuf *, FILE *);
420
421
422 /**
423 * Like `strbuf_getline`, but keeps the trailing terminator (if
424 * any) in the buffer.
425 */
426 extern int strbuf_getwholeline(struct strbuf *, FILE *, int);
427
428 /**
429 * Like `strbuf_getwholeline`, but operates on a file descriptor.
430 * It reads one character at a time, so it is very slow. Do not
431 * use it unless you need the correct position in the file
432 * descriptor.
433 */
434 extern int strbuf_getwholeline_fd(struct strbuf *, int, int);
435
436 /**
437 * Set the buffer to the path of the current working directory.
438 */
439 extern int strbuf_getcwd(struct strbuf *sb);
440
441 /**
442 * Add a path to a buffer, converting a relative path to an
443 * absolute one in the process. Symbolic links are not
444 * resolved.
445 */
446 extern void strbuf_add_absolute_path(struct strbuf *sb, const char *path);
447
448 /**
449 * Canonize `path` (make it absolute, resolve symlinks, remove extra
450 * slashes) and append it to `sb`. Die with an informative error
451 * message if there is a problem.
452 *
453 * The directory part of `path` (i.e., everything up to the last
454 * dir_sep) must denote a valid, existing directory, but the last
455 * component need not exist.
456 *
457 * Callers that don't mind links should use the more lightweight
458 * strbuf_add_absolute_path() instead.
459 */
460 extern void strbuf_add_real_path(struct strbuf *sb, const char *path);
461
462
463 /**
464 * Normalize in-place the path contained in the strbuf. See
465 * normalize_path_copy() for details. If an error occurs, the contents of "sb"
466 * are left untouched, and -1 is returned.
467 */
468 extern int strbuf_normalize_path(struct strbuf *sb);
469
470 /**
471 * Strip whitespace from a buffer. The second parameter controls if
472 * comments are considered contents to be removed or not.
473 */
474 extern void strbuf_stripspace(struct strbuf *buf, int skip_comments);
475
476 /**
477 * Temporary alias until all topic branches have switched to use
478 * strbuf_stripspace directly.
479 */
480 static inline void stripspace(struct strbuf *buf, int skip_comments)
481 {
482 strbuf_stripspace(buf, skip_comments);
483 }
484
485 static inline int strbuf_strip_suffix(struct strbuf *sb, const char *suffix)
486 {
487 if (strip_suffix_mem(sb->buf, &sb->len, suffix)) {
488 strbuf_setlen(sb, sb->len);
489 return 1;
490 } else
491 return 0;
492 }
493
494 /**
495 * Split str (of length slen) at the specified terminator character.
496 * Return a null-terminated array of pointers to strbuf objects
497 * holding the substrings. The substrings include the terminator,
498 * except for the last substring, which might be unterminated if the
499 * original string did not end with a terminator. If max is positive,
500 * then split the string into at most max substrings (with the last
501 * substring containing everything following the (max-1)th terminator
502 * character).
503 *
504 * The most generic form is `strbuf_split_buf`, which takes an arbitrary
505 * pointer/len buffer. The `_str` variant takes a NUL-terminated string,
506 * the `_max` variant takes a strbuf, and just `strbuf_split` is a convenience
507 * wrapper to drop the `max` parameter.
508 *
509 * For lighter-weight alternatives, see string_list_split() and
510 * string_list_split_in_place().
511 */
512 extern struct strbuf **strbuf_split_buf(const char *, size_t,
513 int terminator, int max);
514
515 static inline struct strbuf **strbuf_split_str(const char *str,
516 int terminator, int max)
517 {
518 return strbuf_split_buf(str, strlen(str), terminator, max);
519 }
520
521 static inline struct strbuf **strbuf_split_max(const struct strbuf *sb,
522 int terminator, int max)
523 {
524 return strbuf_split_buf(sb->buf, sb->len, terminator, max);
525 }
526
527 static inline struct strbuf **strbuf_split(const struct strbuf *sb,
528 int terminator)
529 {
530 return strbuf_split_max(sb, terminator, 0);
531 }
532
533 /**
534 * Free a NULL-terminated list of strbufs (for example, the return
535 * values of the strbuf_split*() functions).
536 */
537 extern void strbuf_list_free(struct strbuf **);
538
539 /**
540 * Add the abbreviation, as generated by find_unique_abbrev, of `sha1` to
541 * the strbuf `sb`.
542 */
543 extern void strbuf_add_unique_abbrev(struct strbuf *sb,
544 const unsigned char *sha1,
545 int abbrev_len);
546
547 /**
548 * Launch the user preferred editor to edit a file and fill the buffer
549 * with the file's contents upon the user completing their editing. The
550 * third argument can be used to set the environment which the editor is
551 * run in. If the buffer is NULL the editor is launched as usual but the
552 * file's contents are not read into the buffer upon completion.
553 */
554 extern int launch_editor(const char *path, struct strbuf *buffer, const char *const *env);
555
556 extern void strbuf_add_lines(struct strbuf *sb, const char *prefix, const char *buf, size_t size);
557
558 /**
559 * Append s to sb, with the characters '<', '>', '&' and '"' converted
560 * into XML entities.
561 */
562 extern void strbuf_addstr_xml_quoted(struct strbuf *sb, const char *s);
563
564 /**
565 * "Complete" the contents of `sb` by ensuring that either it ends with the
566 * character `term`, or it is empty. This can be used, for example,
567 * to ensure that text ends with a newline, but without creating an empty
568 * blank line if there is no content in the first place.
569 */
570 static inline void strbuf_complete(struct strbuf *sb, char term)
571 {
572 if (sb->len && sb->buf[sb->len - 1] != term)
573 strbuf_addch(sb, term);
574 }
575
576 static inline void strbuf_complete_line(struct strbuf *sb)
577 {
578 strbuf_complete(sb, '\n');
579 }
580
581 /*
582 * Copy "name" to "sb", expanding any special @-marks as handled by
583 * interpret_branch_name(). The result is a non-qualified branch name
584 * (so "foo" or "origin/master" instead of "refs/heads/foo" or
585 * "refs/remotes/origin/master").
586 *
587 * Note that the resulting name may not be a syntactically valid refname.
588 *
589 * If "allowed" is non-zero, restrict the set of allowed expansions. See
590 * interpret_branch_name() for details.
591 */
592 extern void strbuf_branchname(struct strbuf *sb, const char *name,
593 unsigned allowed);
594
595 /*
596 * Like strbuf_branchname() above, but confirm that the result is
597 * syntactically valid to be used as a local branch name in refs/heads/.
598 *
599 * The return value is "0" if the result is valid, and "-1" otherwise.
600 */
601 extern int strbuf_check_branch_ref(struct strbuf *sb, const char *name);
602
603 extern void strbuf_addstr_urlencode(struct strbuf *, const char *,
604 int reserved);
605
606 __attribute__((format (printf,1,2)))
607 extern int printf_ln(const char *fmt, ...);
608 __attribute__((format (printf,2,3)))
609 extern int fprintf_ln(FILE *fp, const char *fmt, ...);
610
611 char *xstrdup_tolower(const char *);
612
613 /**
614 * Create a newly allocated string using printf format. You can do this easily
615 * with a strbuf, but this provides a shortcut to save a few lines.
616 */
617 __attribute__((format (printf, 1, 0)))
618 char *xstrvfmt(const char *fmt, va_list ap);
619 __attribute__((format (printf, 1, 2)))
620 char *xstrfmt(const char *fmt, ...);
621
622 #endif /* STRBUF_H */