1#ifndef STRBUF_H
2#define STRBUF_H
3
4struct string_list;
5
6/**
7 * strbuf's are meant to be used with all the usual C string and memory
8 * APIs. Given that the length of the buffer is known, it's often better to
9 * use the mem* functions than a str* one (memchr vs. strchr e.g.).
10 * Though, one has to be careful about the fact that str* functions often
11 * stop on NULs and that strbufs may have embedded NULs.
12 *
13 * A strbuf is NUL terminated for convenience, but no function in the
14 * strbuf API actually relies on the string being free of NULs.
15 *
16 * strbufs have some invariants that are very important to keep in mind:
17 *
18 * - The `buf` member is never NULL, so it can be used in any usual C
19 * string operations safely. strbuf's _have_ to be initialized either by
20 * `strbuf_init()` or by `= STRBUF_INIT` before the invariants, though.
21 *
22 * Do *not* assume anything on what `buf` really is (e.g. if it is
23 * allocated memory or not), use `strbuf_detach()` to unwrap a memory
24 * buffer from its strbuf shell in a safe way. That is the sole supported
25 * way. This will give you a malloced buffer that you can later `free()`.
26 *
27 * However, it is totally safe to modify anything in the string pointed by
28 * the `buf` member, between the indices `0` and `len-1` (inclusive).
29 *
30 * - The `buf` member is a byte array that has at least `len + 1` bytes
31 * allocated. The extra byte is used to store a `'\0'`, allowing the
32 * `buf` member to be a valid C-string. Every strbuf function ensure this
33 * invariant is preserved.
34 *
35 * NOTE: It is OK to "play" with the buffer directly if you work it this
36 * way:
37 *
38 * strbuf_grow(sb, SOME_SIZE); <1>
39 * strbuf_setlen(sb, sb->len + SOME_OTHER_SIZE);
40 *
41 * <1> Here, the memory array starting at `sb->buf`, and of length
42 * `strbuf_avail(sb)` is all yours, and you can be sure that
43 * `strbuf_avail(sb)` is at least `SOME_SIZE`.
44 *
45 * NOTE: `SOME_OTHER_SIZE` must be smaller or equal to `strbuf_avail(sb)`.
46 *
47 * Doing so is safe, though if it has to be done in many places, adding the
48 * missing API to the strbuf module is the way to go.
49 *
50 * WARNING: Do _not_ assume that the area that is yours is of size `alloc
51 * - 1` even if it's true in the current implementation. Alloc is somehow a
52 * "private" member that should not be messed with. Use `strbuf_avail()`
53 * instead.
54*/
55
56/**
57 * Data Structures
58 * ---------------
59 */
60
61/**
62 * This is the string buffer structure. The `len` member can be used to
63 * determine the current length of the string, and `buf` member provides
64 * access to the string itself.
65 */
66struct strbuf {
67 size_t alloc;
68 size_t len;
69 char *buf;
70};
71
72extern char strbuf_slopbuf[];
73#define STRBUF_INIT { .alloc = 0, .len = 0, .buf = strbuf_slopbuf }
74
75/*
76 * Predeclare this here, since cache.h includes this file before it defines the
77 * struct.
78 */
79struct object_id;
80
81/**
82 * Life Cycle Functions
83 * --------------------
84 */
85
86/**
87 * Initialize the structure. The second parameter can be zero or a bigger
88 * number to allocate memory, in case you want to prevent further reallocs.
89 */
90extern void strbuf_init(struct strbuf *, size_t);
91
92/**
93 * Release a string buffer and the memory it used. After this call, the
94 * strbuf points to an empty string that does not need to be free()ed, as
95 * if it had been set to `STRBUF_INIT` and never modified.
96 *
97 * To clear a strbuf in preparation for further use without the overhead
98 * of free()ing and malloc()ing again, use strbuf_reset() instead.
99 */
100extern void strbuf_release(struct strbuf *);
101
102/**
103 * Detach the string from the strbuf and returns it; you now own the
104 * storage the string occupies and it is your responsibility from then on
105 * to release it with `free(3)` when you are done with it.
106 *
107 * The strbuf that previously held the string is reset to `STRBUF_INIT` so
108 * it can be reused after calling this function.
109 */
110extern char *strbuf_detach(struct strbuf *, size_t *);
111
112/**
113 * Attach a string to a buffer. You should specify the string to attach,
114 * the current length of the string and the amount of allocated memory.
115 * The amount must be larger than the string length, because the string you
116 * pass is supposed to be a NUL-terminated string. This string _must_ be
117 * malloc()ed, and after attaching, the pointer cannot be relied upon
118 * anymore, and neither be free()d directly.
119 */
120extern void strbuf_attach(struct strbuf *, void *, size_t, size_t);
121
122/**
123 * Swap the contents of two string buffers.
124 */
125static inline void strbuf_swap(struct strbuf *a, struct strbuf *b)
126{
127 SWAP(*a, *b);
128}
129
130
131/**
132 * Functions related to the size of the buffer
133 * -------------------------------------------
134 */
135
136/**
137 * Determine the amount of allocated but unused memory.
138 */
139static inline size_t strbuf_avail(const struct strbuf *sb)
140{
141 return sb->alloc ? sb->alloc - sb->len - 1 : 0;
142}
143
144/**
145 * Ensure that at least this amount of unused memory is available after
146 * `len`. This is used when you know a typical size for what you will add
147 * and want to avoid repetitive automatic resizing of the underlying buffer.
148 * This is never a needed operation, but can be critical for performance in
149 * some cases.
150 */
151extern void strbuf_grow(struct strbuf *, size_t);
152
153/**
154 * Set the length of the buffer to a given value. This function does *not*
155 * allocate new memory, so you should not perform a `strbuf_setlen()` to a
156 * length that is larger than `len + strbuf_avail()`. `strbuf_setlen()` is
157 * just meant as a 'please fix invariants from this strbuf I just messed
158 * with'.
159 */
160static inline void strbuf_setlen(struct strbuf *sb, size_t len)
161{
162 if (len > (sb->alloc ? sb->alloc - 1 : 0))
163 die("BUG: strbuf_setlen() beyond buffer");
164 sb->len = len;
165 if (sb->buf != strbuf_slopbuf)
166 sb->buf[len] = '\0';
167 else
168 assert(!strbuf_slopbuf[0]);
169}
170
171/**
172 * Empty the buffer by setting the size of it to zero.
173 */
174#define strbuf_reset(sb) strbuf_setlen(sb, 0)
175
176
177/**
178 * Functions related to the contents of the buffer
179 * -----------------------------------------------
180 */
181
182/**
183 * Strip whitespace from the beginning (`ltrim`), end (`rtrim`), or both side
184 * (`trim`) of a string.
185 */
186extern void strbuf_trim(struct strbuf *);
187extern void strbuf_rtrim(struct strbuf *);
188extern void strbuf_ltrim(struct strbuf *);
189
190/* Strip trailing directory separators */
191extern void strbuf_trim_trailing_dir_sep(struct strbuf *);
192
193/**
194 * Replace the contents of the strbuf with a reencoded form. Returns -1
195 * on error, 0 on success.
196 */
197extern int strbuf_reencode(struct strbuf *sb, const char *from, const char *to);
198
199/**
200 * Lowercase each character in the buffer using `tolower`.
201 */
202extern void strbuf_tolower(struct strbuf *sb);
203
204/**
205 * Compare two buffers. Returns an integer less than, equal to, or greater
206 * than zero if the first buffer is found, respectively, to be less than,
207 * to match, or be greater than the second buffer.
208 */
209extern int strbuf_cmp(const struct strbuf *, const struct strbuf *);
210
211
212/**
213 * Adding data to the buffer
214 * -------------------------
215 *
216 * NOTE: All of the functions in this section will grow the buffer as
217 * necessary. If they fail for some reason other than memory shortage and the
218 * buffer hadn't been allocated before (i.e. the `struct strbuf` was set to
219 * `STRBUF_INIT`), then they will free() it.
220 */
221
222/**
223 * Add a single character to the buffer.
224 */
225static inline void strbuf_addch(struct strbuf *sb, int c)
226{
227 if (!strbuf_avail(sb))
228 strbuf_grow(sb, 1);
229 sb->buf[sb->len++] = c;
230 sb->buf[sb->len] = '\0';
231}
232
233/**
234 * Add a character the specified number of times to the buffer.
235 */
236extern void strbuf_addchars(struct strbuf *sb, int c, size_t n);
237
238/**
239 * Insert data to the given position of the buffer. The remaining contents
240 * will be shifted, not overwritten.
241 */
242extern void strbuf_insert(struct strbuf *, size_t pos, const void *, size_t);
243
244/**
245 * Remove given amount of data from a given position of the buffer.
246 */
247extern void strbuf_remove(struct strbuf *, size_t pos, size_t len);
248
249/**
250 * Remove the bytes between `pos..pos+len` and replace it with the given
251 * data.
252 */
253extern void strbuf_splice(struct strbuf *, size_t pos, size_t len,
254 const void *, size_t);
255
256/**
257 * Add a NUL-terminated string to the buffer. Each line will be prepended
258 * by a comment character and a blank.
259 */
260extern void strbuf_add_commented_lines(struct strbuf *out, const char *buf, size_t size);
261
262
263/**
264 * Add data of given length to the buffer.
265 */
266extern void strbuf_add(struct strbuf *, const void *, size_t);
267
268/**
269 * Add a NUL-terminated string to the buffer.
270 *
271 * NOTE: This function will *always* be implemented as an inline or a macro
272 * using strlen, meaning that this is efficient to write things like:
273 *
274 * strbuf_addstr(sb, "immediate string");
275 *
276 */
277static inline void strbuf_addstr(struct strbuf *sb, const char *s)
278{
279 strbuf_add(sb, s, strlen(s));
280}
281
282/**
283 * Copy the contents of another buffer at the end of the current one.
284 */
285extern void strbuf_addbuf(struct strbuf *sb, const struct strbuf *sb2);
286
287/**
288 * This function can be used to expand a format string containing
289 * placeholders. To that end, it parses the string and calls the specified
290 * function for every percent sign found.
291 *
292 * The callback function is given a pointer to the character after the `%`
293 * and a pointer to the struct strbuf. It is expected to add the expanded
294 * version of the placeholder to the strbuf, e.g. to add a newline
295 * character if the letter `n` appears after a `%`. The function returns
296 * the length of the placeholder recognized and `strbuf_expand()` skips
297 * over it.
298 *
299 * The format `%%` is automatically expanded to a single `%` as a quoting
300 * mechanism; callers do not need to handle the `%` placeholder themselves,
301 * and the callback function will not be invoked for this placeholder.
302 *
303 * All other characters (non-percent and not skipped ones) are copied
304 * verbatim to the strbuf. If the callback returned zero, meaning that the
305 * placeholder is unknown, then the percent sign is copied, too.
306 *
307 * In order to facilitate caching and to make it possible to give
308 * parameters to the callback, `strbuf_expand()` passes a context pointer,
309 * which can be used by the programmer of the callback as she sees fit.
310 */
311typedef size_t (*expand_fn_t) (struct strbuf *sb, const char *placeholder, void *context);
312extern void strbuf_expand(struct strbuf *sb, const char *format, expand_fn_t fn, void *context);
313
314/**
315 * Used as callback for `strbuf_expand()`, expects an array of
316 * struct strbuf_expand_dict_entry as context, i.e. pairs of
317 * placeholder and replacement string. The array needs to be
318 * terminated by an entry with placeholder set to NULL.
319 */
320struct strbuf_expand_dict_entry {
321 const char *placeholder;
322 const char *value;
323};
324extern size_t strbuf_expand_dict_cb(struct strbuf *sb, const char *placeholder, void *context);
325
326/**
327 * Append the contents of one strbuf to another, quoting any
328 * percent signs ("%") into double-percents ("%%") in the
329 * destination. This is useful for literal data to be fed to either
330 * strbuf_expand or to the *printf family of functions.
331 */
332extern void strbuf_addbuf_percentquote(struct strbuf *dst, const struct strbuf *src);
333
334/**
335 * Append the given byte size as a human-readable string (i.e. 12.23 KiB,
336 * 3.50 MiB).
337 */
338extern void strbuf_humanise_bytes(struct strbuf *buf, off_t bytes);
339
340/**
341 * Add a formatted string to the buffer.
342 */
343__attribute__((format (printf,2,3)))
344extern void strbuf_addf(struct strbuf *sb, const char *fmt, ...);
345
346/**
347 * Add a formatted string prepended by a comment character and a
348 * blank to the buffer.
349 */
350__attribute__((format (printf, 2, 3)))
351extern void strbuf_commented_addf(struct strbuf *sb, const char *fmt, ...);
352
353__attribute__((format (printf,2,0)))
354extern void strbuf_vaddf(struct strbuf *sb, const char *fmt, va_list ap);
355
356/**
357 * Add the time specified by `tm`, as formatted by `strftime`.
358 * `tz_offset` is in decimal hhmm format, e.g. -600 means six hours west
359 * of Greenwich, and it's used to expand %z internally. However, tokens
360 * with modifiers (e.g. %Ez) are passed to `strftime`.
361 * `suppress_tz_name`, when set, expands %Z internally to the empty
362 * string rather than passing it to `strftime`.
363 */
364extern void strbuf_addftime(struct strbuf *sb, const char *fmt,
365 const struct tm *tm, int tz_offset,
366 int suppress_tz_name);
367
368/**
369 * Read a given size of data from a FILE* pointer to the buffer.
370 *
371 * NOTE: The buffer is rewound if the read fails. If -1 is returned,
372 * `errno` must be consulted, like you would do for `read(3)`.
373 * `strbuf_read()`, `strbuf_read_file()` and `strbuf_getline_*()`
374 * family of functions have the same behaviour as well.
375 */
376extern size_t strbuf_fread(struct strbuf *, size_t, FILE *);
377
378/**
379 * Read the contents of a given file descriptor. The third argument can be
380 * used to give a hint about the file size, to avoid reallocs. If read fails,
381 * any partial read is undone.
382 */
383extern ssize_t strbuf_read(struct strbuf *, int fd, size_t hint);
384
385/**
386 * Read the contents of a given file descriptor partially by using only one
387 * attempt of xread. The third argument can be used to give a hint about the
388 * file size, to avoid reallocs. Returns the number of new bytes appended to
389 * the sb.
390 */
391extern ssize_t strbuf_read_once(struct strbuf *, int fd, size_t hint);
392
393/**
394 * Read the contents of a file, specified by its path. The third argument
395 * can be used to give a hint about the file size, to avoid reallocs.
396 * Return the number of bytes read or a negative value if some error
397 * occurred while opening or reading the file.
398 */
399extern ssize_t strbuf_read_file(struct strbuf *sb, const char *path, size_t hint);
400
401/**
402 * Read the target of a symbolic link, specified by its path. The third
403 * argument can be used to give a hint about the size, to avoid reallocs.
404 */
405extern int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint);
406
407/**
408 * Write the whole content of the strbuf to the stream not stopping at
409 * NUL bytes.
410 */
411extern ssize_t strbuf_write(struct strbuf *sb, FILE *stream);
412
413/**
414 * Read a line from a FILE *, overwriting the existing contents of
415 * the strbuf. The strbuf_getline*() family of functions share
416 * this signature, but have different line termination conventions.
417 *
418 * Reading stops after the terminator or at EOF. The terminator
419 * is removed from the buffer before returning. Returns 0 unless
420 * there was nothing left before EOF, in which case it returns `EOF`.
421 */
422typedef int (*strbuf_getline_fn)(struct strbuf *, FILE *);
423
424/* Uses LF as the line terminator */
425extern int strbuf_getline_lf(struct strbuf *sb, FILE *fp);
426
427/* Uses NUL as the line terminator */
428extern int strbuf_getline_nul(struct strbuf *sb, FILE *fp);
429
430/*
431 * Similar to strbuf_getline_lf(), but additionally treats a CR that
432 * comes immediately before the LF as part of the terminator.
433 * This is the most friendly version to be used to read "text" files
434 * that can come from platforms whose native text format is CRLF
435 * terminated.
436 */
437extern int strbuf_getline(struct strbuf *, FILE *);
438
439
440/**
441 * Like `strbuf_getline`, but keeps the trailing terminator (if
442 * any) in the buffer.
443 */
444extern int strbuf_getwholeline(struct strbuf *, FILE *, int);
445
446/**
447 * Like `strbuf_getwholeline`, but operates on a file descriptor.
448 * It reads one character at a time, so it is very slow. Do not
449 * use it unless you need the correct position in the file
450 * descriptor.
451 */
452extern int strbuf_getwholeline_fd(struct strbuf *, int, int);
453
454/**
455 * Set the buffer to the path of the current working directory.
456 */
457extern int strbuf_getcwd(struct strbuf *sb);
458
459/**
460 * Add a path to a buffer, converting a relative path to an
461 * absolute one in the process. Symbolic links are not
462 * resolved.
463 */
464extern void strbuf_add_absolute_path(struct strbuf *sb, const char *path);
465
466/**
467 * Canonize `path` (make it absolute, resolve symlinks, remove extra
468 * slashes) and append it to `sb`. Die with an informative error
469 * message if there is a problem.
470 *
471 * The directory part of `path` (i.e., everything up to the last
472 * dir_sep) must denote a valid, existing directory, but the last
473 * component need not exist.
474 *
475 * Callers that don't mind links should use the more lightweight
476 * strbuf_add_absolute_path() instead.
477 */
478extern void strbuf_add_real_path(struct strbuf *sb, const char *path);
479
480
481/**
482 * Normalize in-place the path contained in the strbuf. See
483 * normalize_path_copy() for details. If an error occurs, the contents of "sb"
484 * are left untouched, and -1 is returned.
485 */
486extern int strbuf_normalize_path(struct strbuf *sb);
487
488/**
489 * Strip whitespace from a buffer. The second parameter controls if
490 * comments are considered contents to be removed or not.
491 */
492extern void strbuf_stripspace(struct strbuf *buf, int skip_comments);
493
494static inline int strbuf_strip_suffix(struct strbuf *sb, const char *suffix)
495{
496 if (strip_suffix_mem(sb->buf, &sb->len, suffix)) {
497 strbuf_setlen(sb, sb->len);
498 return 1;
499 } else
500 return 0;
501}
502
503/**
504 * Split str (of length slen) at the specified terminator character.
505 * Return a null-terminated array of pointers to strbuf objects
506 * holding the substrings. The substrings include the terminator,
507 * except for the last substring, which might be unterminated if the
508 * original string did not end with a terminator. If max is positive,
509 * then split the string into at most max substrings (with the last
510 * substring containing everything following the (max-1)th terminator
511 * character).
512 *
513 * The most generic form is `strbuf_split_buf`, which takes an arbitrary
514 * pointer/len buffer. The `_str` variant takes a NUL-terminated string,
515 * the `_max` variant takes a strbuf, and just `strbuf_split` is a convenience
516 * wrapper to drop the `max` parameter.
517 *
518 * For lighter-weight alternatives, see string_list_split() and
519 * string_list_split_in_place().
520 */
521extern struct strbuf **strbuf_split_buf(const char *, size_t,
522 int terminator, int max);
523
524static inline struct strbuf **strbuf_split_str(const char *str,
525 int terminator, int max)
526{
527 return strbuf_split_buf(str, strlen(str), terminator, max);
528}
529
530static inline struct strbuf **strbuf_split_max(const struct strbuf *sb,
531 int terminator, int max)
532{
533 return strbuf_split_buf(sb->buf, sb->len, terminator, max);
534}
535
536static inline struct strbuf **strbuf_split(const struct strbuf *sb,
537 int terminator)
538{
539 return strbuf_split_max(sb, terminator, 0);
540}
541
542/*
543 * Adds all strings of a string list to the strbuf, separated by the given
544 * separator. For example, if sep is
545 * ', '
546 * and slist contains
547 * ['element1', 'element2', ..., 'elementN'],
548 * then write:
549 * 'element1, element2, ..., elementN'
550 * to str. If only one element, just write "element1" to str.
551 */
552extern void strbuf_add_separated_string_list(struct strbuf *str,
553 const char *sep,
554 struct string_list *slist);
555
556/**
557 * Free a NULL-terminated list of strbufs (for example, the return
558 * values of the strbuf_split*() functions).
559 */
560extern void strbuf_list_free(struct strbuf **);
561
562/**
563 * Add the abbreviation, as generated by find_unique_abbrev, of `sha1` to
564 * the strbuf `sb`.
565 */
566extern void strbuf_add_unique_abbrev(struct strbuf *sb,
567 const struct object_id *oid,
568 int abbrev_len);
569
570/**
571 * Launch the user preferred editor to edit a file and fill the buffer
572 * with the file's contents upon the user completing their editing. The
573 * third argument can be used to set the environment which the editor is
574 * run in. If the buffer is NULL the editor is launched as usual but the
575 * file's contents are not read into the buffer upon completion.
576 */
577extern int launch_editor(const char *path, struct strbuf *buffer, const char *const *env);
578
579extern void strbuf_add_lines(struct strbuf *sb, const char *prefix, const char *buf, size_t size);
580
581/**
582 * Append s to sb, with the characters '<', '>', '&' and '"' converted
583 * into XML entities.
584 */
585extern void strbuf_addstr_xml_quoted(struct strbuf *sb, const char *s);
586
587/**
588 * "Complete" the contents of `sb` by ensuring that either it ends with the
589 * character `term`, or it is empty. This can be used, for example,
590 * to ensure that text ends with a newline, but without creating an empty
591 * blank line if there is no content in the first place.
592 */
593static inline void strbuf_complete(struct strbuf *sb, char term)
594{
595 if (sb->len && sb->buf[sb->len - 1] != term)
596 strbuf_addch(sb, term);
597}
598
599static inline void strbuf_complete_line(struct strbuf *sb)
600{
601 strbuf_complete(sb, '\n');
602}
603
604/*
605 * Copy "name" to "sb", expanding any special @-marks as handled by
606 * interpret_branch_name(). The result is a non-qualified branch name
607 * (so "foo" or "origin/master" instead of "refs/heads/foo" or
608 * "refs/remotes/origin/master").
609 *
610 * Note that the resulting name may not be a syntactically valid refname.
611 *
612 * If "allowed" is non-zero, restrict the set of allowed expansions. See
613 * interpret_branch_name() for details.
614 */
615extern void strbuf_branchname(struct strbuf *sb, const char *name,
616 unsigned allowed);
617
618/*
619 * Like strbuf_branchname() above, but confirm that the result is
620 * syntactically valid to be used as a local branch name in refs/heads/.
621 *
622 * The return value is "0" if the result is valid, and "-1" otherwise.
623 */
624extern int strbuf_check_branch_ref(struct strbuf *sb, const char *name);
625
626extern void strbuf_addstr_urlencode(struct strbuf *, const char *,
627 int reserved);
628
629__attribute__((format (printf,1,2)))
630extern int printf_ln(const char *fmt, ...);
631__attribute__((format (printf,2,3)))
632extern int fprintf_ln(FILE *fp, const char *fmt, ...);
633
634char *xstrdup_tolower(const char *);
635char *xstrdup_toupper(const char *);
636
637/**
638 * Create a newly allocated string using printf format. You can do this easily
639 * with a strbuf, but this provides a shortcut to save a few lines.
640 */
641__attribute__((format (printf, 1, 0)))
642char *xstrvfmt(const char *fmt, va_list ap);
643__attribute__((format (printf, 1, 2)))
644char *xstrfmt(const char *fmt, ...);
645
646#endif /* STRBUF_H */
647