t0080: mark as leak-free
[alt-git.git] / strbuf.h
blobe959caca876ac759debb0c77cd7c6d1046cbc25d
1 #ifndef STRBUF_H
2 #define STRBUF_H
4 /*
5 * NOTE FOR STRBUF DEVELOPERS
7 * strbuf is a low-level primitive; as such it should interact only
8 * with other low-level primitives. Do not introduce new functions
9 * which interact with higher-level APIs.
12 struct string_list;
14 /**
15 * strbufs are meant to be used with all the usual C string and memory
16 * APIs. Given that the length of the buffer is known, it's often better to
17 * use the mem* functions than a str* one (e.g., memchr vs. strchr).
18 * Though, one has to be careful about the fact that str* functions often
19 * stop on NULs and that strbufs may have embedded NULs.
21 * A strbuf is NUL terminated for convenience, but no function in the
22 * strbuf API actually relies on the string being free of NULs.
24 * strbufs have some invariants that are very important to keep in mind:
26 * - The `buf` member is never NULL, so it can be used in any usual C
27 * string operations safely. strbufs _have_ to be initialized either by
28 * `strbuf_init()` or by `= STRBUF_INIT` before the invariants, though.
30 * Do *not* assume anything on what `buf` really is (e.g. if it is
31 * allocated memory or not), use `strbuf_detach()` to unwrap a memory
32 * buffer from its strbuf shell in a safe way. That is the sole supported
33 * way. This will give you a malloced buffer that you can later `free()`.
35 * However, it is totally safe to modify anything in the string pointed by
36 * the `buf` member, between the indices `0` and `len-1` (inclusive).
38 * - The `buf` member is a byte array that has at least `len + 1` bytes
39 * allocated. The extra byte is used to store a `'\0'`, allowing the
40 * `buf` member to be a valid C-string. All strbuf functions ensure this
41 * invariant is preserved.
43 * NOTE: It is OK to "play" with the buffer directly if you work it this
44 * way:
46 * strbuf_grow(sb, SOME_SIZE); <1>
47 * strbuf_setlen(sb, sb->len + SOME_OTHER_SIZE);
49 * <1> Here, the memory array starting at `sb->buf`, and of length
50 * `strbuf_avail(sb)` is all yours, and you can be sure that
51 * `strbuf_avail(sb)` is at least `SOME_SIZE`.
53 * NOTE: `SOME_OTHER_SIZE` must be smaller or equal to `strbuf_avail(sb)`.
55 * Doing so is safe, though if it has to be done in many places, adding the
56 * missing API to the strbuf module is the way to go.
58 * WARNING: Do _not_ assume that the area that is yours is of size `alloc
59 * - 1` even if it's true in the current implementation. Alloc is somehow a
60 * "private" member that should not be messed with. Use `strbuf_avail()`
61 * instead.
64 /**
65 * Data Structures
66 * ---------------
69 /**
70 * This is the string buffer structure. The `len` member can be used to
71 * determine the current length of the string, and `buf` member provides
72 * access to the string itself.
74 struct strbuf {
75 size_t alloc;
76 size_t len;
77 char *buf;
80 extern char strbuf_slopbuf[];
81 #define STRBUF_INIT { .buf = strbuf_slopbuf }
83 struct object_id;
85 /**
86 * Life Cycle Functions
87 * --------------------
90 /**
91 * Initialize the structure. The second parameter can be zero or a bigger
92 * number to allocate memory, in case you want to prevent further reallocs.
94 void strbuf_init(struct strbuf *sb, size_t alloc);
96 /**
97 * Release a string buffer and the memory it used. After this call, the
98 * strbuf points to an empty string that does not need to be free()ed, as
99 * if it had been set to `STRBUF_INIT` and never modified.
101 * To clear a strbuf in preparation for further use without the overhead
102 * of free()ing and malloc()ing again, use strbuf_reset() instead.
104 void strbuf_release(struct strbuf *sb);
107 * Detach the string from the strbuf and returns it; you now own the
108 * storage the string occupies and it is your responsibility from then on
109 * to release it with `free(3)` when you are done with it.
111 * The strbuf that previously held the string is reset to `STRBUF_INIT` so
112 * it can be reused after calling this function.
114 char *strbuf_detach(struct strbuf *sb, size_t *sz);
117 * Attach a string to a buffer. You should specify the string to attach,
118 * the current length of the string and the amount of allocated memory.
119 * The amount must be larger than the string length, because the string you
120 * pass is supposed to be a NUL-terminated string. This string _must_ be
121 * malloc()ed, and after attaching, the pointer cannot be relied upon
122 * anymore, and neither be free()d directly.
124 void strbuf_attach(struct strbuf *sb, void *str, size_t len, size_t mem);
127 * Swap the contents of two string buffers.
129 static inline void strbuf_swap(struct strbuf *a, struct strbuf *b)
131 SWAP(*a, *b);
136 * Functions related to the size of the buffer
137 * -------------------------------------------
141 * Determine the amount of allocated but unused memory.
143 static inline size_t strbuf_avail(const struct strbuf *sb)
145 return sb->alloc ? sb->alloc - sb->len - 1 : 0;
149 * Ensure that at least this amount of unused memory is available after
150 * `len`. This is used when you know a typical size for what you will add
151 * and want to avoid repetitive automatic resizing of the underlying buffer.
152 * This is never a needed operation, but can be critical for performance in
153 * some cases.
155 void strbuf_grow(struct strbuf *sb, size_t amount);
158 * Set the length of the buffer to a given value. This function does *not*
159 * allocate new memory, so you should not perform a `strbuf_setlen()` to a
160 * length that is larger than `len + strbuf_avail()`. `strbuf_setlen()` is
161 * just meant as a 'please fix invariants from this strbuf I just messed
162 * with'.
164 static inline void strbuf_setlen(struct strbuf *sb, size_t len)
166 if (len > (sb->alloc ? sb->alloc - 1 : 0))
167 BUG("strbuf_setlen() beyond buffer");
168 sb->len = len;
169 if (sb->buf != strbuf_slopbuf)
170 sb->buf[len] = '\0';
171 else
172 assert(!strbuf_slopbuf[0]);
176 * Empty the buffer by setting the size of it to zero.
178 #define strbuf_reset(sb) strbuf_setlen(sb, 0)
182 * Functions related to the contents of the buffer
183 * -----------------------------------------------
187 * Strip whitespace from the beginning (`ltrim`), end (`rtrim`), or both side
188 * (`trim`) of a string.
190 void strbuf_trim(struct strbuf *sb);
191 void strbuf_rtrim(struct strbuf *sb);
192 void strbuf_ltrim(struct strbuf *sb);
194 /* Strip trailing directory separators */
195 void strbuf_trim_trailing_dir_sep(struct strbuf *sb);
197 /* Strip trailing LF or CR/LF */
198 void strbuf_trim_trailing_newline(struct strbuf *sb);
201 * Replace the contents of the strbuf with a reencoded form. Returns -1
202 * on error, 0 on success.
204 int strbuf_reencode(struct strbuf *sb, const char *from, const char *to);
207 * Lowercase each character in the buffer using `tolower`.
209 void strbuf_tolower(struct strbuf *sb);
212 * Compare two buffers. Returns an integer less than, equal to, or greater
213 * than zero if the first buffer is found, respectively, to be less than,
214 * to match, or be greater than the second buffer.
216 int strbuf_cmp(const struct strbuf *first, const struct strbuf *second);
220 * Adding data to the buffer
221 * -------------------------
223 * NOTE: All of the functions in this section will grow the buffer as
224 * necessary. If they fail for some reason other than memory shortage and the
225 * buffer hadn't been allocated before (i.e. the `struct strbuf` was set to
226 * `STRBUF_INIT`), then they will free() it.
230 * Add a single character to the buffer.
232 static inline void strbuf_addch(struct strbuf *sb, int c)
234 if (!strbuf_avail(sb))
235 strbuf_grow(sb, 1);
236 sb->buf[sb->len++] = c;
237 sb->buf[sb->len] = '\0';
241 * Add a character the specified number of times to the buffer.
243 void strbuf_addchars(struct strbuf *sb, int c, size_t n);
246 * Insert data to the given position of the buffer. The remaining contents
247 * will be shifted, not overwritten.
249 void strbuf_insert(struct strbuf *sb, size_t pos, const void *, size_t);
252 * Insert a NUL-terminated string to the given position of the buffer.
253 * The remaining contents will be shifted, not overwritten. It's an
254 * inline function to allow the compiler to resolve strlen() calls on
255 * constants at compile time.
257 static inline void strbuf_insertstr(struct strbuf *sb, size_t pos,
258 const char *s)
260 strbuf_insert(sb, pos, s, strlen(s));
264 * Insert data to the given position of the buffer giving a printf format
265 * string. The contents will be shifted, not overwritten.
267 void strbuf_vinsertf(struct strbuf *sb, size_t pos, const char *fmt,
268 va_list ap);
270 __attribute__((format (printf, 3, 4)))
271 void strbuf_insertf(struct strbuf *sb, size_t pos, const char *fmt, ...);
274 * Remove given amount of data from a given position of the buffer.
276 void strbuf_remove(struct strbuf *sb, size_t pos, size_t len);
279 * Remove the bytes between `pos..pos+len` and replace it with the given
280 * data.
282 void strbuf_splice(struct strbuf *sb, size_t pos, size_t len,
283 const void *data, size_t data_len);
286 * Add a NUL-terminated string to the buffer. Each line will be prepended
287 * by a comment character and a blank.
289 void strbuf_add_commented_lines(struct strbuf *out,
290 const char *buf, size_t size,
291 char comment_line_char);
295 * Add data of given length to the buffer.
297 void strbuf_add(struct strbuf *sb, const void *data, size_t len);
300 * Add a NUL-terminated string to the buffer.
302 * NOTE: This function will *always* be implemented as an inline or a macro
303 * using strlen, meaning that this is efficient to write things like:
305 * strbuf_addstr(sb, "immediate string");
308 static inline void strbuf_addstr(struct strbuf *sb, const char *s)
310 strbuf_add(sb, s, strlen(s));
314 * Copy the contents of another buffer at the end of the current one.
316 void strbuf_addbuf(struct strbuf *sb, const struct strbuf *sb2);
319 * Join the arguments into a buffer. `delim` is put between every
320 * two arguments.
322 const char *strbuf_join_argv(struct strbuf *buf, int argc,
323 const char **argv, char delim);
326 * Used with `strbuf_expand_step` to expand the literals %n and %x
327 * followed by two hexadecimal digits. Returns the number of recognized
328 * characters.
330 size_t strbuf_expand_literal(struct strbuf *sb, const char *placeholder);
333 * If the string pointed to by `formatp` contains a percent sign ("%"),
334 * advance it to point to the character following the next one and
335 * return 1, otherwise return 0. Append the substring before that
336 * percent sign to `sb`, or the whole string if there is none.
338 int strbuf_expand_step(struct strbuf *sb, const char **formatp);
341 * Append the contents of one strbuf to another, quoting any
342 * percent signs ("%") into double-percents ("%%") in the
343 * destination. This is useful for literal data to be fed to either
344 * strbuf_expand or to the *printf family of functions.
346 void strbuf_addbuf_percentquote(struct strbuf *dst, const struct strbuf *src);
348 #define STRBUF_ENCODE_SLASH 1
351 * Append the contents of a string to a strbuf, percent-encoding any characters
352 * that are needed to be encoded for a URL.
354 * If STRBUF_ENCODE_SLASH is set in flags, percent-encode slashes. Otherwise,
355 * slashes are not percent-encoded.
357 void strbuf_add_percentencode(struct strbuf *dst, const char *src, int flags);
360 * Append the given byte size as a human-readable string (i.e. 12.23 KiB,
361 * 3.50 MiB).
363 void strbuf_humanise_bytes(struct strbuf *buf, off_t bytes);
366 * Append the given byte rate as a human-readable string (i.e. 12.23 KiB/s,
367 * 3.50 MiB/s).
369 void strbuf_humanise_rate(struct strbuf *buf, off_t bytes);
372 * Add a formatted string to the buffer.
374 __attribute__((format (printf,2,3)))
375 void strbuf_addf(struct strbuf *sb, const char *fmt, ...);
378 * Add a formatted string prepended by a comment character and a
379 * blank to the buffer.
381 __attribute__((format (printf, 3, 4)))
382 void strbuf_commented_addf(struct strbuf *sb, char comment_line_char, const char *fmt, ...);
384 __attribute__((format (printf,2,0)))
385 void strbuf_vaddf(struct strbuf *sb, const char *fmt, va_list ap);
388 * Add the time specified by `tm`, as formatted by `strftime`.
389 * `tz_offset` is in decimal hhmm format, e.g. -600 means six hours west
390 * of Greenwich, and it's used to expand %z internally. However, tokens
391 * with modifiers (e.g. %Ez) are passed to `strftime`.
392 * `suppress_tz_name`, when set, expands %Z internally to the empty
393 * string rather than passing it to `strftime`.
395 void strbuf_addftime(struct strbuf *sb, const char *fmt,
396 const struct tm *tm, int tz_offset,
397 int suppress_tz_name);
400 * Read a given size of data from a FILE* pointer to the buffer.
402 * NOTE: The buffer is rewound if the read fails. If -1 is returned,
403 * `errno` must be consulted, like you would do for `read(3)`.
404 * `strbuf_read()`, `strbuf_read_file()` and `strbuf_getline_*()`
405 * family of functions have the same behaviour as well.
407 size_t strbuf_fread(struct strbuf *sb, size_t size, FILE *file);
410 * Read the contents of a given file descriptor. The third argument can be
411 * used to give a hint about the file size, to avoid reallocs. If read fails,
412 * any partial read is undone.
414 ssize_t strbuf_read(struct strbuf *sb, int fd, size_t hint);
417 * Read the contents of a given file descriptor partially by using only one
418 * attempt of xread. The third argument can be used to give a hint about the
419 * file size, to avoid reallocs. Returns the number of new bytes appended to
420 * the sb.
422 ssize_t strbuf_read_once(struct strbuf *sb, int fd, size_t hint);
425 * Read the contents of a file, specified by its path. The third argument
426 * can be used to give a hint about the file size, to avoid reallocs.
427 * Return the number of bytes read or a negative value if some error
428 * occurred while opening or reading the file.
430 ssize_t strbuf_read_file(struct strbuf *sb, const char *path, size_t hint);
433 * Read the target of a symbolic link, specified by its path. The third
434 * argument can be used to give a hint about the size, to avoid reallocs.
436 int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint);
439 * Write the whole content of the strbuf to the stream not stopping at
440 * NUL bytes.
442 ssize_t strbuf_write(struct strbuf *sb, FILE *stream);
445 * Read from a FILE * until the specified terminator is encountered,
446 * overwriting the existing contents of the strbuf.
448 * Reading stops after the terminator or at EOF. The terminator is
449 * removed from the buffer before returning. If the terminator is LF
450 * and if it is preceded by a CR, then the whole CRLF is stripped.
451 * Returns 0 unless there was nothing left before EOF, in which case
452 * it returns `EOF`.
454 int strbuf_getdelim_strip_crlf(struct strbuf *sb, FILE *fp, int term);
457 * Read a line from a FILE *, overwriting the existing contents of
458 * the strbuf. The strbuf_getline*() family of functions share
459 * this signature, but have different line termination conventions.
461 * Reading stops after the terminator or at EOF. The terminator
462 * is removed from the buffer before returning. Returns 0 unless
463 * there was nothing left before EOF, in which case it returns `EOF`.
465 typedef int (*strbuf_getline_fn)(struct strbuf *, FILE *);
467 /* Uses LF as the line terminator */
468 int strbuf_getline_lf(struct strbuf *sb, FILE *fp);
470 /* Uses NUL as the line terminator */
471 int strbuf_getline_nul(struct strbuf *sb, FILE *fp);
474 * Similar to strbuf_getline_lf(), but additionally treats a CR that
475 * comes immediately before the LF as part of the terminator.
476 * This is the most friendly version to be used to read "text" files
477 * that can come from platforms whose native text format is CRLF
478 * terminated.
480 int strbuf_getline(struct strbuf *sb, FILE *file);
484 * Like `strbuf_getline`, but keeps the trailing terminator (if
485 * any) in the buffer.
487 int strbuf_getwholeline(struct strbuf *sb, FILE *file, int term);
490 * Like `strbuf_getwholeline`, but appends the line instead of
491 * resetting the buffer first.
493 int strbuf_appendwholeline(struct strbuf *sb, FILE *file, int term);
496 * Like `strbuf_getwholeline`, but operates on a file descriptor.
497 * It reads one character at a time, so it is very slow. Do not
498 * use it unless you need the correct position in the file
499 * descriptor.
501 int strbuf_getwholeline_fd(struct strbuf *sb, int fd, int term);
504 * Set the buffer to the path of the current working directory.
506 int strbuf_getcwd(struct strbuf *sb);
509 * Normalize in-place the path contained in the strbuf. See
510 * normalize_path_copy() for details. If an error occurs, the contents of "sb"
511 * are left untouched, and -1 is returned.
513 int strbuf_normalize_path(struct strbuf *sb);
516 * Strip whitespace from a buffer. If comment_line_char is non-NUL,
517 * then lines beginning with that character are considered comments,
518 * thus removed.
520 void strbuf_stripspace(struct strbuf *buf, char comment_line_char);
522 static inline int strbuf_strip_suffix(struct strbuf *sb, const char *suffix)
524 if (strip_suffix_mem(sb->buf, &sb->len, suffix)) {
525 strbuf_setlen(sb, sb->len);
526 return 1;
527 } else
528 return 0;
532 * Split str (of length slen) at the specified terminator character.
533 * Return a null-terminated array of pointers to strbuf objects
534 * holding the substrings. The substrings include the terminator,
535 * except for the last substring, which might be unterminated if the
536 * original string did not end with a terminator. If max is positive,
537 * then split the string into at most max substrings (with the last
538 * substring containing everything following the (max-1)th terminator
539 * character).
541 * The most generic form is `strbuf_split_buf`, which takes an arbitrary
542 * pointer/len buffer. The `_str` variant takes a NUL-terminated string,
543 * the `_max` variant takes a strbuf, and just `strbuf_split` is a convenience
544 * wrapper to drop the `max` parameter.
546 * For lighter-weight alternatives, see string_list_split() and
547 * string_list_split_in_place().
549 struct strbuf **strbuf_split_buf(const char *str, size_t len,
550 int terminator, int max);
552 static inline struct strbuf **strbuf_split_str(const char *str,
553 int terminator, int max)
555 return strbuf_split_buf(str, strlen(str), terminator, max);
558 static inline struct strbuf **strbuf_split_max(const struct strbuf *sb,
559 int terminator, int max)
561 return strbuf_split_buf(sb->buf, sb->len, terminator, max);
564 static inline struct strbuf **strbuf_split(const struct strbuf *sb,
565 int terminator)
567 return strbuf_split_max(sb, terminator, 0);
571 * Adds all strings of a string list to the strbuf, separated by the given
572 * separator. For example, if sep is
573 * ', '
574 * and slist contains
575 * ['element1', 'element2', ..., 'elementN'],
576 * then write:
577 * 'element1, element2, ..., elementN'
578 * to str. If only one element, just write "element1" to str.
580 void strbuf_add_separated_string_list(struct strbuf *str,
581 const char *sep,
582 struct string_list *slist);
585 * Free a NULL-terminated list of strbufs (for example, the return
586 * values of the strbuf_split*() functions).
588 void strbuf_list_free(struct strbuf **list);
591 * Remove the filename from the provided path string. If the path
592 * contains a trailing separator, then the path is considered a directory
593 * and nothing is modified.
595 * Examples:
596 * - "/path/to/file" -> "/path/to/"
597 * - "/path/to/dir/" -> "/path/to/dir/"
599 void strbuf_strip_file_from_path(struct strbuf *sb);
601 void strbuf_add_lines(struct strbuf *sb,
602 const char *prefix,
603 const char *buf,
604 size_t size);
607 * Append s to sb, with the characters '<', '>', '&' and '"' converted
608 * into XML entities.
610 void strbuf_addstr_xml_quoted(struct strbuf *sb,
611 const char *s);
614 * "Complete" the contents of `sb` by ensuring that either it ends with the
615 * character `term`, or it is empty. This can be used, for example,
616 * to ensure that text ends with a newline, but without creating an empty
617 * blank line if there is no content in the first place.
619 static inline void strbuf_complete(struct strbuf *sb, char term)
621 if (sb->len && sb->buf[sb->len - 1] != term)
622 strbuf_addch(sb, term);
625 static inline void strbuf_complete_line(struct strbuf *sb)
627 strbuf_complete(sb, '\n');
631 * Copy "name" to "sb", expanding any special @-marks as handled by
632 * repo_interpret_branch_name(). The result is a non-qualified branch name
633 * (so "foo" or "origin/master" instead of "refs/heads/foo" or
634 * "refs/remotes/origin/master").
636 * Note that the resulting name may not be a syntactically valid refname.
638 * If "allowed" is non-zero, restrict the set of allowed expansions. See
639 * repo_interpret_branch_name() for details.
641 void strbuf_branchname(struct strbuf *sb, const char *name,
642 unsigned allowed);
645 * Like strbuf_branchname() above, but confirm that the result is
646 * syntactically valid to be used as a local branch name in refs/heads/.
648 * The return value is "0" if the result is valid, and "-1" otherwise.
650 int strbuf_check_branch_ref(struct strbuf *sb, const char *name);
652 typedef int (*char_predicate)(char ch);
654 void strbuf_addstr_urlencode(struct strbuf *sb, const char *name,
655 char_predicate allow_unencoded_fn);
657 __attribute__((format (printf,1,2)))
658 int printf_ln(const char *fmt, ...);
659 __attribute__((format (printf,2,3)))
660 int fprintf_ln(FILE *fp, const char *fmt, ...);
662 char *xstrdup_tolower(const char *);
663 char *xstrdup_toupper(const char *);
666 * Create a newly allocated string using printf format. You can do this easily
667 * with a strbuf, but this provides a shortcut to save a few lines.
669 __attribute__((format (printf, 1, 0)))
670 char *xstrvfmt(const char *fmt, va_list ap);
671 __attribute__((format (printf, 1, 2)))
672 char *xstrfmt(const char *fmt, ...);
674 int starts_with(const char *str, const char *prefix);
675 int istarts_with(const char *str, const char *prefix);
678 * If the string "str" is the same as the string in "prefix", then the "arg"
679 * parameter is set to the "def" parameter and 1 is returned.
680 * If the string "str" begins with the string found in "prefix" and then a
681 * "=" sign, then the "arg" parameter is set to "str + strlen(prefix) + 1"
682 * (i.e., to the point in the string right after the prefix and the "=" sign),
683 * and 1 is returned.
685 * Otherwise, return 0 and leave "arg" untouched.
687 * When we accept both a "--key" and a "--key=<val>" option, this function
688 * can be used instead of !strcmp(arg, "--key") and then
689 * skip_prefix(arg, "--key=", &arg) to parse such an option.
691 int skip_to_optional_arg_default(const char *str, const char *prefix,
692 const char **arg, const char *def);
694 static inline int skip_to_optional_arg(const char *str, const char *prefix,
695 const char **arg)
697 return skip_to_optional_arg_default(str, prefix, arg, "");
700 static inline int ends_with(const char *str, const char *suffix)
702 size_t len;
703 return strip_suffix(str, suffix, &len);
706 #endif /* STRBUF_H */