strbuf.hon commit Merge branch 'nd/worktree-kill-parse-ref' into maint (5253ad1)
   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 */
  64struct strbuf {
  65        size_t alloc;
  66        size_t len;
  67        char *buf;
  68};
  69
  70extern 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 */
  82extern void strbuf_init(struct strbuf *, size_t);
  83
  84/**
  85 * Release a string buffer and the memory it used. After this call, the
  86 * strbuf points to an empty string that does not need to be free()ed, as
  87 * if it had been set to `STRBUF_INIT` and never modified.
  88 *
  89 * To clear a strbuf in preparation for further use without the overhead
  90 * of free()ing and malloc()ing again, use strbuf_reset() instead.
  91 */
  92extern void strbuf_release(struct strbuf *);
  93
  94/**
  95 * Detach the string from the strbuf and returns it; you now own the
  96 * storage the string occupies and it is your responsibility from then on
  97 * to release it with `free(3)` when you are done with it.
  98 *
  99 * The strbuf that previously held the string is reset to `STRBUF_INIT` so
 100 * it can be reused after calling this function.
 101 */
 102extern char *strbuf_detach(struct strbuf *, size_t *);
 103
 104/**
 105 * Attach a string to a buffer. You should specify the string to attach,
 106 * the current length of the string and the amount of allocated memory.
 107 * The amount must be larger than the string length, because the string you
 108 * pass is supposed to be a NUL-terminated string.  This string _must_ be
 109 * malloc()ed, and after attaching, the pointer cannot be relied upon
 110 * anymore, and neither be free()d directly.
 111 */
 112extern void strbuf_attach(struct strbuf *, void *, size_t, size_t);
 113
 114/**
 115 * Swap the contents of two string buffers.
 116 */
 117static inline void strbuf_swap(struct strbuf *a, struct strbuf *b)
 118{
 119        SWAP(*a, *b);
 120}
 121
 122
 123/**
 124 * Functions related to the size of the buffer
 125 * -------------------------------------------
 126 */
 127
 128/**
 129 * Determine the amount of allocated but unused memory.
 130 */
 131static inline size_t strbuf_avail(const struct strbuf *sb)
 132{
 133        return sb->alloc ? sb->alloc - sb->len - 1 : 0;
 134}
 135
 136/**
 137 * Ensure that at least this amount of unused memory is available after
 138 * `len`. This is used when you know a typical size for what you will add
 139 * and want to avoid repetitive automatic resizing of the underlying buffer.
 140 * This is never a needed operation, but can be critical for performance in
 141 * some cases.
 142 */
 143extern void strbuf_grow(struct strbuf *, size_t);
 144
 145/**
 146 * Set the length of the buffer to a given value. This function does *not*
 147 * allocate new memory, so you should not perform a `strbuf_setlen()` to a
 148 * length that is larger than `len + strbuf_avail()`. `strbuf_setlen()` is
 149 * just meant as a 'please fix invariants from this strbuf I just messed
 150 * with'.
 151 */
 152static inline void strbuf_setlen(struct strbuf *sb, size_t len)
 153{
 154        if (len > (sb->alloc ? sb->alloc - 1 : 0))
 155                die("BUG: strbuf_setlen() beyond buffer");
 156        sb->len = len;
 157        sb->buf[len] = '\0';
 158}
 159
 160/**
 161 * Empty the buffer by setting the size of it to zero.
 162 */
 163#define strbuf_reset(sb)  strbuf_setlen(sb, 0)
 164
 165
 166/**
 167 * Functions related to the contents of the buffer
 168 * -----------------------------------------------
 169 */
 170
 171/**
 172 * Strip whitespace from the beginning (`ltrim`), end (`rtrim`), or both side
 173 * (`trim`) of a string.
 174 */
 175extern void strbuf_trim(struct strbuf *);
 176extern void strbuf_rtrim(struct strbuf *);
 177extern void strbuf_ltrim(struct strbuf *);
 178
 179/**
 180 * Replace the contents of the strbuf with a reencoded form.  Returns -1
 181 * on error, 0 on success.
 182 */
 183extern int strbuf_reencode(struct strbuf *sb, const char *from, const char *to);
 184
 185/**
 186 * Lowercase each character in the buffer using `tolower`.
 187 */
 188extern void strbuf_tolower(struct strbuf *sb);
 189
 190/**
 191 * Compare two buffers. Returns an integer less than, equal to, or greater
 192 * than zero if the first buffer is found, respectively, to be less than,
 193 * to match, or be greater than the second buffer.
 194 */
 195extern int strbuf_cmp(const struct strbuf *, const struct strbuf *);
 196
 197
 198/**
 199 * Adding data to the buffer
 200 * -------------------------
 201 *
 202 * NOTE: All of the functions in this section will grow the buffer as
 203 * necessary.  If they fail for some reason other than memory shortage and the
 204 * buffer hadn't been allocated before (i.e. the `struct strbuf` was set to
 205 * `STRBUF_INIT`), then they will free() it.
 206 */
 207
 208/**
 209 * Add a single character to the buffer.
 210 */
 211static inline void strbuf_addch(struct strbuf *sb, int c)
 212{
 213        if (!strbuf_avail(sb))
 214                strbuf_grow(sb, 1);
 215        sb->buf[sb->len++] = c;
 216        sb->buf[sb->len] = '\0';
 217}
 218
 219/**
 220 * Add a character the specified number of times to the buffer.
 221 */
 222extern void strbuf_addchars(struct strbuf *sb, int c, size_t n);
 223
 224/**
 225 * Insert data to the given position of the buffer. The remaining contents
 226 * will be shifted, not overwritten.
 227 */
 228extern void strbuf_insert(struct strbuf *, size_t pos, const void *, size_t);
 229
 230/**
 231 * Remove given amount of data from a given position of the buffer.
 232 */
 233extern void strbuf_remove(struct strbuf *, size_t pos, size_t len);
 234
 235/**
 236 * Remove the bytes between `pos..pos+len` and replace it with the given
 237 * data.
 238 */
 239extern void strbuf_splice(struct strbuf *, size_t pos, size_t len,
 240                          const void *, size_t);
 241
 242/**
 243 * Add a NUL-terminated string to the buffer. Each line will be prepended
 244 * by a comment character and a blank.
 245 */
 246extern void strbuf_add_commented_lines(struct strbuf *out, const char *buf, size_t size);
 247
 248
 249/**
 250 * Add data of given length to the buffer.
 251 */
 252extern void strbuf_add(struct strbuf *, const void *, size_t);
 253
 254/**
 255 * Add a NUL-terminated string to the buffer.
 256 *
 257 * NOTE: This function will *always* be implemented as an inline or a macro
 258 * using strlen, meaning that this is efficient to write things like:
 259 *
 260 *     strbuf_addstr(sb, "immediate string");
 261 *
 262 */
 263static inline void strbuf_addstr(struct strbuf *sb, const char *s)
 264{
 265        strbuf_add(sb, s, strlen(s));
 266}
 267
 268/**
 269 * Copy the contents of another buffer at the end of the current one.
 270 */
 271extern void strbuf_addbuf(struct strbuf *sb, const struct strbuf *sb2);
 272
 273/**
 274 * This function can be used to expand a format string containing
 275 * placeholders. To that end, it parses the string and calls the specified
 276 * function for every percent sign found.
 277 *
 278 * The callback function is given a pointer to the character after the `%`
 279 * and a pointer to the struct strbuf.  It is expected to add the expanded
 280 * version of the placeholder to the strbuf, e.g. to add a newline
 281 * character if the letter `n` appears after a `%`.  The function returns
 282 * the length of the placeholder recognized and `strbuf_expand()` skips
 283 * over it.
 284 *
 285 * The format `%%` is automatically expanded to a single `%` as a quoting
 286 * mechanism; callers do not need to handle the `%` placeholder themselves,
 287 * and the callback function will not be invoked for this placeholder.
 288 *
 289 * All other characters (non-percent and not skipped ones) are copied
 290 * verbatim to the strbuf.  If the callback returned zero, meaning that the
 291 * placeholder is unknown, then the percent sign is copied, too.
 292 *
 293 * In order to facilitate caching and to make it possible to give
 294 * parameters to the callback, `strbuf_expand()` passes a context pointer,
 295 * which can be used by the programmer of the callback as she sees fit.
 296 */
 297typedef size_t (*expand_fn_t) (struct strbuf *sb, const char *placeholder, void *context);
 298extern void strbuf_expand(struct strbuf *sb, const char *format, expand_fn_t fn, void *context);
 299
 300/**
 301 * Used as callback for `strbuf_expand()`, expects an array of
 302 * struct strbuf_expand_dict_entry as context, i.e. pairs of
 303 * placeholder and replacement string.  The array needs to be
 304 * terminated by an entry with placeholder set to NULL.
 305 */
 306struct strbuf_expand_dict_entry {
 307        const char *placeholder;
 308        const char *value;
 309};
 310extern size_t strbuf_expand_dict_cb(struct strbuf *sb, const char *placeholder, void *context);
 311
 312/**
 313 * Append the contents of one strbuf to another, quoting any
 314 * percent signs ("%") into double-percents ("%%") in the
 315 * destination. This is useful for literal data to be fed to either
 316 * strbuf_expand or to the *printf family of functions.
 317 */
 318extern void strbuf_addbuf_percentquote(struct strbuf *dst, const struct strbuf *src);
 319
 320/**
 321 * Append the given byte size as a human-readable string (i.e. 12.23 KiB,
 322 * 3.50 MiB).
 323 */
 324extern void strbuf_humanise_bytes(struct strbuf *buf, off_t bytes);
 325
 326/**
 327 * Add a formatted string to the buffer.
 328 */
 329__attribute__((format (printf,2,3)))
 330extern void strbuf_addf(struct strbuf *sb, const char *fmt, ...);
 331
 332/**
 333 * Add a formatted string prepended by a comment character and a
 334 * blank to the buffer.
 335 */
 336__attribute__((format (printf, 2, 3)))
 337extern void strbuf_commented_addf(struct strbuf *sb, const char *fmt, ...);
 338
 339__attribute__((format (printf,2,0)))
 340extern void strbuf_vaddf(struct strbuf *sb, const char *fmt, va_list ap);
 341
 342/**
 343 * Add the time specified by `tm`, as formatted by `strftime`.
 344 * `tz_offset` is in decimal hhmm format, e.g. -600 means six hours west
 345 * of Greenwich, and it's used to expand %z internally.  However, tokens
 346 * with modifiers (e.g. %Ez) are passed to `strftime`.
 347 * `suppress_tz_name`, when set, expands %Z internally to the empty
 348 * string rather than passing it to `strftime`.
 349 */
 350extern void strbuf_addftime(struct strbuf *sb, const char *fmt,
 351                            const struct tm *tm, int tz_offset,
 352                            int suppress_tz_name);
 353
 354/**
 355 * Read a given size of data from a FILE* pointer to the buffer.
 356 *
 357 * NOTE: The buffer is rewound if the read fails. If -1 is returned,
 358 * `errno` must be consulted, like you would do for `read(3)`.
 359 * `strbuf_read()`, `strbuf_read_file()` and `strbuf_getline_*()`
 360 * family of functions have the same behaviour as well.
 361 */
 362extern size_t strbuf_fread(struct strbuf *, size_t, FILE *);
 363
 364/**
 365 * Read the contents of a given file descriptor. The third argument can be
 366 * used to give a hint about the file size, to avoid reallocs.  If read fails,
 367 * any partial read is undone.
 368 */
 369extern ssize_t strbuf_read(struct strbuf *, int fd, size_t hint);
 370
 371/**
 372 * Read the contents of a given file descriptor partially by using only one
 373 * attempt of xread. The third argument can be used to give a hint about the
 374 * file size, to avoid reallocs. Returns the number of new bytes appended to
 375 * the sb.
 376 */
 377extern ssize_t strbuf_read_once(struct strbuf *, int fd, size_t hint);
 378
 379/**
 380 * Read the contents of a file, specified by its path. The third argument
 381 * can be used to give a hint about the file size, to avoid reallocs.
 382 * Return the number of bytes read or a negative value if some error
 383 * occurred while opening or reading the file.
 384 */
 385extern ssize_t strbuf_read_file(struct strbuf *sb, const char *path, size_t hint);
 386
 387/**
 388 * Read the target of a symbolic link, specified by its path.  The third
 389 * argument can be used to give a hint about the size, to avoid reallocs.
 390 */
 391extern int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint);
 392
 393/**
 394 * Write the whole content of the strbuf to the stream not stopping at
 395 * NUL bytes.
 396 */
 397extern ssize_t strbuf_write(struct strbuf *sb, FILE *stream);
 398
 399/**
 400 * Read a line from a FILE *, overwriting the existing contents of
 401 * the strbuf.  The strbuf_getline*() family of functions share
 402 * this signature, but have different line termination conventions.
 403 *
 404 * Reading stops after the terminator or at EOF.  The terminator
 405 * is removed from the buffer before returning.  Returns 0 unless
 406 * there was nothing left before EOF, in which case it returns `EOF`.
 407 */
 408typedef int (*strbuf_getline_fn)(struct strbuf *, FILE *);
 409
 410/* Uses LF as the line terminator */
 411extern int strbuf_getline_lf(struct strbuf *sb, FILE *fp);
 412
 413/* Uses NUL as the line terminator */
 414extern int strbuf_getline_nul(struct strbuf *sb, FILE *fp);
 415
 416/*
 417 * Similar to strbuf_getline_lf(), but additionally treats a CR that
 418 * comes immediately before the LF as part of the terminator.
 419 * This is the most friendly version to be used to read "text" files
 420 * that can come from platforms whose native text format is CRLF
 421 * terminated.
 422 */
 423extern int strbuf_getline(struct strbuf *, FILE *);
 424
 425
 426/**
 427 * Like `strbuf_getline`, but keeps the trailing terminator (if
 428 * any) in the buffer.
 429 */
 430extern int strbuf_getwholeline(struct strbuf *, FILE *, int);
 431
 432/**
 433 * Like `strbuf_getwholeline`, but operates on a file descriptor.
 434 * It reads one character at a time, so it is very slow.  Do not
 435 * use it unless you need the correct position in the file
 436 * descriptor.
 437 */
 438extern int strbuf_getwholeline_fd(struct strbuf *, int, int);
 439
 440/**
 441 * Set the buffer to the path of the current working directory.
 442 */
 443extern int strbuf_getcwd(struct strbuf *sb);
 444
 445/**
 446 * Add a path to a buffer, converting a relative path to an
 447 * absolute one in the process.  Symbolic links are not
 448 * resolved.
 449 */
 450extern void strbuf_add_absolute_path(struct strbuf *sb, const char *path);
 451
 452/**
 453 * Canonize `path` (make it absolute, resolve symlinks, remove extra
 454 * slashes) and append it to `sb`.  Die with an informative error
 455 * message if there is a problem.
 456 *
 457 * The directory part of `path` (i.e., everything up to the last
 458 * dir_sep) must denote a valid, existing directory, but the last
 459 * component need not exist.
 460 *
 461 * Callers that don't mind links should use the more lightweight
 462 * strbuf_add_absolute_path() instead.
 463 */
 464extern void strbuf_add_real_path(struct strbuf *sb, const char *path);
 465
 466
 467/**
 468 * Normalize in-place the path contained in the strbuf. See
 469 * normalize_path_copy() for details. If an error occurs, the contents of "sb"
 470 * are left untouched, and -1 is returned.
 471 */
 472extern int strbuf_normalize_path(struct strbuf *sb);
 473
 474/**
 475 * Strip whitespace from a buffer. The second parameter controls if
 476 * comments are considered contents to be removed or not.
 477 */
 478extern void strbuf_stripspace(struct strbuf *buf, int skip_comments);
 479
 480/**
 481 * Temporary alias until all topic branches have switched to use
 482 * strbuf_stripspace directly.
 483 */
 484static inline void stripspace(struct strbuf *buf, int skip_comments)
 485{
 486        strbuf_stripspace(buf, skip_comments);
 487}
 488
 489static inline int strbuf_strip_suffix(struct strbuf *sb, const char *suffix)
 490{
 491        if (strip_suffix_mem(sb->buf, &sb->len, suffix)) {
 492                strbuf_setlen(sb, sb->len);
 493                return 1;
 494        } else
 495                return 0;
 496}
 497
 498/**
 499 * Split str (of length slen) at the specified terminator character.
 500 * Return a null-terminated array of pointers to strbuf objects
 501 * holding the substrings.  The substrings include the terminator,
 502 * except for the last substring, which might be unterminated if the
 503 * original string did not end with a terminator.  If max is positive,
 504 * then split the string into at most max substrings (with the last
 505 * substring containing everything following the (max-1)th terminator
 506 * character).
 507 *
 508 * The most generic form is `strbuf_split_buf`, which takes an arbitrary
 509 * pointer/len buffer. The `_str` variant takes a NUL-terminated string,
 510 * the `_max` variant takes a strbuf, and just `strbuf_split` is a convenience
 511 * wrapper to drop the `max` parameter.
 512 *
 513 * For lighter-weight alternatives, see string_list_split() and
 514 * string_list_split_in_place().
 515 */
 516extern struct strbuf **strbuf_split_buf(const char *, size_t,
 517                                        int terminator, int max);
 518
 519static inline struct strbuf **strbuf_split_str(const char *str,
 520                                               int terminator, int max)
 521{
 522        return strbuf_split_buf(str, strlen(str), terminator, max);
 523}
 524
 525static inline struct strbuf **strbuf_split_max(const struct strbuf *sb,
 526                                                int terminator, int max)
 527{
 528        return strbuf_split_buf(sb->buf, sb->len, terminator, max);
 529}
 530
 531static inline struct strbuf **strbuf_split(const struct strbuf *sb,
 532                                           int terminator)
 533{
 534        return strbuf_split_max(sb, terminator, 0);
 535}
 536
 537/**
 538 * Free a NULL-terminated list of strbufs (for example, the return
 539 * values of the strbuf_split*() functions).
 540 */
 541extern void strbuf_list_free(struct strbuf **);
 542
 543/**
 544 * Add the abbreviation, as generated by find_unique_abbrev, of `sha1` to
 545 * the strbuf `sb`.
 546 */
 547extern void strbuf_add_unique_abbrev(struct strbuf *sb,
 548                                     const unsigned char *sha1,
 549                                     int abbrev_len);
 550
 551/**
 552 * Launch the user preferred editor to edit a file and fill the buffer
 553 * with the file's contents upon the user completing their editing. The
 554 * third argument can be used to set the environment which the editor is
 555 * run in. If the buffer is NULL the editor is launched as usual but the
 556 * file's contents are not read into the buffer upon completion.
 557 */
 558extern int launch_editor(const char *path, struct strbuf *buffer, const char *const *env);
 559
 560extern void strbuf_add_lines(struct strbuf *sb, const char *prefix, const char *buf, size_t size);
 561
 562/**
 563 * Append s to sb, with the characters '<', '>', '&' and '"' converted
 564 * into XML entities.
 565 */
 566extern void strbuf_addstr_xml_quoted(struct strbuf *sb, const char *s);
 567
 568/**
 569 * "Complete" the contents of `sb` by ensuring that either it ends with the
 570 * character `term`, or it is empty.  This can be used, for example,
 571 * to ensure that text ends with a newline, but without creating an empty
 572 * blank line if there is no content in the first place.
 573 */
 574static inline void strbuf_complete(struct strbuf *sb, char term)
 575{
 576        if (sb->len && sb->buf[sb->len - 1] != term)
 577                strbuf_addch(sb, term);
 578}
 579
 580static inline void strbuf_complete_line(struct strbuf *sb)
 581{
 582        strbuf_complete(sb, '\n');
 583}
 584
 585/*
 586 * Copy "name" to "sb", expanding any special @-marks as handled by
 587 * interpret_branch_name(). The result is a non-qualified branch name
 588 * (so "foo" or "origin/master" instead of "refs/heads/foo" or
 589 * "refs/remotes/origin/master").
 590 *
 591 * Note that the resulting name may not be a syntactically valid refname.
 592 *
 593 * If "allowed" is non-zero, restrict the set of allowed expansions. See
 594 * interpret_branch_name() for details.
 595 */
 596extern void strbuf_branchname(struct strbuf *sb, const char *name,
 597                              unsigned allowed);
 598
 599/*
 600 * Like strbuf_branchname() above, but confirm that the result is
 601 * syntactically valid to be used as a local branch name in refs/heads/.
 602 *
 603 * The return value is "0" if the result is valid, and "-1" otherwise.
 604 */
 605extern int strbuf_check_branch_ref(struct strbuf *sb, const char *name);
 606
 607extern void strbuf_addstr_urlencode(struct strbuf *, const char *,
 608                                    int reserved);
 609
 610__attribute__((format (printf,1,2)))
 611extern int printf_ln(const char *fmt, ...);
 612__attribute__((format (printf,2,3)))
 613extern int fprintf_ln(FILE *fp, const char *fmt, ...);
 614
 615char *xstrdup_tolower(const char *);
 616
 617/**
 618 * Create a newly allocated string using printf format. You can do this easily
 619 * with a strbuf, but this provides a shortcut to save a few lines.
 620 */
 621__attribute__((format (printf, 1, 0)))
 622char *xstrvfmt(const char *fmt, va_list ap);
 623__attribute__((format (printf, 1, 2)))
 624char *xstrfmt(const char *fmt, ...);
 625
 626#endif /* STRBUF_H */