strbuf.hon commit Merge branch 'pc/submodule-helper' (ead8dbe)
   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  { .alloc = 0, .len = 0, .buf = 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        if (sb->buf != strbuf_slopbuf)
 158                sb->buf[len] = '\0';
 159        else
 160                assert(!strbuf_slopbuf[0]);
 161}
 162
 163/**
 164 * Empty the buffer by setting the size of it to zero.
 165 */
 166#define strbuf_reset(sb)  strbuf_setlen(sb, 0)
 167
 168
 169/**
 170 * Functions related to the contents of the buffer
 171 * -----------------------------------------------
 172 */
 173
 174/**
 175 * Strip whitespace from the beginning (`ltrim`), end (`rtrim`), or both side
 176 * (`trim`) of a string.
 177 */
 178extern void strbuf_trim(struct strbuf *);
 179extern void strbuf_rtrim(struct strbuf *);
 180extern void strbuf_ltrim(struct strbuf *);
 181
 182/**
 183 * Replace the contents of the strbuf with a reencoded form.  Returns -1
 184 * on error, 0 on success.
 185 */
 186extern int strbuf_reencode(struct strbuf *sb, const char *from, const char *to);
 187
 188/**
 189 * Lowercase each character in the buffer using `tolower`.
 190 */
 191extern void strbuf_tolower(struct strbuf *sb);
 192
 193/**
 194 * Compare two buffers. Returns an integer less than, equal to, or greater
 195 * than zero if the first buffer is found, respectively, to be less than,
 196 * to match, or be greater than the second buffer.
 197 */
 198extern int strbuf_cmp(const struct strbuf *, const struct strbuf *);
 199
 200
 201/**
 202 * Adding data to the buffer
 203 * -------------------------
 204 *
 205 * NOTE: All of the functions in this section will grow the buffer as
 206 * necessary.  If they fail for some reason other than memory shortage and the
 207 * buffer hadn't been allocated before (i.e. the `struct strbuf` was set to
 208 * `STRBUF_INIT`), then they will free() it.
 209 */
 210
 211/**
 212 * Add a single character to the buffer.
 213 */
 214static inline void strbuf_addch(struct strbuf *sb, int c)
 215{
 216        if (!strbuf_avail(sb))
 217                strbuf_grow(sb, 1);
 218        sb->buf[sb->len++] = c;
 219        sb->buf[sb->len] = '\0';
 220}
 221
 222/**
 223 * Add a character the specified number of times to the buffer.
 224 */
 225extern void strbuf_addchars(struct strbuf *sb, int c, size_t n);
 226
 227/**
 228 * Insert data to the given position of the buffer. The remaining contents
 229 * will be shifted, not overwritten.
 230 */
 231extern void strbuf_insert(struct strbuf *, size_t pos, const void *, size_t);
 232
 233/**
 234 * Remove given amount of data from a given position of the buffer.
 235 */
 236extern void strbuf_remove(struct strbuf *, size_t pos, size_t len);
 237
 238/**
 239 * Remove the bytes between `pos..pos+len` and replace it with the given
 240 * data.
 241 */
 242extern void strbuf_splice(struct strbuf *, size_t pos, size_t len,
 243                          const void *, size_t);
 244
 245/**
 246 * Add a NUL-terminated string to the buffer. Each line will be prepended
 247 * by a comment character and a blank.
 248 */
 249extern void strbuf_add_commented_lines(struct strbuf *out, const char *buf, size_t size);
 250
 251
 252/**
 253 * Add data of given length to the buffer.
 254 */
 255extern void strbuf_add(struct strbuf *, const void *, size_t);
 256
 257/**
 258 * Add a NUL-terminated string to the buffer.
 259 *
 260 * NOTE: This function will *always* be implemented as an inline or a macro
 261 * using strlen, meaning that this is efficient to write things like:
 262 *
 263 *     strbuf_addstr(sb, "immediate string");
 264 *
 265 */
 266static inline void strbuf_addstr(struct strbuf *sb, const char *s)
 267{
 268        strbuf_add(sb, s, strlen(s));
 269}
 270
 271/**
 272 * Copy the contents of another buffer at the end of the current one.
 273 */
 274extern void strbuf_addbuf(struct strbuf *sb, const struct strbuf *sb2);
 275
 276/**
 277 * This function can be used to expand a format string containing
 278 * placeholders. To that end, it parses the string and calls the specified
 279 * function for every percent sign found.
 280 *
 281 * The callback function is given a pointer to the character after the `%`
 282 * and a pointer to the struct strbuf.  It is expected to add the expanded
 283 * version of the placeholder to the strbuf, e.g. to add a newline
 284 * character if the letter `n` appears after a `%`.  The function returns
 285 * the length of the placeholder recognized and `strbuf_expand()` skips
 286 * over it.
 287 *
 288 * The format `%%` is automatically expanded to a single `%` as a quoting
 289 * mechanism; callers do not need to handle the `%` placeholder themselves,
 290 * and the callback function will not be invoked for this placeholder.
 291 *
 292 * All other characters (non-percent and not skipped ones) are copied
 293 * verbatim to the strbuf.  If the callback returned zero, meaning that the
 294 * placeholder is unknown, then the percent sign is copied, too.
 295 *
 296 * In order to facilitate caching and to make it possible to give
 297 * parameters to the callback, `strbuf_expand()` passes a context pointer,
 298 * which can be used by the programmer of the callback as she sees fit.
 299 */
 300typedef size_t (*expand_fn_t) (struct strbuf *sb, const char *placeholder, void *context);
 301extern void strbuf_expand(struct strbuf *sb, const char *format, expand_fn_t fn, void *context);
 302
 303/**
 304 * Used as callback for `strbuf_expand()`, expects an array of
 305 * struct strbuf_expand_dict_entry as context, i.e. pairs of
 306 * placeholder and replacement string.  The array needs to be
 307 * terminated by an entry with placeholder set to NULL.
 308 */
 309struct strbuf_expand_dict_entry {
 310        const char *placeholder;
 311        const char *value;
 312};
 313extern size_t strbuf_expand_dict_cb(struct strbuf *sb, const char *placeholder, void *context);
 314
 315/**
 316 * Append the contents of one strbuf to another, quoting any
 317 * percent signs ("%") into double-percents ("%%") in the
 318 * destination. This is useful for literal data to be fed to either
 319 * strbuf_expand or to the *printf family of functions.
 320 */
 321extern void strbuf_addbuf_percentquote(struct strbuf *dst, const struct strbuf *src);
 322
 323/**
 324 * Append the given byte size as a human-readable string (i.e. 12.23 KiB,
 325 * 3.50 MiB).
 326 */
 327extern void strbuf_humanise_bytes(struct strbuf *buf, off_t bytes);
 328
 329/**
 330 * Add a formatted string to the buffer.
 331 */
 332__attribute__((format (printf,2,3)))
 333extern void strbuf_addf(struct strbuf *sb, const char *fmt, ...);
 334
 335/**
 336 * Add a formatted string prepended by a comment character and a
 337 * blank to the buffer.
 338 */
 339__attribute__((format (printf, 2, 3)))
 340extern void strbuf_commented_addf(struct strbuf *sb, const char *fmt, ...);
 341
 342__attribute__((format (printf,2,0)))
 343extern void strbuf_vaddf(struct strbuf *sb, const char *fmt, va_list ap);
 344
 345/**
 346 * Add the time specified by `tm`, as formatted by `strftime`.
 347 * `tz_offset` is in decimal hhmm format, e.g. -600 means six hours west
 348 * of Greenwich, and it's used to expand %z internally.  However, tokens
 349 * with modifiers (e.g. %Ez) are passed to `strftime`.
 350 * `suppress_tz_name`, when set, expands %Z internally to the empty
 351 * string rather than passing it to `strftime`.
 352 */
 353extern void strbuf_addftime(struct strbuf *sb, const char *fmt,
 354                            const struct tm *tm, int tz_offset,
 355                            int suppress_tz_name);
 356
 357/**
 358 * Read a given size of data from a FILE* pointer to the buffer.
 359 *
 360 * NOTE: The buffer is rewound if the read fails. If -1 is returned,
 361 * `errno` must be consulted, like you would do for `read(3)`.
 362 * `strbuf_read()`, `strbuf_read_file()` and `strbuf_getline_*()`
 363 * family of functions have the same behaviour as well.
 364 */
 365extern size_t strbuf_fread(struct strbuf *, size_t, FILE *);
 366
 367/**
 368 * Read the contents of a given file descriptor. The third argument can be
 369 * used to give a hint about the file size, to avoid reallocs.  If read fails,
 370 * any partial read is undone.
 371 */
 372extern ssize_t strbuf_read(struct strbuf *, int fd, size_t hint);
 373
 374/**
 375 * Read the contents of a given file descriptor partially by using only one
 376 * attempt of xread. The third argument can be used to give a hint about the
 377 * file size, to avoid reallocs. Returns the number of new bytes appended to
 378 * the sb.
 379 */
 380extern ssize_t strbuf_read_once(struct strbuf *, int fd, size_t hint);
 381
 382/**
 383 * Read the contents of a file, specified by its path. The third argument
 384 * can be used to give a hint about the file size, to avoid reallocs.
 385 * Return the number of bytes read or a negative value if some error
 386 * occurred while opening or reading the file.
 387 */
 388extern ssize_t strbuf_read_file(struct strbuf *sb, const char *path, size_t hint);
 389
 390/**
 391 * Read the target of a symbolic link, specified by its path.  The third
 392 * argument can be used to give a hint about the size, to avoid reallocs.
 393 */
 394extern int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint);
 395
 396/**
 397 * Write the whole content of the strbuf to the stream not stopping at
 398 * NUL bytes.
 399 */
 400extern ssize_t strbuf_write(struct strbuf *sb, FILE *stream);
 401
 402/**
 403 * Read a line from a FILE *, overwriting the existing contents of
 404 * the strbuf.  The strbuf_getline*() family of functions share
 405 * this signature, but have different line termination conventions.
 406 *
 407 * Reading stops after the terminator or at EOF.  The terminator
 408 * is removed from the buffer before returning.  Returns 0 unless
 409 * there was nothing left before EOF, in which case it returns `EOF`.
 410 */
 411typedef int (*strbuf_getline_fn)(struct strbuf *, FILE *);
 412
 413/* Uses LF as the line terminator */
 414extern int strbuf_getline_lf(struct strbuf *sb, FILE *fp);
 415
 416/* Uses NUL as the line terminator */
 417extern int strbuf_getline_nul(struct strbuf *sb, FILE *fp);
 418
 419/*
 420 * Similar to strbuf_getline_lf(), but additionally treats a CR that
 421 * comes immediately before the LF as part of the terminator.
 422 * This is the most friendly version to be used to read "text" files
 423 * that can come from platforms whose native text format is CRLF
 424 * terminated.
 425 */
 426extern int strbuf_getline(struct strbuf *, FILE *);
 427
 428
 429/**
 430 * Like `strbuf_getline`, but keeps the trailing terminator (if
 431 * any) in the buffer.
 432 */
 433extern int strbuf_getwholeline(struct strbuf *, FILE *, int);
 434
 435/**
 436 * Like `strbuf_getwholeline`, but operates on a file descriptor.
 437 * It reads one character at a time, so it is very slow.  Do not
 438 * use it unless you need the correct position in the file
 439 * descriptor.
 440 */
 441extern int strbuf_getwholeline_fd(struct strbuf *, int, int);
 442
 443/**
 444 * Set the buffer to the path of the current working directory.
 445 */
 446extern int strbuf_getcwd(struct strbuf *sb);
 447
 448/**
 449 * Add a path to a buffer, converting a relative path to an
 450 * absolute one in the process.  Symbolic links are not
 451 * resolved.
 452 */
 453extern void strbuf_add_absolute_path(struct strbuf *sb, const char *path);
 454
 455/**
 456 * Canonize `path` (make it absolute, resolve symlinks, remove extra
 457 * slashes) and append it to `sb`.  Die with an informative error
 458 * message if there is a problem.
 459 *
 460 * The directory part of `path` (i.e., everything up to the last
 461 * dir_sep) must denote a valid, existing directory, but the last
 462 * component need not exist.
 463 *
 464 * Callers that don't mind links should use the more lightweight
 465 * strbuf_add_absolute_path() instead.
 466 */
 467extern void strbuf_add_real_path(struct strbuf *sb, const char *path);
 468
 469
 470/**
 471 * Normalize in-place the path contained in the strbuf. See
 472 * normalize_path_copy() for details. If an error occurs, the contents of "sb"
 473 * are left untouched, and -1 is returned.
 474 */
 475extern int strbuf_normalize_path(struct strbuf *sb);
 476
 477/**
 478 * Strip whitespace from a buffer. The second parameter controls if
 479 * comments are considered contents to be removed or not.
 480 */
 481extern void strbuf_stripspace(struct strbuf *buf, int skip_comments);
 482
 483static inline int strbuf_strip_suffix(struct strbuf *sb, const char *suffix)
 484{
 485        if (strip_suffix_mem(sb->buf, &sb->len, suffix)) {
 486                strbuf_setlen(sb, sb->len);
 487                return 1;
 488        } else
 489                return 0;
 490}
 491
 492/**
 493 * Split str (of length slen) at the specified terminator character.
 494 * Return a null-terminated array of pointers to strbuf objects
 495 * holding the substrings.  The substrings include the terminator,
 496 * except for the last substring, which might be unterminated if the
 497 * original string did not end with a terminator.  If max is positive,
 498 * then split the string into at most max substrings (with the last
 499 * substring containing everything following the (max-1)th terminator
 500 * character).
 501 *
 502 * The most generic form is `strbuf_split_buf`, which takes an arbitrary
 503 * pointer/len buffer. The `_str` variant takes a NUL-terminated string,
 504 * the `_max` variant takes a strbuf, and just `strbuf_split` is a convenience
 505 * wrapper to drop the `max` parameter.
 506 *
 507 * For lighter-weight alternatives, see string_list_split() and
 508 * string_list_split_in_place().
 509 */
 510extern struct strbuf **strbuf_split_buf(const char *, size_t,
 511                                        int terminator, int max);
 512
 513static inline struct strbuf **strbuf_split_str(const char *str,
 514                                               int terminator, int max)
 515{
 516        return strbuf_split_buf(str, strlen(str), terminator, max);
 517}
 518
 519static inline struct strbuf **strbuf_split_max(const struct strbuf *sb,
 520                                                int terminator, int max)
 521{
 522        return strbuf_split_buf(sb->buf, sb->len, terminator, max);
 523}
 524
 525static inline struct strbuf **strbuf_split(const struct strbuf *sb,
 526                                           int terminator)
 527{
 528        return strbuf_split_max(sb, terminator, 0);
 529}
 530
 531/**
 532 * Free a NULL-terminated list of strbufs (for example, the return
 533 * values of the strbuf_split*() functions).
 534 */
 535extern void strbuf_list_free(struct strbuf **);
 536
 537/**
 538 * Add the abbreviation, as generated by find_unique_abbrev, of `sha1` to
 539 * the strbuf `sb`.
 540 */
 541extern void strbuf_add_unique_abbrev(struct strbuf *sb,
 542                                     const unsigned char *sha1,
 543                                     int abbrev_len);
 544
 545/**
 546 * Launch the user preferred editor to edit a file and fill the buffer
 547 * with the file's contents upon the user completing their editing. The
 548 * third argument can be used to set the environment which the editor is
 549 * run in. If the buffer is NULL the editor is launched as usual but the
 550 * file's contents are not read into the buffer upon completion.
 551 */
 552extern int launch_editor(const char *path, struct strbuf *buffer, const char *const *env);
 553
 554extern void strbuf_add_lines(struct strbuf *sb, const char *prefix, const char *buf, size_t size);
 555
 556/**
 557 * Append s to sb, with the characters '<', '>', '&' and '"' converted
 558 * into XML entities.
 559 */
 560extern void strbuf_addstr_xml_quoted(struct strbuf *sb, const char *s);
 561
 562/**
 563 * "Complete" the contents of `sb` by ensuring that either it ends with the
 564 * character `term`, or it is empty.  This can be used, for example,
 565 * to ensure that text ends with a newline, but without creating an empty
 566 * blank line if there is no content in the first place.
 567 */
 568static inline void strbuf_complete(struct strbuf *sb, char term)
 569{
 570        if (sb->len && sb->buf[sb->len - 1] != term)
 571                strbuf_addch(sb, term);
 572}
 573
 574static inline void strbuf_complete_line(struct strbuf *sb)
 575{
 576        strbuf_complete(sb, '\n');
 577}
 578
 579/*
 580 * Copy "name" to "sb", expanding any special @-marks as handled by
 581 * interpret_branch_name(). The result is a non-qualified branch name
 582 * (so "foo" or "origin/master" instead of "refs/heads/foo" or
 583 * "refs/remotes/origin/master").
 584 *
 585 * Note that the resulting name may not be a syntactically valid refname.
 586 *
 587 * If "allowed" is non-zero, restrict the set of allowed expansions. See
 588 * interpret_branch_name() for details.
 589 */
 590extern void strbuf_branchname(struct strbuf *sb, const char *name,
 591                              unsigned allowed);
 592
 593/*
 594 * Like strbuf_branchname() above, but confirm that the result is
 595 * syntactically valid to be used as a local branch name in refs/heads/.
 596 *
 597 * The return value is "0" if the result is valid, and "-1" otherwise.
 598 */
 599extern int strbuf_check_branch_ref(struct strbuf *sb, const char *name);
 600
 601extern void strbuf_addstr_urlencode(struct strbuf *, const char *,
 602                                    int reserved);
 603
 604__attribute__((format (printf,1,2)))
 605extern int printf_ln(const char *fmt, ...);
 606__attribute__((format (printf,2,3)))
 607extern int fprintf_ln(FILE *fp, const char *fmt, ...);
 608
 609char *xstrdup_tolower(const char *);
 610
 611/**
 612 * Create a newly allocated string using printf format. You can do this easily
 613 * with a strbuf, but this provides a shortcut to save a few lines.
 614 */
 615__attribute__((format (printf, 1, 0)))
 616char *xstrvfmt(const char *fmt, va_list ap);
 617__attribute__((format (printf, 1, 2)))
 618char *xstrfmt(const char *fmt, ...);
 619
 620#endif /* STRBUF_H */