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