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 { 65size_t alloc; 66size_t len; 67char*buf; 68}; 69 70externchar 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 */ 82externvoidstrbuf_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 */ 92externvoidstrbuf_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 */ 102externchar*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 */ 112externvoidstrbuf_attach(struct strbuf *,void*,size_t,size_t); 113 114/** 115 * Swap the contents of two string buffers. 116 */ 117staticinlinevoidstrbuf_swap(struct strbuf *a,struct strbuf *b) 118{ 119SWAP(*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 */ 131staticinlinesize_tstrbuf_avail(const struct strbuf *sb) 132{ 133return 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 */ 143externvoidstrbuf_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 */ 152staticinlinevoidstrbuf_setlen(struct strbuf *sb,size_t len) 153{ 154if(len > (sb->alloc ? sb->alloc -1:0)) 155die("BUG: strbuf_setlen() beyond buffer"); 156 sb->len = len; 157if(sb->buf != strbuf_slopbuf) 158 sb->buf[len] ='\0'; 159else 160assert(!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 */ 178externvoidstrbuf_trim(struct strbuf *); 179externvoidstrbuf_rtrim(struct strbuf *); 180externvoidstrbuf_ltrim(struct strbuf *); 181 182/* Strip trailing directory separators */ 183externvoidstrbuf_trim_trailing_dir_sep(struct strbuf *); 184 185/** 186 * Replace the contents of the strbuf with a reencoded form. Returns -1 187 * on error, 0 on success. 188 */ 189externintstrbuf_reencode(struct strbuf *sb,const char*from,const char*to); 190 191/** 192 * Lowercase each character in the buffer using `tolower`. 193 */ 194externvoidstrbuf_tolower(struct strbuf *sb); 195 196/** 197 * Compare two buffers. Returns an integer less than, equal to, or greater 198 * than zero if the first buffer is found, respectively, to be less than, 199 * to match, or be greater than the second buffer. 200 */ 201externintstrbuf_cmp(const struct strbuf *,const struct strbuf *); 202 203 204/** 205 * Adding data to the buffer 206 * ------------------------- 207 * 208 * NOTE: All of the functions in this section will grow the buffer as 209 * necessary. If they fail for some reason other than memory shortage and the 210 * buffer hadn't been allocated before (i.e. the `struct strbuf` was set to 211 * `STRBUF_INIT`), then they will free() it. 212 */ 213 214/** 215 * Add a single character to the buffer. 216 */ 217staticinlinevoidstrbuf_addch(struct strbuf *sb,int c) 218{ 219if(!strbuf_avail(sb)) 220strbuf_grow(sb,1); 221 sb->buf[sb->len++] = c; 222 sb->buf[sb->len] ='\0'; 223} 224 225/** 226 * Add a character the specified number of times to the buffer. 227 */ 228externvoidstrbuf_addchars(struct strbuf *sb,int c,size_t n); 229 230/** 231 * Insert data to the given position of the buffer. The remaining contents 232 * will be shifted, not overwritten. 233 */ 234externvoidstrbuf_insert(struct strbuf *,size_t pos,const void*,size_t); 235 236/** 237 * Remove given amount of data from a given position of the buffer. 238 */ 239externvoidstrbuf_remove(struct strbuf *,size_t pos,size_t len); 240 241/** 242 * Remove the bytes between `pos..pos+len` and replace it with the given 243 * data. 244 */ 245externvoidstrbuf_splice(struct strbuf *,size_t pos,size_t len, 246const void*,size_t); 247 248/** 249 * Add a NUL-terminated string to the buffer. Each line will be prepended 250 * by a comment character and a blank. 251 */ 252externvoidstrbuf_add_commented_lines(struct strbuf *out,const char*buf,size_t size); 253 254 255/** 256 * Add data of given length to the buffer. 257 */ 258externvoidstrbuf_add(struct strbuf *,const void*,size_t); 259 260/** 261 * Add a NUL-terminated string to the buffer. 262 * 263 * NOTE: This function will *always* be implemented as an inline or a macro 264 * using strlen, meaning that this is efficient to write things like: 265 * 266 * strbuf_addstr(sb, "immediate string"); 267 * 268 */ 269staticinlinevoidstrbuf_addstr(struct strbuf *sb,const char*s) 270{ 271strbuf_add(sb, s,strlen(s)); 272} 273 274/** 275 * Copy the contents of another buffer at the end of the current one. 276 */ 277externvoidstrbuf_addbuf(struct strbuf *sb,const struct strbuf *sb2); 278 279/** 280 * This function can be used to expand a format string containing 281 * placeholders. To that end, it parses the string and calls the specified 282 * function for every percent sign found. 283 * 284 * The callback function is given a pointer to the character after the `%` 285 * and a pointer to the struct strbuf. It is expected to add the expanded 286 * version of the placeholder to the strbuf, e.g. to add a newline 287 * character if the letter `n` appears after a `%`. The function returns 288 * the length of the placeholder recognized and `strbuf_expand()` skips 289 * over it. 290 * 291 * The format `%%` is automatically expanded to a single `%` as a quoting 292 * mechanism; callers do not need to handle the `%` placeholder themselves, 293 * and the callback function will not be invoked for this placeholder. 294 * 295 * All other characters (non-percent and not skipped ones) are copied 296 * verbatim to the strbuf. If the callback returned zero, meaning that the 297 * placeholder is unknown, then the percent sign is copied, too. 298 * 299 * In order to facilitate caching and to make it possible to give 300 * parameters to the callback, `strbuf_expand()` passes a context pointer, 301 * which can be used by the programmer of the callback as she sees fit. 302 */ 303typedefsize_t(*expand_fn_t) (struct strbuf *sb,const char*placeholder,void*context); 304externvoidstrbuf_expand(struct strbuf *sb,const char*format, expand_fn_t fn,void*context); 305 306/** 307 * Used as callback for `strbuf_expand()`, expects an array of 308 * struct strbuf_expand_dict_entry as context, i.e. pairs of 309 * placeholder and replacement string. The array needs to be 310 * terminated by an entry with placeholder set to NULL. 311 */ 312struct strbuf_expand_dict_entry { 313const char*placeholder; 314const char*value; 315}; 316externsize_tstrbuf_expand_dict_cb(struct strbuf *sb,const char*placeholder,void*context); 317 318/** 319 * Append the contents of one strbuf to another, quoting any 320 * percent signs ("%") into double-percents ("%%") in the 321 * destination. This is useful for literal data to be fed to either 322 * strbuf_expand or to the *printf family of functions. 323 */ 324externvoidstrbuf_addbuf_percentquote(struct strbuf *dst,const struct strbuf *src); 325 326/** 327 * Append the given byte size as a human-readable string (i.e. 12.23 KiB, 328 * 3.50 MiB). 329 */ 330externvoidstrbuf_humanise_bytes(struct strbuf *buf, off_t bytes); 331 332/** 333 * Add a formatted string to the buffer. 334 */ 335__attribute__((format(printf,2,3))) 336externvoidstrbuf_addf(struct strbuf *sb,const char*fmt, ...); 337 338/** 339 * Add a formatted string prepended by a comment character and a 340 * blank to the buffer. 341 */ 342__attribute__((format(printf,2,3))) 343externvoidstrbuf_commented_addf(struct strbuf *sb,const char*fmt, ...); 344 345__attribute__((format(printf,2,0))) 346externvoidstrbuf_vaddf(struct strbuf *sb,const char*fmt,va_list ap); 347 348/** 349 * Add the time specified by `tm`, as formatted by `strftime`. 350 * `tz_offset` is in decimal hhmm format, e.g. -600 means six hours west 351 * of Greenwich, and it's used to expand %z internally. However, tokens 352 * with modifiers (e.g. %Ez) are passed to `strftime`. 353 * `suppress_tz_name`, when set, expands %Z internally to the empty 354 * string rather than passing it to `strftime`. 355 */ 356externvoidstrbuf_addftime(struct strbuf *sb,const char*fmt, 357const struct tm *tm,int tz_offset, 358int suppress_tz_name); 359 360/** 361 * Read a given size of data from a FILE* pointer to the buffer. 362 * 363 * NOTE: The buffer is rewound if the read fails. If -1 is returned, 364 * `errno` must be consulted, like you would do for `read(3)`. 365 * `strbuf_read()`, `strbuf_read_file()` and `strbuf_getline_*()` 366 * family of functions have the same behaviour as well. 367 */ 368externsize_tstrbuf_fread(struct strbuf *,size_t,FILE*); 369 370/** 371 * Read the contents of a given file descriptor. The third argument can be 372 * used to give a hint about the file size, to avoid reallocs. If read fails, 373 * any partial read is undone. 374 */ 375extern ssize_t strbuf_read(struct strbuf *,int fd,size_t hint); 376 377/** 378 * Read the contents of a given file descriptor partially by using only one 379 * attempt of xread. The third argument can be used to give a hint about the 380 * file size, to avoid reallocs. Returns the number of new bytes appended to 381 * the sb. 382 */ 383extern ssize_t strbuf_read_once(struct strbuf *,int fd,size_t hint); 384 385/** 386 * Read the contents of a file, specified by its path. The third argument 387 * can be used to give a hint about the file size, to avoid reallocs. 388 * Return the number of bytes read or a negative value if some error 389 * occurred while opening or reading the file. 390 */ 391extern ssize_t strbuf_read_file(struct strbuf *sb,const char*path,size_t hint); 392 393/** 394 * Read the target of a symbolic link, specified by its path. The third 395 * argument can be used to give a hint about the size, to avoid reallocs. 396 */ 397externintstrbuf_readlink(struct strbuf *sb,const char*path,size_t hint); 398 399/** 400 * Write the whole content of the strbuf to the stream not stopping at 401 * NUL bytes. 402 */ 403extern ssize_t strbuf_write(struct strbuf *sb,FILE*stream); 404 405/** 406 * Read a line from a FILE *, overwriting the existing contents of 407 * the strbuf. The strbuf_getline*() family of functions share 408 * this signature, but have different line termination conventions. 409 * 410 * Reading stops after the terminator or at EOF. The terminator 411 * is removed from the buffer before returning. Returns 0 unless 412 * there was nothing left before EOF, in which case it returns `EOF`. 413 */ 414typedefint(*strbuf_getline_fn)(struct strbuf *,FILE*); 415 416/* Uses LF as the line terminator */ 417externintstrbuf_getline_lf(struct strbuf *sb,FILE*fp); 418 419/* Uses NUL as the line terminator */ 420externintstrbuf_getline_nul(struct strbuf *sb,FILE*fp); 421 422/* 423 * Similar to strbuf_getline_lf(), but additionally treats a CR that 424 * comes immediately before the LF as part of the terminator. 425 * This is the most friendly version to be used to read "text" files 426 * that can come from platforms whose native text format is CRLF 427 * terminated. 428 */ 429externintstrbuf_getline(struct strbuf *,FILE*); 430 431 432/** 433 * Like `strbuf_getline`, but keeps the trailing terminator (if 434 * any) in the buffer. 435 */ 436externintstrbuf_getwholeline(struct strbuf *,FILE*,int); 437 438/** 439 * Like `strbuf_getwholeline`, but operates on a file descriptor. 440 * It reads one character at a time, so it is very slow. Do not 441 * use it unless you need the correct position in the file 442 * descriptor. 443 */ 444externintstrbuf_getwholeline_fd(struct strbuf *,int,int); 445 446/** 447 * Set the buffer to the path of the current working directory. 448 */ 449externintstrbuf_getcwd(struct strbuf *sb); 450 451/** 452 * Add a path to a buffer, converting a relative path to an 453 * absolute one in the process. Symbolic links are not 454 * resolved. 455 */ 456externvoidstrbuf_add_absolute_path(struct strbuf *sb,const char*path); 457 458/** 459 * Canonize `path` (make it absolute, resolve symlinks, remove extra 460 * slashes) and append it to `sb`. Die with an informative error 461 * message if there is a problem. 462 * 463 * The directory part of `path` (i.e., everything up to the last 464 * dir_sep) must denote a valid, existing directory, but the last 465 * component need not exist. 466 * 467 * Callers that don't mind links should use the more lightweight 468 * strbuf_add_absolute_path() instead. 469 */ 470externvoidstrbuf_add_real_path(struct strbuf *sb,const char*path); 471 472 473/** 474 * Normalize in-place the path contained in the strbuf. See 475 * normalize_path_copy() for details. If an error occurs, the contents of "sb" 476 * are left untouched, and -1 is returned. 477 */ 478externintstrbuf_normalize_path(struct strbuf *sb); 479 480/** 481 * Strip whitespace from a buffer. The second parameter controls if 482 * comments are considered contents to be removed or not. 483 */ 484externvoidstrbuf_stripspace(struct strbuf *buf,int skip_comments); 485 486staticinlineintstrbuf_strip_suffix(struct strbuf *sb,const char*suffix) 487{ 488if(strip_suffix_mem(sb->buf, &sb->len, suffix)) { 489strbuf_setlen(sb, sb->len); 490return1; 491}else 492return0; 493} 494 495/** 496 * Split str (of length slen) at the specified terminator character. 497 * Return a null-terminated array of pointers to strbuf objects 498 * holding the substrings. The substrings include the terminator, 499 * except for the last substring, which might be unterminated if the 500 * original string did not end with a terminator. If max is positive, 501 * then split the string into at most max substrings (with the last 502 * substring containing everything following the (max-1)th terminator 503 * character). 504 * 505 * The most generic form is `strbuf_split_buf`, which takes an arbitrary 506 * pointer/len buffer. The `_str` variant takes a NUL-terminated string, 507 * the `_max` variant takes a strbuf, and just `strbuf_split` is a convenience 508 * wrapper to drop the `max` parameter. 509 * 510 * For lighter-weight alternatives, see string_list_split() and 511 * string_list_split_in_place(). 512 */ 513externstruct strbuf **strbuf_split_buf(const char*,size_t, 514int terminator,int max); 515 516staticinlinestruct strbuf **strbuf_split_str(const char*str, 517int terminator,int max) 518{ 519returnstrbuf_split_buf(str,strlen(str), terminator, max); 520} 521 522staticinlinestruct strbuf **strbuf_split_max(const struct strbuf *sb, 523int terminator,int max) 524{ 525returnstrbuf_split_buf(sb->buf, sb->len, terminator, max); 526} 527 528staticinlinestruct strbuf **strbuf_split(const struct strbuf *sb, 529int terminator) 530{ 531returnstrbuf_split_max(sb, terminator,0); 532} 533 534/** 535 * Free a NULL-terminated list of strbufs (for example, the return 536 * values of the strbuf_split*() functions). 537 */ 538externvoidstrbuf_list_free(struct strbuf **); 539 540/** 541 * Add the abbreviation, as generated by find_unique_abbrev, of `sha1` to 542 * the strbuf `sb`. 543 */ 544externvoidstrbuf_add_unique_abbrev(struct strbuf *sb, 545const unsigned char*sha1, 546int abbrev_len); 547 548/** 549 * Launch the user preferred editor to edit a file and fill the buffer 550 * with the file's contents upon the user completing their editing. The 551 * third argument can be used to set the environment which the editor is 552 * run in. If the buffer is NULL the editor is launched as usual but the 553 * file's contents are not read into the buffer upon completion. 554 */ 555externintlaunch_editor(const char*path,struct strbuf *buffer,const char*const*env); 556 557externvoidstrbuf_add_lines(struct strbuf *sb,const char*prefix,const char*buf,size_t size); 558 559/** 560 * Append s to sb, with the characters '<', '>', '&' and '"' converted 561 * into XML entities. 562 */ 563externvoidstrbuf_addstr_xml_quoted(struct strbuf *sb,const char*s); 564 565/** 566 * "Complete" the contents of `sb` by ensuring that either it ends with the 567 * character `term`, or it is empty. This can be used, for example, 568 * to ensure that text ends with a newline, but without creating an empty 569 * blank line if there is no content in the first place. 570 */ 571staticinlinevoidstrbuf_complete(struct strbuf *sb,char term) 572{ 573if(sb->len && sb->buf[sb->len -1] != term) 574strbuf_addch(sb, term); 575} 576 577staticinlinevoidstrbuf_complete_line(struct strbuf *sb) 578{ 579strbuf_complete(sb,'\n'); 580} 581 582/* 583 * Copy "name" to "sb", expanding any special @-marks as handled by 584 * interpret_branch_name(). The result is a non-qualified branch name 585 * (so "foo" or "origin/master" instead of "refs/heads/foo" or 586 * "refs/remotes/origin/master"). 587 * 588 * Note that the resulting name may not be a syntactically valid refname. 589 * 590 * If "allowed" is non-zero, restrict the set of allowed expansions. See 591 * interpret_branch_name() for details. 592 */ 593externvoidstrbuf_branchname(struct strbuf *sb,const char*name, 594unsigned allowed); 595 596/* 597 * Like strbuf_branchname() above, but confirm that the result is 598 * syntactically valid to be used as a local branch name in refs/heads/. 599 * 600 * The return value is "0" if the result is valid, and "-1" otherwise. 601 */ 602externintstrbuf_check_branch_ref(struct strbuf *sb,const char*name); 603 604externvoidstrbuf_addstr_urlencode(struct strbuf *,const char*, 605int reserved); 606 607__attribute__((format(printf,1,2))) 608externintprintf_ln(const char*fmt, ...); 609__attribute__((format(printf,2,3))) 610externintfprintf_ln(FILE*fp,const char*fmt, ...); 611 612char*xstrdup_tolower(const char*); 613 614/** 615 * Create a newly allocated string using printf format. You can do this easily 616 * with a strbuf, but this provides a shortcut to save a few lines. 617 */ 618__attribute__((format(printf,1,0))) 619char*xstrvfmt(const char*fmt,va_list ap); 620__attribute__((format(printf,1,2))) 621char*xstrfmt(const char*fmt, ...); 622 623#endif/* STRBUF_H */