1#ifndef STRBUF_H 2#define STRBUF_H 3 4struct string_list; 5 6/** 7 * strbuf's are meant to be used with all the usual C string and memory 8 * APIs. Given that the length of the buffer is known, it's often better to 9 * use the mem* functions than a str* one (memchr vs. strchr e.g.). 10 * Though, one has to be careful about the fact that str* functions often 11 * stop on NULs and that strbufs may have embedded NULs. 12 * 13 * A strbuf is NUL terminated for convenience, but no function in the 14 * strbuf API actually relies on the string being free of NULs. 15 * 16 * strbufs have some invariants that are very important to keep in mind: 17 * 18 * - The `buf` member is never NULL, so it can be used in any usual C 19 * string operations safely. strbuf's _have_ to be initialized either by 20 * `strbuf_init()` or by `= STRBUF_INIT` before the invariants, though. 21 * 22 * Do *not* assume anything on what `buf` really is (e.g. if it is 23 * allocated memory or not), use `strbuf_detach()` to unwrap a memory 24 * buffer from its strbuf shell in a safe way. That is the sole supported 25 * way. This will give you a malloced buffer that you can later `free()`. 26 * 27 * However, it is totally safe to modify anything in the string pointed by 28 * the `buf` member, between the indices `0` and `len-1` (inclusive). 29 * 30 * - The `buf` member is a byte array that has at least `len + 1` bytes 31 * allocated. The extra byte is used to store a `'\0'`, allowing the 32 * `buf` member to be a valid C-string. Every strbuf function ensure this 33 * invariant is preserved. 34 * 35 * NOTE: It is OK to "play" with the buffer directly if you work it this 36 * way: 37 * 38 * strbuf_grow(sb, SOME_SIZE); <1> 39 * strbuf_setlen(sb, sb->len + SOME_OTHER_SIZE); 40 * 41 * <1> Here, the memory array starting at `sb->buf`, and of length 42 * `strbuf_avail(sb)` is all yours, and you can be sure that 43 * `strbuf_avail(sb)` is at least `SOME_SIZE`. 44 * 45 * NOTE: `SOME_OTHER_SIZE` must be smaller or equal to `strbuf_avail(sb)`. 46 * 47 * Doing so is safe, though if it has to be done in many places, adding the 48 * missing API to the strbuf module is the way to go. 49 * 50 * WARNING: Do _not_ assume that the area that is yours is of size `alloc 51 * - 1` even if it's true in the current implementation. Alloc is somehow a 52 * "private" member that should not be messed with. Use `strbuf_avail()` 53 * instead. 54*/ 55 56/** 57 * Data Structures 58 * --------------- 59 */ 60 61/** 62 * This is the string buffer structure. The `len` member can be used to 63 * determine the current length of the string, and `buf` member provides 64 * access to the string itself. 65 */ 66struct strbuf { 67size_t alloc; 68size_t len; 69char*buf; 70}; 71 72externchar strbuf_slopbuf[]; 73#define STRBUF_INIT { .alloc = 0, .len = 0, .buf = strbuf_slopbuf } 74 75/** 76 * Life Cycle Functions 77 * -------------------- 78 */ 79 80/** 81 * Initialize the structure. The second parameter can be zero or a bigger 82 * number to allocate memory, in case you want to prevent further reallocs. 83 */ 84externvoidstrbuf_init(struct strbuf *,size_t); 85 86/** 87 * Release a string buffer and the memory it used. After this call, the 88 * strbuf points to an empty string that does not need to be free()ed, as 89 * if it had been set to `STRBUF_INIT` and never modified. 90 * 91 * To clear a strbuf in preparation for further use without the overhead 92 * of free()ing and malloc()ing again, use strbuf_reset() instead. 93 */ 94externvoidstrbuf_release(struct strbuf *); 95 96/** 97 * Detach the string from the strbuf and returns it; you now own the 98 * storage the string occupies and it is your responsibility from then on 99 * to release it with `free(3)` when you are done with it. 100 * 101 * The strbuf that previously held the string is reset to `STRBUF_INIT` so 102 * it can be reused after calling this function. 103 */ 104externchar*strbuf_detach(struct strbuf *,size_t*); 105 106/** 107 * Attach a string to a buffer. You should specify the string to attach, 108 * the current length of the string and the amount of allocated memory. 109 * The amount must be larger than the string length, because the string you 110 * pass is supposed to be a NUL-terminated string. This string _must_ be 111 * malloc()ed, and after attaching, the pointer cannot be relied upon 112 * anymore, and neither be free()d directly. 113 */ 114externvoidstrbuf_attach(struct strbuf *,void*,size_t,size_t); 115 116/** 117 * Swap the contents of two string buffers. 118 */ 119staticinlinevoidstrbuf_swap(struct strbuf *a,struct strbuf *b) 120{ 121SWAP(*a, *b); 122} 123 124 125/** 126 * Functions related to the size of the buffer 127 * ------------------------------------------- 128 */ 129 130/** 131 * Determine the amount of allocated but unused memory. 132 */ 133staticinlinesize_tstrbuf_avail(const struct strbuf *sb) 134{ 135return sb->alloc ? sb->alloc - sb->len -1:0; 136} 137 138/** 139 * Ensure that at least this amount of unused memory is available after 140 * `len`. This is used when you know a typical size for what you will add 141 * and want to avoid repetitive automatic resizing of the underlying buffer. 142 * This is never a needed operation, but can be critical for performance in 143 * some cases. 144 */ 145externvoidstrbuf_grow(struct strbuf *,size_t); 146 147/** 148 * Set the length of the buffer to a given value. This function does *not* 149 * allocate new memory, so you should not perform a `strbuf_setlen()` to a 150 * length that is larger than `len + strbuf_avail()`. `strbuf_setlen()` is 151 * just meant as a 'please fix invariants from this strbuf I just messed 152 * with'. 153 */ 154staticinlinevoidstrbuf_setlen(struct strbuf *sb,size_t len) 155{ 156if(len > (sb->alloc ? sb->alloc -1:0)) 157die("BUG: strbuf_setlen() beyond buffer"); 158 sb->len = len; 159if(sb->buf != strbuf_slopbuf) 160 sb->buf[len] ='\0'; 161else 162assert(!strbuf_slopbuf[0]); 163} 164 165/** 166 * Empty the buffer by setting the size of it to zero. 167 */ 168#define strbuf_reset(sb) strbuf_setlen(sb, 0) 169 170 171/** 172 * Functions related to the contents of the buffer 173 * ----------------------------------------------- 174 */ 175 176/** 177 * Strip whitespace from the beginning (`ltrim`), end (`rtrim`), or both side 178 * (`trim`) of a string. 179 */ 180externvoidstrbuf_trim(struct strbuf *); 181externvoidstrbuf_rtrim(struct strbuf *); 182externvoidstrbuf_ltrim(struct strbuf *); 183 184/* Strip trailing directory separators */ 185externvoidstrbuf_trim_trailing_dir_sep(struct strbuf *); 186 187/** 188 * Replace the contents of the strbuf with a reencoded form. Returns -1 189 * on error, 0 on success. 190 */ 191externintstrbuf_reencode(struct strbuf *sb,const char*from,const char*to); 192 193/** 194 * Lowercase each character in the buffer using `tolower`. 195 */ 196externvoidstrbuf_tolower(struct strbuf *sb); 197 198/** 199 * Compare two buffers. Returns an integer less than, equal to, or greater 200 * than zero if the first buffer is found, respectively, to be less than, 201 * to match, or be greater than the second buffer. 202 */ 203externintstrbuf_cmp(const struct strbuf *,const struct strbuf *); 204 205 206/** 207 * Adding data to the buffer 208 * ------------------------- 209 * 210 * NOTE: All of the functions in this section will grow the buffer as 211 * necessary. If they fail for some reason other than memory shortage and the 212 * buffer hadn't been allocated before (i.e. the `struct strbuf` was set to 213 * `STRBUF_INIT`), then they will free() it. 214 */ 215 216/** 217 * Add a single character to the buffer. 218 */ 219staticinlinevoidstrbuf_addch(struct strbuf *sb,int c) 220{ 221if(!strbuf_avail(sb)) 222strbuf_grow(sb,1); 223 sb->buf[sb->len++] = c; 224 sb->buf[sb->len] ='\0'; 225} 226 227/** 228 * Add a character the specified number of times to the buffer. 229 */ 230externvoidstrbuf_addchars(struct strbuf *sb,int c,size_t n); 231 232/** 233 * Insert data to the given position of the buffer. The remaining contents 234 * will be shifted, not overwritten. 235 */ 236externvoidstrbuf_insert(struct strbuf *,size_t pos,const void*,size_t); 237 238/** 239 * Remove given amount of data from a given position of the buffer. 240 */ 241externvoidstrbuf_remove(struct strbuf *,size_t pos,size_t len); 242 243/** 244 * Remove the bytes between `pos..pos+len` and replace it with the given 245 * data. 246 */ 247externvoidstrbuf_splice(struct strbuf *,size_t pos,size_t len, 248const void*,size_t); 249 250/** 251 * Add a NUL-terminated string to the buffer. Each line will be prepended 252 * by a comment character and a blank. 253 */ 254externvoidstrbuf_add_commented_lines(struct strbuf *out,const char*buf,size_t size); 255 256 257/** 258 * Add data of given length to the buffer. 259 */ 260externvoidstrbuf_add(struct strbuf *,const void*,size_t); 261 262/** 263 * Add a NUL-terminated string to the buffer. 264 * 265 * NOTE: This function will *always* be implemented as an inline or a macro 266 * using strlen, meaning that this is efficient to write things like: 267 * 268 * strbuf_addstr(sb, "immediate string"); 269 * 270 */ 271staticinlinevoidstrbuf_addstr(struct strbuf *sb,const char*s) 272{ 273strbuf_add(sb, s,strlen(s)); 274} 275 276/** 277 * Copy the contents of another buffer at the end of the current one. 278 */ 279externvoidstrbuf_addbuf(struct strbuf *sb,const struct strbuf *sb2); 280 281/** 282 * This function can be used to expand a format string containing 283 * placeholders. To that end, it parses the string and calls the specified 284 * function for every percent sign found. 285 * 286 * The callback function is given a pointer to the character after the `%` 287 * and a pointer to the struct strbuf. It is expected to add the expanded 288 * version of the placeholder to the strbuf, e.g. to add a newline 289 * character if the letter `n` appears after a `%`. The function returns 290 * the length of the placeholder recognized and `strbuf_expand()` skips 291 * over it. 292 * 293 * The format `%%` is automatically expanded to a single `%` as a quoting 294 * mechanism; callers do not need to handle the `%` placeholder themselves, 295 * and the callback function will not be invoked for this placeholder. 296 * 297 * All other characters (non-percent and not skipped ones) are copied 298 * verbatim to the strbuf. If the callback returned zero, meaning that the 299 * placeholder is unknown, then the percent sign is copied, too. 300 * 301 * In order to facilitate caching and to make it possible to give 302 * parameters to the callback, `strbuf_expand()` passes a context pointer, 303 * which can be used by the programmer of the callback as she sees fit. 304 */ 305typedefsize_t(*expand_fn_t) (struct strbuf *sb,const char*placeholder,void*context); 306externvoidstrbuf_expand(struct strbuf *sb,const char*format, expand_fn_t fn,void*context); 307 308/** 309 * Used as callback for `strbuf_expand()`, expects an array of 310 * struct strbuf_expand_dict_entry as context, i.e. pairs of 311 * placeholder and replacement string. The array needs to be 312 * terminated by an entry with placeholder set to NULL. 313 */ 314struct strbuf_expand_dict_entry { 315const char*placeholder; 316const char*value; 317}; 318externsize_tstrbuf_expand_dict_cb(struct strbuf *sb,const char*placeholder,void*context); 319 320/** 321 * Append the contents of one strbuf to another, quoting any 322 * percent signs ("%") into double-percents ("%%") in the 323 * destination. This is useful for literal data to be fed to either 324 * strbuf_expand or to the *printf family of functions. 325 */ 326externvoidstrbuf_addbuf_percentquote(struct strbuf *dst,const struct strbuf *src); 327 328/** 329 * Append the given byte size as a human-readable string (i.e. 12.23 KiB, 330 * 3.50 MiB). 331 */ 332externvoidstrbuf_humanise_bytes(struct strbuf *buf, off_t bytes); 333 334/** 335 * Add a formatted string to the buffer. 336 */ 337__attribute__((format(printf,2,3))) 338externvoidstrbuf_addf(struct strbuf *sb,const char*fmt, ...); 339 340/** 341 * Add a formatted string prepended by a comment character and a 342 * blank to the buffer. 343 */ 344__attribute__((format(printf,2,3))) 345externvoidstrbuf_commented_addf(struct strbuf *sb,const char*fmt, ...); 346 347__attribute__((format(printf,2,0))) 348externvoidstrbuf_vaddf(struct strbuf *sb,const char*fmt,va_list ap); 349 350/** 351 * Add the time specified by `tm`, as formatted by `strftime`. 352 * `tz_offset` is in decimal hhmm format, e.g. -600 means six hours west 353 * of Greenwich, and it's used to expand %z internally. However, tokens 354 * with modifiers (e.g. %Ez) are passed to `strftime`. 355 * `suppress_tz_name`, when set, expands %Z internally to the empty 356 * string rather than passing it to `strftime`. 357 */ 358externvoidstrbuf_addftime(struct strbuf *sb,const char*fmt, 359const struct tm *tm,int tz_offset, 360int suppress_tz_name); 361 362/** 363 * Read a given size of data from a FILE* pointer to the buffer. 364 * 365 * NOTE: The buffer is rewound if the read fails. If -1 is returned, 366 * `errno` must be consulted, like you would do for `read(3)`. 367 * `strbuf_read()`, `strbuf_read_file()` and `strbuf_getline_*()` 368 * family of functions have the same behaviour as well. 369 */ 370externsize_tstrbuf_fread(struct strbuf *,size_t,FILE*); 371 372/** 373 * Read the contents of a given file descriptor. The third argument can be 374 * used to give a hint about the file size, to avoid reallocs. If read fails, 375 * any partial read is undone. 376 */ 377extern ssize_t strbuf_read(struct strbuf *,int fd,size_t hint); 378 379/** 380 * Read the contents of a given file descriptor partially by using only one 381 * attempt of xread. The third argument can be used to give a hint about the 382 * file size, to avoid reallocs. Returns the number of new bytes appended to 383 * the sb. 384 */ 385extern ssize_t strbuf_read_once(struct strbuf *,int fd,size_t hint); 386 387/** 388 * Read the contents of a file, specified by its path. The third argument 389 * can be used to give a hint about the file size, to avoid reallocs. 390 * Return the number of bytes read or a negative value if some error 391 * occurred while opening or reading the file. 392 */ 393extern ssize_t strbuf_read_file(struct strbuf *sb,const char*path,size_t hint); 394 395/** 396 * Read the target of a symbolic link, specified by its path. The third 397 * argument can be used to give a hint about the size, to avoid reallocs. 398 */ 399externintstrbuf_readlink(struct strbuf *sb,const char*path,size_t hint); 400 401/** 402 * Write the whole content of the strbuf to the stream not stopping at 403 * NUL bytes. 404 */ 405extern ssize_t strbuf_write(struct strbuf *sb,FILE*stream); 406 407/** 408 * Read a line from a FILE *, overwriting the existing contents of 409 * the strbuf. The strbuf_getline*() family of functions share 410 * this signature, but have different line termination conventions. 411 * 412 * Reading stops after the terminator or at EOF. The terminator 413 * is removed from the buffer before returning. Returns 0 unless 414 * there was nothing left before EOF, in which case it returns `EOF`. 415 */ 416typedefint(*strbuf_getline_fn)(struct strbuf *,FILE*); 417 418/* Uses LF as the line terminator */ 419externintstrbuf_getline_lf(struct strbuf *sb,FILE*fp); 420 421/* Uses NUL as the line terminator */ 422externintstrbuf_getline_nul(struct strbuf *sb,FILE*fp); 423 424/* 425 * Similar to strbuf_getline_lf(), but additionally treats a CR that 426 * comes immediately before the LF as part of the terminator. 427 * This is the most friendly version to be used to read "text" files 428 * that can come from platforms whose native text format is CRLF 429 * terminated. 430 */ 431externintstrbuf_getline(struct strbuf *,FILE*); 432 433 434/** 435 * Like `strbuf_getline`, but keeps the trailing terminator (if 436 * any) in the buffer. 437 */ 438externintstrbuf_getwholeline(struct strbuf *,FILE*,int); 439 440/** 441 * Like `strbuf_getwholeline`, but operates on a file descriptor. 442 * It reads one character at a time, so it is very slow. Do not 443 * use it unless you need the correct position in the file 444 * descriptor. 445 */ 446externintstrbuf_getwholeline_fd(struct strbuf *,int,int); 447 448/** 449 * Set the buffer to the path of the current working directory. 450 */ 451externintstrbuf_getcwd(struct strbuf *sb); 452 453/** 454 * Add a path to a buffer, converting a relative path to an 455 * absolute one in the process. Symbolic links are not 456 * resolved. 457 */ 458externvoidstrbuf_add_absolute_path(struct strbuf *sb,const char*path); 459 460/** 461 * Canonize `path` (make it absolute, resolve symlinks, remove extra 462 * slashes) and append it to `sb`. Die with an informative error 463 * message if there is a problem. 464 * 465 * The directory part of `path` (i.e., everything up to the last 466 * dir_sep) must denote a valid, existing directory, but the last 467 * component need not exist. 468 * 469 * Callers that don't mind links should use the more lightweight 470 * strbuf_add_absolute_path() instead. 471 */ 472externvoidstrbuf_add_real_path(struct strbuf *sb,const char*path); 473 474 475/** 476 * Normalize in-place the path contained in the strbuf. See 477 * normalize_path_copy() for details. If an error occurs, the contents of "sb" 478 * are left untouched, and -1 is returned. 479 */ 480externintstrbuf_normalize_path(struct strbuf *sb); 481 482/** 483 * Strip whitespace from a buffer. The second parameter controls if 484 * comments are considered contents to be removed or not. 485 */ 486externvoidstrbuf_stripspace(struct strbuf *buf,int skip_comments); 487 488staticinlineintstrbuf_strip_suffix(struct strbuf *sb,const char*suffix) 489{ 490if(strip_suffix_mem(sb->buf, &sb->len, suffix)) { 491strbuf_setlen(sb, sb->len); 492return1; 493}else 494return0; 495} 496 497/** 498 * Split str (of length slen) at the specified terminator character. 499 * Return a null-terminated array of pointers to strbuf objects 500 * holding the substrings. The substrings include the terminator, 501 * except for the last substring, which might be unterminated if the 502 * original string did not end with a terminator. If max is positive, 503 * then split the string into at most max substrings (with the last 504 * substring containing everything following the (max-1)th terminator 505 * character). 506 * 507 * The most generic form is `strbuf_split_buf`, which takes an arbitrary 508 * pointer/len buffer. The `_str` variant takes a NUL-terminated string, 509 * the `_max` variant takes a strbuf, and just `strbuf_split` is a convenience 510 * wrapper to drop the `max` parameter. 511 * 512 * For lighter-weight alternatives, see string_list_split() and 513 * string_list_split_in_place(). 514 */ 515externstruct strbuf **strbuf_split_buf(const char*,size_t, 516int terminator,int max); 517 518staticinlinestruct strbuf **strbuf_split_str(const char*str, 519int terminator,int max) 520{ 521returnstrbuf_split_buf(str,strlen(str), terminator, max); 522} 523 524staticinlinestruct strbuf **strbuf_split_max(const struct strbuf *sb, 525int terminator,int max) 526{ 527returnstrbuf_split_buf(sb->buf, sb->len, terminator, max); 528} 529 530staticinlinestruct strbuf **strbuf_split(const struct strbuf *sb, 531int terminator) 532{ 533returnstrbuf_split_max(sb, terminator,0); 534} 535 536/* 537 * Adds all strings of a string list to the strbuf, separated by the given 538 * separator. For example, if sep is 539 * ', ' 540 * and slist contains 541 * ['element1', 'element2', ..., 'elementN'], 542 * then write: 543 * 'element1, element2, ..., elementN' 544 * to str. If only one element, just write "element1" to str. 545 */ 546externvoidstrbuf_add_separated_string_list(struct strbuf *str, 547const char*sep, 548struct string_list *slist); 549 550/** 551 * Free a NULL-terminated list of strbufs (for example, the return 552 * values of the strbuf_split*() functions). 553 */ 554externvoidstrbuf_list_free(struct strbuf **); 555 556/** 557 * Add the abbreviation, as generated by find_unique_abbrev, of `sha1` to 558 * the strbuf `sb`. 559 */ 560externvoidstrbuf_add_unique_abbrev(struct strbuf *sb, 561const unsigned char*sha1, 562int abbrev_len); 563 564/** 565 * Launch the user preferred editor to edit a file and fill the buffer 566 * with the file's contents upon the user completing their editing. The 567 * third argument can be used to set the environment which the editor is 568 * run in. If the buffer is NULL the editor is launched as usual but the 569 * file's contents are not read into the buffer upon completion. 570 */ 571externintlaunch_editor(const char*path,struct strbuf *buffer,const char*const*env); 572 573externvoidstrbuf_add_lines(struct strbuf *sb,const char*prefix,const char*buf,size_t size); 574 575/** 576 * Append s to sb, with the characters '<', '>', '&' and '"' converted 577 * into XML entities. 578 */ 579externvoidstrbuf_addstr_xml_quoted(struct strbuf *sb,const char*s); 580 581/** 582 * "Complete" the contents of `sb` by ensuring that either it ends with the 583 * character `term`, or it is empty. This can be used, for example, 584 * to ensure that text ends with a newline, but without creating an empty 585 * blank line if there is no content in the first place. 586 */ 587staticinlinevoidstrbuf_complete(struct strbuf *sb,char term) 588{ 589if(sb->len && sb->buf[sb->len -1] != term) 590strbuf_addch(sb, term); 591} 592 593staticinlinevoidstrbuf_complete_line(struct strbuf *sb) 594{ 595strbuf_complete(sb,'\n'); 596} 597 598/* 599 * Copy "name" to "sb", expanding any special @-marks as handled by 600 * interpret_branch_name(). The result is a non-qualified branch name 601 * (so "foo" or "origin/master" instead of "refs/heads/foo" or 602 * "refs/remotes/origin/master"). 603 * 604 * Note that the resulting name may not be a syntactically valid refname. 605 * 606 * If "allowed" is non-zero, restrict the set of allowed expansions. See 607 * interpret_branch_name() for details. 608 */ 609externvoidstrbuf_branchname(struct strbuf *sb,const char*name, 610unsigned allowed); 611 612/* 613 * Like strbuf_branchname() above, but confirm that the result is 614 * syntactically valid to be used as a local branch name in refs/heads/. 615 * 616 * The return value is "0" if the result is valid, and "-1" otherwise. 617 */ 618externintstrbuf_check_branch_ref(struct strbuf *sb,const char*name); 619 620externvoidstrbuf_addstr_urlencode(struct strbuf *,const char*, 621int reserved); 622 623__attribute__((format(printf,1,2))) 624externintprintf_ln(const char*fmt, ...); 625__attribute__((format(printf,2,3))) 626externintfprintf_ln(FILE*fp,const char*fmt, ...); 627 628char*xstrdup_tolower(const char*); 629 630/** 631 * Create a newly allocated string using printf format. You can do this easily 632 * with a strbuf, but this provides a shortcut to save a few lines. 633 */ 634__attribute__((format(printf,1,0))) 635char*xstrvfmt(const char*fmt,va_list ap); 636__attribute__((format(printf,1,2))) 637char*xstrfmt(const char*fmt, ...); 638 639#endif/* STRBUF_H */