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