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 * ---- 37 * strbuf_grow(sb, SOME_SIZE); <1> 38 * strbuf_setlen(sb, sb->len + SOME_OTHER_SIZE); 39 * ---- 40 * <1> Here, the memory array starting at `sb->buf`, and of length 41 * `strbuf_avail(sb)` is all yours, and you can be sure that 42 * `strbuf_avail(sb)` is at least `SOME_SIZE`. 43 * + 44 * NOTE: `SOME_OTHER_SIZE` must be smaller or equal to `strbuf_avail(sb)`. 45 * + 46 * Doing so is safe, though if it has to be done in many places, adding the 47 * missing API to the strbuf module is the way to go. 48 * + 49 * WARNING: Do _not_ assume that the area that is yours is of size `alloc 50 * - 1` even if it's true in the current implementation. Alloc is somehow a 51 * "private" member that should not be messed with. Use `strbuf_avail()` 52 * instead. 53 */ 54 55/** 56 * Data Structures 57 * --------------- 58 */ 59 60/** 61 * This is the string buffer structure. The `len` member can be used to 62 * determine the current length of the string, and `buf` member provides 63 * access to the string itself. 64 */ 65struct strbuf { 66 size_t alloc; 67 size_t len; 68 char *buf; 69}; 70 71extern char strbuf_slopbuf[]; 72#define STRBUF_INIT { 0, 0, strbuf_slopbuf } 73 74/** 75 * Functions 76 * --------- 77 */ 78 79/** 80 * * Life Cycle 81 */ 82 83/** 84 * Initialize the structure. The second parameter can be zero or a bigger 85 * number to allocate memory, in case you want to prevent further reallocs. 86 */ 87extern void strbuf_init(struct strbuf *, size_t); 88 89/** 90 * Release a string buffer and the memory it used. You should not use the 91 * string buffer after using this function, unless you initialize it again. 92 */ 93extern void strbuf_release(struct strbuf *); 94 95/** 96 * Detach the string from the strbuf and returns it; you now own the 97 * storage the string occupies and it is your responsibility from then on 98 * to release it with `free(3)` when you are done with it. 99 */ 100extern char *strbuf_detach(struct strbuf *, size_t *); 101 102/** 103 * Attach a string to a buffer. You should specify the string to attach, 104 * the current length of the string and the amount of allocated memory. 105 * The amount must be larger than the string length, because the string you 106 * pass is supposed to be a NUL-terminated string. This string _must_ be 107 * malloc()ed, and after attaching, the pointer cannot be relied upon 108 * anymore, and neither be free()d directly. 109 */ 110extern void strbuf_attach(struct strbuf *, void *, size_t, size_t); 111 112/** 113 * Swap the contents of two string buffers. 114 */ 115static inline void strbuf_swap(struct strbuf *a, struct strbuf *b) 116{ 117 struct strbuf tmp = *a; 118 *a = *b; 119 *b = tmp; 120} 121 122 123/** 124 * * Related to the size of the buffer 125 */ 126 127/** 128 * Determine the amount of allocated but unused memory. 129 */ 130static inline size_t strbuf_avail(const struct strbuf *sb) 131{ 132 return sb->alloc ? sb->alloc - sb->len - 1 : 0; 133} 134 135/** 136 * Ensure that at least this amount of unused memory is available after 137 * `len`. This is used when you know a typical size for what you will add 138 * and want to avoid repetitive automatic resizing of the underlying buffer. 139 * This is never a needed operation, but can be critical for performance in 140 * some cases. 141 */ 142extern void strbuf_grow(struct strbuf *, size_t); 143 144/** 145 * Set the length of the buffer to a given value. This function does *not* 146 * allocate new memory, so you should not perform a `strbuf_setlen()` to a 147 * length that is larger than `len + strbuf_avail()`. `strbuf_setlen()` is 148 * just meant as a 'please fix invariants from this strbuf I just messed 149 * with'. 150 */ 151static inline void strbuf_setlen(struct strbuf *sb, size_t len) 152{ 153 if (len > (sb->alloc ? sb->alloc - 1 : 0)) 154 die("BUG: strbuf_setlen() beyond buffer"); 155 sb->len = len; 156 sb->buf[len] = '\0'; 157} 158 159/** 160 * Empty the buffer by setting the size of it to zero. 161 */ 162#define strbuf_reset(sb) strbuf_setlen(sb, 0) 163 164 165/** 166 * * Related to the contents of the buffer 167 */ 168 169/** 170 * Strip whitespace from the beginning and end of a string. 171 * Equivalent to performing `strbuf_rtrim()` followed by `strbuf_ltrim()`. 172 */ 173extern void strbuf_trim(struct strbuf *); 174 175/** 176 * Strip whitespace from the end of a string. 177 */ 178extern void strbuf_rtrim(struct strbuf *); 179 180/** 181 * Strip whitespace from the beginning of a string. 182 */ 183extern void strbuf_ltrim(struct strbuf *); 184 185/** 186 * Replace the contents of the strbuf with a reencoded form. Returns -1 187 * on error, 0 on success. 188 */ 189extern int strbuf_reencode(struct strbuf *sb, const char *from, const char *to); 190 191/** 192 * Lowercase each character in the buffer using `tolower`. 193 */ 194extern void strbuf_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 */ 201extern int strbuf_cmp(const struct strbuf *, const struct strbuf *); 202 203 204/** 205 * * Adding data to the buffer 206 * 207 * NOTE: All of the functions in this section will grow the buffer as 208 * necessary. If they fail for some reason other than memory shortage and the 209 * buffer hadn't been allocated before (i.e. the `struct strbuf` was set to 210 * `STRBUF_INIT`), then they will free() it. 211 */ 212 213/** 214 * Add a single character to the buffer. 215 */ 216static inline void strbuf_addch(struct strbuf *sb, int c) 217{ 218 strbuf_grow(sb, 1); 219 sb->buf[sb->len++] = c; 220 sb->buf[sb->len] = '\0'; 221} 222 223/** 224 * Add a character the specified number of times to the buffer. 225 */ 226extern void strbuf_addchars(struct strbuf *sb, int c, size_t n); 227 228/** 229 * Insert data to the given position of the buffer. The remaining contents 230 * will be shifted, not overwritten. 231 */ 232extern void strbuf_insert(struct strbuf *, size_t pos, const void *, size_t); 233 234/** 235 * Remove given amount of data from a given position of the buffer. 236 */ 237extern void strbuf_remove(struct strbuf *, size_t pos, size_t len); 238 239/** 240 * Remove the bytes between `pos..pos+len` and replace it with the given 241 * data. 242 */ 243extern void strbuf_splice(struct strbuf *, size_t pos, size_t len, 244 const void *, size_t); 245 246/** 247 * Add a NUL-terminated string to the buffer. Each line will be prepended 248 * by a comment character and a blank. 249 */ 250extern void strbuf_add_commented_lines(struct strbuf *out, const char *buf, size_t size); 251 252 253/** 254 * Add data of given length to the buffer. 255 */ 256extern void strbuf_add(struct strbuf *, const void *, size_t); 257 258/** 259 * Add a NUL-terminated string to the buffer. 260 * 261 * NOTE: This function will *always* be implemented as an inline or a macro 262 * using strlen, meaning that this is efficient to write things like: 263 * 264 * ---- 265 * strbuf_addstr(sb, "immediate string"); 266 * ---- 267 * 268 */ 269static inline void strbuf_addstr(struct strbuf *sb, const char *s) 270{ 271 strbuf_add(sb, s, strlen(s)); 272} 273 274/** 275 * Copy the contents of another buffer at the end of the current one. 276 */ 277static inline void strbuf_addbuf(struct strbuf *sb, const struct strbuf *sb2) 278{ 279 strbuf_grow(sb, sb2->len); 280 strbuf_add(sb, sb2->buf, sb2->len); 281} 282 283/** 284 * Copy part of the buffer from a given position till a given length to the 285 * end of the buffer. 286 */ 287extern void strbuf_adddup(struct strbuf *sb, size_t pos, size_t len); 288 289/** 290 * This function can be used to expand a format string containing 291 * placeholders. To that end, it parses the string and calls the specified 292 * function for every percent sign found. 293 * 294 * The callback function is given a pointer to the character after the `%` 295 * and a pointer to the struct strbuf. It is expected to add the expanded 296 * version of the placeholder to the strbuf, e.g. to add a newline 297 * character if the letter `n` appears after a `%`. The function returns 298 * the length of the placeholder recognized and `strbuf_expand()` skips 299 * over it. 300 * 301 * The format `%%` is automatically expanded to a single `%` as a quoting 302 * mechanism; callers do not need to handle the `%` placeholder themselves, 303 * and the callback function will not be invoked for this placeholder. 304 * 305 * All other characters (non-percent and not skipped ones) are copied 306 * verbatim to the strbuf. If the callback returned zero, meaning that the 307 * placeholder is unknown, then the percent sign is copied, too. 308 * 309 * In order to facilitate caching and to make it possible to give 310 * parameters to the callback, `strbuf_expand()` passes a context pointer, 311 * which can be used by the programmer of the callback as she sees fit. 312 */ 313typedef size_t (*expand_fn_t) (struct strbuf *sb, const char *placeholder, void *context); 314extern void strbuf_expand(struct strbuf *sb, const char *format, expand_fn_t fn, void *context); 315 316/** 317 * Used as callback for `strbuf_expand()`, expects an array of 318 * struct strbuf_expand_dict_entry as context, i.e. pairs of 319 * placeholder and replacement string. The array needs to be 320 * terminated by an entry with placeholder set to NULL. 321 */ 322struct strbuf_expand_dict_entry { 323 const char *placeholder; 324 const char *value; 325}; 326extern size_t strbuf_expand_dict_cb(struct strbuf *sb, const char *placeholder, void *context); 327 328/** 329 * Append the contents of one strbuf to another, quoting any 330 * percent signs ("%") into double-percents ("%%") in the 331 * destination. This is useful for literal data to be fed to either 332 * strbuf_expand or to the *printf family of functions. 333 */ 334extern void strbuf_addbuf_percentquote(struct strbuf *dst, const struct strbuf *src); 335 336/** 337 * Append the given byte size as a human-readable string (i.e. 12.23 KiB, 338 * 3.50 MiB). 339 */ 340extern void strbuf_humanise_bytes(struct strbuf *buf, off_t bytes); 341 342/** 343 * Add a formatted string to the buffer. 344 */ 345__attribute__((format (printf,2,3))) 346extern void strbuf_addf(struct strbuf *sb, const char *fmt, ...); 347 348/** 349 * Add a formatted string prepended by a comment character and a 350 * blank to the buffer. 351 */ 352__attribute__((format (printf, 2, 3))) 353extern void strbuf_commented_addf(struct strbuf *sb, const char *fmt, ...); 354 355__attribute__((format (printf,2,0))) 356extern void strbuf_vaddf(struct strbuf *sb, const char *fmt, va_list ap); 357 358/** 359 * Read a given size of data from a FILE* pointer to the buffer. 360 * 361 * NOTE: The buffer is rewound if the read fails. If -1 is returned, 362 * `errno` must be consulted, like you would do for `read(3)`. 363 * `strbuf_read()`, `strbuf_read_file()` and `strbuf_getline()` has the 364 * same behaviour as well. 365 */ 366extern size_t strbuf_fread(struct strbuf *, size_t, FILE *); 367 368/** 369 * Read the contents of a given file descriptor. The third argument can be 370 * used to give a hint about the file size, to avoid reallocs. If read fails, 371 * any partial read is undone. 372 */ 373extern ssize_t strbuf_read(struct strbuf *, int fd, size_t hint); 374 375/** 376 * Read the contents of a file, specified by its path. The third argument 377 * can be used to give a hint about the file size, to avoid reallocs. 378 */ 379extern int strbuf_read_file(struct strbuf *sb, const char *path, size_t hint); 380 381/** 382 * Read the target of a symbolic link, specified by its path. The third 383 * argument can be used to give a hint about the size, to avoid reallocs. 384 */ 385extern int strbuf_readlink(struct strbuf *sb, const char *path, size_t hint); 386 387/** 388 * Read a line from a FILE *, overwriting the existing contents 389 * of the strbuf. The second argument specifies the line 390 * terminator character, typically `'\n'`. 391 * Reading stops after the terminator or at EOF. The terminator 392 * is removed from the buffer before returning. Returns 0 unless 393 * there was nothing left before EOF, in which case it returns `EOF`. 394 */ 395extern int strbuf_getline(struct strbuf *, FILE *, int); 396 397/** 398 * Like `strbuf_getline`, but keeps the trailing terminator (if 399 * any) in the buffer. 400 */ 401extern int strbuf_getwholeline(struct strbuf *, FILE *, int); 402 403/** 404 * Like `strbuf_getwholeline`, but operates on a file descriptor. 405 * It reads one character at a time, so it is very slow. Do not 406 * use it unless you need the correct position in the file 407 * descriptor. 408 */ 409extern int strbuf_getwholeline_fd(struct strbuf *, int, int); 410 411/** 412 * Set the buffer to the path of the current working directory. 413 */ 414extern int strbuf_getcwd(struct strbuf *sb); 415 416/** 417 * Add a path to a buffer, converting a relative path to an 418 * absolute one in the process. Symbolic links are not 419 * resolved. 420 */ 421extern void strbuf_add_absolute_path(struct strbuf *sb, const char *path); 422 423/** 424 * Strip whitespace from a buffer. The second parameter controls if 425 * comments are considered contents to be removed or not. 426 */ 427extern void stripspace(struct strbuf *buf, int skip_comments); 428 429static inline int strbuf_strip_suffix(struct strbuf *sb, const char *suffix) 430{ 431 if (strip_suffix_mem(sb->buf, &sb->len, suffix)) { 432 strbuf_setlen(sb, sb->len); 433 return 1; 434 } else 435 return 0; 436} 437 438/** 439 * Split str (of length slen) at the specified terminator character. 440 * Return a null-terminated array of pointers to strbuf objects 441 * holding the substrings. The substrings include the terminator, 442 * except for the last substring, which might be unterminated if the 443 * original string did not end with a terminator. If max is positive, 444 * then split the string into at most max substrings (with the last 445 * substring containing everything following the (max-1)th terminator 446 * character). 447 * 448 * For lighter-weight alternatives, see string_list_split() and 449 * string_list_split_in_place(). 450 */ 451extern struct strbuf **strbuf_split_buf(const char *, size_t, 452 int terminator, int max); 453 454/** 455 * Split a NUL-terminated string at the specified terminator 456 * character. See strbuf_split_buf() for more information. 457 */ 458static inline struct strbuf **strbuf_split_str(const char *str, 459 int terminator, int max) 460{ 461 return strbuf_split_buf(str, strlen(str), terminator, max); 462} 463 464/** 465 * Split a strbuf at the specified terminator character. See 466 * strbuf_split_buf() for more information. 467 */ 468static inline struct strbuf **strbuf_split_max(const struct strbuf *sb, 469 int terminator, int max) 470{ 471 return strbuf_split_buf(sb->buf, sb->len, terminator, max); 472} 473 474/** 475 * Split a strbuf at the specified terminator character. See 476 * strbuf_split_buf() for more information. 477 */ 478static inline struct strbuf **strbuf_split(const struct strbuf *sb, 479 int terminator) 480{ 481 return strbuf_split_max(sb, terminator, 0); 482} 483 484/** 485 * Free a NULL-terminated list of strbufs (for example, the return 486 * values of the strbuf_split*() functions). 487 */ 488extern void strbuf_list_free(struct strbuf **); 489 490/** 491 * Launch the user preferred editor to edit a file and fill the buffer 492 * with the file's contents upon the user completing their editing. The 493 * third argument can be used to set the environment which the editor is 494 * run in. If the buffer is NULL the editor is launched as usual but the 495 * file's contents are not read into the buffer upon completion. 496 */ 497extern int launch_editor(const char *path, struct strbuf *buffer, const char *const *env); 498 499extern void strbuf_add_lines(struct strbuf *sb, const char *prefix, const char *buf, size_t size); 500 501/** 502 * Append s to sb, with the characters '<', '>', '&' and '"' converted 503 * into XML entities. 504 */ 505extern void strbuf_addstr_xml_quoted(struct strbuf *sb, const char *s); 506 507static inline void strbuf_complete_line(struct strbuf *sb) 508{ 509 if (sb->len && sb->buf[sb->len - 1] != '\n') 510 strbuf_addch(sb, '\n'); 511} 512 513extern int strbuf_branchname(struct strbuf *sb, const char *name); 514extern int strbuf_check_branch_ref(struct strbuf *sb, const char *name); 515 516extern void strbuf_addstr_urlencode(struct strbuf *, const char *, 517 int reserved); 518 519__attribute__((format (printf,1,2))) 520extern int printf_ln(const char *fmt, ...); 521__attribute__((format (printf,2,3))) 522extern int fprintf_ln(FILE *fp, const char *fmt, ...); 523 524char *xstrdup_tolower(const char *); 525 526/** 527 * Create a newly allocated string using printf format. You can do this easily 528 * with a strbuf, but this provides a shortcut to save a few lines. 529 */ 530__attribute__((format (printf, 1, 0))) 531char *xstrvfmt(const char *fmt, va_list ap); 532__attribute__((format (printf, 1, 2))) 533char *xstrfmt(const char *fmt, ...); 534 535#endif /* STRBUF_H */