| Jari Aalto | ccc6cda | 1996-12-23 17:02:34 +0000 | [diff] [blame^] | 1 | /* stringlib.c - Miscellaneous string functions. */ |
| 2 | |
| 3 | /* Copyright (C) 1996 |
| 4 | Free Software Foundation, Inc. |
| 5 | |
| 6 | This file is part of GNU Bash, the Bourne Again SHell. |
| 7 | |
| 8 | Bash is free software; you can redistribute it and/or modify it under |
| 9 | the terms of the GNU General Public License as published by the Free |
| 10 | Software Foundation; either version 2, or (at your option) any later |
| 11 | version. |
| 12 | |
| 13 | Bash is distributed in the hope that it will be useful, but WITHOUT ANY |
| 14 | WARRANTY; without even the implied warranty of MERCHANTABILITY or |
| 15 | FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
| 16 | for more details. |
| 17 | |
| 18 | You should have received a copy of the GNU General Public License along |
| 19 | with Bash; see the file COPYING. If not, write to the Free Software |
| 20 | Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. */ |
| 21 | |
| 22 | #include "config.h" |
| 23 | |
| 24 | #include "bashtypes.h" |
| 25 | |
| 26 | #if defined (HAVE_UNISTD_H) |
| 27 | # include <unistd.h> |
| 28 | #endif |
| 29 | |
| 30 | #include "bashansi.h" |
| 31 | #include <stdio.h> |
| 32 | #include <ctype.h> |
| 33 | |
| 34 | #include "shell.h" |
| 35 | |
| 36 | #ifndef to_upper |
| 37 | # define to_upper(c) (islower(c) ? toupper(c) : (c)) |
| 38 | # define to_lower(c) (isupper(c) ? tolower(c) : (c)) |
| 39 | #endif |
| 40 | |
| 41 | /* Convert STRING by expanding the escape sequences specified by the |
| 42 | ANSI C standard. If SAWC is non-null, recognize `\c' and use that |
| 43 | as a string terminator. If we see \c, set *SAWC to 1 before |
| 44 | returning. LEN is the length of STRING. */ |
| 45 | char * |
| 46 | ansicstr (string, len, sawc) |
| 47 | char *string; |
| 48 | int len, *sawc; |
| 49 | { |
| 50 | int c; |
| 51 | char *ret, *r, *s; |
| 52 | |
| 53 | if (string == 0 || *string == '\0') |
| 54 | return ((char *)NULL); |
| 55 | |
| 56 | ret = xmalloc (len + 1); |
| 57 | for (r = ret, s = string; s && *s; ) |
| 58 | { |
| 59 | c = *s++; |
| 60 | if (c != '\\' || *s == '\0') |
| 61 | *r++ = c; |
| 62 | else |
| 63 | { |
| 64 | switch (c = *s++) |
| 65 | { |
| 66 | #if defined (__STDC__) |
| 67 | case 'a': c = '\a'; break; |
| 68 | case 'v': c = '\v'; break; |
| 69 | #else |
| 70 | case 'a': c = '\007'; break; |
| 71 | case 'v': c = (int) 0x0B; break; |
| 72 | #endif |
| 73 | case 'b': c = '\b'; break; |
| 74 | case 'e': c = '\033'; break; /* ESC -- non-ANSI */ |
| 75 | case 'E': c = '\033'; break; /* ESC -- non-ANSI */ |
| 76 | case 'f': c = '\f'; break; |
| 77 | case 'n': c = '\n'; break; |
| 78 | case 'r': c = '\r'; break; |
| 79 | case 't': c = '\t'; break; |
| 80 | case '0': case '1': case '2': case '3': |
| 81 | case '4': case '5': case '6': case '7': |
| 82 | c -= '0'; |
| 83 | if (*s >= '0' && *s <= '7') |
| 84 | c = c * 8 + (*s++ - '0'); |
| 85 | if (*s >= '0' && *s <= '7') |
| 86 | c = c * 8 + (*s++ - '0'); |
| 87 | break; |
| 88 | case '\\': |
| 89 | case '\'': |
| 90 | break; |
| 91 | case 'c': |
| 92 | if (sawc) |
| 93 | { |
| 94 | *sawc = 1; |
| 95 | *r = '\0'; |
| 96 | return ret; |
| 97 | } |
| 98 | default: *r++ = '\\'; break; |
| 99 | } |
| 100 | *r++ = c; |
| 101 | } |
| 102 | } |
| 103 | *r = '\0'; |
| 104 | return ret; |
| 105 | } |
| 106 | |
| 107 | /* **************************************************************** */ |
| 108 | /* */ |
| 109 | /* Functions to manage arrays of strings */ |
| 110 | /* */ |
| 111 | /* **************************************************************** */ |
| 112 | |
| 113 | /* Find NAME in ARRAY. Return the index of NAME, or -1 if not present. |
| 114 | ARRAY should be NULL terminated. */ |
| 115 | int |
| 116 | find_name_in_array (name, array) |
| 117 | char *name, **array; |
| 118 | { |
| 119 | int i; |
| 120 | |
| 121 | for (i = 0; array[i]; i++) |
| 122 | if (STREQ (name, array[i])) |
| 123 | return (i); |
| 124 | |
| 125 | return (-1); |
| 126 | } |
| 127 | |
| 128 | /* Return the length of ARRAY, a NULL terminated array of char *. */ |
| 129 | int |
| 130 | array_len (array) |
| 131 | char **array; |
| 132 | { |
| 133 | register int i; |
| 134 | |
| 135 | for (i = 0; array[i]; i++); |
| 136 | return (i); |
| 137 | } |
| 138 | |
| 139 | /* Free the contents of ARRAY, a NULL terminated array of char *. */ |
| 140 | void |
| 141 | free_array_members (array) |
| 142 | char **array; |
| 143 | { |
| 144 | register int i; |
| 145 | |
| 146 | if (array == 0) |
| 147 | return; |
| 148 | |
| 149 | for (i = 0; array[i]; i++) |
| 150 | free (array[i]); |
| 151 | } |
| 152 | |
| 153 | void |
| 154 | free_array (array) |
| 155 | char **array; |
| 156 | { |
| 157 | if (array == 0) |
| 158 | return; |
| 159 | |
| 160 | free_array_members (array); |
| 161 | free (array); |
| 162 | } |
| 163 | |
| 164 | /* Allocate and return a new copy of ARRAY and its contents. */ |
| 165 | char ** |
| 166 | copy_array (array) |
| 167 | char **array; |
| 168 | { |
| 169 | register int i; |
| 170 | int len; |
| 171 | char **new_array; |
| 172 | |
| 173 | len = array_len (array); |
| 174 | |
| 175 | new_array = (char **)xmalloc ((len + 1) * sizeof (char *)); |
| 176 | for (i = 0; array[i]; i++) |
| 177 | new_array[i] = savestring (array[i]); |
| 178 | new_array[i] = (char *)NULL; |
| 179 | |
| 180 | return (new_array); |
| 181 | } |
| 182 | |
| 183 | /* Comparison routine for use with qsort() on arrays of strings. Uses |
| 184 | strcoll(3) if available, otherwise it uses strcmp(3). */ |
| 185 | int |
| 186 | qsort_string_compare (s1, s2) |
| 187 | register char **s1, **s2; |
| 188 | { |
| 189 | #if defined (HAVE_STRCOLL) |
| 190 | return (strcoll (*s1, *s2)); |
| 191 | #else /* !HAVE_STRCOLL */ |
| 192 | int result; |
| 193 | |
| 194 | if ((result = **s1 - **s2) == 0) |
| 195 | result = strcmp (*s1, *s2); |
| 196 | |
| 197 | return (result); |
| 198 | #endif /* !HAVE_STRCOLL */ |
| 199 | } |
| 200 | |
| 201 | /* Sort ARRAY, a null terminated array of pointers to strings. */ |
| 202 | void |
| 203 | sort_char_array (array) |
| 204 | char **array; |
| 205 | { |
| 206 | qsort (array, array_len (array), sizeof (char *), |
| 207 | (Function *)qsort_string_compare); |
| 208 | } |
| 209 | |
| 210 | /* Cons up a new array of words. The words are taken from LIST, |
| 211 | which is a WORD_LIST *. If COPY is true, everything is malloc'ed, |
| 212 | so you should free everything in this array when you are done. |
| 213 | The array is NULL terminated. If IP is non-null, it gets the |
| 214 | number of words in the returned array. STARTING_INDEX says where |
| 215 | to start filling in the returned array; it can be used to reserve |
| 216 | space at the beginning of the array. */ |
| 217 | char ** |
| 218 | word_list_to_argv (list, copy, starting_index, ip) |
| 219 | WORD_LIST *list; |
| 220 | int copy, starting_index, *ip; |
| 221 | { |
| 222 | int count; |
| 223 | char **array; |
| 224 | |
| 225 | count = list_length (list); |
| 226 | array = (char **)xmalloc ((1 + count + starting_index) * sizeof (char *)); |
| 227 | |
| 228 | for (count = 0; count < starting_index; count++) |
| 229 | array[count] = (char *)NULL; |
| 230 | for (count = starting_index; list; count++, list = list->next) |
| 231 | array[count] = copy ? savestring (list->word->word) : list->word->word; |
| 232 | array[count] = (char *)NULL; |
| 233 | |
| 234 | if (ip) |
| 235 | *ip = count; |
| 236 | return (array); |
| 237 | } |
| 238 | |
| 239 | /* Convert an array of strings into the form used internally by the shell. |
| 240 | COPY means to copy the values in ARRAY into the returned list rather |
| 241 | than allocate new storage. STARTING_INDEX says where in ARRAY to begin. */ |
| 242 | WORD_LIST * |
| 243 | argv_to_word_list (array, copy, starting_index) |
| 244 | char **array; |
| 245 | int copy, starting_index; |
| 246 | { |
| 247 | WORD_LIST *list; |
| 248 | WORD_DESC *w; |
| 249 | int i, count; |
| 250 | |
| 251 | if (array == 0 || array[0] == 0) |
| 252 | return (WORD_LIST *)NULL; |
| 253 | |
| 254 | for (count = 0; array[count]; count++) |
| 255 | ; |
| 256 | |
| 257 | for (i = starting_index, list = (WORD_LIST *)NULL; i < count; i++) |
| 258 | { |
| 259 | w = make_bare_word (copy ? "" : array[i]); |
| 260 | if (copy) |
| 261 | { |
| 262 | free (w->word); |
| 263 | w->word = array[i]; |
| 264 | } |
| 265 | list = make_word_list (w, list); |
| 266 | } |
| 267 | return (REVERSE_LIST(list, WORD_LIST *)); |
| 268 | } |
| 269 | |
| 270 | /* **************************************************************** */ |
| 271 | /* */ |
| 272 | /* String Management Functions */ |
| 273 | /* */ |
| 274 | /* **************************************************************** */ |
| 275 | |
| 276 | /* Replace occurrences of PAT with REP in STRING. If GLOBAL is non-zero, |
| 277 | replace all occurrences, otherwise replace only the first. |
| 278 | This returns a new string; the caller should free it. */ |
| 279 | char * |
| 280 | strsub (string, pat, rep, global) |
| 281 | char *string, *pat, *rep; |
| 282 | int global; |
| 283 | { |
| 284 | int patlen, templen, tempsize, repl, i; |
| 285 | char *temp, *r; |
| 286 | |
| 287 | patlen = strlen (pat); |
| 288 | for (temp = (char *)NULL, i = templen = tempsize = 0, repl = 1; string[i]; ) |
| 289 | { |
| 290 | if (repl && STREQN (string + i, pat, patlen)) |
| 291 | { |
| 292 | RESIZE_MALLOCED_BUFFER (temp, templen, patlen, tempsize, (patlen * 2)); |
| 293 | |
| 294 | for (r = rep; *r; ) |
| 295 | temp[templen++] = *r++; |
| 296 | |
| 297 | i += patlen; |
| 298 | repl = global != 0; |
| 299 | } |
| 300 | else |
| 301 | { |
| 302 | RESIZE_MALLOCED_BUFFER (temp, templen, 1, tempsize, 16); |
| 303 | temp[templen++] = string[i++]; |
| 304 | } |
| 305 | } |
| 306 | temp[templen] = 0; |
| 307 | return (temp); |
| 308 | } |
| 309 | |
| 310 | /* Remove all leading whitespace from STRING. This includes |
| 311 | newlines. STRING should be terminated with a zero. */ |
| 312 | void |
| 313 | strip_leading (string) |
| 314 | char *string; |
| 315 | { |
| 316 | char *start = string; |
| 317 | |
| 318 | while (*string && (whitespace (*string) || *string == '\n')) |
| 319 | string++; |
| 320 | |
| 321 | if (string != start) |
| 322 | { |
| 323 | int len = strlen (string); |
| 324 | FASTCOPY (string, start, len); |
| 325 | start[len] = '\0'; |
| 326 | } |
| 327 | } |
| 328 | |
| 329 | /* Remove all trailing whitespace from STRING. This includes |
| 330 | newlines. If NEWLINES_ONLY is non-zero, only trailing newlines |
| 331 | are removed. STRING should be terminated with a zero. */ |
| 332 | void |
| 333 | strip_trailing (string, newlines_only) |
| 334 | char *string; |
| 335 | int newlines_only; |
| 336 | { |
| 337 | int len = strlen (string) - 1; |
| 338 | |
| 339 | while (len >= 0) |
| 340 | { |
| 341 | if ((newlines_only && string[len] == '\n') || |
| 342 | (!newlines_only && whitespace (string[len]))) |
| 343 | len--; |
| 344 | else |
| 345 | break; |
| 346 | } |
| 347 | string[len + 1] = '\0'; |
| 348 | } |
| 349 | |
| 350 | /* Determine if s2 occurs in s1. If so, return a pointer to the |
| 351 | match in s1. The compare is case insensitive. This is a |
| 352 | case-insensitive strstr(3). */ |
| 353 | char * |
| 354 | strindex (s1, s2) |
| 355 | char *s1, *s2; |
| 356 | { |
| 357 | register int i, l, len, c; |
| 358 | |
| 359 | c = to_upper (s2[0]); |
| 360 | for (i = 0, len = strlen (s1), l = strlen (s2); (len - i) >= l; i++) |
| 361 | if ((to_upper (s1[i]) == c) && (strncasecmp (s1 + i, s2, l) == 0)) |
| 362 | return (s1 + i); |
| 363 | return ((char *)NULL); |
| 364 | } |
| 365 | |
| 366 | /* A wrapper for bcopy that can be prototyped in general.h */ |
| 367 | void |
| 368 | xbcopy (s, d, n) |
| 369 | char *s, *d; |
| 370 | int n; |
| 371 | { |
| 372 | FASTCOPY (s, d, n); |
| 373 | } |