1 /* $NetBSD: ascmagic.c,v 1.1.1.11 2019/05/22 17:19:56 christos Exp $ */ 2 3 /* 4 * Copyright (c) Ian F. Darwin 1986-1995. 5 * Software written by Ian F. Darwin and others; 6 * maintained 1995-present by Christos Zoulas and others. 7 * 8 * Redistribution and use in source and binary forms, with or without 9 * modification, are permitted provided that the following conditions 10 * are met: 11 * 1. Redistributions of source code must retain the above copyright 12 * notice immediately at the beginning of the file, without modification, 13 * this list of conditions, and the following disclaimer. 14 * 2. Redistributions in binary form must reproduce the above copyright 15 * notice, this list of conditions and the following disclaimer in the 16 * documentation and/or other materials provided with the distribution. 17 * 18 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 19 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 20 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 21 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR 22 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 23 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 24 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 25 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 26 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 27 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 28 * SUCH DAMAGE. 29 */ 30 /* 31 * ASCII magic -- try to detect text encoding. 32 * 33 * Extensively modified by Eric Fischer <enf@pobox.com> in July, 2000, 34 * to handle character codes other than ASCII on a unified basis. 35 */ 36 37 #include "file.h" 38 39 #ifndef lint 40 #if 0 41 FILE_RCSID("@(#)$File: ascmagic.c,v 1.104 2019/05/07 02:27:11 christos Exp $") 42 #else 43 __RCSID("$NetBSD: ascmagic.c,v 1.1.1.11 2019/05/22 17:19:56 christos Exp $"); 44 #endif 45 #endif /* lint */ 46 47 #include "magic.h" 48 #include <string.h> 49 #include <memory.h> 50 #include <ctype.h> 51 #include <stdlib.h> 52 #ifdef HAVE_UNISTD_H 53 #include <unistd.h> 54 #endif 55 56 #define MAXLINELEN 300 /* longest sane line length */ 57 #define ISSPC(x) ((x) == ' ' || (x) == '\t' || (x) == '\r' || (x) == '\n' \ 58 || (x) == 0x85 || (x) == '\f') 59 60 private unsigned char *encode_utf8(unsigned char *, size_t, unichar *, size_t); 61 private size_t trim_nuls(const unsigned char *, size_t); 62 63 /* 64 * Undo the NUL-termination kindly provided by process() 65 * but leave at least one byte to look at 66 */ 67 private size_t 68 trim_nuls(const unsigned char *buf, size_t nbytes) 69 { 70 while (nbytes > 1 && buf[nbytes - 1] == '\0') 71 nbytes--; 72 73 return nbytes; 74 } 75 76 protected int 77 file_ascmagic(struct magic_set *ms, const struct buffer *b, int text) 78 { 79 unichar *ubuf = NULL; 80 size_t ulen = 0; 81 int rv = 1; 82 struct buffer bb; 83 84 const char *code = NULL; 85 const char *code_mime = NULL; 86 const char *type = NULL; 87 88 bb = *b; 89 bb.flen = trim_nuls(CAST(const unsigned char *, b->fbuf), b->flen); 90 /* 91 * Avoid trimming at an odd byte if the original buffer was evenly 92 * sized; this avoids losing the last character on UTF-16 LE text 93 */ 94 if ((bb.flen & 1) && !(b->flen & 1)) 95 bb.flen++; 96 97 /* If file doesn't look like any sort of text, give up. */ 98 if (file_encoding(ms, &bb, &ubuf, &ulen, &code, &code_mime, 99 &type) == 0) 100 rv = 0; 101 else 102 rv = file_ascmagic_with_encoding(ms, &bb, 103 ubuf, ulen, code, type, text); 104 105 free(ubuf); 106 107 return rv; 108 } 109 110 protected int 111 file_ascmagic_with_encoding(struct magic_set *ms, 112 const struct buffer *b, unichar *ubuf, size_t ulen, const char *code, 113 const char *type, int text) 114 { 115 struct buffer bb; 116 const unsigned char *buf = CAST(const unsigned char *, b->fbuf); 117 size_t nbytes = b->flen; 118 unsigned char *utf8_buf = NULL, *utf8_end; 119 size_t mlen, i, len; 120 int rv = -1; 121 int mime = ms->flags & MAGIC_MIME; 122 int need_separator = 0; 123 124 const char *subtype = NULL; 125 const char *subtype_mime = NULL; 126 127 int has_escapes = 0; 128 int has_backspace = 0; 129 int seen_cr = 0; 130 131 int n_crlf = 0; 132 int n_lf = 0; 133 int n_cr = 0; 134 int n_nel = 0; 135 int executable = 0; 136 137 size_t last_line_end = CAST(size_t, -1); 138 int has_long_lines = 0; 139 140 nbytes = trim_nuls(buf, nbytes); 141 142 /* If we have fewer than 2 bytes, give up. */ 143 if (nbytes <= 1) { 144 rv = 0; 145 goto done; 146 } 147 148 if (ulen > 0 && (ms->flags & MAGIC_NO_CHECK_SOFT) == 0) { 149 /* Convert ubuf to UTF-8 and try text soft magic */ 150 /* malloc size is a conservative overestimate; could be 151 improved, or at least realloced after conversion. */ 152 mlen = ulen * 6; 153 if ((utf8_buf = CAST(unsigned char *, malloc(mlen))) == NULL) { 154 file_oomem(ms, mlen); 155 goto done; 156 } 157 if ((utf8_end = encode_utf8(utf8_buf, mlen, ubuf, ulen)) 158 == NULL) 159 goto done; 160 buffer_init(&bb, b->fd, &b->st, utf8_buf, 161 CAST(size_t, utf8_end - utf8_buf)); 162 163 if ((rv = file_softmagic(ms, &bb, NULL, NULL, 164 TEXTTEST, text)) == 0) 165 rv = -1; 166 else 167 need_separator = 1; 168 buffer_fini(&bb); 169 if ((ms->flags & (MAGIC_APPLE|MAGIC_EXTENSION))) { 170 rv = rv == -1 ? 0 : 1; 171 goto done; 172 } 173 } 174 if ((ms->flags & (MAGIC_APPLE|MAGIC_EXTENSION))) 175 return 0; 176 177 /* Now try to discover other details about the file. */ 178 for (i = 0; i < ulen; i++) { 179 if (ubuf[i] == '\n') { 180 if (seen_cr) 181 n_crlf++; 182 else 183 n_lf++; 184 last_line_end = i; 185 } else if (seen_cr) 186 n_cr++; 187 188 seen_cr = (ubuf[i] == '\r'); 189 if (seen_cr) 190 last_line_end = i; 191 192 if (ubuf[i] == 0x85) { /* X3.64/ECMA-43 "next line" character */ 193 n_nel++; 194 last_line_end = i; 195 } 196 197 /* If this line is _longer_ than MAXLINELEN, remember it. */ 198 if (i > last_line_end + MAXLINELEN) 199 has_long_lines = 1; 200 201 if (ubuf[i] == '\033') 202 has_escapes = 1; 203 if (ubuf[i] == '\b') 204 has_backspace = 1; 205 } 206 207 /* Beware, if the data has been truncated, the final CR could have 208 been followed by a LF. If we have ms->bytes_max bytes, it indicates 209 that the data might have been truncated, probably even before 210 this function was called. */ 211 if (seen_cr && nbytes < ms->bytes_max) 212 n_cr++; 213 214 if (strcmp(type, "binary") == 0) { 215 rv = 0; 216 goto done; 217 } 218 len = file_printedlen(ms); 219 if (mime) { 220 if ((mime & MAGIC_MIME_TYPE) != 0) { 221 if (len) { 222 /* 223 * Softmagic printed something, we 224 * are either done, or we need a separator 225 */ 226 if ((ms->flags & MAGIC_CONTINUE) == 0) { 227 rv = 1; 228 goto done; 229 } 230 if (need_separator && file_separator(ms) == -1) 231 goto done; 232 } 233 if (subtype_mime) { 234 if (file_printf(ms, "%s", subtype_mime) == -1) 235 goto done; 236 } else { 237 if (file_printf(ms, "text/plain") == -1) 238 goto done; 239 } 240 } 241 } else { 242 if (len) { 243 switch (file_replace(ms, " text$", ", ")) { 244 case 0: 245 switch (file_replace(ms, " text executable$", 246 ", ")) { 247 case 0: 248 if (file_printf(ms, ", ") == -1) 249 goto done; 250 break; 251 case -1: 252 goto done; 253 default: 254 executable = 1; 255 break; 256 } 257 break; 258 case -1: 259 goto done; 260 default: 261 break; 262 } 263 } 264 265 if (file_printf(ms, "%s", code) == -1) 266 goto done; 267 268 if (subtype) { 269 if (file_printf(ms, " %s", subtype) == -1) 270 goto done; 271 } 272 273 if (file_printf(ms, " %s", type) == -1) 274 goto done; 275 276 if (executable) 277 if (file_printf(ms, " executable") == -1) 278 goto done; 279 280 if (has_long_lines) 281 if (file_printf(ms, ", with very long lines") == -1) 282 goto done; 283 284 /* 285 * Only report line terminators if we find one other than LF, 286 * or if we find none at all. 287 */ 288 if ((n_crlf == 0 && n_cr == 0 && n_nel == 0 && n_lf == 0) || 289 (n_crlf != 0 || n_cr != 0 || n_nel != 0)) { 290 if (file_printf(ms, ", with") == -1) 291 goto done; 292 293 if (n_crlf == 0 && n_cr == 0 && n_nel == 0 && n_lf == 0) { 294 if (file_printf(ms, " no") == -1) 295 goto done; 296 } else { 297 if (n_crlf) { 298 if (file_printf(ms, " CRLF") == -1) 299 goto done; 300 if (n_cr || n_lf || n_nel) 301 if (file_printf(ms, ",") == -1) 302 goto done; 303 } 304 if (n_cr) { 305 if (file_printf(ms, " CR") == -1) 306 goto done; 307 if (n_lf || n_nel) 308 if (file_printf(ms, ",") == -1) 309 goto done; 310 } 311 if (n_lf) { 312 if (file_printf(ms, " LF") == -1) 313 goto done; 314 if (n_nel) 315 if (file_printf(ms, ",") == -1) 316 goto done; 317 } 318 if (n_nel) 319 if (file_printf(ms, " NEL") == -1) 320 goto done; 321 } 322 323 if (file_printf(ms, " line terminators") == -1) 324 goto done; 325 } 326 327 if (has_escapes) 328 if (file_printf(ms, ", with escape sequences") == -1) 329 goto done; 330 if (has_backspace) 331 if (file_printf(ms, ", with overstriking") == -1) 332 goto done; 333 } 334 rv = 1; 335 done: 336 free(utf8_buf); 337 338 return rv; 339 } 340 341 /* 342 * Encode Unicode string as UTF-8, returning pointer to character 343 * after end of string, or NULL if an invalid character is found. 344 */ 345 private unsigned char * 346 encode_utf8(unsigned char *buf, size_t len, unichar *ubuf, size_t ulen) 347 { 348 size_t i; 349 unsigned char *end = buf + len; 350 351 for (i = 0; i < ulen; i++) { 352 if (ubuf[i] <= 0x7f) { 353 if (end - buf < 1) 354 return NULL; 355 *buf++ = CAST(unsigned char, ubuf[i]); 356 } else if (ubuf[i] <= 0x7ff) { 357 if (end - buf < 2) 358 return NULL; 359 *buf++ = CAST(unsigned char, (ubuf[i] >> 6) + 0xc0); 360 *buf++ = CAST(unsigned char, (ubuf[i] & 0x3f) + 0x80); 361 } else if (ubuf[i] <= 0xffff) { 362 if (end - buf < 3) 363 return NULL; 364 *buf++ = CAST(unsigned char, (ubuf[i] >> 12) + 0xe0); 365 *buf++ = CAST(unsigned char, ((ubuf[i] >> 6) & 0x3f) + 0x80); 366 *buf++ = CAST(unsigned char, (ubuf[i] & 0x3f) + 0x80); 367 } else if (ubuf[i] <= 0x1fffff) { 368 if (end - buf < 4) 369 return NULL; 370 *buf++ = CAST(unsigned char, (ubuf[i] >> 18) + 0xf0); 371 *buf++ = CAST(unsigned char, ((ubuf[i] >> 12) & 0x3f) + 0x80); 372 *buf++ = CAST(unsigned char, ((ubuf[i] >> 6) & 0x3f) + 0x80); 373 *buf++ = CAST(unsigned char, (ubuf[i] & 0x3f) + 0x80); 374 } else if (ubuf[i] <= 0x3ffffff) { 375 if (end - buf < 5) 376 return NULL; 377 *buf++ = CAST(unsigned char, (ubuf[i] >> 24) + 0xf8); 378 *buf++ = CAST(unsigned char, ((ubuf[i] >> 18) & 0x3f) + 0x80); 379 *buf++ = CAST(unsigned char, ((ubuf[i] >> 12) & 0x3f) + 0x80); 380 *buf++ = CAST(unsigned char, ((ubuf[i] >> 6) & 0x3f) + 0x80); 381 *buf++ = CAST(unsigned char, (ubuf[i] & 0x3f) + 0x80); 382 } else if (ubuf[i] <= 0x7fffffff) { 383 if (end - buf < 6) 384 return NULL; 385 *buf++ = CAST(unsigned char, (ubuf[i] >> 30) + 0xfc); 386 *buf++ = CAST(unsigned char, ((ubuf[i] >> 24) & 0x3f) + 0x80); 387 *buf++ = CAST(unsigned char, ((ubuf[i] >> 18) & 0x3f) + 0x80); 388 *buf++ = CAST(unsigned char, ((ubuf[i] >> 12) & 0x3f) + 0x80); 389 *buf++ = CAST(unsigned char, ((ubuf[i] >> 6) & 0x3f) + 0x80); 390 *buf++ = CAST(unsigned char, (ubuf[i] & 0x3f) + 0x80); 391 } else /* Invalid character */ 392 return NULL; 393 } 394 395 return buf; 396 } 397