e375fe33 |
1 | /* -*-c-*- |
2 | * |
b817bfc6 |
3 | * $Id: hashsum.c,v 1.10 2004/04/08 01:36:15 mdw Exp $ |
e375fe33 |
4 | * |
5 | * Hash files using some secure hash function |
6 | * |
7 | * (c) 2000 Straylight/Edgeware |
8 | */ |
9 | |
10 | /*----- Licensing notice --------------------------------------------------* |
11 | * |
12 | * This file is part of Catacomb. |
13 | * |
14 | * Catacomb is free software; you can redistribute it and/or modify |
15 | * it under the terms of the GNU Library General Public License as |
16 | * published by the Free Software Foundation; either version 2 of the |
17 | * License, or (at your option) any later version. |
18 | * |
19 | * Catacomb is distributed in the hope that it will be useful, |
20 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
21 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
22 | * GNU Library General Public License for more details. |
23 | * |
24 | * You should have received a copy of the GNU Library General Public |
25 | * License along with Catacomb; if not, write to the Free |
26 | * Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, |
27 | * MA 02111-1307, USA. |
28 | */ |
29 | |
e375fe33 |
30 | /*----- Header files ------------------------------------------------------*/ |
31 | |
32 | #include "config.h" |
33 | |
34 | #include <ctype.h> |
35 | #include <errno.h> |
36 | #include <stdio.h> |
37 | #include <stdlib.h> |
38 | #include <string.h> |
39 | |
40 | #include <mLib/alloc.h> |
41 | #include <mLib/dstr.h> |
42 | #include <mLib/mdwopt.h> |
43 | #include <mLib/quis.h> |
44 | #include <mLib/report.h> |
45 | #include <mLib/sub.h> |
46 | #include <mLib/str.h> |
47 | |
48 | #include "ghash.h" |
49 | |
e375fe33 |
50 | /*----- Static variables --------------------------------------------------*/ |
51 | |
16efd15b |
52 | #define f_binary 1u |
53 | #define f_bogus 2u |
54 | #define f_verbose 4u |
55 | #define f_check 8u |
56 | #define f_files 16u |
57 | #define f_raw 32u |
58 | #define f_oddhash 64u |
59 | #define f_escape 128u |
e375fe33 |
60 | |
61 | /*----- Support functions -------------------------------------------------*/ |
62 | |
63 | /* --- @fhash@ --- * |
64 | * |
65 | * Arguments: @const char *file@ = file name to be hashed (null for stdin) |
66 | * @unsigned f@ = flags to set |
67 | * @const gchash *gch@ = pointer to hash function to use |
68 | * @void *buf@ = pointer to hash output buffer |
69 | * |
70 | * Returns: Zero if it worked, nonzero on error. |
71 | * |
72 | * Use: Hashes a file. |
73 | */ |
74 | |
75 | static int fhash(const char *file, unsigned f, const gchash *gch, void *buf) |
76 | { |
77 | FILE *fp; |
78 | char fbuf[BUFSIZ]; |
79 | size_t sz; |
80 | ghash *h; |
81 | int e; |
82 | |
83 | if (!file) |
84 | fp = stdin; |
85 | else if ((fp = fopen(file, f & f_binary ? "rb" : "r")) == 0) |
86 | return (-1); |
87 | |
b817bfc6 |
88 | h = GH_INIT(gch); |
e375fe33 |
89 | while ((sz = fread(fbuf, 1, sizeof(fbuf), fp)) > 0) |
b817bfc6 |
90 | GH_HASH(h, fbuf, sz); |
91 | GH_DONE(h, buf); |
92 | GH_DESTROY(h); |
e375fe33 |
93 | e = ferror(fp); |
94 | if (file) |
95 | fclose(fp); |
96 | return (e ? -1 : 0); |
97 | } |
98 | |
99 | /* --- @puthex@ --- * |
100 | * |
101 | * Arguments: @const octet *buf@ = pointer to a binary buffer |
102 | * @size_t sz@ = size of the buffer |
103 | * @FILE *fp@ = pointer to output file handle |
104 | * |
105 | * Returns: --- |
106 | * |
107 | * Use: Writes a hex dump of a block of memory. |
108 | */ |
109 | |
110 | static void puthex(const octet *buf, size_t sz, FILE *fp) |
111 | { |
112 | while (sz) { |
113 | fprintf(fp, "%02x", *buf++); |
114 | sz--; |
115 | } |
116 | } |
117 | |
118 | /* --- @gethex@ --- * |
119 | * |
120 | * Arguments: @const char *p@ = pointer to input string |
121 | * @octet *q@ = pointer to output buffer |
122 | * @size_t sz@ = size of the output buffer |
123 | * @char **pp@ = where to put the end pointer |
124 | * |
125 | * Returns: The number of bytes written to the buffer. |
126 | * |
127 | * Use: Reads hex dumps from the input string. |
128 | */ |
129 | |
130 | static size_t gethex(const char *p, octet *q, size_t sz, char **pp) |
131 | { |
132 | size_t i = 0; |
133 | while (sz > 0 && |
134 | isxdigit((unsigned char)p[0]) && |
135 | isxdigit((unsigned char)p[1])) { |
136 | char buf[3]; |
137 | buf[0] = p[0]; |
138 | buf[1] = p[1]; |
139 | buf[2] = 0; |
140 | *q++ = strtoul(buf, 0, 16); |
141 | sz--; |
142 | p += 2; |
143 | i++; |
144 | } |
145 | if (pp) |
146 | *pp = (char *)p; |
147 | return (i); |
148 | } |
149 | |
150 | /* --- @gethash@ --- * |
151 | * |
152 | * Arguments: @const char *name@ = pointer to name string |
153 | * |
154 | * Returns: Pointer to appropriate hash class. |
155 | * |
156 | * Use: Chooses a hash function by name. |
157 | */ |
158 | |
159 | static const gchash *gethash(const char *name) |
160 | { |
e9026a0a |
161 | const gchash *const *g, *gg = 0; |
e375fe33 |
162 | size_t sz = strlen(name); |
e9026a0a |
163 | for (g = ghashtab; *g; g++) { |
e375fe33 |
164 | if (strncmp(name, (*g)->name, sz) == 0) { |
165 | if ((*g)->name[sz] == 0) { |
166 | gg = *g; |
167 | break; |
168 | } else if (gg) |
169 | return (0); |
170 | else |
171 | gg = *g; |
172 | } |
173 | } |
174 | return (gg); |
175 | } |
176 | |
177 | /* --- @getstring@ --- * |
178 | * |
179 | * Arguments: @FILE *fp@ = stream from which to read |
180 | * @const char *p@ = string to read from instead |
181 | * @dstr *d@ = destination string |
182 | * @unsigned raw@ = raw or cooked read |
183 | * |
184 | * Returns: Zero if OK, nonzero on end-of-file. |
185 | * |
186 | * Use: Reads a filename (or something similar) from a stream. |
187 | */ |
188 | |
189 | static int getstring(FILE *fp, const char *p, dstr *d, unsigned raw) |
190 | { |
191 | int ch; |
192 | int q = 0; |
193 | |
194 | /* --- Raw: just read exactly what's written up to a null byte --- */ |
195 | |
d470270a |
196 | #define NEXTCH (fp ? getc(fp) : (unsigned char)*p++) |
e375fe33 |
197 | #define EOFCH (fp ? EOF : 0) |
198 | |
199 | if (raw) { |
200 | if ((ch = NEXTCH) == EOFCH) |
201 | return (EOF); |
202 | for (;;) { |
203 | if (!ch) |
204 | break; |
205 | DPUTC(d, ch); |
206 | if ((ch = NEXTCH) == EOFCH) |
207 | break; |
208 | } |
209 | DPUTZ(d); |
210 | return (0); |
211 | } |
212 | |
213 | /* --- Skip as far as whitespace --- * |
214 | * |
215 | * Also skip past comments. |
216 | */ |
217 | |
218 | again: |
219 | ch = NEXTCH; |
d470270a |
220 | while (isspace(ch)) |
e375fe33 |
221 | ch = NEXTCH; |
222 | if (ch == '#') { |
223 | do ch = NEXTCH; while (ch != '\n' && ch != EOFCH); |
224 | goto again; |
225 | } |
226 | if (ch == EOFCH) |
227 | return (EOF); |
228 | |
229 | /* --- If the character is a quote then read a quoted string --- */ |
230 | |
231 | switch (ch) { |
232 | case '`': |
233 | ch = '\''; |
234 | case '\'': |
235 | case '\"': |
236 | q = ch; |
237 | ch = NEXTCH; |
238 | break; |
239 | } |
240 | |
241 | /* --- Now read all sorts of interesting things --- */ |
242 | |
243 | for (;;) { |
244 | |
245 | /* --- Handle an escaped thing --- */ |
246 | |
247 | if (ch == '\\') { |
248 | ch = NEXTCH; |
249 | if (ch == EOFCH) |
250 | break; |
251 | switch (ch) { |
252 | case 'a': ch = '\a'; break; |
253 | case 'b': ch = '\b'; break; |
254 | case 'f': ch = '\f'; break; |
255 | case 'n': ch = '\n'; break; |
256 | case 'r': ch = '\r'; break; |
257 | case 't': ch = '\t'; break; |
258 | case 'v': ch = '\v'; break; |
259 | } |
260 | DPUTC(d, ch); |
261 | ch = NEXTCH; |
262 | continue; |
263 | } |
264 | |
265 | /* --- If it's a quote or some other end marker then stop --- */ |
266 | |
267 | if (ch == q) |
268 | break; |
d470270a |
269 | if (!q && isspace(ch)) |
e375fe33 |
270 | break; |
271 | |
272 | /* --- Otherwise contribute and continue --- */ |
273 | |
274 | DPUTC(d, ch); |
275 | if ((ch = NEXTCH) == EOFCH) |
276 | break; |
277 | } |
278 | |
279 | /* --- Done --- */ |
280 | |
281 | DPUTZ(d); |
282 | return (0); |
283 | |
284 | #undef NEXTCH |
285 | #undef EOFCH |
286 | } |
287 | |
288 | /* --- @putstring@ --- * |
289 | * |
290 | * Arguments: @FILE *fp@ = stream to write on |
291 | * @const char *p@ = pointer to text |
292 | * @unsigned raw@ = whether the string is to be written raw |
293 | * |
294 | * Returns: --- |
295 | * |
296 | * Use: Emits a string to a stream. |
297 | */ |
298 | |
299 | static void putstring(FILE *fp, const char *p, unsigned raw) |
300 | { |
301 | size_t sz = strlen(p); |
302 | unsigned qq; |
303 | const char *q; |
304 | |
305 | /* --- Just write the string null terminated if raw --- */ |
306 | |
307 | if (raw) { |
308 | fwrite(p, 1, sz + 1, fp); |
309 | return; |
310 | } |
311 | |
312 | /* --- Check for any dodgy characters --- */ |
313 | |
314 | qq = 0; |
315 | for (q = p; *q; q++) { |
316 | if (isspace((unsigned char)*q)) { |
317 | qq = '\"'; |
318 | break; |
319 | } |
320 | } |
321 | |
322 | if (qq) |
323 | putc(qq, fp); |
324 | |
325 | /* --- Emit the string --- */ |
326 | |
327 | for (q = p; *q; q++) { |
328 | switch (*q) { |
329 | case '\a': fputc('\\', fp); fputc('a', fp); break; |
330 | case '\b': fputc('\\', fp); fputc('b', fp); break; |
331 | case '\f': fputc('\\', fp); fputc('f', fp); break; |
332 | case '\n': fputc('\\', fp); fputc('n', fp); break; |
333 | case '\r': fputc('\\', fp); fputc('r', fp); break; |
334 | case '\t': fputc('\\', fp); fputc('t', fp); break; |
335 | case '\v': fputc('\\', fp); fputc('v', fp); break; |
336 | case '`': fputc('\\', fp); fputc('`', fp); break; |
337 | case '\'': fputc('\\', fp); fputc('\'', fp); break; |
338 | case '\"': fputc('\\', fp); fputc('\"', fp); break; |
339 | case '#': fputc('\\', fp); fputc('#', fp); break; |
340 | default: |
341 | putc(*q, fp); |
342 | break; |
343 | } |
344 | } |
345 | |
346 | /* --- Done --- */ |
347 | |
348 | if (qq) |
349 | putc(qq, fp); |
350 | } |
351 | |
352 | /*----- Guts --------------------------------------------------------------*/ |
353 | |
354 | static int checkhash(const char *file, unsigned f, const gchash *gch) |
355 | { |
356 | int rc; |
357 | FILE *fp; |
358 | dstr d = DSTR_INIT; |
359 | dstr dd = DSTR_INIT; |
360 | unsigned long n = 0, nfail = 0; |
361 | octet *buf = xmalloc(2 * gch->hashsz); |
362 | |
363 | if (!file) |
364 | fp = stdin; |
365 | else if ((fp = fopen(file, f & f_raw ? "r" : "rb")) == 0) { |
366 | moan("couldn't open `%s': %s", file, strerror(errno)); |
367 | return (EXIT_FAILURE); |
368 | } |
369 | |
370 | while (DRESET(&d), dstr_putline(&d, fp) != EOF) { |
371 | char *p = d.buf; |
372 | char *q; |
373 | unsigned ff = f; |
374 | |
375 | /* --- Handle a directive --- */ |
376 | |
377 | if (*p == '#') { |
378 | p++; |
379 | if ((q = str_getword(&p)) == 0) |
380 | continue; |
381 | if (strcmp(q, "hash") == 0) { |
382 | const gchash *g; |
383 | if ((q = str_getword(&p)) == 0) |
384 | continue; |
385 | if ((g = gethash(q)) == 0) |
386 | continue; |
387 | gch = g; |
388 | xfree(buf); |
389 | buf = xmalloc(2 * gch->hashsz); |
390 | } else if (strcmp(q, "escape") == 0) |
391 | f |= f_escape; |
392 | continue; |
393 | } |
394 | |
395 | /* --- Otherwise it's a hex thing --- */ |
396 | |
12902a5c |
397 | q = p; |
398 | while (*p && *p != ' ') |
399 | p++; |
400 | if (!*p) |
e375fe33 |
401 | continue; |
12902a5c |
402 | *p++ = 0; |
e375fe33 |
403 | if (gethex(q, buf, gch->hashsz, 0) < gch->hashsz) |
404 | continue; |
12902a5c |
405 | if (*p == '*') |
e375fe33 |
406 | ff |= f_binary; |
12902a5c |
407 | else if (*p != ' ') |
e375fe33 |
408 | continue; |
12902a5c |
409 | p++; |
e375fe33 |
410 | |
411 | if (f & f_escape) { |
412 | DRESET(&dd); |
413 | getstring(0, p, &dd, 0); |
414 | p = dd.buf; |
415 | } |
416 | |
417 | if (fhash(p, ff, gch, buf + gch->hashsz)) { |
418 | moan("couldn't read `%s': %s", p, strerror(errno)); |
419 | rc = EXIT_FAILURE; |
420 | continue; |
421 | } |
422 | if (memcmp(buf, buf + gch->hashsz, gch->hashsz) != 0) { |
423 | if (ff & f_verbose) |
424 | fprintf(stderr, "FAIL %s\n", p); |
425 | else |
426 | moan("%s check failed for `%s'", gch->name, p); |
427 | nfail++; |
428 | rc = EXIT_FAILURE; |
429 | } else { |
430 | if (ff & f_verbose) |
431 | fprintf(stderr, "OK %s\n", p); |
432 | } |
433 | n++; |
434 | } |
435 | |
436 | dstr_destroy(&d); |
437 | dstr_destroy(&dd); |
438 | xfree(buf); |
439 | if ((f & f_verbose) && nfail) |
440 | moan("%lu of %lu file(s) failed %s check", nfail, n, gch->name); |
441 | else if (!n) |
442 | moan("no files checked"); |
443 | return (0); |
444 | } |
445 | |
446 | static int dohash(const char *file, unsigned f, const gchash *gch) |
447 | { |
448 | int rc = 0; |
449 | octet *p = xmalloc(gch->hashsz); |
450 | |
451 | if (fhash(file, f, gch, p)) { |
452 | moan("couldn't read `%s': %s", file ? file : "<stdin>", strerror(errno)); |
453 | rc = EXIT_FAILURE; |
454 | } else { |
455 | puthex(p, gch->hashsz, stdout); |
456 | if (file) { |
457 | fputc(' ', stdout); |
458 | fputc(f & f_binary ? '*' : ' ', stdout); |
459 | if (f & f_escape) |
460 | putstring(stdout, file, 0); |
461 | else |
462 | fputs(file, stdout); |
463 | } |
464 | fputc('\n', stdout); |
465 | } |
466 | |
467 | xfree(p); |
468 | return (rc); |
469 | } |
470 | |
12902a5c |
471 | static int dofile(const char *file, unsigned f, const gchash *gch) |
472 | { |
473 | return (f & f_check ? checkhash : dohash)(file, f, gch); |
474 | } |
475 | |
e375fe33 |
476 | static int hashfiles(const char *file, unsigned f, const gchash *gch) |
477 | { |
478 | FILE *fp; |
479 | dstr d = DSTR_INIT; |
480 | int rc = 0; |
481 | int rrc; |
482 | |
483 | if (!file) |
484 | fp = stdin; |
485 | else if ((fp = fopen(file, f & f_raw ? "r" : "rb")) == 0) { |
486 | moan("couldn't open `%s': %s", file, strerror(errno)); |
487 | return (EXIT_FAILURE); |
488 | } |
489 | |
490 | for (;;) { |
491 | DRESET(&d); |
492 | if (getstring(fp, 0, &d, f & f_raw)) |
493 | break; |
12902a5c |
494 | if ((rrc = dofile(d.buf, f, gch)) != 0) |
e375fe33 |
495 | rc = rrc; |
496 | } |
497 | |
498 | return (rc); |
499 | } |
500 | |
501 | static int hashsum(const char *file, unsigned f, const gchash *gch) |
502 | { |
12902a5c |
503 | return (f & f_files ? hashfiles : dofile)(file, f, gch); |
e375fe33 |
504 | } |
505 | |
506 | /*----- Main driver -------------------------------------------------------*/ |
507 | |
508 | static void version(FILE *fp) |
509 | { |
510 | pquis(fp, "$, Catacomb version " VERSION "\n"); |
511 | } |
512 | |
513 | static void usage(FILE *fp) |
514 | { |
5af0f9cb |
515 | pquis(fp, "Usage: $ [-f0ebcv] [-a algorithm] [files...]\n"); |
e375fe33 |
516 | } |
517 | |
518 | static void help(FILE *fp, const gchash *gch) |
519 | { |
520 | version(fp); |
521 | fputc('\n', fp); |
522 | usage(fp); |
523 | pquis(fp, "\n\ |
524 | Generates or checks message digests on files. Options available:\n\ |
525 | \n\ |
526 | -h, --help Display this help message.\n\ |
527 | -V, --version Display program's version number.\n\ |
528 | -u, --usage Display a terse usage message.\n\ |
529 | \n\ |
530 | -a, --algorithm=ALG Use the message digest algorithm ALG.\n\ |
531 | \n\ |
532 | -f, --files Read a list of file names from standard input.\n\ |
533 | -0, --null File names are null terminated, not plain text.\n\ |
534 | \n\ |
535 | -e, --escape Escape funny characters in filenames.\n\ |
536 | -c, --check Check message digests rather than emitting them.\n\ |
537 | -b, --binary When reading files, treat them as binary.\n\ |
538 | -v, --verbose Be verbose when checking digests.\n\ |
539 | \n\ |
540 | For a list of supported message digest algorithms, type `$ --list'.\n\ |
541 | "); |
542 | if (gch) |
543 | fprintf(fp, "The default message digest algorithm is %s.\n", gch->name); |
544 | } |
545 | |
546 | int main(int argc, char *argv[]) |
547 | { |
548 | unsigned f = 0; |
549 | const gchash *gch = 0; |
550 | int rc; |
551 | |
552 | /* --- Initialization --- */ |
553 | |
554 | ego(argv[0]); |
555 | sub_init(); |
556 | |
557 | /* --- Choose a hash function from the name --- */ |
558 | |
559 | { |
560 | char *q = xstrdup(QUIS); |
561 | size_t len = strlen(q); |
562 | if (len > 3 && strcmp(q + len - 3, "sum") == 0) { |
563 | q[len - 3] = 0; |
564 | gch = gethash(q); |
565 | } |
566 | if (!gch) |
e9026a0a |
567 | gch = gethash("md5"); |
e375fe33 |
568 | xfree(q); |
569 | } |
570 | |
571 | /* --- Read options --- */ |
572 | |
573 | for (;;) { |
574 | static struct option opts[] = { |
575 | { "help", 0, 0, 'h' }, |
576 | { "verbose", 0, 0, 'V' }, |
577 | { "usage", 0, 0, 'u' }, |
578 | |
579 | { "algorithm", OPTF_ARGREQ, 0, 'a' }, |
580 | { "hash", OPTF_ARGREQ, 0, 'a' }, |
581 | { "list", 0, 0, 'l' }, |
582 | |
583 | { "files", 0, 0, 'f' }, |
584 | { "find", 0, 0, 'f' }, |
585 | { "null", 0, 0, '0' }, |
586 | |
587 | { "escape", 0, 0, 'e' }, |
588 | { "check", 0, 0, 'c' }, |
589 | { "binary", 0, 0, 'b' }, |
590 | { "verbose", 0, 0, 'v' }, |
591 | |
592 | { 0, 0, 0, 0 } |
593 | }; |
594 | int i = mdwopt(argc, argv, "hVu a:l f0 ecbv", opts, 0, 0, 0); |
595 | if (i < 0) |
596 | break; |
597 | |
598 | switch (i) { |
599 | case 'h': |
600 | help(stdout, gch); |
601 | exit(0); |
602 | case 'V': |
603 | version(stdout); |
604 | exit(0); |
605 | case 'u': |
606 | usage(stdout); |
607 | exit(0); |
608 | case 'a': |
609 | if ((gch = gethash(optarg)) == 0) |
610 | die(EXIT_FAILURE, "unknown hash algorithm `%s'", optarg); |
611 | f |= f_oddhash; |
612 | break; |
613 | case 'l': { |
614 | unsigned j; |
e9026a0a |
615 | for (j = 0; ghashtab[j]; j++) { |
e375fe33 |
616 | if (j) |
617 | fputc(' ', stdout); |
e9026a0a |
618 | printf("%s", ghashtab[j]->name); |
e375fe33 |
619 | } |
620 | fputc('\n', stdout); |
621 | exit(0); |
622 | } break; |
623 | case 'f': |
624 | f |= f_files; |
625 | break; |
626 | case '0': |
627 | f |= f_raw; |
628 | break; |
629 | case 'e': |
630 | f |= f_escape; |
631 | break; |
632 | case 'c': |
633 | f |= f_check; |
634 | break; |
635 | case 'b': |
636 | f |= f_binary; |
637 | break; |
638 | case 'v': |
639 | f |= f_verbose; |
640 | break; |
641 | default: |
642 | f |= f_bogus; |
643 | break; |
644 | } |
645 | } |
646 | |
647 | if (f & f_bogus) { |
648 | usage(stderr); |
649 | exit(EXIT_FAILURE); |
650 | } |
651 | argv += optind; |
652 | argc -= optind; |
653 | |
654 | /* --- Generate output --- */ |
655 | |
656 | if (!(f & f_check)) { |
657 | if (f & f_oddhash) |
658 | printf("#hash %s\n", gch->name); |
659 | if (f & f_escape) |
660 | fputs("#escape\n", stdout); |
661 | } |
662 | |
663 | if (argc) { |
664 | int i; |
665 | int rrc; |
666 | rc = 0; |
667 | for (i = 0; i < argc; i++) { |
668 | if ((rrc = hashsum(argv[i], f, gch)) != 0) |
669 | rc = rrc; |
670 | } |
671 | } else |
672 | rc = hashsum(0, f, gch); |
673 | |
674 | return (rc); |
675 | } |
676 | |
677 | /*----- That's all, folks -------------------------------------------------*/ |