3 * Common functions for hashing utilities
5 * (c) 2011 Straylight/Edgeware
8 /*----- Licensing notice --------------------------------------------------*
10 * This file is part of Catacomb.
12 * Catacomb is free software; you can redistribute it and/or modify
13 * it under the terms of the GNU Library General Public License as
14 * published by the Free Software Foundation; either version 2 of the
15 * License, or (at your option) any later version.
17 * Catacomb is distributed in the hope that it will be useful,
18 * but WITHOUT ANY WARRANTY; without even the implied warranty of
19 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
20 * GNU Library General Public License for more details.
22 * You should have received a copy of the GNU Library General Public
23 * License along with Catacomb; if not, write to the Free
24 * Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
28 /*----- Header files ------------------------------------------------------*/
30 #define _FILE_OFFSET_BITS 64
40 #include <sys/types.h>
43 #include <mLib/alloc.h>
44 #include <mLib/dstr.h>
45 #include <mLib/macros.h>
46 #include <mLib/report.h>
50 #include <mLib/base32.h>
51 #include <mLib/base64.h>
56 /*----- Encoding and decoding ---------------------------------------------*/
58 /* --- Hex encoding --- */
60 static void puthex(const octet *buf, size_t sz, FILE *fp)
63 fprintf(fp, "%02x", *buf++);
68 static size_t gethex(const char *p, octet *q, size_t sz, char **pp)
71 while (sz > 0 && ISXDIGIT(p[0]) && ISXDIGIT(p[1])) {
76 *q++ = strtoul(buf, 0, 16);
86 /* --- Base64 encoding --- */
88 static void putbase64(const octet *buf, size_t sz, FILE *fp)
96 base64_encode(&b, buf, sz, &d);
97 base64_encode(&b, 0, 0, &d);
102 static size_t getbase64(const char *p, octet *q, size_t sz, char **pp)
106 size_t n = strlen(p);
109 base64_decode(&b, p, n, &d);
110 if (pp) *pp = (/*unconst*/ char *)p + n;
111 base64_decode(&b, 0, 0, &d);
113 memcpy(q, d.buf, sz);
119 /* --- Base32 encoding --- */
121 static void putbase32(const octet *buf, size_t sz, FILE *fp)
129 base32_encode(&b, buf, sz, &d);
130 base32_encode(&b, 0, 0, &d);
135 static size_t getbase32(const char *p, octet *q, size_t sz, char **pp)
139 size_t n = strlen(p);
142 base32_decode(&b, p, n, &d);
143 if (pp) *pp = (/*unconst*/ char *)p + n;
144 base32_decode(&b, 0, 0, &d);
146 memcpy(q, d.buf, sz);
154 const encodeops encodingtab[] = {
155 #define TAB(tag, name) { #name, put##name, get##name },
161 const encodeops *getencoding(const char *ename)
165 for (e = encodingtab; e->name; e++) {
166 if (STRCMP(ename, ==, e->name))
172 /*----- File hashing ------------------------------------------------------*/
179 typedef struct fhent {
186 /* --- @gethash@ --- *
188 * Arguments: @const char *name@ = pointer to name string
190 * Returns: Pointer to appropriate hash class.
192 * Use: Chooses a hash function by name.
195 const gchash *gethash(const char *name)
197 const gchash *const *g, *gg = 0;
198 size_t sz = strlen(name);
199 for (g = ghashtab; *g; g++) {
200 if (STRNCMP(name, ==, (*g)->name, sz)) {
201 if ((*g)->name[sz] == 0) {
213 /* --- @describefile@ --- *
215 * Arguments: @const struct stat *st@ = pointer to file state
217 * Returns: A snappy one-word description of the file.
220 const char *describefile(const struct stat *st)
222 switch (st->st_mode & S_IFMT) {
223 case S_IFBLK: return ("block-special");
224 case S_IFCHR: return ("char-special");
225 case S_IFIFO: return ("fifo");
226 case S_IFREG: return ("file");
227 case S_IFLNK: return ("symlink");
228 case S_IFDIR: return ("directory");
229 case S_IFSOCK: return ("socket");
230 default: return ("unknown");
234 /* --- @fhash_init@ ---*
236 * Arguments: @fhashstate *fh@ = pointer to fhash state to initialize
237 * @const gchash *gch@ = hash class to set
238 * @unsigned f@ initial flags to set
242 * Use: Initializes an @fhashstate@ structure.
245 void fhash_init(fhashstate *fh, const gchash *gch, unsigned f)
246 { fh->f = f; fh->gch = gch; fh->ents = 0; }
248 /* --- @fhash_free@ --- *
250 * Arguments: @fhashstate *fh@ = pointer to fhash state to free
254 * Use: Frees an fhash state.
257 static void freefhents(struct fhent *fhe)
261 for (; fhe; fhe = ffhe) {
263 freefhents(fhe->sub);
268 void fhash_free(fhashstate *fh)
269 { freefhents(fh->ents); }
273 * Arguments: @fhashstate *fh@ = pointer to fhash state
274 * @const char *file@ = file name to be hashed (null for stdin)
275 * @void *buf@ = pointer to hash output buffer
277 * Returns: Zero if it worked, nonzero on error.
279 * Use: Hashes a file.
282 int fhash(fhashstate *fh, const char *file, void *buf)
285 char fbuf[1024 * 128];
289 struct fhent *fhe, **ffhe;
294 if (!file || STRCMP(file, ==, "-"))
296 else if ((fp = fopen(file, fh->f & FHF_BINARY ? "rb" : "r")) == 0)
299 if (fh->f & FHF_PROGRESS) {
300 if (fprogress_init(&ff, file ? file : "<stdin>", fp)) return (-1);
303 if (fh->f & FHF_JUNK) {
305 if (STRNCMP(p, ==, "./", 2)) p += 2;
309 if (*q == '/' || *q == 0) {
311 for (; *ffhe; ffhe = &(*ffhe)->next) {
313 if (STRNCMP(p, ==, fhe->name, n) && fhe->name[n] == 0)
316 fhe = xmalloc(offsetof(struct fhent, name) + n + 1);
318 fhe->ty = *q == '/' ? FHETY_DIR : FHETY_FILE;
321 memcpy(fhe->name, p, n); fhe->name[n] = 0;
332 h = GH_INIT(fh->gch);
333 while ((sz = fread(fbuf, 1, sizeof(fbuf), fp)) > 0) {
334 GH_HASH(h, fbuf, sz);
335 if (fh->f & FHF_PROGRESS) fprogress_update(&ff, sz);
337 if (ferror(fp)) rc = -1;
338 if (fp != stdin) fclose(fp);
339 if (fh->f & FHF_PROGRESS) fprogress_done(&ff);
345 /* --- @fhash_junk@ --- *
347 * Arguments: @fhashstate *fh@ = pointer to fhash state
348 * @void (*func)(const char *, const struct stat *, void *)@
349 * @void *p@ = pointer to pass to function
351 * Returns: Positive if any junk was found, negative on error, zero if
352 * everything was fine.
354 * Use: Reports junk files in any directories covered by the hash
359 int (*func)(const char *, const struct stat *, void *);
364 static int fhjunk(struct fhjunk *fhj, struct fhent *ents)
371 size_t n = fhj->d->len;
374 dname = n ? fhj->d->buf : ".";
375 if ((dp = opendir(dname)) == 0) {
376 moan("failed to open directory `%s': %s", dname, strerror(errno));
381 dstr_putc(fhj->d, '/');
384 while (errno = 0, (d = readdir(dp)) != 0) {
385 if (STRCMP(d->d_name, ==, ".") || STRCMP(d->d_name, ==, ".."))
387 for (fhe = ents; fhe; fhe = fhe->next) {
388 if (STRCMP(d->d_name, ==, fhe->name)) goto found;
391 dstr_puts(fhj->d, d->d_name);
392 if (!lstat(fhj->d->buf, &st)) {
394 rrc = fhj->func(fhj->d->buf, &st, fhj->p);
397 rrc = fhj->func(fhj->d->buf, 0, fhj->p);
399 if (rrc < 0) rc = -1;
404 moan("failed to read directory `%s': %s", dname, strerror(errno));
409 for (fhe = ents; fhe; fhe = fhe->next) {
410 if (fhe->ty == FHETY_DIR) {
412 dstr_puts(fhj->d, fhe->name);
413 rrc = fhjunk(fhj, fhe->sub);
414 if (rrc < 0) rc = -1;
415 else if (!rc) rc = rrc;
422 int fhash_junk(fhashstate *fh,
423 int (*func)(const char *, const struct stat *, void *),
433 rc = fhjunk(&fhj, fh->ents);
438 /* --- @hfparse@ --- *
440 * Arguments: @hfpctx *hfp@ = pointer to the context structure
442 * Returns: A code indicating what happened.
444 * Use: Parses a line from the input file.
447 int hfparse(hfpctx *hfp)
452 dstr *d = hfp->dline;
455 /* --- Fetch the input line and get ready to parse --- */
458 if (dstr_putline(d, hfp->fp) == EOF) return (HF_EOF);
461 /* --- Parse magic comments --- */
465 if ((q = str_getword(&p)) == 0) return (HF_BAD);
466 if (STRCMP(q, ==, "hash")) {
467 if ((q = str_getword(&p)) == 0) return (HF_BAD);
468 if ((gch = gethash(q)) == 0) return (HF_BAD);
471 } else if (STRCMP(q, ==, "encoding")) {
472 if ((q = str_getword(&p)) == 0) return (HF_BAD);
473 if ((ee = getencoding(q)) == 0) return (HF_BAD);
476 } else if (STRCMP(q, ==, "escape")) {
477 hfp->f |= HFF_ESCAPE;
483 /* --- Otherwise it's a file line --- */
486 while (*p && *p != ' ') p++;
487 if (!*p) return (HF_BAD);
489 hsz = hfp->gch->hashsz;
490 if (hfp->ee->get(q, hfp->hbuf, hsz, 0) < hsz) return (HF_BAD);
492 case '*': hfp->f |= FHF_BINARY; break;
493 case ' ': hfp->f &= ~FHF_BINARY; break;
494 default: return (HF_BAD);
499 if (hfp->f & HFF_ESCAPE)
500 getstring(&p, hfp->dfile, GSF_STRING);
502 dstr_putm(hfp->dfile, p, d->len - (p - d->buf));
503 dstr_putz(hfp->dfile);
509 /*----- String I/O --------------------------------------------------------*/
511 /* --- @getstring@ --- *
513 * Arguments: @void *in@ = input source
514 * @dstr *d@ = destination string
515 * @unsigned f@ = input flags
517 * Returns: Zero if OK, nonzero on end-of-file.
519 * Use: Reads a filename (or something similar) from a stream.
522 static int nextch_file(void *in)
523 { FILE *fp = in; return (getc(fp)); }
525 static int nextch_string(void *in)
526 { const unsigned char **p = in; return (*(*p)++); }
528 int getstring(void *in, dstr *d, unsigned f)
531 int eofch = (f & GSF_STRING) ? 0 : EOF;
532 int (*nextch)(void *) = (f & GSF_STRING) ? nextch_string : nextch_file;
535 /* --- Raw: just read exactly what's written up to a null byte --- */
538 if ((ch = nextch(in)) == eofch)
544 if ((ch = nextch(in)) == eofch)
551 /* --- Skip as far as whitespace --- *
553 * Also skip past comments.
561 do ch = nextch(in); while (ch != '\n' && ch != eofch);
567 /* --- If the character is a quote then read a quoted string --- */
579 /* --- Now read all sorts of interesting things --- */
583 /* --- Handle an escaped thing --- */
590 case 'a': ch = '\a'; break;
591 case 'b': ch = '\b'; break;
592 case 'f': ch = '\f'; break;
593 case 'n': ch = '\n'; break;
594 case 'r': ch = '\r'; break;
595 case 't': ch = '\t'; break;
596 case 'v': ch = '\v'; break;
603 /* --- If it's a quote or some other end marker then stop --- */
607 if (!q && ISSPACE(ch))
610 /* --- Otherwise contribute and continue --- */
613 if ((ch = nextch(in)) == eofch)
623 /* --- @putstring@ --- *
625 * Arguments: @FILE *fp@ = stream to write on
626 * @const char *p@ = pointer to text
627 * @unsigned f@ = output flags
631 * Use: Emits a string to a stream.
634 void putstring(FILE *fp, const char *p, unsigned f)
636 size_t sz = strlen(p);
640 /* --- Just write the string null terminated if raw --- */
643 fwrite(p, 1, sz + 1, fp);
647 /* --- Check for any dodgy characters --- */
650 for (q = p; *q; q++) {
660 /* --- Emit the string --- */
662 for (q = p; *q; q++) {
664 case '\a': fputc('\\', fp); fputc('a', fp); break;
665 case '\b': fputc('\\', fp); fputc('b', fp); break;
666 case '\f': fputc('\\', fp); fputc('f', fp); break;
667 case '\n': fputc('\\', fp); fputc('n', fp); break;
668 case '\r': fputc('\\', fp); fputc('r', fp); break;
669 case '\t': fputc('\\', fp); fputc('t', fp); break;
670 case '\v': fputc('\\', fp); fputc('v', fp); break;
671 case '`': fputc('\\', fp); fputc('`', fp); break;
672 case '\'': fputc('\\', fp); fputc('\'', fp); break;
673 case '\"': fputc('\\', fp); fputc('\"', fp); break;
686 /*----- That's all, folks -------------------------------------------------*/