X-Git-Url: https://git.distorted.org.uk/u/mdw/catacomb/blobdiff_plain/5685a696f1a1d1944b6eb1326e0df9ed84128ebd..097fb6f2f97575ce17738b4afb3216e9492de2b4:/hashsum.c diff --git a/hashsum.c b/hashsum.c index dbb6662..8a14a76 100644 --- a/hashsum.c +++ b/hashsum.c @@ -7,7 +7,7 @@ * (c) 2000 Straylight/Edgeware */ -/*----- Licensing notice --------------------------------------------------* +/*----- Licensing notice --------------------------------------------------* * * This file is part of Catacomb. * @@ -15,12 +15,12 @@ * it under the terms of the GNU Library General Public License as * published by the Free Software Foundation; either version 2 of the * License, or (at your option) any later version. - * + * * Catacomb is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU Library General Public License for more details. - * + * * You should have received a copy of the GNU Library General Public * License along with Catacomb; if not, write to the Free * Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, @@ -29,10 +29,11 @@ /*----- Header files ------------------------------------------------------*/ +#define _FILE_OFFSET_BITS 64 + #include "config.h" #include -#include #include #include #include @@ -44,507 +45,131 @@ #include #include #include -#include - -#include -#include -#include #include "ghash.h" +#include "cc.h" /*----- Static variables --------------------------------------------------*/ -#define f_binary 1u -#define f_bogus 2u -#define f_verbose 4u -#define f_check 8u -#define f_files 16u -#define f_raw 32u -#define f_oddhash 64u -#define f_escape 128u -#define f_oddenc 256u - -/*----- Encoding and decoding ---------------------------------------------*/ - -/* --- Hex encoding --- */ - -static void puthex(const octet *buf, size_t sz, FILE *fp) -{ - while (sz) { - fprintf(fp, "%02x", *buf++); - sz--; - } -} - -static size_t gethex(const char *p, octet *q, size_t sz, char **pp) -{ - size_t i = 0; - while (sz > 0 && - isxdigit((unsigned char)p[0]) && - isxdigit((unsigned char)p[1])) { - char buf[3]; - buf[0] = p[0]; - buf[1] = p[1]; - buf[2] = 0; - *q++ = strtoul(buf, 0, 16); - sz--; - p += 2; - i++; - } - if (pp) - *pp = (char *)p; - return (i); -} - -/* --- Base64 encoding --- */ - -static void putb64(const octet *buf, size_t sz, FILE *fp) -{ - base64_ctx b; - dstr d = DSTR_INIT; - - base64_init(&b); - b.indent = ""; - b.maxline = 0; - base64_encode(&b, buf, sz, &d); - base64_encode(&b, 0, 0, &d); - dstr_write(&d, fp); - dstr_destroy(&d); -} - -static size_t getb64(const char *p, octet *q, size_t sz, char **pp) -{ - base64_ctx b; - dstr d = DSTR_INIT; - size_t n = strlen(p); - - base64_init(&b); - base64_decode(&b, p, n, &d); - if (pp) *pp = (/*unconst*/ char *)p + n; - base64_decode(&b, 0, 0, &d); - assert(d.len <= sz); - memcpy(q, d.buf, sz); - n = d.len; - dstr_destroy(&d); - return (n); -} - -/* --- Base32 encoding --- */ - -static void putb32(const octet *buf, size_t sz, FILE *fp) -{ - base32_ctx b; - dstr d = DSTR_INIT; - - base32_init(&b); - b.indent = ""; - b.maxline = 0; - base32_encode(&b, buf, sz, &d); - base32_encode(&b, 0, 0, &d); - dstr_write(&d, fp); - dstr_destroy(&d); -} - -static size_t getb32(const char *p, octet *q, size_t sz, char **pp) -{ - base32_ctx b; - dstr d = DSTR_INIT; - size_t n = strlen(p); - - base32_init(&b); - base32_decode(&b, p, n, &d); - if (pp) *pp = (/*unconst*/ char *)p + n; - base32_decode(&b, 0, 0, &d); - assert(d.len <= sz); - memcpy(q, d.buf, sz); - n = d.len; - dstr_destroy(&d); - return (n); -} - -/* --- Table --- */ +#define f_bogus 1u +#define f_verbose 2u +#define f_check 4u +#define f_files 8u +#define f_oddhash 16u +#define f_escape 32u +#define f_oddenc 64u -typedef struct encops { - const char *name; - void (*put)(const octet *, size_t, FILE *); - size_t (*get)(const char *, octet *, size_t, char **); -} encops; - -static const encops enctab[] = { - { "hex", puthex, gethex }, - { "base64", putb64, getb64 }, - { "base32", putb32, getb32 }, - { 0, 0, 0 } -}; +/*----- Guts --------------------------------------------------------------*/ -static const encops *getenc(const char *ename) +static int checkjunk(const char *path, const struct stat *st, void *p) { - const encops *e; - - for (e = enctab; e->name; e++) { - if (strcmp(ename, e->name) == 0) - return (e); + const char *what; + fhashstate *fh = p; + + if (!st) { + if (fh->f & f_verbose) + fprintf(stderr, "JUNK (error %s) %s\n", strerror(errno), path); + else + moan("couldn't stat junk file `%s': %s", path, strerror(errno)); + } else { + what = describefile(st); + if (fh->f & f_verbose) + fprintf(stderr, "JUNK %s %s\n", what, path); + else + moan("found junk %s `%s'", what, path); } return (0); } -/*----- Support functions -------------------------------------------------*/ - -/* --- @fhash@ --- * - * - * Arguments: @const char *file@ = file name to be hashed (null for stdin) - * @unsigned f@ = flags to set - * @const gchash *gch@ = pointer to hash function to use - * @void *buf@ = pointer to hash output buffer - * - * Returns: Zero if it worked, nonzero on error. - * - * Use: Hashes a file. - */ - -static int fhash(const char *file, unsigned f, const gchash *gch, void *buf) -{ - FILE *fp; - char fbuf[BUFSIZ]; - size_t sz; - ghash *h; - int e; - - if (!file) - fp = stdin; - else if ((fp = fopen(file, f & f_binary ? "rb" : "r")) == 0) - return (-1); - - h = GH_INIT(gch); - while ((sz = fread(fbuf, 1, sizeof(fbuf), fp)) > 0) - GH_HASH(h, fbuf, sz); - GH_DONE(h, buf); - GH_DESTROY(h); - e = ferror(fp); - if (file) - fclose(fp); - return (e ? -1 : 0); -} - -/* --- @gethash@ --- * - * - * Arguments: @const char *name@ = pointer to name string - * - * Returns: Pointer to appropriate hash class. - * - * Use: Chooses a hash function by name. - */ - -static const gchash *gethash(const char *name) -{ - const gchash *const *g, *gg = 0; - size_t sz = strlen(name); - for (g = ghashtab; *g; g++) { - if (strncmp(name, (*g)->name, sz) == 0) { - if ((*g)->name[sz] == 0) { - gg = *g; - break; - } else if (gg) - return (0); - else - gg = *g; - } - } - return (gg); -} - -/* --- @getstring@ --- * - * - * Arguments: @FILE *fp@ = stream from which to read - * @const char *p@ = string to read from instead - * @dstr *d@ = destination string - * @unsigned raw@ = raw or cooked read - * - * Returns: Zero if OK, nonzero on end-of-file. - * - * Use: Reads a filename (or something similar) from a stream. - */ - -static int getstring(FILE *fp, const char *p, dstr *d, unsigned raw) +static int warnjunk(const char *path, const struct stat *st, void *p) { - int ch; - int q = 0; - - /* --- Raw: just read exactly what's written up to a null byte --- */ - -#define NEXTCH (fp ? getc(fp) : (unsigned char)*p++) -#define EOFCH (fp ? EOF : 0) - - if (raw) { - if ((ch = NEXTCH) == EOFCH) - return (EOF); - for (;;) { - if (!ch) - break; - DPUTC(d, ch); - if ((ch = NEXTCH) == EOFCH) - break; - } - DPUTZ(d); - return (0); - } - - /* --- Skip as far as whitespace --- * - * - * Also skip past comments. - */ - -again: - ch = NEXTCH; - while (isspace(ch)) - ch = NEXTCH; - if (ch == '#') { - do ch = NEXTCH; while (ch != '\n' && ch != EOFCH); - goto again; - } - if (ch == EOFCH) - return (EOF); - - /* --- If the character is a quote then read a quoted string --- */ - - switch (ch) { - case '`': - ch = '\''; - case '\'': - case '\"': - q = ch; - ch = NEXTCH; - break; - } - - /* --- Now read all sorts of interesting things --- */ - - for (;;) { - - /* --- Handle an escaped thing --- */ - - if (ch == '\\') { - ch = NEXTCH; - if (ch == EOFCH) - break; - switch (ch) { - case 'a': ch = '\a'; break; - case 'b': ch = '\b'; break; - case 'f': ch = '\f'; break; - case 'n': ch = '\n'; break; - case 'r': ch = '\r'; break; - case 't': ch = '\t'; break; - case 'v': ch = '\v'; break; - } - DPUTC(d, ch); - ch = NEXTCH; - continue; - } - - /* --- If it's a quote or some other end marker then stop --- */ - - if (ch == q) - break; - if (!q && isspace(ch)) - break; - - /* --- Otherwise contribute and continue --- */ - - DPUTC(d, ch); - if ((ch = NEXTCH) == EOFCH) - break; - } - - /* --- Done --- */ - - DPUTZ(d); + if (st) + moan("unexpected %s `%s'", describefile(st), path); + else + moan("couldn't stat unexpected file `%s': %s", path, strerror(errno)); return (0); - -#undef NEXTCH -#undef EOFCH -} - -/* --- @putstring@ --- * - * - * Arguments: @FILE *fp@ = stream to write on - * @const char *p@ = pointer to text - * @unsigned raw@ = whether the string is to be written raw - * - * Returns: --- - * - * Use: Emits a string to a stream. - */ - -static void putstring(FILE *fp, const char *p, unsigned raw) -{ - size_t sz = strlen(p); - unsigned qq; - const char *q; - - /* --- Just write the string null terminated if raw --- */ - - if (raw) { - fwrite(p, 1, sz + 1, fp); - return; - } - - /* --- Check for any dodgy characters --- */ - - qq = 0; - for (q = p; *q; q++) { - if (isspace((unsigned char)*q)) { - qq = '\"'; - break; - } - } - - if (qq) - putc(qq, fp); - - /* --- Emit the string --- */ - - for (q = p; *q; q++) { - switch (*q) { - case '\a': fputc('\\', fp); fputc('a', fp); break; - case '\b': fputc('\\', fp); fputc('b', fp); break; - case '\f': fputc('\\', fp); fputc('f', fp); break; - case '\n': fputc('\\', fp); fputc('n', fp); break; - case '\r': fputc('\\', fp); fputc('r', fp); break; - case '\t': fputc('\\', fp); fputc('t', fp); break; - case '\v': fputc('\\', fp); fputc('v', fp); break; - case '`': fputc('\\', fp); fputc('`', fp); break; - case '\'': fputc('\\', fp); fputc('\'', fp); break; - case '\"': fputc('\\', fp); fputc('\"', fp); break; - case '#': fputc('\\', fp); fputc('#', fp); break; - default: - putc(*q, fp); - break; - } - } - - /* --- Done --- */ - - if (qq) - putc(qq, fp); } -/*----- Guts --------------------------------------------------------------*/ - -static int checkhash(const char *file, unsigned f, - const gchash *gch, const encops *e) +static int checkhash(fhashstate *fh, const char *file, const encodeops *e) { int rc; - FILE *fp; - dstr d = DSTR_INIT; - dstr dd = DSTR_INIT; + hfpctx hfp; + dstr dl = DSTR_INIT; + dstr df = DSTR_INIT; unsigned long n = 0, nfail = 0; - octet *buf = xmalloc(2 * gch->hashsz); + int hf; - if (!file) - fp = stdin; - else if ((fp = fopen(file, f & f_raw ? "r" : "rb")) == 0) { + if (!file || strcmp(file, "-") == 0) + hfp.fp = stdin; + else if ((hfp.fp = fopen(file, fh->f & GSF_RAW ? "r" : "rb")) == 0) { moan("couldn't open `%s': %s", file, strerror(errno)); return (EXIT_FAILURE); } - while (DRESET(&d), dstr_putline(&d, fp) != EOF) { - char *p = d.buf; - char *q; - unsigned ff = f; - - /* --- Handle a directive --- */ - - if (*p == '#') { - p++; - if ((q = str_getword(&p)) == 0) - continue; - if (strcmp(q, "hash") == 0) { - const gchash *g; - if ((q = str_getword(&p)) == 0) - continue; - if ((g = gethash(q)) == 0) - continue; - gch = g; - xfree(buf); - buf = xmalloc(2 * gch->hashsz); - } else if (strcmp(q, "encoding") == 0) { - const encops *ee; - if ((q = str_getword(&p)) == 0) - continue; - if ((ee = getenc(q)) == 0) + hfp.dline = &dl; + hfp.dfile = &df; + hfp.hbuf = xmalloc(2 * fh->gch->hashsz); + hfp.gch = fh->gch; + hfp.ee = e; + hfp.f = fh->f; + + while ((hf = hfparse(&hfp)) != HF_EOF) { + switch (hf) { + case HF_HASH: + xfree(hfp.hbuf); + hfp.hbuf = xmalloc(2 * hfp.gch->hashsz); + break; + case HF_FILE: + if (fhash(fh, df.buf, hfp.hbuf + hfp.gch->hashsz)) { + moan("couldn't read `%s': %s", df.buf, strerror(errno)); + rc = EXIT_FAILURE; continue; - e = ee; - } else if (strcmp(q, "escape") == 0) - f |= f_escape; - continue; - } - - /* --- Otherwise it's a hex thing --- */ - - q = p; - while (*p && *p != ' ') - p++; - if (!*p) - continue; - *p++ = 0; - if (e->get(q, buf, gch->hashsz, 0) < gch->hashsz) - continue; - if (*p == '*') - ff |= f_binary; - else if (*p != ' ') - continue; - p++; - - if (f & f_escape) { - DRESET(&dd); - getstring(0, p, &dd, 0); - p = dd.buf; - } - - if (fhash(p, ff, gch, buf + gch->hashsz)) { - moan("couldn't read `%s': %s", p, strerror(errno)); - rc = EXIT_FAILURE; - continue; - } - if (memcmp(buf, buf + gch->hashsz, gch->hashsz) != 0) { - if (ff & f_verbose) - fprintf(stderr, "FAIL %s\n", p); - else - moan("%s check failed for `%s'", gch->name, p); - nfail++; - rc = EXIT_FAILURE; - } else { - if (ff & f_verbose) - fprintf(stderr, "OK %s\n", p); + } + if (memcmp(hfp.hbuf, hfp.hbuf + hfp.gch->hashsz, + hfp.gch->hashsz) != 0) { + if (hfp.f & f_verbose) + fprintf(stderr, "FAIL %s\n", df.buf); + else + moan("%s check failed for `%s'", hfp.gch->name, df.buf); + nfail++; + rc = EXIT_FAILURE; + } else { + if (hfp.f & f_verbose) + fprintf(stderr, "OK %s\n", df.buf); + } + n++; } - n++; } - dstr_destroy(&d); - dstr_destroy(&dd); - xfree(buf); - if ((f & f_verbose) && nfail) - moan("%lu of %lu file(s) failed %s check", nfail, n, gch->name); + if (ferror(hfp.fp)) { + moan("error reading input `%s': %s", + file ? file : "", strerror(errno)); + rc = EXIT_FAILURE; + } + dstr_destroy(&dl); + dstr_destroy(&df); + xfree(hfp.hbuf); + if ((fh->f & f_verbose) && nfail) + moan("%lu of %lu file(s) failed %s check", nfail, n, hfp.gch->name); else if (!n) moan("no files checked"); - return (0); + return (rc); } -static int dohash(const char *file, unsigned f, - const gchash *gch, const encops *e) +static int dohash(fhashstate *fh, const char *file, const encodeops *e) { int rc = 0; - octet *p = xmalloc(gch->hashsz); + octet *p = xmalloc(fh->gch->hashsz); - if (fhash(file, f, gch, p)) { + if (fhash(fh, file, p)) { moan("couldn't read `%s': %s", file ? file : "", strerror(errno)); rc = EXIT_FAILURE; } else { - e->put(p, gch->hashsz, stdout); + e->put(p, fh->gch->hashsz, stdout); if (file) { fputc(' ', stdout); - fputc(f & f_binary ? '*' : ' ', stdout); - if (f & f_escape) + fputc(fh->f & FHF_BINARY ? '*' : ' ', stdout); + if (fh->f & f_escape) putstring(stdout, file, 0); else fputs(file, stdout); @@ -556,54 +181,47 @@ static int dohash(const char *file, unsigned f, return (rc); } -static int dofile(const char *file, unsigned f, - const gchash *gch, const encops *e) -{ - return (f & f_check ? checkhash : dohash)(file, f, gch, e); -} +static int dofile(fhashstate *fh, const char *file, const encodeops *e) + { return (fh->f & f_check ? checkhash : dohash)(fh, file, e); } -static int hashfiles(const char *file, unsigned f, - const gchash *gch, const encops *e) +static int hashfiles(fhashstate *fh, const char *file, const encodeops *e) { FILE *fp; dstr d = DSTR_INIT; int rc = 0; int rrc; - if (!file) + if (!file || strcmp(file, "-") == 0) fp = stdin; - else if ((fp = fopen(file, f & f_raw ? "r" : "rb")) == 0) { + else if ((fp = fopen(file, fh->f & GSF_RAW ? "r" : "rb")) == 0) { moan("couldn't open `%s': %s", file, strerror(errno)); return (EXIT_FAILURE); } for (;;) { DRESET(&d); - if (getstring(fp, 0, &d, f & f_raw)) + if (getstring(fp, &d, GSF_FILE | fh->f)) break; - if ((rrc = dofile(d.buf, f, gch, e)) != 0) + if ((rrc = dofile(fh, d.buf, e)) != 0) rc = rrc; } return (rc); } -static int hashsum(const char *file, unsigned f, - const gchash *gch, const encops *e) -{ - return (f & f_files ? hashfiles : dofile)(file, f, gch, e); -} +static int hashsum(fhashstate *fh, const char *file, const encodeops *e) + { return (fh->f & f_files ? hashfiles : dofile)(fh, file, e); } /*----- Main driver -------------------------------------------------------*/ -static void version(FILE *fp) +void version(FILE *fp) { pquis(fp, "$, Catacomb version " VERSION "\n"); } static void usage(FILE *fp) { - pquis(fp, "Usage: $ [-f0ebcv] [-a algorithm] [files...]\n"); + pquis(fp, "Usage: $ [-f0ebcv] [-a ALGORITHM] [-E ENC] [FILES...]\n"); } static void help(FILE *fp, const gchash *gch) @@ -617,8 +235,10 @@ Generates or checks message digests on files. Options available:\n\ -h, --help Display this help message.\n\ -V, --version Display program's version number.\n\ -u, --usage Display a terse usage message.\n\ +-l, --list [ITEM...] Show known hash functions and/or encodings.\n\ \n\ -a, --algorithm=ALG Use the message digest algorithm ALG.\n\ +-E, --encoding=ENC Represent hashes using encoding ENC.\n\ \n\ -f, --files Read a list of file names from standard input.\n\ -0, --null File names are null terminated, not plain text.\n\ @@ -628,23 +248,30 @@ Generates or checks message digests on files. Options available:\n\ -b, --binary When reading files, treat them as binary.\n\ -v, --verbose Be verbose when checking digests.\n\ \n\ -For a list of supported message digest algorithms, type `$ --list'.\n\ +For a list of hashing algorithms and encodings, type `$ --list'.\n\ "); if (gch) fprintf(fp, "The default message digest algorithm is %s.\n", gch->name); } +#define LISTS(LI) \ + LI("Lists", list, listtab[i].name, listtab[i].name) \ + LI("Hash functions", hash, ghashtab[i], ghashtab[i]->name) \ + LI("Encodings", enc, encodingtab[i].name, encodingtab[i].name) + +MAKELISTTAB(listtab, LISTS) + int main(int argc, char *argv[]) { - unsigned f = 0; - const gchash *gch = 0; - const encops *e = &enctab[0]; + fhashstate fh; + const encodeops *e = &encodingtab[ENC_HEX]; int rc; /* --- Initialization --- */ ego(argv[0]); sub_init(); + fhash_init(&fh, 0, 0); /* --- Choose a hash function from the name --- */ @@ -653,10 +280,10 @@ int main(int argc, char *argv[]) size_t len = strlen(q); if (len > 3 && strcmp(q + len - 3, "sum") == 0) { q[len - 3] = 0; - gch = gethash(q); + fh.gch = gethash(q); } - if (!gch) - gch = gethash("md5"); + if (!fh.gch) + fh.gch = gethash("md5"); xfree(q); } @@ -679,18 +306,20 @@ int main(int argc, char *argv[]) { "escape", 0, 0, 'e' }, { "check", 0, 0, 'c' }, + { "junk", 0, 0, 'j' }, { "binary", 0, 0, 'b' }, { "verbose", 0, 0, 'v' }, + { "progress", 0, 0, 'p' }, { 0, 0, 0, 0 } }; - int i = mdwopt(argc, argv, "hVu a:E:l f0 ecbv", opts, 0, 0, 0); + int i = mdwopt(argc, argv, "hVu a:E:l f0 ecjbvp", opts, 0, 0, 0); if (i < 0) break; switch (i) { case 'h': - help(stdout, gch); + help(stdout, fh.gch); exit(0); case 'V': version(stdout); @@ -698,51 +327,49 @@ int main(int argc, char *argv[]) case 'u': usage(stdout); exit(0); + case 'l': + exit(displaylists(listtab, argv + optind)); case 'a': - if ((gch = gethash(optarg)) == 0) + if ((fh.gch = gethash(optarg)) == 0) die(EXIT_FAILURE, "unknown hash algorithm `%s'", optarg); - f |= f_oddhash; + fh.f |= f_oddhash; break; - case 'l': { - unsigned j; - for (j = 0; ghashtab[j]; j++) { - if (j) - fputc(' ', stdout); - printf("%s", ghashtab[j]->name); - } - fputc('\n', stdout); - exit(0); - } break; case 'E': - if ((e = getenc(optarg)) == 0) + if ((e = getencoding(optarg)) == 0) die(EXIT_FAILURE, "unknown encoding `%s'", optarg); - f |= f_oddenc; + fh.f |= f_oddenc; break; case 'f': - f |= f_files; + fh.f |= f_files; break; case '0': - f |= f_raw; + fh.f |= GSF_RAW; break; case 'e': - f |= f_escape; + fh.f |= f_escape; break; case 'c': - f |= f_check; + fh.f |= f_check; + break; + case 'j': + fh.f |= FHF_JUNK; break; case 'b': - f |= f_binary; + fh.f |= FHF_BINARY; break; case 'v': - f |= f_verbose; + fh.f |= f_verbose; + break; + case 'p': + fh.f |= FHF_PROGRESS; break; default: - f |= f_bogus; + fh.f |= f_bogus; break; } } - if (f & f_bogus) { + if (fh.f & f_bogus) { usage(stderr); exit(EXIT_FAILURE); } @@ -751,22 +378,32 @@ int main(int argc, char *argv[]) /* --- Generate output --- */ - if (!(f & f_check)) { - if (f & f_oddhash) printf("#hash %s\n", gch->name); - if (f & f_oddenc) printf("#encoding %s\n", e->name); - if (f & f_escape) fputs("#escape\n", stdout); + if (!(fh.f & f_check) && (argc || (fh.f & f_files))) { + if (fh.f & f_oddhash) printf("#hash %s\n", fh.gch->name); + if (fh.f & f_oddenc) printf("#encoding %s\n", e->name); + if (fh.f & f_escape) fputs("#escape\n", stdout); } - - if (argc) { + if (!argc) + rc = hashsum(&fh, 0, e); + else { int i; int rrc; + rc = 0; for (i = 0; i < argc; i++) { - if ((rrc = hashsum(argv[i], f, gch, e)) != 0) + if ((rrc = hashsum(&fh, argv[i], e)) != 0) rc = rrc; } - } else - rc = hashsum(0, f, gch, e); + } + + if (fh.f & FHF_JUNK) { + if (fh.f & f_check) { + if (fhash_junk(&fh, checkjunk, &fh)) rc = EXIT_FAILURE; + } else { + if (fhash_junk(&fh, warnjunk, 0) < 0) rc = EXIT_FAILURE; + } + } + fhash_free(&fh); return (rc); }