/* -*-c-*-
*
- * $Id: hashsum.c,v 1.8 2001/04/19 18:26:33 mdw Exp $
+ * $Id$
*
* Hash files using some secure hash function
*
* MA 02111-1307, USA.
*/
-/*----- Revision history --------------------------------------------------*
- *
- * $Log: hashsum.c,v $
- * Revision 1.8 2001/04/19 18:26:33 mdw
- * Add CRC as another hash function.
- *
- * Revision 1.7 2001/02/21 20:03:22 mdw
- * Added support for MD2 hash function.
- *
- * Revision 1.6 2001/01/25 21:40:14 mdw
- * Support for new SHA variants added.
- *
- * Revision 1.5 2000/12/06 20:33:27 mdw
- * Make flags be macros rather than enumerations, to ensure that they're
- * unsigned.
- *
- * Revision 1.4 2000/08/04 23:23:44 mdw
- * Various <ctype.h> fixes.
- *
- * Revision 1.3 2000/07/29 17:02:43 mdw
- * (checkhash): Be pettier about spaces between the hash and filename, for
- * compatiblity with `md5sum'.
- *
- * Revision 1.2 2000/07/15 21:14:05 mdw
- * Missed `-e' out of the usage string.
- *
- * Revision 1.1 2000/07/15 20:52:34 mdw
- * Useful replacement for `md5sum' with support for many different hash
- * functions and for reading filename lists from `find'.
- *
- */
-
/*----- Header files ------------------------------------------------------*/
#include "config.h"
+#include <assert.h>
#include <ctype.h>
#include <errno.h>
#include <stdio.h>
#include <mLib/sub.h>
#include <mLib/str.h>
-#include "ghash.h"
+#include <mLib/hex.h>
+#include <mLib/base32.h>
+#include <mLib/base64.h>
-#include "crc32.h"
-#include "md2.h"
-#include "md4.h"
-#include "md5.h"
-#include "rmd128.h"
-#include "rmd160.h"
-#include "rmd256.h"
-#include "rmd320.h"
-#include "sha.h"
-#include "sha256.h"
-#include "sha384.h"
-#include "sha512.h"
-#include "tiger.h"
+#include "ghash.h"
+#include "cc.h"
/*----- Static variables --------------------------------------------------*/
-static const gchash *hashtab[] = {
- &md5, &md4, &md2,
- &sha, &sha256, &sha384, &sha512,
- &rmd128, &rmd160, &rmd256, &rmd320,
- &tiger, &gcrc32,
- 0
-};
-
#define f_binary 1u
#define f_bogus 2u
#define f_verbose 4u
#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 --- */
+
+typedef struct encodeops {
+ const char *name;
+ void (*put)(const octet *, size_t, FILE *);
+ size_t (*get)(const char *, octet *, size_t, char **);
+} encodeops;
+
+static const encodeops encodingtab[] = {
+ { "hex", puthex, gethex },
+ { "base64", putb64, getb64 },
+ { "base32", putb32, getb32 },
+ { 0, 0, 0 }
+};
+
+static const encodeops *getencoding(const char *ename)
+{
+ const encodeops *e;
+
+ for (e = encodingtab; e->name; e++) {
+ if (strcmp(ename, e->name) == 0)
+ return (e);
+ }
+ return (0);
+}
/*----- Support functions -------------------------------------------------*/
ghash *h;
int e;
- if (!file)
+ if (!file || strcmp(file, "-") == 0)
fp = stdin;
else if ((fp = fopen(file, f & f_binary ? "rb" : "r")) == 0)
return (-1);
- h = gch->init();
+ h = GH_INIT(gch);
while ((sz = fread(fbuf, 1, sizeof(fbuf), fp)) > 0)
- h->ops->hash(h, fbuf, sz);
- h->ops->done(h, buf);
- h->ops->destroy(h);
+ GH_HASH(h, fbuf, sz);
+ GH_DONE(h, buf);
+ GH_DESTROY(h);
e = ferror(fp);
if (file)
fclose(fp);
return (e ? -1 : 0);
}
-/* --- @puthex@ --- *
- *
- * Arguments: @const octet *buf@ = pointer to a binary buffer
- * @size_t sz@ = size of the buffer
- * @FILE *fp@ = pointer to output file handle
- *
- * Returns: ---
- *
- * Use: Writes a hex dump of a block of memory.
- */
-
-static void puthex(const octet *buf, size_t sz, FILE *fp)
-{
- while (sz) {
- fprintf(fp, "%02x", *buf++);
- sz--;
- }
-}
-
-/* --- @gethex@ --- *
- *
- * Arguments: @const char *p@ = pointer to input string
- * @octet *q@ = pointer to output buffer
- * @size_t sz@ = size of the output buffer
- * @char **pp@ = where to put the end pointer
- *
- * Returns: The number of bytes written to the buffer.
- *
- * Use: Reads hex dumps from the input string.
- */
-
-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);
-}
-
/* --- @gethash@ --- *
*
* Arguments: @const char *name@ = pointer to name string
static const gchash *gethash(const char *name)
{
- const gchash **g, *gg = 0;
+ const gchash *const *g, *gg = 0;
size_t sz = strlen(name);
- for (g = hashtab; *g; g++) {
+ for (g = ghashtab; *g; g++) {
if (strncmp(name, (*g)->name, sz) == 0) {
if ((*g)->name[sz] == 0) {
gg = *g;
/*----- Guts --------------------------------------------------------------*/
-static int checkhash(const char *file, unsigned f, const gchash *gch)
+static int checkhash(const char *file, unsigned f,
+ const gchash *gch, const encodeops *e)
{
int rc;
FILE *fp;
unsigned long n = 0, nfail = 0;
octet *buf = xmalloc(2 * gch->hashsz);
- if (!file)
+ if (!file || strcmp(file, "-") == 0)
fp = stdin;
else if ((fp = fopen(file, f & f_raw ? "r" : "rb")) == 0) {
moan("couldn't open `%s': %s", file, strerror(errno));
gch = g;
xfree(buf);
buf = xmalloc(2 * gch->hashsz);
+ } else if (strcmp(q, "encoding") == 0) {
+ const encodeops *ee;
+ if ((q = str_getword(&p)) == 0)
+ continue;
+ if ((ee = getencoding(q)) == 0)
+ continue;
+ e = ee;
} else if (strcmp(q, "escape") == 0)
f |= f_escape;
continue;
if (!*p)
continue;
*p++ = 0;
- if (gethex(q, buf, gch->hashsz, 0) < gch->hashsz)
+ if (e->get(q, buf, gch->hashsz, 0) < gch->hashsz)
continue;
if (*p == '*')
ff |= f_binary;
return (0);
}
-static int dohash(const char *file, unsigned f, const gchash *gch)
+static int dohash(const char *file, unsigned f,
+ const gchash *gch, const encodeops *e)
{
int rc = 0;
octet *p = xmalloc(gch->hashsz);
moan("couldn't read `%s': %s", file ? file : "<stdin>", strerror(errno));
rc = EXIT_FAILURE;
} else {
- puthex(p, gch->hashsz, stdout);
+ e->put(p, gch->hashsz, stdout);
if (file) {
fputc(' ', stdout);
fputc(f & f_binary ? '*' : ' ', stdout);
return (rc);
}
-static int dofile(const char *file, unsigned f, const gchash *gch)
+static int dofile(const char *file, unsigned f,
+ const gchash *gch, const encodeops *e)
{
- return (f & f_check ? checkhash : dohash)(file, f, gch);
+ return (f & f_check ? checkhash : dohash)(file, f, gch, e);
}
-static int hashfiles(const char *file, unsigned f, const gchash *gch)
+static int hashfiles(const char *file, unsigned f,
+ const gchash *gch, 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) {
moan("couldn't open `%s': %s", file, strerror(errno));
DRESET(&d);
if (getstring(fp, 0, &d, f & f_raw))
break;
- if ((rrc = dofile(d.buf, f, gch)) != 0)
+ if ((rrc = dofile(d.buf, f, gch, e)) != 0)
rc = rrc;
}
return (rc);
}
-static int hashsum(const char *file, unsigned f, const gchash *gch)
+static int hashsum(const char *file, unsigned f,
+ const gchash *gch, const encodeops *e)
{
- return (f & f_files ? hashfiles : dofile)(file, f, gch);
+ return (f & f_files ? hashfiles : dofile)(file, f, gch, 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)
-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\
-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 encodeops *e = &encodingtab[0];
int rc;
/* --- Initialization --- */
gch = gethash(q);
}
if (!gch)
- gch = hashtab[0];
+ gch = gethash("md5");
xfree(q);
}
{ "algorithm", OPTF_ARGREQ, 0, 'a' },
{ "hash", OPTF_ARGREQ, 0, 'a' },
+ { "encoding", OPTF_ARGREQ, 0, 'E' },
{ "list", 0, 0, 'l' },
{ "files", 0, 0, 'f' },
{ 0, 0, 0, 0 }
};
- int i = mdwopt(argc, argv, "hVu a:l f0 ecbv", opts, 0, 0, 0);
+ int i = mdwopt(argc, argv, "hVu a:E:l f0 ecbv", opts, 0, 0, 0);
if (i < 0)
break;
case 'u':
usage(stdout);
exit(0);
+ case 'l':
+ exit(displaylists(listtab, argv + optind));
case 'a':
if ((gch = gethash(optarg)) == 0)
die(EXIT_FAILURE, "unknown hash algorithm `%s'", optarg);
f |= f_oddhash;
break;
- case 'l': {
- unsigned j;
- for (j = 0; hashtab[j]; j++) {
- if (j)
- fputc(' ', stdout);
- printf("%s", hashtab[j]->name);
- }
- fputc('\n', stdout);
- exit(0);
- } break;
+ case 'E':
+ if ((e = getencoding(optarg)) == 0)
+ die(EXIT_FAILURE, "unknown encoding `%s'", optarg);
+ f |= f_oddenc;
+ break;
case 'f':
f |= f_files;
break;
/* --- Generate output --- */
- if (!(f & f_check)) {
- if (f & f_oddhash)
- printf("#hash %s\n", gch->name);
- if (f & f_escape)
- fputs("#escape\n", stdout);
- }
-
- if (argc) {
+ if (!argc)
+ rc = hashsum(0, f, gch, e);
+ else {
int i;
int rrc;
+
rc = 0;
+ 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);
+ }
for (i = 0; i < argc; i++) {
- if ((rrc = hashsum(argv[i], f, gch)) != 0)
+ if ((rrc = hashsum(argv[i], f, gch, e)) != 0)
rc = rrc;
}
- } else
- rc = hashsum(0, f, gch);
+ }
return (rc);
}