void media_identifier::identify_file(const char *name)
{
std::vector<file_info> info;
digest_file(info, name);
match_hashes(info);
print_results(info);
}
int
main(int argc, char **argv)
{
alg_t *alg;
int rval;
int i;
#ifdef HAVE_SETLOCALE
(void) setlocale(LC_ALL, "");
#endif
while ((i = getopt(argc, argv, "V")) != -1) {
switch(i) {
case 'V':
printf("%s\n", VERSION);
return EXIT_SUCCESS;
}
}
argc -= optind;
argv += optind;
if (argc == 0) {
(void) fprintf(stderr, "Usage: %s algorithm [file...]\n",
argv[-optind]);
return EXIT_FAILURE;
}
if ((alg = find_algorithm(argv[0])) == NULL) {
(void) fprintf(stderr, "No such algorithm `%s'\n", argv[0]);
exit(EXIT_FAILURE);
}
argc--;
argv++;
rval = EXIT_SUCCESS;
if (argc == 0) {
if (!digest_file(NULL, alg)) {
(void) fprintf(stderr, "stdin\n");
rval = EXIT_FAILURE;
}
} else {
for (i = 0 ; i < argc ; i++) {
if (!digest_file(argv[i], alg)) {
(void) fprintf(stderr, "%s\n", argv[i]);
rval = EXIT_FAILURE;
}
}
}
return rval;
}
JNIEXPORT jbyteArray JNICALL Java_org_devtcg_demo_jnitest_NativeMD5_digestFile
(JNIEnv *env, jclass c, jstring in)
{
const char *inname;
jbyteArray ret = NULL;
if ((inname = (*env)->GetStringUTFChars(env, in, NULL)) != NULL)
{
ret = digest_file(env, inname);
(*env)->ReleaseStringUTFChars(env, in, inname);
}
return ret;
}
int
main (int argc, char **argv)
{
if (argc != 2) {
printf ("usage: gen-md.test file_to_process\n");
return 1;
}
mono_init_version ("gen-md-test", "v2.0.50727");
mono_marshal_init ();
digest_file (argv [1]);
return 0;
}
//Digests the file then encrypts the digest with the supplied key
//length of verification will be stored in len
unsigned char *create_file_verification(unsigned char *key, char *filename, int *len)
{
//printf("Creating file verification for %s\n", filename);
unsigned char *digest = digest_file(filename, len);
unsigned char *out = NULL;
//printf("%s DIGEST: ", filename);
print_hash(digest, *len);
if(digest != NULL) {
out = blowfish_enc(key, digest, *len);
//printf("Enc Digest: ");
print_hash(out, *len);
free(digest);
}
return out;
}
/*
* Calculate the chksum of a whole file and updates:
* - digest
* - digest_stream
* - digest_buffer
* - digest_name
*
* Returns: true if digest calculation succeeded.
* false if digest calculation failed.
*/
static bool calculate_file_chksum(JCR *jcr, FF_PKT *ff_pkt, DIGEST **digest,
int *digest_stream, char **digest_buf, const char **digest_name)
{
/*
* Create our digest context.
* If this fails, the digest will be set to NULL and not used.
*/
if (ff_pkt->flags & FO_MD5) {
*digest = crypto_digest_new(jcr, CRYPTO_DIGEST_MD5);
*digest_stream = STREAM_MD5_DIGEST;
} else if (ff_pkt->flags & FO_SHA1) {
*digest = crypto_digest_new(jcr, CRYPTO_DIGEST_SHA1);
*digest_stream = STREAM_SHA1_DIGEST;
} else if (ff_pkt->flags & FO_SHA256) {
*digest = crypto_digest_new(jcr, CRYPTO_DIGEST_SHA256);
*digest_stream = STREAM_SHA256_DIGEST;
} else if (ff_pkt->flags & FO_SHA512) {
*digest = crypto_digest_new(jcr, CRYPTO_DIGEST_SHA512);
*digest_stream = STREAM_SHA512_DIGEST;
}
/*
* compute MD5 or SHA1 hash
*/
if (*digest) {
uint32_t size;
char md[CRYPTO_DIGEST_MAX_SIZE];
size = sizeof(md);
if (digest_file(jcr, ff_pkt, *digest) != 0) {
jcr->JobErrors++;
return false;
}
if (crypto_digest_finalize(*digest, (uint8_t *)md, &size)) {
*digest_buf = (char *)malloc(BASE64_SIZE(size));
*digest_name = crypto_digest_name(*digest);
bin_to_base64(*digest_buf, BASE64_SIZE(size), md, size, true);
}
}
return true;
}
/*
* Called here by find() for each file.
*
* Find the file, compute the MD5 or SHA1 and send it back to the Director
*/
static int verify_file(JCR *jcr, FF_PKT *ff_pkt, bool top_level)
{
POOL_MEM attribs(PM_NAME),
attribsEx(PM_NAME);
int digest_stream = STREAM_NONE;
int status;
DIGEST *digest = NULL;
BSOCK *dir;
if (job_canceled(jcr)) {
return 0;
}
dir = jcr->dir_bsock;
jcr->num_files_examined++; /* bump total file count */
switch (ff_pkt->type) {
case FT_LNKSAVED: /* Hard linked, file already saved */
Dmsg2(30, "FT_LNKSAVED saving: %s => %s\n", ff_pkt->fname, ff_pkt->link);
break;
case FT_REGE:
Dmsg1(30, "FT_REGE saving: %s\n", ff_pkt->fname);
break;
case FT_REG:
Dmsg1(30, "FT_REG saving: %s\n", ff_pkt->fname);
break;
case FT_LNK:
Dmsg2(30, "FT_LNK saving: %s -> %s\n", ff_pkt->fname, ff_pkt->link);
break;
case FT_DIRBEGIN:
jcr->num_files_examined--; /* correct file count */
return 1; /* ignored */
case FT_REPARSE:
case FT_JUNCTION:
case FT_DIREND:
Dmsg1(30, "FT_DIR saving: %s\n", ff_pkt->fname);
break;
case FT_SPEC:
Dmsg1(30, "FT_SPEC saving: %s\n", ff_pkt->fname);
break;
case FT_RAW:
Dmsg1(30, "FT_RAW saving: %s\n", ff_pkt->fname);
break;
case FT_FIFO:
Dmsg1(30, "FT_FIFO saving: %s\n", ff_pkt->fname);
break;
case FT_NOACCESS: {
berrno be;
be.set_errno(ff_pkt->ff_errno);
Jmsg(jcr, M_NOTSAVED, 1, _(" Could not access %s: ERR=%s\n"), ff_pkt->fname, be.bstrerror());
jcr->JobErrors++;
return 1;
}
case FT_NOFOLLOW: {
berrno be;
be.set_errno(ff_pkt->ff_errno);
Jmsg(jcr, M_NOTSAVED, 1, _(" Could not follow link %s: ERR=%s\n"), ff_pkt->fname, be.bstrerror());
jcr->JobErrors++;
return 1;
}
case FT_NOSTAT: {
berrno be;
be.set_errno(ff_pkt->ff_errno);
Jmsg(jcr, M_NOTSAVED, 1, _(" Could not stat %s: ERR=%s\n"), ff_pkt->fname, be.bstrerror());
jcr->JobErrors++;
return 1;
}
case FT_DIRNOCHG:
case FT_NOCHG:
Jmsg(jcr, M_SKIPPED, 1, _(" Unchanged file skipped: %s\n"), ff_pkt->fname);
return 1;
case FT_ISARCH:
Jmsg(jcr, M_SKIPPED, 1, _(" Archive file skipped: %s\n"), ff_pkt->fname);
return 1;
case FT_NORECURSE:
Jmsg(jcr, M_SKIPPED, 1, _(" Recursion turned off. Directory skipped: %s\n"), ff_pkt->fname);
ff_pkt->type = FT_DIREND; /* directory entry was backed up */
break;
case FT_NOFSCHG:
Jmsg(jcr, M_SKIPPED, 1, _(" File system change prohibited. Directory skipped: %s\n"), ff_pkt->fname);
return 1;
case FT_PLUGIN_CONFIG:
case FT_RESTORE_FIRST:
return 1; /* silently skip */
case FT_NOOPEN: {
berrno be;
be.set_errno(ff_pkt->ff_errno);
Jmsg(jcr, M_NOTSAVED, 1, _(" Could not open directory %s: ERR=%s\n"), ff_pkt->fname, be.bstrerror());
jcr->JobErrors++;
return 1;
}
default:
Jmsg(jcr, M_NOTSAVED, 0, _(" Unknown file type %d: %s\n"), ff_pkt->type, ff_pkt->fname);
jcr->JobErrors++;
return 1;
}
/* Encode attributes and possibly extend them */
encode_stat(attribs.c_str(), &ff_pkt->statp, sizeof(ff_pkt->statp), ff_pkt->LinkFI, 0);
encode_attribsEx(jcr, attribsEx.c_str(), ff_pkt);
jcr->lock();
jcr->JobFiles++; /* increment number of files sent */
pm_strcpy(jcr->last_fname, ff_pkt->fname);
jcr->unlock();
/*
* Send file attributes to Director
* File_index
* Stream
* Verify Options
* Filename (full path)
* Encoded attributes
* Link name (if type==FT_LNK)
* For a directory, link is the same as fname, but with trailing
* slash. For a linked file, link is the link.
*/
/* Send file attributes to Director (note different format than for Storage) */
Dmsg2(400, "send ATTR inx=%d fname=%s\n", jcr->JobFiles, ff_pkt->fname);
if (ff_pkt->type == FT_LNK || ff_pkt->type == FT_LNKSAVED) {
status = dir->fsend("%d %d %s %s%c%s%c%s%c", jcr->JobFiles,
STREAM_UNIX_ATTRIBUTES, ff_pkt->VerifyOpts, ff_pkt->fname,
0, attribs.c_str(), 0, ff_pkt->link, 0);
} else if (ff_pkt->type == FT_DIREND || ff_pkt->type == FT_REPARSE ||
ff_pkt->type == FT_JUNCTION) {
/* Here link is the canonical filename (i.e. with trailing slash) */
status = dir->fsend("%d %d %s %s%c%s%c%c", jcr->JobFiles,
STREAM_UNIX_ATTRIBUTES, ff_pkt->VerifyOpts, ff_pkt->link,
0, attribs.c_str(), 0, 0);
} else {
status = dir->fsend("%d %d %s %s%c%s%c%c", jcr->JobFiles,
STREAM_UNIX_ATTRIBUTES, ff_pkt->VerifyOpts, ff_pkt->fname,
0, attribs.c_str(), 0, 0);
}
Dmsg2(20, "filed>dir: attribs len=%d: msg=%s\n", dir->msglen, dir->msg);
if (!status) {
Jmsg(jcr, M_FATAL, 0, _("Network error in send to Director: ERR=%s\n"), bnet_strerror(dir));
return 0;
}
/*
* The remainder of the function is all about getting the checksum.
* First we initialise, then we read files, other streams and Finder Info.
*/
if (ff_pkt->type != FT_LNKSAVED && (S_ISREG(ff_pkt->statp.st_mode) &&
ff_pkt->flags & (FO_MD5|FO_SHA1|FO_SHA256|FO_SHA512))) {
/*
* Create our digest context. If this fails, the digest will be set to NULL
* and not used.
*/
if (ff_pkt->flags & FO_MD5) {
digest = crypto_digest_new(jcr, CRYPTO_DIGEST_MD5);
digest_stream = STREAM_MD5_DIGEST;
} else if (ff_pkt->flags & FO_SHA1) {
digest = crypto_digest_new(jcr, CRYPTO_DIGEST_SHA1);
digest_stream = STREAM_SHA1_DIGEST;
} else if (ff_pkt->flags & FO_SHA256) {
digest = crypto_digest_new(jcr, CRYPTO_DIGEST_SHA256);
digest_stream = STREAM_SHA256_DIGEST;
} else if (ff_pkt->flags & FO_SHA512) {
digest = crypto_digest_new(jcr, CRYPTO_DIGEST_SHA512);
digest_stream = STREAM_SHA512_DIGEST;
}
/* Did digest initialization fail? */
if (digest_stream != STREAM_NONE && digest == NULL) {
Jmsg(jcr, M_WARNING, 0, _("%s digest initialization failed\n"),
stream_to_ascii(digest_stream));
}
/* compute MD5 or SHA1 hash */
if (digest) {
char md[CRYPTO_DIGEST_MAX_SIZE];
uint32_t size;
size = sizeof(md);
if (digest_file(jcr, ff_pkt, digest) != 0) {
jcr->JobErrors++;
goto good_rtn;
}
if (crypto_digest_finalize(digest, (uint8_t *)md, &size)) {
char *digest_buf;
const char *digest_name;
digest_buf = (char *)malloc(BASE64_SIZE(size));
digest_name = crypto_digest_name(digest);
bin_to_base64(digest_buf, BASE64_SIZE(size), md, size, true);
Dmsg3(400, "send inx=%d %s=%s\n", jcr->JobFiles, digest_name, digest_buf);
dir->fsend("%d %d %s *%s-%d*", jcr->JobFiles, digest_stream, digest_buf,
digest_name, jcr->JobFiles);
Dmsg3(20, "filed>dir: %s len=%d: msg=%s\n", digest_name,
dir->msglen, dir->msg);
free(digest_buf);
}
}
}
good_rtn:
if (digest) {
crypto_digest_free(digest);
}
return 1;
}
void media_identifier::collect_files(std::vector<file_info> &info, char const *path)
{
// first try to open as a directory
osd::directory::ptr const directory = osd::directory::open(path);
if (directory)
{
// iterate over all files in the directory
for (osd::directory::entry const *entry = directory->read(); entry; entry = directory->read())
{
if (entry->type == osd::directory::entry::entry_type::FILE)
{
std::string const curfile = std::string(path).append(PATH_SEPARATOR).append(entry->name);
collect_files(info, curfile.c_str());
}
}
}
else if (core_filename_ends_with(path, ".7z") || core_filename_ends_with(path, ".zip"))
{
// first attempt to examine it as a valid zip/7z file
util::archive_file::ptr archive;
util::archive_file::error err;
if (core_filename_ends_with(path, ".7z"))
err = util::archive_file::open_7z(path, archive);
else
err = util::archive_file::open_zip(path, archive);
if ((util::archive_file::error::NONE == err) && archive)
{
std::vector<std::uint8_t> data;
// loop over entries in the .7z, skipping empty files and directories
for (int i = archive->first_file(); i >= 0; i = archive->next_file())
{
std::uint64_t const length(archive->current_uncompressed_length());
if (!archive->current_is_directory() && length)
{
std::string const curfile = std::string(path).append(PATH_SEPARATOR).append(archive->current_name());
if (std::uint32_t(length) == length)
{
// decompress data into RAM and identify it
try
{
data.resize(std::size_t(length));
err = archive->decompress(&data[0], std::uint32_t(length));
if (util::archive_file::error::NONE == err)
digest_data(info, curfile.c_str(), &data[0], length);
else
osd_printf_error("%s: error decompressing file\n", curfile.c_str());
}
catch (...)
{
// resizing the buffer could cause a bad_alloc if archive contains large files
osd_printf_error("%s: error decompressing file\n", curfile.c_str());
}
data.clear();
}
else
{
osd_printf_error("%s: file too large to decompress into memory\n", curfile.c_str());
}
}
}
}
else
{
osd_printf_error("%s: error opening archive\n", path);
}
// clear out any cached files
util::archive_file::cache_clear();
}
else
{
// otherwise, identify as a raw file
digest_file(info, path);
}
}