def test_compute(self):
raw_hdr, hdr = mp3_header_test.VALID_MP3_HEADERS.items()[0]
frame_data = raw_hdr + ("x" * (hdr.frame_size - len(raw_hdr)))
for i, seq in enumerate((
[ frame_data ],
[ "junk", frame_data ],
[ frame_data, "junk" ],
[ "junk", frame_data, "junk" ],
[ frame_data, frame_data ],
[ "junk", frame_data, frame_data ],
[ frame_data, "junk", frame_data ],
[ frame_data, frame_data, "junk" ],
[ "junk", frame_data, "junk", frame_data ],
[ frame_data, "junk", frame_data, "junk" ],
[ "junk", frame_data, frame_data, "junk" ],
[ "junk", frame_data, "junk", frame_data, "junk" ],
)):
sha1_calc = hashlib.sha1()
for data in seq:
if data == frame_data:
sha1_calc.update(data)
expected_fingerprint = sha1_calc.hexdigest()
stream = cStringIO.StringIO(''.join(seq))
actual_fingerprint = fingerprint.compute(stream)
self.assertEqual(expected_fingerprint, actual_fingerprint,
msg="Case #%d failed" % i)
self.assertTrue(fingerprint.is_valid(actual_fingerprint))
# We return None if we cannot find any valid frames.
stream = cStringIO.StringIO('no valid MPEG frames')
self.assertTrue(fingerprint.compute(stream) is None)
def test_crawling(self):
crawl = crawler.Crawler()
crawl.add_root(os.path.join(TESTDATA, "test_tree"))
seen_files = set()
for au_file in crawl:
# Each file has the filename in TPE1.
tpe1 = au_file.mutagen_id3["TPE1"]
self.assertTrue(au_file.path.endswith(tpe1.text[0]))
# There should be a header.
self.assertTrue(au_file.mp3_header is not None)
# Make sure the fingerprint returned by the crawler matches
# what is returned by fingerprint.compute().
f_in = open(au_file.path)
computed_fp = fingerprint.compute(f_in)
self.assertEqual(computed_fp, au_file.fingerprint)
# Each test file has the fingerprint in a UFID tag.
# Make sure it matches.
ufid = au_file.mutagen_id3["UFID:test"]
self.assertEqual(ufid.data, au_file.fingerprint)
# TODO(trow): We should have a better test of the frame count
# and frame size.
self.assertTrue(au_file.frame_count > 0)
self.assertTrue(au_file.frame_size > 0)
# Remember the basename of each file that we see.
seen_files.add(os.path.basename(au_file.path))
# Make sure we've seen all of the files we expect.
self.assertEqual(15, len(seen_files))
for a in ("A", "B", "C"):
for b in range(5):
self.assertTrue("%s%d.mp3" % (a, b) in seen_files)
# Make sure we've seen the directories we expect.
self.assertEqual(
[os.path.join(TESTDATA, "test_tree", x) for x in ("A", "B", "C")],
sorted(crawl.directories_seen))
# We should have skipped one file.
self.assertEqual(1, len(crawl.skipped_files))
self.assertEqual(
os.path.join(TESTDATA, "test_tree/A/invalid_file.mp3"),
crawl.skipped_files[0][0])
def test_crawling(self):
crawl = crawler.Crawler()
crawl.add_root(os.path.join(TESTDATA, "test_tree"))
seen_files = set()
for au_file in crawl:
# Each file has the filename in TPE1.
tpe1 = au_file.mutagen_id3["TPE1"]
self.assertTrue(au_file.path.endswith(tpe1.text[0]))
# There should be a header.
self.assertTrue(au_file.mp3_header is not None)
# Make sure the fingerprint returned by the crawler matches
# what is returned by fingerprint.compute().
f_in = open(au_file.path)
computed_fp = fingerprint.compute(f_in)
self.assertEqual(computed_fp, au_file.fingerprint)
# Each test file has the fingerprint in a UFID tag.
# Make sure it matches.
ufid = au_file.mutagen_id3["UFID:test"]
self.assertEqual(ufid.data, au_file.fingerprint)
# TODO(trow): We should have a better test of the frame count
# and frame size.
self.assertTrue(au_file.frame_count > 0)
self.assertTrue(au_file.frame_size > 0)
# Remember the basename of each file that we see.
seen_files.add(os.path.basename(au_file.path))
# Make sure we've seen all of the files we expect.
self.assertEqual(15, len(seen_files))
for a in ("A", "B", "C"):
for b in range(5):
self.assertTrue("%s%d.mp3" % (a, b) in seen_files)
# Make sure we've seen the directories we expect.
self.assertEqual(
[os.path.join(TESTDATA, "test_tree", x) for x in ("A", "B", "C")], sorted(crawl.directories_seen)
)
# We should have skipped one file.
self.assertEqual(1, len(crawl.skipped_files))
self.assertEqual(os.path.join(TESTDATA, "test_tree/A/invalid_file.mp3"), crawl.skipped_files[0][0])