def test_content_id_text():
cid_t_np = iscc.content_id_text("")
assert len(cid_t_np) == 13
assert cid_t_np == "CT7A4zpmccuEv"
cid_t_p = iscc.content_id_text("", partial=True)
assert cid_t_p == "Ct7A4zpmccuEv"
assert 0 == iscc.distance(cid_t_p, cid_t_np)
cid_t_a = iscc.content_id_text(TEXT_A)
cid_t_b = iscc.content_id_text(TEXT_B)
assert iscc.distance(cid_t_a, cid_t_b) == 2
def test_hamming_distance():
a = 0b0001111
b = 0b1000111
assert iscc.distance(a, b) == 2
mid1 = iscc.meta_id("Die Unendliche Geschichte", "von Michael Ende")[0]
# Change one Character
mid2 = iscc.meta_id("Die UnXndliche Geschichte", "von Michael Ende")[0]
assert iscc.distance(mid1, mid2) <= 10
# Delete one Character
mid2 = iscc.meta_id("Die nendliche Geschichte", "von Michael Ende")[0]
assert iscc.distance(mid1, mid2) <= 14
# Add one Character
mid2 = iscc.meta_id("Die UnendlicheX Geschichte", "von Michael Ende")[0]
assert iscc.distance(mid1, mid2) <= 13
# Add, change, delete
mid2 = iscc.meta_id("Diex Unandlische Geschiche", "von Michael Ende")[0]
assert iscc.distance(mid1, mid2) <= 22
# Change Word order
mid2 = iscc.meta_id("Unendliche Geschichte, Die", "von Michael Ende")[0]
assert iscc.distance(mid1, mid2) <= 13
# Totaly different
mid2 = iscc.meta_id("Now for something different")[0]
assert iscc.distance(mid1, mid2) >= 24
def sim(a, b):
"""Estimate Similarity of ISCC Codes A & B.
Example:
$ iscc sim CCUcKwdQc1jUM CCjMmrCsKWu1D
You may also compare fully qualified ISCC Codes with each other.
"""
try:
iscc_verify(a)
iscc_verify(b)
except ValueError as e:
click.echo(str(e))
sys.exit(1)
# Fully Qualified ISCC Code Similarity
avg_msg = None
if len(iscc_clean(a)) == 52 and len(iscc_clean(b)) == 52:
digest_a = b"".join(iscc.decode(code)[1:] for code in iscc_split(a))
digest_b = b"".join(iscc.decode(code)[1:] for code in iscc_split(b))
int_a = int.from_bytes(digest_a, "big", signed=False)
int_b = int.from_bytes(digest_b, "big", signed=False)
dist = bin(int_a ^ int_b).count("1")
similarity = ((192 - dist) / 192) * 100
avg_msg = "Average Estimated Similarity: {:.2f} % ({} of 192 bits differnt)".format(
similarity, dist)
# Per Component Similarity
a = iscc_split(a)
b = iscc_split(b)
if len(a) == 1 and len(b) == 1:
type_a = ISCC_COMPONENT_CODES.get(a[0][:2])["name"]
type_b = ISCC_COMPONENT_CODES.get(b[0][:2])["name"]
if type_a != type_b:
click.echo("Incompatible component types ({} & {}).".format(
type_a, type_b))
for ca in a:
for cb in b:
type_a = ISCC_COMPONENT_CODES.get(ca[:2])["name"]
type_b = ISCC_COMPONENT_CODES.get(cb[:2])["name"]
if type_a == type_b and type_a != "Instance-ID":
hamming_dist = iscc.distance(ca, cb)
hamming_sim = 64 - hamming_dist
similarity = round(hamming_sim / (2 * 64 - hamming_sim) * 100)
click.echo(
"Estimated Similarity of {}: {:.2f} % ({} of 64 bits match)"
.format(type_a, similarity, hamming_sim))
if type_a == "Instance-ID" and type_b == "Instance-ID":
if ca == cb:
click.echo("Identical Instance-ID")
if avg_msg:
click.echo(avg_msg)
def test_meta_id():
mid1, _, _ = iscc.meta_id("ISCC Content Identifiers")
assert mid1 == "CCDFPFc87MhdT"
mid1, _, _ = iscc.meta_id(b"ISCC Content Identifiers")
assert mid1 == "CCDFPFc87MhdT"
mid1, title, extra = iscc.meta_id("Die Unendliche Geschichte")
assert mid1 == "CCAKevDpE1eEL"
assert title == "die unendliche geschichte"
assert extra == ""
mid2 = iscc.meta_id(" Die unéndlÃche, Geschichte ")[0]
assert mid1 == mid2
mid3 = iscc.meta_id("Die Unentliche Geschichte")[0]
assert iscc.distance(mid1, mid3) == 8
mid4 = iscc.meta_id("Geschichte, Die Unendliche")[0]
assert iscc.distance(mid1, mid4) == 9
with pytest.raises(UnicodeDecodeError):
iscc.meta_id(b"\xc3\x28")
def test_meta_id():
mid1, _, _ = iscc.meta_id('ISCC Content Identifiers')
assert mid1 == 'CCDGhLx6tREif'
mid1, _, _ = iscc.meta_id(b'ISCC Content Identifiers')
assert mid1 == 'CCDGhLx6tREif'
mid1, title, extra = iscc.meta_id('Die Unendliche Geschichte')
assert mid1 == "CCAZF4K1bBv8i"
assert title == 'die unendliche geschichte'
assert extra == ''
mid2 = iscc.meta_id(' Die unéndlÃche, Geschichte ')[0]
assert mid1 == mid2
mid3 = iscc.meta_id('Die Unentliche Geschichte')[0]
assert iscc.distance(mid1, mid3) == 12
mid4 = iscc.meta_id('Geschichte, Die Unendliche')[0]
assert iscc.distance(mid1, mid4) == 7
with pytest.raises(UnicodeDecodeError):
iscc.meta_id(b"\xc3\x28")
def test_content_id_image():
cid_i = iscc.content_id_image("lenna.jpg")
assert len(cid_i) == 13
assert cid_i == "CYmLoqBRgV32u"
data = BytesIO(open("lenna.jpg", "rb").read())
cid_i = iscc.content_id_image(data, partial=True)
assert len(cid_i) == 13
assert cid_i == "CimLoqBRgV32u"
img1 = Image.open("lenna.jpg")
img2 = img1.filter(ImageFilter.GaussianBlur(10))
img3 = ImageEnhance.Brightness(img1).enhance(1.4)
img4 = ImageEnhance.Contrast(img1).enhance(1.2)
cid1 = iscc.content_id_image(img1)
cid2 = iscc.content_id_image(img2)
cid3 = iscc.content_id_image(img3)
cid4 = iscc.content_id_image(img4)
assert iscc.distance(cid1, cid2) == 0
assert iscc.distance(cid1, cid3) == 2
assert iscc.distance(cid1, cid4) == 0
def test_data_id():
random.seed(1)
data = bytearray([random.getrandbits(8) for _ in range(1000000)]) # 1 mb
did_a = iscc.data_id(data)
assert did_a == "CDK2KdVAz5XTs"
data.insert(500000, 1)
data.insert(500001, 2)
data.insert(500002, 3)
did_b = iscc.data_id(data)
assert did_b == did_b
for x in range(100): # insert 100 bytes random noise
data.insert(random.randint(0, 1000000), random.randint(0, 255))
did_c = iscc.data_id(data)
assert iscc.distance(did_a, did_c) == 17