def from_bytes(cidbytes):
"""
Creates a CID object from a encoded form
:param bytes cidbytes: can be
- base58-encoded multihash
- multihash
- multibase-encoded multihash
:return: a CID object
:rtype: :py:class:`cid.CIDv0` or :py:class:`cid.CIDv1`
"""
if multibase.is_encoded(cidbytes):
# if the bytestream is multibase encoded
cid = multibase.decode(cidbytes)
data = cid[1:]
version = int(cid[0])
codec = multicodec.get_codec(data)
multihash = multicodec.remove_prefix(data)
return make_cid(version, codec, multihash)
elif cidbytes[0] in (0, 1):
# if the bytestream is a CID
version = cidbytes[0]
data = cidbytes[1:]
codec = multicodec.get_codec(data)
multihash = multicodec.remove_prefix(data)
return make_cid(version, codec, multihash)
else:
# otherwise its just base58-encoded multihash
return make_cid(0, CIDv0.CODEC, base58.b58decode(cidbytes))
def from_bytes(cidbytes):
"""
Creates a CID object from a encoded form
:param bytes cidbytes: can be
- base58-encoded multihash
- multihash
- multibase-encoded multihash
:return: a CID object
:rtype: :py:class:`cid.CIDv0` or :py:class:`cid.CIDv1`
:raises: `ValueError` if the base58-encoded string is not a valid string
:raises: `ValueError` if the length of the argument is zero
:raises: `ValueError` if the length of decoded CID is invalid
"""
if len(cidbytes) < 2:
raise ValueError('argument length can not be zero')
# first byte for identity multibase and CIDv0 is 0x00
# putting in assumption that multibase for CIDv0 can not be identity
# refer: https://github.com/ipld/cid/issues/13#issuecomment-326490275
if cidbytes[0] != 0 and multibase.is_encoded(cidbytes):
# if the bytestream is multibase encoded
cid = multibase.decode(cidbytes)
if len(cid) < 2:
raise ValueError('cid length is invalid')
data = cid[1:]
version = int(cid[0])
codec = multicodec.get_codec(data)
multihash = multicodec.remove_prefix(data)
elif cidbytes[0] in (0, 1):
# if the bytestream is a CID
version = cidbytes[0]
data = cidbytes[1:]
codec = multicodec.get_codec(data)
multihash = multicodec.remove_prefix(data)
else:
# otherwise its just base58-encoded multihash
try:
version = 0
codec = CIDv0.CODEC
multihash = base58.b58decode(cidbytes)
except ValueError:
raise ValueError(
'multihash is not a valid base58 encoded multihash')
try:
mh.decode(multihash)
except ValueError:
raise
return make_cid(version, codec, multihash)
def test_verify_prefix_complete(multicodec, prefix):
data = b'testbytesbuffer'
prefix_int = prefix['prefix']
prefixed_data = add_prefix(multicodec, data)
assert is_codec(multicodec)
assert get_codec(prefixed_data) == multicodec
assert remove_prefix(prefixed_data) == data
assert extract_prefix(prefixed_data) == prefix_int
def from_bytes(cidbytes):
"""
Creates a CID object from a encoded form
:param bytes cidbytes: can be
- base58-encoded multihash
- multihash
- multibase-encoded multihash
:return: a CID object
:rtype: :py:class:`cid.CIDv0` or :py:class:`cid.CIDv1`
:raises: `ValueError` if the base58-encoded string is not a valid string
"""
if multibase.is_encoded(cidbytes):
# if the bytestream is multibase encoded
cid = multibase.decode(cidbytes)
data = cid[1:]
version = int(cid[0])
codec = multicodec.get_codec(data)
multihash = multicodec.remove_prefix(data)
elif cidbytes[0] in (b'0', b'1'):
# if the bytestream is a CID
version = cidbytes[0]
data = cidbytes[1:]
codec = multicodec.get_codec(data)
multihash = multicodec.remove_prefix(data)
else:
# otherwise its just base58-encoded multihash
try:
version = 0
codec = CIDv0.CODEC
multihash = base58.b58decode(cidbytes)
except ValueError:
raise ValueError(
'multihash is not a valid base58 encoded multihash')
try:
mh.decode(multihash)
except ValueError:
raise
return make_cid(version, codec, multihash)
def get_codec(chash):
"""
Extract the codec of a content hash
:param str hash: a hex string containing a content hash
:return: the extracted codec
:rtype: str
"""
buffer = multihash.from_hex_string(chash.lstrip('0x'))
return multicodec.get_codec(buffer)
def decode(chash):
"""
Decode a content hash.
:param str hash: a hex string containing a content hash
:return: the decoded content
:rtype: str
"""
buffer = multihash.from_hex_string(chash.lstrip('0x'))
codec = multicodec.get_codec(buffer)
value = multicodec.remove_prefix(buffer)
profile = get_profile(codec)
return profile.decode(value)
def test_get_codec_invalid_prefix(_, prefix):
prefix_bytes = varint.encode(prefix)
with pytest.raises(ValueError) as excinfo:
get_codec(prefix_bytes)
assert 'not present in the lookup table' in str(excinfo.value)