Api implementation #2
Conversation
peerdid/peer_did.py
Outdated
| if peer_did[9] == '0': | ||
| inception_key = peer_did[11:] | ||
| decoded_encnumbasis = decode_encnumbasis(inception_key, peer_did) | ||
| did_doc = { |
There was a problem hiding this comment.
I would move DID DOC building into a separate helper method
| self.did_doc_file_extension = did_doc_file_extension or self.DEFAULT_PEERDID_FILE_EXTENSION | ||
|
|
||
| def save(self, data: bytes): | ||
| """ | ||
| Saves data to a storage | ||
| """ | ||
| pass | ||
| file_path = os.path.join(self.peer_did_storage_folder, self.peer_did_filename + self.did_doc_file_extension) | ||
| with open(file_path, "bw+") as f: |
There was a problem hiding this comment.
Do we expect to append or override existing file?
| decoded_encnumbasises = [decode_encnumbasis(key, peer_did) for key in keys_without_purpose_code] | ||
| decoded_service = [decode_service(service, peer_did) for service in services] | ||
| did_doc = { | ||
| 'id': peer_did, |
There was a problem hiding this comment.
It makes sense to create a special entity DIDDoc (in did_doc.py) as follows
class DIDDoc(NamedTuple):
id: str
authentication: List[PublicKeyTypeAuthentication]
key_agreement: List[PublicKeyTypeAgreement]
service: str
def to_json(self) -> str:
....
| """ | ||
| if peer_did is None: | ||
| return False | ||
| peer_did_pattern = re.compile(r'^did:peer:[01](z)([1-9a-km-zA-HJ-NP-Z]{46,47})$') | ||
| peer_did_pattern = re.compile( | ||
| r"^did:peer:(([0](z)([1-9a-km-zA-HJ-NP-Z]{46,47}))|(2((\.[AEVID](z)([1-9a-km-zA-HJ-NP-Z]{46,47}))+" |
There was a problem hiding this comment.
sorry if I missed this discussion but is there a reason to change from the regex listed in the spec?
There was a problem hiding this comment.
The regexp in the spec appeared to be incorrect (not full) as it doesn't support numalgo2 methods for peer did creation.
| :return: encoded service | ||
| """ | ||
| service_to_encode = re.sub(r"[\n\t\s]*", "", service) \ | ||
| .replace('\'', '\"') \ |
There was a problem hiding this comment.
A bit of an edge case, but could this have implications on something like
{
"serviceEndpoint": "https://myurl.com?q=some 'thing'"
}
Perhaps they should be encoded beforehand, but I think https://www.rfc-editor.org/rfc/rfc3986#section-2 -> "2.2. Reserved Characters" allows it
There was a problem hiding this comment.
Thanks, i will add encoding and decoding of url special characters
No description provided.