I created a couple of dataclasses (similar to Go's structs) and I want to put my response/data inside of the dataclass. I haven't been able to find wether or not the json package supports this out of the box.
Dataclasses
from dataclasses import dataclass
from typing import List
@dataclass
class TakerPays:
currency: str
issuer: str
value: str
@dataclass
class Offers:
account: str
book_directory: str
book_node: str
flags: int
ledger_entry_type: str
owner_node: str
previous_tx_id: str
previous_tx_lgr_seq: int
sequence: int
taker_gets: str
taker_pays: TakerPays
index: str
owner_funds: str
quality: str
@dataclass
class Warnings:
id: int
message: str
@dataclass
class Result:
ledger_hash: str
ledger_index: int
offers: List[Offers]
validated: bool
warnings: List[Warnings]
@dataclass
class Response:
id: int
result: Result
status: str
type: str
Preview of the json that needs to be put into the Response dataclass
{
"id": 4,
"result": {
"ledger_hash": "5848C7DB5024EC3B532AC2F93BA8086A3D6281D3C0746BFE62E7E3CF4853F663",
"ledger_index": 68379996,
"offers": [
{
"Account": "rPbMHxs7vy5t6e19tYfqG7XJ6Fog8EPZLk",
"BookDirectory": "DFA3B6DDAB58C7E8E5D944E736DA4B7046C30E4F460FD9DE4E1D157637A1048F",
"BookNode": "0",
"Flags": 0,
"LedgerEntryType": "Offer",
"OwnerNode": "0",
"PreviousTxnID": "72B8928E31DF89223C7ADE0030685289BAD772C72DF23DDFFB92FF7B48BAC622",
"PreviousTxnLgrSeq": 68379985,
"Sequence": 386826,
"TakerGets": "789784836",
"TakerPays": {
"currency": "USD",
"issuer": "rvYAfWj5gh67oV6fW32ZzP3Aw4Eubs59B",
"value": "646.5472316"
},
"index": "82F565EDEF8661D7D9C92A75E2F0F5DBF2BAAFAE96A5A5A768AD76B933016031",
"owner_funds": "4587408572",
"quality": "0.0000008186371808232591"
}
],
"validated": true,
"warnings": [
{
"id": 1004,
"message": "This is a reporting server. The default behavior of a reporting server is to only return validated data. If you are looking for not yet validated data, include \"ledger_index : current\" in your request, which will cause this server to forward the request to a p2p node. If the forward is successful the response will include \"forwarded\" : \"true\""
}
]
},
"status": "success",
"type": "response"
}
I've not been able to find support to put the json into the dataclass similar to Go's json.Unmarshal and I'm curious to the best way to do this
CodePudding user response：
Unfortunately the builtin modules in Python such as json don't support de-serializing JSON into a nested dataclass model as in this case.
If you're on board with using third-party libraries, a solid option is to leverage the dataclass-wizard library for this task, as shown below; one advantage that it offers - which really helps in this particular example - is auto key casing transforms, such as from "camelCase" and "TitleCase".
Note that if you prefer not to subclass from any Mixin class, you can opt to use the helper functions fromdict and asdict to convert data from/to Python dict objects instead.
from dataclasses import dataclass
from typing import List
from dataclass_wizard import JSONWizard
@dataclass
class TakerPays:
currency: str
issuer: str
value: str
@dataclass
class Offers:
account: str
book_directory: str
book_node: str
flags: int
ledger_entry_type: str
owner_node: str
previous_txn_id: str
previous_txn_lgr_seq: int
sequence: int
taker_gets: str
taker_pays: TakerPays
index: str
owner_funds: str
quality: str
@dataclass
class Warnings:
id: int
message: str
@dataclass
class Result:
ledger_hash: str
ledger_index: int
offers: List[Offers]
validated: bool
warnings: List[Warnings]
@dataclass
class Response(JSONWizard):
id: int
result: Result
status: str
type: str
def main():
string = r"""
{
"id": 4,
"result": {
"ledger_hash": "5848C7DB5024EC3B532AC2F93BA8086A3D6281D3C0746BFE62E7E3CF4853F663",
"ledger_index": 68379996,
"offers": [
{
"Account": "rPbMHxs7vy5t6e19tYfqG7XJ6Fog8EPZLk",
"BookDirectory": "DFA3B6DDAB58C7E8E5D944E736DA4B7046C30E4F460FD9DE4E1D157637A1048F",
"BookNode": "0",
"Flags": 0,
"LedgerEntryType": "Offer",
"OwnerNode": "0",
"PreviousTxnID": "72B8928E31DF89223C7ADE0030685289BAD772C72DF23DDFFB92FF7B48BAC622",
"PreviousTxnLgrSeq": 68379985,
"Sequence": 386826,
"TakerGets": "789784836",
"TakerPays": {
"currency": "USD",
"issuer": "rvYAfWj5gh67oV6fW32ZzP3Aw4Eubs59B",
"value": "646.5472316"
},
"index": "82F565EDEF8661D7D9C92A75E2F0F5DBF2BAAFAE96A5A5A768AD76B933016031",
"owner_funds": "4587408572",
"quality": "0.0000008186371808232591"
}
],
"validated": true,
"warnings": [
{
"id": 1004,
"message": "This is a reporting server. The default behavior of a reporting server is to only return validated data. If you are looking for not yet validated data, include \"ledger_index : current\" in your request, which will cause this server to forward the request to a p2p node. If the forward is successful the response will include \"forwarded\" : \"true\""
}
]
},
"status": "success",
"type": "response"
}
"""
r = Response.from_json(string)
print(repr(r))
if __name__ == '__main__':
main()
Output:
Response(id=4, result=Result(ledger_hash='5848C7DB5024EC3B532AC2F93BA8086A3D6281D3C0746BFE62E7E3CF4853F663', ledger_index=68379996, offers=[Offers(account='rPbMHxs7vy5t6e19tYfqG7XJ6Fog8EPZLk', book_directory='DFA3B6DDAB58C7E8E5D944E736DA4B7046C30E4F460FD9DE4E1D157637A1048F', book_node='0', flags=0, ledger_entry_type='Offer', owner_node='0', previous_txn_id='72B8928E31DF89223C7ADE0030685289BAD772C72DF23DDFFB92FF7B48BAC622', previous_txn_lgr_seq=68379985, sequence=386826, taker_gets='789784836', taker_pays=TakerPays(currency='USD', issuer='rvYAfWj5gh67oV6fW32ZzP3Aw4Eubs59B', value='646.5472316'), index='82F565EDEF8661D7D9C92A75E2F0F5DBF2BAAFAE96A5A5A768AD76B933016031', owner_funds='4587408572', quality='0.0000008186371808232591')], validated=True, warnings=[Warnings(id=1004, message='This is a reporting server. The default behavior of a reporting server is to only return validated data. If you are looking for not yet validated data, include "ledger_index : current" in your request, which will cause this server to forward the request to a p2p node. If the forward is successful the response will include "forwarded" : "true"')]), status='success', type='response')
NB: I noted that two fields in the Offers dataclass have slightly different names than the fields in the JSON object. For example, the field previous_tx_id is associated with the key PreviousTxnID in the JSON object.
Assuming this was intentional, you could easily work around this by defining a field alias mapping, as shown below:
from dataclass_wizard import json_key
# Note: In Python 3.9 you can import this from `typing` instead
from typing_extensions import Annotated
@dataclass
class Offers:
...
previous_tx_id: Annotated[str, json_key('PreviousTxnID')]
previous_tx_lgr_seq: Annotated[int, json_key('PreviousTxnLgrSeq')]
...