Config.json_loads is not used with BaseModel.parse_file #1067
Description
Bug
BaseModel.parse_file does not use the custom JSON decoder as specified in Config.json_loads.
Please complete:
- OS: Ubuntu 18.04.3 LTS
- Python version
import sys; print(sys.version): 3.8.0 (default, Oct 28 2019, 16:14:01) - Pydantic version
import pydantic; print(pydantic.VERSION): 1.2
Please read the docs and search through issues to
confirm your bug hasn't already been reported.
Where possible please include a self contained code snippet describing your bug:
import json
import tempfile
from pydantic import BaseModel
def name_overwriting_decoder(*args, **kwargs):
"""A silly JSON decoder that hard codes a `name` value"""
data = json.loads(*args, **kwargs)
data['name'] = "Overwritten Name"
return data
class Example(BaseModel):
name = "Default Name"
class Config:
json_loads = name_overwriting_decoder
raw_example = Example.parse_raw(b'{"name": "Name To Overwrite"}')
assert raw_example.dict() == {"name": "Overwritten Name"}
# Create a temporary JSON file and write some JSON
tmp_file = tempfile.NamedTemporaryFile(delete=False)
tmp_file.write(b'{"name": "Name From Temp File"}')
tmp_file.close()
file_example = Example.parse_file(tmp_file.name)
assert file_example.dict() == {"name": "Overwritten Name"}I would have expected Example.parse_file to use the same decoder as Example.parse_raw
FYI: I have a use-case where I parse a Hjson file with hjson.loads.
I am happy to produce a PR for this, if you think it is actually a bug :)