when using the code below and attached file <a href="https://github.com/amirziai/f

unflatten does not work when importing JSON from file about flatten HOT 3 OPEN

amirziai commented on July 17, 2024 1

unflatten does not work when importing JSON from file

from flatten.

Comments (3)

amirziai commented on July 17, 2024

do you have the latest version of flatten_json? try upgrading and try again

pip install --upgrade flatten_json

i'm get this output with your example:

{'seo': {'test': 'hello'}}

from flatten.

alexabbott commented on July 17, 2024

I have the latest version of flatten_json installed. I'm running Python 2.7, pip 9.0.1

Still getting the same error...

Uncommenting these lines from the library allows me to run the command without error:

def _unflatten_asserts(flat_dict, separator):
    assert isinstance(flat_dict, dict), "un_flatten requires a dictionary input"
    assert isinstance(separator, str), "separator must be a string"
    # assert all((not isinstance(value, Iterable) or isinstance(value, str)
                # for value in flat_dict.values())), "provided dictionary is not flat"

from flatten.

stevegoossens commented on July 17, 2024

Is this a unicode and Python 2 issue? Looks like it to me. My guess would be that people who don't experience this issue are using Python 3....

e.g. on python 2.7

>>> config = {u'mfa': u'off', u'datadog.apm_config.enabled': u'true'}
>>> unflatten(config)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python2.7/site-packages/flatten_json.py", line 79, in unflatten
    _unflatten_asserts(flat_dict, separator)
  File "/usr/local/lib/python2.7/site-packages/flatten_json.py", line 68, in _unflatten_asserts
    for value in flat_dict.values())), "provided dictionary is not flat"
AssertionError: provided dictionary is not flat
>>> config['mfa']
u'off'
>>> import json
>>> json.dumps(config, indent=2)
'{\n  "mfa": "off", \n  "datadog.apm_config.enabled": "true"\n}'
>>> print json.dumps(config, indent=2)
{
  "mfa": "off",
  "datadog.apm_config.enabled": "true"
}

a workaround in my case is to cast the keys and values to strings (they're both always strings in my scenario) and then the assertion passes:

>>> config_as_str = { str(k): str(v) for k, v in config.iteritems() }
>>> config_as_str
{'mfa': 'off', 'datadog.apm_config.enabled': 'true'}
>>> unflatten(config_as_str, separator='.')
{'mfa': 'off', 'datadog': {'apm_config': {'enabled': 'true'}}}

from flatten.

Recommend Projects