As the title suggests, how can I convert response objects from the new Python SDK into a pandas dataframe? Or can they be converted into dictionaries?
Using pd.DataFrame() on the objects seems to just put the whole object in a single column.
Thanks!
Managed to find a workaround!
list_of_dicts = [vars(obj) for obj in response]
2 Likes
Hi @anon25483135,
Glad you found a work around.
And yes, you can convert the responses from the new python client library into dictionaries using the class method to_dict().
NOTE: this only works at the top layer so if you do like api_response.data[0].to_dict() it will not work
Here’s an example of converting a response object to a python dict:
import asana
from asana.rest import ApiException
from pprint import pprint
# Configure OAuth2 access token for authorization: oauth2
configuration = asana.Configuration()
configuration.access_token = "<YOUR_ASANA_PERSONAL_ACCESS_TOKEN>"
# create an instance of the API class
task_api_instance = asana.TasksApi(asana.ApiClient(configuration))
project = '<PROJECT_GID>' # str | The project to filter tasks on. (optional)
try:
# Get multiple tasks
task_api_response_dict = task_api_instance.get_tasks(project=project).to_dict()
# Access name of first task in api_response_dict
pprint(task_api_response_dict['data'][0]['name'])
except ApiException as e:
print("Exception when calling TasksApi->get_tasks: %s\n" % e)
WARNING: You might want to be careful in converting the response object into a dict because it will contain all the properties of a resource schema.
Scenario:
Let’s say you make a request to Get multiple tasks without specify any opt_fields → in our docs we say this returns an array of compact task schema in the 200 repsonse
EX:
api_response = task_api_instance.get_tasks(project=project)
What you might expect:
{'data': [{'gid': 'TASK_GID_1',
'name': 'Task 1',
'resource_subtype': 'default_task',
'resource_type': 'task'},
{'gid': '<TASK_GID_2>',
'name': 'Task 2',
'resource_subtype': 'default_task',
'resource_type': 'task'},],
'next_page': None}
Actual result:
{'data': [{'actual_time_minutes': None,
'approval_status': None,
'assignee': None,
'assignee_section': None,
'assignee_status': None,
'completed': None,
'completed_at': None,
'completed_by': None,
'created_at': None,
'created_by': None,
'custom_fields': None,
'dependencies': None,
'dependents': None,
'due_at': None,
'due_on': None,
'external': None,
'followers': None,
'gid': 'TASK_GID_1',
'hearted': None,
'hearts': None,
'html_notes': None,
'is_rendered_as_separator': None,
'liked': None,
'likes': None,
'memberships': None,
'modified_at': None,
'name': 'Task 1',
'notes': None,
'num_hearts': None,
'num_likes': None,
'num_subtasks': None,
'parent': None,
'permalink_url': None,
'projects': None,
'resource_subtype': 'default_task',
'resource_type': 'task',
'start_at': None,
'start_on': None,
'tags': None,
'workspace': None},
{'actual_time_minutes': None,
'approval_status': None,
'assignee': None,
'assignee_section': None,
'assignee_status': None,
'completed': None,
'completed_at': None,
'completed_by': None,
'created_at': None,
'created_by': None,
'custom_fields': None,
'dependencies': None,
'dependents': None,
'due_at': None,
'due_on': None,
'external': None,
'followers': None,
'gid': '<TASK_GID_2>',
'hearted': None,
'hearts': None,
'html_notes': None,
'is_rendered_as_separator': None,
'liked': None,
'likes': None,
'memberships': None,
'modified_at': None,
'name': 'Task 2',
'notes': None,
'num_hearts': None,
'num_likes': None,
'num_subtasks': None,
'parent': None,
'permalink_url': None,
'projects': None,
'resource_subtype': 'default_task',
'resource_type': 'task',
'start_at': None,
'start_on': None,
'tags': None,
'workspace': None},],
'next_page': None}
Notice how all the other properties are “None”
It might be safer to use the call_api option if you don’t want excess data form the model.
Here’s an example request:
import asana
from asana.rest import ApiException
from pprint import pprint
# Configure OAuth2 access token for authorization:
configuration = asana.Configuration()
configuration.access_token = '<YOUR_ASANA_PERSONAL_ACCESS_TOKEN>'
api_client = asana.ApiClient(configuration)
# GET - get a task
api_response = api_client.call_api(
"/tasks",
"GET",
path_params={},
query_params=[('project', '<PROJECT_GID>')],
header_params={"Accept": "application/json; charset=utf-8"},
body=None,
post_params=[],
files={},
response_type=object, # If there is an existing response model for the resource you can use that EX: "TaskResponseData" or you can specify one of the following types: float, bool, bytes, str, object
auth_settings=["oauth2"],
async_req=None,
_return_http_data_only=True,
_preload_content=True,
_request_timeout=None,
collection_formats={},
)
pprint(api_response)
Here’s what the response would look like:
{'data': [{'gid': '<TASK_GID_1>',
'name': 'Task 1',
'resource_subtype': 'default_task',
'resource_type': 'task'},
{'gid': '<TASK_GID_2>',
'name': 'Task 2',
'resource_subtype': 'default_task',
'resource_type': 'task'}]}
2 Likes
Thanks for the detailed response @John_Vu I will give that a go! Noted about the excess columns as well - I was wondering about that.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.