Unable to fill an editable dataset with write_with_schema

dmer
Level 1
Unable to fill an editable dataset with write_with_schema

Hi,
I created an editable dataset and i tried to fill it in a Python recipe but i got this exception :

ExceptionTraceback (most recent call last)
<ipython-input-16-e040560bd627> in <module>()
----> 1 test.write_schema_from_dataframe(df_INDEX_EOD)

/opt/dataiku-dss-5.1.3/python/dataiku/core/dataset.py in write_schema_from_dataframe(self, df, dropAndCreate)
853
854 def write_schema_from_dataframe(self, df, dropAndCreate=False):
--> 855 self.write_schema(schema_handling.get_schema_from_df(df), dropAndCreate)
856
857

/opt/dataiku-dss-5.1.3/python/dataiku/core/dataset.py in write_schema(self, columns, dropAndCreate)
849 "columns": columns
850 }),
--> 851 "dropAndCreate" : dropAndCreate
852 })
853

/opt/dataiku-dss-5.1.3/python/dataiku/core/intercom.py in jek_or_backend_void_call(path, data, err_msg, **kwargs)
192 return jek_void_call(path, data, err_msg, **kwargs)
193 else:
--> 194 return backend_void_call(path, data, err_msg, **kwargs)
195
196 def backend_stream_call(path, data=None, err_msg=None, **kwargs):

/opt/dataiku-dss-5.1.3/python/dataiku/core/intercom.py in backend_void_call(path, data, err_msg, **kwargs)
183
184 def backend_void_call(path, data=None, err_msg=None, **kwargs):
--> 185 return _handle_void_resp(backend_api_post_call(path, data, **kwargs), err_msg = err_msg)
186
187 def jek_void_call(path, data=None, err_msg=None, **kwargs):

/opt/dataiku-dss-5.1.3/python/dataiku/core/intercom.py in _handle_void_resp(resp, err_msg)
235 err_data = resp.text
236 if err_data:
--> 237 raise Exception("%s: %s" % (err_msg, _get_error_message(err_data).encode("utf8")))
238 else:
239 raise Exception("%s: %s" % (err_msg, "No details"))

Exception: None: Ticket not given or unrecognized. Unload and reload the notebook.


Thanx for the help

0 Kudos
1 Reply
Alex_Combessie
Dataiker Alumni
Hi,

It is not currently possible to edit an editable dataset using the dataiku Python API. We have logged this request in our backlog.

For now I would advise to use a sync recipe to push it to a regular dataset, and then use the API.

Hope it helps,

Alex
0 Kudos

Labels

?
Labels (1)
A banner prompting to get Dataiku