Note: sintax is the old pycelonis 1.0.

To get tables from the datamodel, something like this (dmtemp is a datamodel object):

from pycelonis.celonis_api.pql.pql import PQLColumn

from pycelonis.celonis_api.pql.pql import PQL, PQLFilter

query=PQL()

tablename=start_table_name

for c in dmtemp.tables.find(tablename).columns:

  c_name=cm'name']

  qval='"%s"."%s"' %(tablename,c_name)

  column=PQLColumn(query=qval, name = c_name)

  query.add(column)

df_final=dmtemp.get_data_frame(query)

To put data you need to update the table in the datamodel, then reload the datamodel

Something like:

pool.append_table(df_or_path=df_final,

         table_name=hist_table_name,

         connection=data_conection

         )

(i think there is also an upsert function)

then reload datamodel:

dm.reload(wait_for_reload=True)

where dm is a datamodel object, got with something like:

  dm=pool.datamodels.find(datamodel_name)

HTH. Happy 2023

Hi Guillermo,

Thank you for the explaination but my intention is to get the tables before they reach the datamodel in the datapool level.

To be exact, i need to get the tables from the ST and make some queries with 2 loops in ML and return it back to datapool for further processing in the datapool

So is there any way for that?

Thank you

Yes, you can run sql from python,

something like:

 job_transformation_min=datajob.transformations.find(trans_job_Min_Notif_Rerouting_Start)

  df_min=job_transformation_min.get_data_frame()

 but the problem is that you will get only 100 records.

So the workaround is to load the tables into a temporary datamodel, then query them using PQL, and rewrite the result back

Hi Guillermo.

Thank you. However in the current PyCelonis version all functions regarding "get_data_frame()" have been removed from PyCelonis. I am also looking for alternative ways to access data via Python.

Hello Max

In the last version, you need to use "data_model.export_data_frame(query)" .