Hi all,
Using the updated PyCelonis version 2.0.0, I'm unable to push Pandas dataframes en parquet files towards the data pool.
In the feedback, it states that it cannot be parsed towards a parquet file, but when I do this via Pandas (pandas.to_parquet()), this works without any problem.
Are people facing similar problems?
Best regards,
Jan-peter
Hmmm... definitely I am not going to update my ML to the new python package... 
Hi Guillermo,
The problem was that one of my columns of type list was. Whereas this is not a problem for pd.to_parquet(), it is for the Celonis API. This makes sense since lists is not a supported column data type in the EMS, but a bit more feedback was welcome.
Well, still, after Mateusz complained about the documentation, I checked and agree with him. I don't see any advantage in the new version, but a lot of problems... as I say "let the last version be tested by another...."
To be honest, I think it is better designed than v1, especially if you read the reasons for changes in the migration docs (Migration from PyCelonis 1.X to PyCelonis 2.X - PyCelonis).
However, it has changed and not all exceptions and use cases are indeed documented. It would be nice to have a feedback button at some points or issue submit form.
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.
