Python script - ascii codec cannot decode input

KNIME Analytics Platform
1

Dear KNIME community,

I cannot read input data given to a python script node (see error attached) when working on windows 10.

Python is working properly and I can read the input file without problem when running directly in system terminal.

Same node works perfectly fine on linux (CentOS).

Could you please help me to fix the issue? I cannot even execute selected lines to find out where the error comes from.

Any help would be much appreciated!

Jose Manuel

2

hi, try replace the special characters (accent marks, etc ), try convert the data format:
 

<'some string acsii'> = <'some string acsii'>.decode('utf8')
3

Hello,

thanks for the tip!

I tried something. In my input column, I have a RDKit molecule (so in RDKit format). If I filter this column, I can execute the node without problem.

Same node with same input table works just fine on my linux desktop,  so I guess this is a windows related issue.

What am I missing?  I attached an example workflow to illustrate the problem.

Thanks for your help!

Jose Manuel

4

Hi all,

I digged a little bit more into the matter.

I updated my KNIME to latest release (see below) and compared with CDK column format (see attached wf).

Python script fails with RDKit column but not with CDK column in input, so I guess this is more of a RDKIt / KNIME Windows issue. Is there a work-around? (preferably without re-converting to SDF column before)

Thanks in advance for your help!

Jose Manuel

KNIME node versions:

KNIME Python Integration          3.3.0.v201611242050     org.knime.features.python.feature.group           KNIME GmbH, Konstanz, Germany

KNIME Python Scripting extension          3.1.0.v201602120920                de.mpicbg.tds.knime.scripting.python.feature.feature.group     Max Planck Institute of Molecular Cell Biology and Genetics (MPI-CBG)

RDKit KNIME integration              3.2.2.v201612161513     org.rdkit.knime.feature.feature.group  NIBR