Feature suggestion: EDI/EDIFACT Nodes

[e2dboden]

Hi there,

Knime is really a jack of all trades for every challenge my workplace throws at me.
The only thing I frequently need to outsource is Problems related to EDI Messages. I also could not find any workarounds on the forum. If anyone has a useful approach, I´m all ears.

The companies we exchange EDI Messages use almost exclusively the EDIFACT Standard and only few document types are getting exchanged (e.g. ORDERS, DESADV, INVOIC,…).

I was thinking about what would help me in my daily business, and probably a lot of others as well:

• A Reader / EDI Encoder node, which could extract the structured data from an EDI Message and output it into a table. From there you could work with existing Knime nodes for further processing.

• A EDI Decoder/ EDI-Mapper (table to EDI-Message) node, which would take a table input and map it to the segments. This would probably require a graphical interface for the user to create the mappings.

Another approach would be to create a node to convert the EDI-Messages to another Format (e.g. JSON) and use existing Knime nodes for further data wrangling. I´m not sure though if this approach is sensible from a technical point of view.

Here is a wikipedia link to the EDIFACT Standard: EDIFACT - Wikipedia

Cheers

[temesgen-dadi]

Hi,

This is an interesting feature request. I will discuss this internally, and create a feature request.

In the meantime, you can use existing libraries (python, R, java) and create a snippet that can do what you want. I, for example, run in to GitHub - nerdocs/pydifact: A python EDIFACT library.

Kind regards,
Temesgen

[mlauber71]

@e2dboden

A Reader / EDI Encoder node, which could extract the structured data from an EDI Message and output it into a table.

@e2dboden I recently tried my hand on combining KNIME and the mentioned pydifact package to parse EDIFACT messages. Maybe you could have a look.

Parsing EDIFACT files

@gentile I tried a first import and split up the messages with the Python library and imported them into KNIME (and Excel to show what they look like). Further work will have to be done to split the lines into the tables you want. I am not sure if the Python package would also offer this options but it might be worth a try (I am not familiar with the format as to be able to feed some sort of pattern into the code). Otherwise you will have to identify the blocks in the data and transform them to…

[e2dboden]

Thanks,

this looks really interesting. Unfortanately my Python skills are close to non-existent.
I’ll give it a try nevertheless.