A script to create a dummy datatable for dev purposes is lacking

[Bachibouzouk]

Problem

After a fresh install it would be nice to run a script which would create a few example table with the test user credentials.

[Bachibouzouk]

@wingechr, @gnn wouldn't you be able to do that in no time :)

[MGlauer]

Data we need for testing:

• Some geo data for the maps (Points, Polygons)
• some time series for the graphs

[wingechr]

@MGlauer, @Bachibouzouk I would be happy to do that.

Unfortunately, in the current development branch (f786c2502bbb), insertion of data via REST API seems to be broken.

The error thrown is 'IntegrityError('null value in column "_autocheck" violates not-null constraint) I am still trying to figure out why. I set up my test environment according to the README.md

The same error also occurs when running python manage.py test

[MGlauer]

@wingechr This is already fixed and in review. See #493

[Bachibouzouk]

@wingechr This is already fixed and in review. See #493

@MGlauer I approved #493 yesterday, it can be merged (just add a line in the changelog before)

[Bachibouzouk]

@wingechr - #493 is merged into develop, you can go ahead :)

[4lm]

Yes, I need this (dev data)! Thanks @wingechr for working on this!

[4lm]

If the test data also could contain geodata, this would be nice!

[Bachibouzouk]

@wingechr - we are now (at least) two people awaiting this feature :), when do you think you can do it?

[Bachibouzouk]

@wingechr, @MGlauer - isn't this something you can do fast? To quickly make some tests it would be very useful (and also very good that we all use the same reference test datasets)

[jh-RLI]

@wingechr @Ludee @chrwm @steull We should create dummy data that can be added to the fresh OEDB install (for local tesing). I would also like to integrate this into the docker container.

Lets collect some usefull data together.

• [ ] timeseries
• [ ] scalars
• [ ] text
• [ ] geo data

I have already created some test data for the oedatamodel tables (data from the open_MODEX project). We could reuse these and create some simple but jet-realistic tables and add them to all available schemas after the database setup is done. It would be useful to add appropriate metadata as well.

[chrwm]

We could reuse these and create some simple but jet-realistic tables

I can do that / prepare a proposal.

[jh-RLI]

Very nice :) I can create a script to create the tables via the API on the local development version of the OEP and then upload the data. I will also help with creating the test data. Perhaps @steull can also help create the test data and create the metadata once the test datasets are available? This data/tables could also be used to test the review process (I think).

I think it is enough to create single tables (tables without forigen keys/relations), the example for the "oedatamodel" already covers tables with relations.

[jh-RLI]

To complete this for our use case, we should try to add data that matches the schemas/data categories on the OEP (which could result in different data sets for each schema). This is also helpful when we use the data to test the review process, and provides good examples of what data to expect (for developers).

[Ludee]

These are suitable for testing:

• model_draft.oep_metadata_table_example_v151
• boundaries.bkg_vg250_1_sta

[chrwm]

This is the data from the open_MODEX project, base scenario 2016:

• oed_scenario.csv
• oed_scalar.csv
• oed_timeseries.csv
• datapackage.txt

and data sources

• modex_openmodex_base_scenario_oep_dummy_data_dev.txt