Wrong input in .ad file of local dataset causes bug in local etmodel

[DorinevanderVlies]

Problem

When I have a random wrong input in the .ad file of a local dataset it causes my complete etmodel.dev not to open. Not even other datasets.

Would that also occur on the beta or live server?

Images

The "some random completely wrong input" I used in the .ad file of a local dataset:

The result when I try to load etmodel.dev

[grdw]

Would that also occur on the beta or live server?

Yes. 👍 Which is exactly why you're doing this locally.

[DorinevanderVlies]

Yes. 👍 Which is exactly why you're doing this locally.

Check, a related question. Would we be able to share local datasets with clients?

[grdw]

On a side note: maybe the error handling can be a little bit more descriptive though 🤔

Would we be able to share local datasets with clients?

Eh.. we could zip them and mail them I guess 😆 but what do you mean by this?

[DorinevanderVlies]

For example if we make a scenario for Drenthe, using a local model, and we want to share it with the other people involved in the project, would that be possible?

[grdw]

For example if we make a scenario for Drenthe, using a local model, and we want to share it with the other people involved in the project, would that be possible?

No, that would not be possible locally. The only wat this can be achieved (as far as I know) is to deploy the Drenthe dataset and create a scenario similar to the one you created locally on either the beta or production server.

[DorinevanderVlies]

The only wat this can be achieved (as far as I know) is to deploy the Drenthe dataset

Would deploying the local dataset (made with the .ad file as input) be possible?

@AlexanderWirtz The question you raised during a daily standup last week was if the methods of the "local datasets project" can be used for Drenthe. As I understand it now, sharing a scenario would not be possible. That seems a problem to me.

[grdw]

Would deploying the local dataset (made with the .ad file as input) be possible?

Not yet. All local-dataset branches need to be merged inside master for:

• etengine
• etmodel
• etsource
• atlas

Once that is done we can actually start deploying datasets like Drenthe :)

[AlexanderWirtz]

Not yet. All local-dataset branches need to be merged inside master for:

@ChaelKruip @ploh @antw

We would like to use the new method, but do of course need to share the created scenarios with our client (the province of Drenthe). Do you foresee such a deploy happening by early February?

Good catch @DorinevanderVlies

[ploh]

Do you foresee such a deploy happening by early February?

Yes. I think we should discuss this in more detail in the meeting tomorrow.

[ploh]

@DorinevanderVlies By now, the existence of the inputs used in the .ad file is verified if you run rspec in your etsource checkout - so you should probably do that before starting up your local server.

@DorinevanderVlies @grdw Do you agree that this can be closed now?

[grdw]

I agree, this can be closed 👍