specify datasetType when creating a dataset

[pdurbin]

What this PR does / why we need it:

At dataset creation time, we need to allow datasetType to be passed so that types other than the default ("dataset") can be persisted to the database.

Which issue(s) this PR closes:

• Intended for this issue but we'll make a fresh PR: allow a dataset to be created with a dataset type other than "dataset" (e.g. "software", "workflow", "review", etc.) #359

Related Dataverse PRs:

• Frontend PR that that uses this PR: Show "New Review" button, when enabled dataverse-frontend#822
• Parent project, Trusted Data: Project: Trusted Data dataverse-pm#425

Special notes for your reviewer:

We also need to support listing dataset types ( https://guides.dataverse.org/en/6.7.1/api/native-api.html#list-dataset-types ) but perhaps that should be a separate issue and pull request?

Suggestions on how to test this:

• On the backend, create a datasetType such as "software" or "review": https://guides.dataverse.org/en/6.7.1/api/native-api.html#add-dataset-type
• Use that type to create a dataset of that type by supplying it in the datasetType field. In the JSON, it looks like this:

{
  "datasetType": "software",
  "datasetVersion": {
    "license": {
      "name": "CC0 1.0",
      "uri": "http://creativecommons.org/publicdomain/zero/1.0"
    },
    "metadataBlocks": {
      "citation": {
        "fields": [
          {
            "value": "pyDataverse",
            "typeClass": "primitive",
            "multiple": false,
            "typeName": "title"
          },
...

Is there a release notes update needed for this change?:

Yes, we should add something like this:

You can now specify a datasetType when creating a dataset.

Additional documentation:

None.

[ChengShi-1]

Looks awesome!

Could you also add or edit test cases in integration test test/integration/datasets/DatasetsRepository.test.ts and functional test test/functional/datasets/CreateDataset.test.ts? it may be harder to test if we don't have datasetType on DatasetDTO and dataset model.
Also, the useCase.md may need to be updated with new parameter.

[ChengShi-1]

Could you add the param to the description above @param {string}...

[pdurbin]

Sure, added in 156963d.

[ChengShi-1]

You could add an additional test to make sure datasetType would be called here.
For example,
you could create a function createDatasetwithDatasetType in DataseHelper and do something similar with this
expect(datasetsRepositoryStub.createDatasetwithDatasetType).toHaveBeenCalledWith( testDataset, testMetadataBlocks, ROOT_COLLECTION_ID, datasetType )

[pdurbin]

Good idea. Added in 48cc5dc. Thanks for the help with this!

[pdurbin]

Closing. As discussed with @ChengShi-1 we're thinking a better order for PRs would be:

• get dataset types: for this I created a new PR: Use Case for Getting Available Dataset Types #366
• add, modify and delete dataset types (I'll work on this next)
• save a dataset with a specific dataset type (this PR, but I'll make a new one after I'm done with the PRs above)