221 - Integration multiple files download

[MellyGray]

What this PR does / why we need it:

This PR adds the integration of the Download multiple files button in the Files Table with the dataverse API. With this integration the SPA user will be able to select multiple files from the dataset table and download them.

Which issue(s) this PR closes:

• Closes Files Table row selection Download Button #221

Special notes for your reviewer:

Suggestions on how to test this:

Step 1: Run the development environment

1. Run npm I
2. cd packages/design-system && npm run build
3. cd ../../
4. Check that you have a .env file such as the .env.example, with the VITE_DATAVERSE_BACKEND_URL=http://localhost:8000 variable
5. cd dev-env
6. ./run-env.sh unstable
7. To check the environment go to http://localhost:8000 and check your local dataverse installation

Step 2: Test Dataset View mode with the implemented changes to download the files

1. Go to http://localhost:8000
2. Login as admin using username: dataverseAdmin and password: admin1
3. Create a new Dataset
4. Upload some files to the dataset
5. From the dataset view mode copy the search parameters from the url. Ex.: ?persistentId=doi:10.5072/FK2/LHGRHP&version=DRAFT
6. Go to http://localhost:8000/spa/datasets and paste the search parameters. Ex.: http://localhost:8000/spa/datasets?persistentId=doi:10.5072/FK2/LHGRHP&version=DRAFT
7. You are currently viewing the same dataset in the SPA. In the files table, locate the "Download" button on the right of the Files Table. Select some files using the checkboxes and then click the Download button to download all of them at once.

Does this PR introduce a user interface change? If mockups are available, please link/include them here:

No

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

No

Additional documentation:

[coveralls]

coverage: 98.173% (-0.03%) from 98.205%
when pulling 66f2331 on feature/221-integration-files-table-row-selection-download
into 14a74ae on develop.

[GPortas]

Can you please resolve conflicts? @MellyGray

[MellyGray]

@GPortas merge conflicts resolved

[GPortas]

Downloading multiple files works fine. The message saying that the zip download limit has been exceeded is confusing, although I understand that it appears because the integration of the zip limit setting is pending, and the limit (1B) is mocked, right?

On the other hand for downloading a single file, a zip should not be generated, and the file must be downloaded individually unzipped, as in JSF.

multfile.mov

[MellyGray]

@GPortas

1. I can increase the mock from 1B zip limit to something higher so it doesn't appear in beta
2. I don't understand why is downloading 1 file as a zip, I saw this comment in JSF

// There's a chance that this is not really a batch download - i.e.,
 // there may only be one file on the downloadable list. But the fileDownloadService
 // method below will check for that, and will redirect to the single download, if
 // that's the case. -- L.A.
 fileDownloadService.writeGuestbookAndStartBatchDownload(guestbookResponse, doNotSaveGuestbookResponse);

If you look at the implementation if (fileIds.length == 1) { it calls redirectToDownloadAPI() for the single file download instead of settingsService.getValueForKey(SettingsServiceBean.Key.CustomZipDownloadServiceUrl); to get the zip. Is the API not using this same fileDownloadService than JSF?

I mean, I can handle it in the frontend and call the single download from the frontend, but isn't the desired API behavior to download a single file if only 1 file is selected? I want to be sure before applying the logic in the UI

[GPortas]

@GPortas

1. I can increase the mock from 1B zip limit to something higher so it doesn't appear in beta
2. I don't understand why is downloading 1 file as a zip, I saw this comment in JSF

// There's a chance that this is not really a batch download - i.e.,
 // there may only be one file on the downloadable list. But the fileDownloadService
 // method below will check for that, and will redirect to the single download, if
 // that's the case. -- L.A.
 fileDownloadService.writeGuestbookAndStartBatchDownload(guestbookResponse, doNotSaveGuestbookResponse);

If you look at the implementation if (fileIds.length == 1) { it calls redirectToDownloadAPI() for the single file download instead of settingsService.getValueForKey(SettingsServiceBean.Key.CustomZipDownloadServiceUrl); to get the zip. Is the API not using this same fileDownloadService than JSF?

I mean, I can handle it in the frontend and call the single download from the frontend, but isn't the desired API behavior to download a single file if only 1 file is selected? I want to be sure before applying the logic in the UI

Is the API not using this same fileDownloadService than JSF? -> No. That logic is only consumed from the JSF UI Helper class.

We need to replicate the same logic on the SPA.

The /datafiles endpoint always create a zipped bundle. The /datafile endpoint returns an individual unzipped file.

[MellyGray]

@GPortas I applied the requested changes

[GPortas]

LGTM