Improve performance of Metadata Uploader

[ksierks]

What needs changed?

The Sample Metadata importer needs to be updated to perform all parsing and validation on Client side instead of Server side. This will enhance performance as we will no longer be sending large payloads to/from the server.

How does it work now?

How does this component work now, and what's wrong with it?

How should it work?

Note: These steps do not necessarily map to Wizard steps.

• 1. User selects a file which is parsed client side (i.e. in Javascript)

• 2. Identifying the sample name column

  • a. Display all the parsed header fields and prompt the user to select the Sample Name column
  • b. Optionally: Preselect the column that matches a list of expected Sample Name column name

• 3. Validate Sample Names using validate endpoint from Unify sample validation/existence endpoints #1349

  • a. Identify Samples that don't exist in the Project and allow for creation
  • b. Validate new Samples Names for proper formatting
  • c. Allow invalid Sample Names to be edited by the user (validate sample existence)
    • Allow mapping corrected Sample Name to existing Sample if it isn't already present in the provided Metadata Moving this to the next release with virtual table enhancement

• 4. Display a preview of the Metadata attempted to be uploaded

  • a. Allow user to deselect/select rows to be uploaded

• 5. Send payload to server to update the metadata
  Process each Sample individually with a progress bar.
  Use endpoints created/updated in Update Project Sample Creation and Add in Partial Project Sample updating #1350

  • Batched endpoint
  • show progress for each batch

• 6. Display a success messaged and then after a timeout redirect to the launch page (i.e. where the metadata importer was initiated from)

Additional information

Add any other context, files, or screenshots about this request here.
LargeProjectImport.xlsx