Implement CSV import

[1michhu1]

Implemented CSV import based on the format of the Blank Membership List for Blueprint.

Response shape:

{
  status: "success" | "partial_success" | "failed",
  summary: {
    totalRows: int,
    parsedSucceeded: int,
    parseFailed: int,
    dbSucceeded: int,
    dbFailed: int
  },
  parseErrors: [
    {
      rowIndex: int,
      column: string* (see Blank Membership List for possible columns),
      message: string (ie. 'Missing required volunteer name')
    }
  ],
  dbErrors: [
    {
      rowIndex: number;
      message: string;
      pg_code?: string (PostgreSQL error code);
    }
  ]
}

Important points about response shape and implementation

• import_csv response returns whether some, all or none of the imported csv rows were created based on the status (success, partial_success, failed)
• This function tries to insert rows on a best effort basis. There are two types of errors which describe why a volunteer could not be inserted:
  1. ParseError: Either Papaparse failed to parse the csv row, or a value(s) in a row did not align with the expected formatting from the Blank Membership List. This is listed in summary.parseErrors
  2. DbError: A row fails to be upserted into Roles, Cohorts, Volunteers, VolunteerRoles or VolunteerCohorts. This could be due to a check constraint violation or some other database error listed in summary.dbErrors. To note, the RPC always attempts to upsert into all of these tables to avoid duplicate key errors.
• rowIndex in errors corresponds to the actual index of that volunteer in the raw csv. rowIndex skips the header and starts counting from the first data row
• The function does not attempt to insert rows that failed the parsing stage
• The package Papaparse was installed to simplify csv parsing
• Created Roles default to is_active = true
• The RPC
  • upsert_volunteer_with_roles_and_cohorts was created to easily rollback db changes on errors. This is because cleanup on up to 5 tables would otherwise be difficult
  • The RPC tries to match csv volunteers to rows in the Volunteers table by email and name. csv data will overwite Volunteer table data on email and name match
• A migration file was manually added to account for changes in the remote schema. I believe it would be helpful to add supabase db pull --db-url [db-url with password] to our workflow and readme in order to sync up schema changes.
• Local database schema changes have not yet been pushed to the remote

Other Notes:

• Another test suite unrelated to this feature was edited to fix a minor database cleanup bug
• A folder for test data was added to /tests
• Currently, the supabase

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Actions	Updated (UTC)
trcc	Ignored		Mar 23, 2026 3:20pm

[notjackl3]

LGTM!