TimeImportAnalysisDto

Result of analysing an uploaded time CSV: the parsed rows for review plus the distinct
team member and client names in the file, each with a suggested match, so the user maps
names once rather than row by row. Returned by the analyze endpoint; the RowIds are the
keys used in the import selections.

Properties

Name Type Description
totalRowsinteger

Number of data rows in the file (usable or not)

Number of data rows in the file (usable or not)

fromDatestring

The earliest entry date in the file; null when no row has a usable date

The earliest entry date in the file; null when no row has a usable date

toDatestring

The latest entry date in the file; null when no row has a usable date

The latest entry date in the file; null when no row has a usable date

rowsarray of TimeImportRowDto

The parsed rows, in file order

The parsed rows, in file order

teamMembersarray of TimeImportTeamMemberDto

The distinct team member names in the file, most rows first

The distinct team member names in the file, most rows first

clientsarray of TimeImportClientDto

The distinct client names in the file, most rows first

The distinct client names in the file, most rows first

JSON Example

{
  "totalRows": 0,
  "fromDate": "2024-01-01",
  "toDate": "2024-01-01",
  "rows": [
    {
      "rowId": "...",
      "entryDate": "...",
      "teamMemberName": "...",
      "clientName": "...",
      "startTime": "..."
    }
  ],
  "teamMembers": [
    {
      "name": "...",
      "rowCount": "...",
      "suggestedUserCode": "..."
    }
  ]
}

Used By Operations