● CSV files must be delimited by commas. CSV files should be provided with a title row detailing the column names. There is no header row and no trailer row.
● The preference would be that data items within the commas are surrounded by quotation marks to ensure that any rogue commas in address data (for example) do not inadvertently move the data to the right.
● Quotations are not required in blank data columns. Thus, data should be presented as: ,,,,,,,,”23, The Street”, ”The Town”,,,etc.
● Data can be supplied with simple comma separation (i.e. no quotation marks), however in doing so it is reliant on source applications to ensure that there are no commas in the data being uploaded, as this will cause these lines of the file to reject as the data will not meet the validation requirements.
● There is no provision for escape characters in the standard where quotation marks are provided in the data. In this scenario – and where the data source is providing data items surrounded by quotation marks then the data will be presented as: ,,,,,,,,”23, “Greenfields” TheStreet”, “The Town”,,,,,,etc.
● No columns should be omitted, even if they are effectively blank.
● Fields marked as mandatory must be completed with valid input otherwise the data row will be rejected.
● Fields marked as optional can be left blank. If completed the input must be valid, otherwise the data in that field will be rejected.
● Fields marked as conditional will be mandatory or optional depending on circumstances.
For further information in relation to the SIF, please refer to the 'SIF guide'.