This document is a reference for the configuration of flat file data sources that are used as input the Connect batch sending system. The information provided here will assist in providing a standardized structure for flat-file inputs to conversions.

As the batch file is highly flexible, this document should be used only as a guide, specific unique requirements should be handled as necessary

We require a Public SSH key from our clients for authentication to the SFTP server. Please provide to your Agent Connect implementation manager, and we will then confirm completion.

If PGP is required, our Agent Connect Public PGP key allows the client to encrypt the files before sending is as follows (please notify your account manager when doing do):

All files delivered should have a name with one of the following structures.

Your "<company_subdomain>" is a unique name for a data source that will not change and can be referenced within the Connect application to uniquely identify the company the batch file is used, if you need help please contact your client services manager.

All files should be in ASCII format. Files should not contain any non-readable characters.

A field delimiter must be chosen. The preferred delimiter is the “|” (‘pipe’) character.

However, to use the “|” as a delimiter, it must be guaranteed that “|” will never occur naturally in any of the data in the file. If this is not possible, another delimiter must be chosen which satisfies this constraint. Once chosen, this character will be known as the ‘field delimiter’.

The string data type should be wrapped in opening and closing quotation marks (“”).

An example of a string data type:

The boolean data type only has valid values of: "true" or "false"'.
They should be wrapped in opening and closing quotation marks (“”).

An example of a boolean data type:

The array data type is a collection of string data types that are separated by a comma (‘,’).

They should be wrapped in opening and closing quotation marks (“”).

They should not contain spaces, leading, trailing, or in between words.
To separate words, dashes (-) or underscores (_) can be used.

An example of an array data type:

The first row of the file should be a header row that denotes field labels. This is a delimiter record that names each field in the Data section. This is the ‘schema’ for the data.

The position (left to right order) of each field label must correspond to the position of the data for that field in the Data section. The field labels are delimited by the chosen field delimiter.

Field labels should be wrapped in opening and closing quotation marks (“”).

The Header row must be delimited by a newline character (‘\n’).

An example of a header row:

The second and all subsequent rows are the data section. Each row must be homogenous, containing the same number of fields in each row, each field having the same meaning in each row and matching the field labels in the header row. Each field is delimited by the chosen field delimiter.

Field labels should be wrapped in opening and closing quotation marks (“”).

Each Data row must be delimited by a newline character (‘\n’).

An example of a data row:

The following fields and rows are suggested in the flat file.

Fields in bold are required for our system to successfully process a valid request.

​Batch File - Sample CSV File

Filename: blancapeak_20180828.csv

Decided field delimiter: “|”

CSV content:

We always advise uploading one sample file first after following all of the above recommendations before going-live. This allows us to map the .csv headers to the relevant data points in Agent Connect to confirm surveys will be processed correctly. We will then re-process the file on our end and then work with you on a go-live date/time.