How can I configure Splunk to read a csv file from a sourcetype

[leandromatperei]

Guys

I have the following .csv file that needs to be captured by Universal Forwarder, but the data is coming in messy. Could you help me how to create a sourcetype so that they are indexed in the order of the first line?

"Data","Site","Tipo","Agencia","Posicao","RCAF","Nome","Status","mes_ano"
"04/03/2021","SP","Agência","1010","AS","TESTE","Claudio A.","OnHook (01:00:00)","03-2021"
"04/03/2021","","Agência","","Consultor","32323232","Claudio A.","OnHook (10:00:41)","03-2021"

[richgalloway]

Please show an example of the "messy" input.  Also, what do you mean by "in the order of the first line"?  The first line is a header that doesn't specify an order.  Besides, the forwarder always reads files from the first line to the last and there is no way to change that.