A teammate and I were discussing our work experience and he said he'd worked in the tool called Informatica. His work involved working on extracting required information from EDI files and transforming them and then loading them into the databases. He said EDI files have unstructured data as opposed to formats like .CSV and .SQL and that in EDI files the columns are separated by the '|' symbol.
In my opinion I do not see any difference between EDI and CSV only the CSV is ',' separated and EDI is '|' separated then why would EDI files be classified as unstructured data?
Traditional EDI is NOT unstructured. EDI typically follows some kind of standard (X12, EDIFACT, TRADACOMS, etc) that specifically define what the structure of the data is. XML, CSV, and delimited files are also structured. They have a defined field delimiter, and a record terminator.
An example of unstructured data would be an Excel file that has multiple pieces of data in a proprietary format. There would be no record identifiers, and a data parser wouldn't be able to understand what the data is. It would appear as a stream of data / text, but wouldn't have any context a mapper would need to translate/integrate. A Word document or PDF could also be considered as "unstructured".