The sample data is currently specified as a set of flat files where each row is a json blob to be copied into the appropriate DB table. This makes it difficult to use/share this data for other purposes, e.g. unit tests, api-tests, etc.
The idea is to split these files up so that each json blob (row/line) exists in its own json file.
- ALL sample data (po_lines, purchase_orders, etc.) is refactored as described above
NOTE: If possible name the json files such that it's clear what the data is for or the sample case they apply to, e.g. as opposed to a UUID.
NOTE: An appropriate directory structure should be used (not flat please)