i'm running issue , cannot find way handle it. i'll provide example.
say i'm making api call amtrack using python requests information tickets. return coming in json format this:
{ "tickets": [ { "id": 12345, "external_id": "abcdefg", "created_at": "2015-04-27t22:55:29z", "updated_at": "2015-04-27t10:38:52z", "type": "commuter", "priority": "high", "status": "open", "recipient": "amanda", "requester_id": 1234567, "route_id": 1234567, "tags": ["round-trip", "commuter"], "via": { "channel": "web" }, "custom_fields": [ { "id": 27642, "value": "commuter ticket" }, { "id": 27648, "value": "reoccuring charge" } ], } } i generating json file 1,000,000 tickets.
i want put vertica table, use built in fjsonparser, , goes great, except custom_fields key. can table, column full of barely readable garbage can't put text here.
i'm trying figure way handle this. need data inside custom_fields, because of formatting, parsing string not feasible. considered adding custom fields own json file when write file after api pull. like:
cf_list = data['tickets']['custom_fields'] open(filepath, mode='w') f: json.dump(cf_list, f) the problem here have no way of tying custom_fields ticket came from. i'm needing help.
thanks!
add flatten_arrays=true fjsonparser. flatten nested arrays.
