Skip to content

Check response validity separate from flight matching in RID tests #1386

@BenjaminPelletier

Description

@BenjaminPelletier

Observed behavior
In certain RID tests, it appears that the conceptually-separate checks of 1) was the API response valid? and 2) was the flight present? are combined into a single check. This means that an invalid API response can lead to a very confusing failed check message, as per this example from scenarios.astm.netrid.v22a.NetworkedUASDisconnect:

Image

In this case, the flight is present (the lat, lng, and alt all exactly match a position for an injected flight in the appropriate time range), but I believe the API response was discarded because it does not validate against the schema due to HELICOPTER instead of Helicopter. Therefore, I would expect a message regarding an invalid API response, not a message regarding a missing flight.

Test check
Missing flight check, though there may be others that improperly combine multiple checks into a single check.

Difference from expected behavior
When an API response is discarded/rejected/ignored due to validation issues, that should be obvious to a reader of the test report. This is probably best accomplished by making checks more fine-grained (checking exactly one thing with narrow focus). In this case, the NET0710,1 and NET0340 requirements relating to the correct API implementation should be performed first in a separate check before we even reach determination of a missing flight.

Metadata

Metadata

Assignees

Labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions