Currently the JSON formatter performs several transformations of the data beyond formatting it as compatible JSON. This can be a problem for debugging or other environments that want to use human readable logs, but test the transformations that make this valid mozlog.
We should separate the two sets of concerns into two or more independent layers that can each be used separately.
Currently the JSON formatter performs several transformations of the data beyond formatting it as compatible JSON. This can be a problem for debugging or other environments that want to use human readable logs, but test the transformations that make this valid mozlog.
We should separate the two sets of concerns into two or more independent layers that can each be used separately.