Problem: both integration and analysis tools have very similar modules that read models from disk and then operate on them. This causes duplication of code and creates two maintaining processes that can be merged into one.
Possible solution: create separate microservice in the root of repository that would operate like a blackbox: input is config dictionary like so:
I the future this microservice might deal with reading of csv files with models to creation tool so might need to account for that in naming and scalability.
Problem: both integration and analysis tools have very similar modules that read models from disk and then operate on them. This causes duplication of code and creates two maintaining processes that can be merged into one.
Possible solution: create separate microservice in the root of repository that would operate like a blackbox: input is config dictionary like so:
and the output is a generator of
Snapshot
objects and probably single method to show the number of snapshots or something like that.