Right now, the pipeline code won't run an ingest (or the ontology transform) if the output file already exists. We should an option (force?) to override that and run the ingests/transform anyway
We should be able to
[x] Run a single ingest (with or without the merge, defaults to force mode?)
[x] Run all ingests (with or without the merge, force mode optional but not default)
[x] Merge (this should still require upstream ingest stuff to happen / files to exist)
[x] Support for writing to a release directory, could just be an output dir argument, or we could go with --release with an optional argument to specify a release name - but by default --release would write into a YYYY-MM-DD directory
Let's make our CLI more friendly & useful
Right now, the pipeline code won't run an ingest (or the ontology transform) if the output file already exists. We should an option (force?) to override that and run the ingests/transform anyway
We should be able to
--release
with an optional argument to specify a release name - but by default --release would write into a YYYY-MM-DD directory