lewismc
commented
1 year ago Use case: say an outside collaborator successfully executes one or more POST /ingest request(s). At some later stage, one of the ICCAT Team wish to see the raw file.
We should allow possibility to export file based on following spec
Change
GET /tags/{tag_id}To
GET /tags/{tag_ids}?export=false&archive=true&format=...Note following changes
- tag_ids: now an array of one or more values
- export: by default false
- archive: if export=true then by default the raw files are compressed using patool which is already packaged with tagbase_server
- format: client can select which archive format to package the files by defining one of patools formats.
The service will then get the raw files for these tags, by default zip them up and send to the client.
Post #152 this issue will provide the Python implementation logic for getting a raw data file. Currently, io_utils#process_get_input_data writes files to
/tmp, which is absolutely fine for prototyping this however we will want to implement a persistent object store for longer term, stable storage. @tagtuna is going to clarify this with the ICCAT Team.