harmonisation pipeline using zooma

[amnonkhen]

start with this. Add final version as a Markdown file in the repo.

sequenceDiagram
    participant User
    box Harmonisation Service
        participant HS as HarmonisationService
        participant JS as JobService
        participant FMS as FieldMatchingService
    end
    box EBI Services
        participant Zooma

    end
    note right of Zooma: "initially use common Zooma, later a hosted"
    User ->> HS: harmonise(cohort dictionary)

    note left of HS: REST API

    HS ->> HS: start harmonisation
    HS ->> JS: create new job
    note right of JS: job should be a REST resource
    HS -->> User: harmonisation job id

    par async harmonisation job
        loop for each field 
            HS ->>  FMS: find closest field (field)
            FMS ->> Zooma: lookup by name in GECKO
            FMS -->> HS: field
            FMS ->> JS: update job document with progress
            HS --> HS: update harmonisation result
        end
    end

    HS -->> User: harmonisation result

[Mil-m]

https://github.com/EBIBioSamples/cohort-atlas-harmonisation/tree/33-zooma_endp

[Mil-m]

https://github.com/EBIBioSamples/cohort-atlas-harmonisation/pull/1/files