make-data-count.uml

@startuml

skinparam BoxPadding 30

title Make Data Count components in Dataverse
autonumber "<font color=blue>"

actor user
box "Dataverse" #DDD
  participant datasetPage as "dataset\npage"
  participant downloadApi as "download\nAPI"
  participant metricsApi as "metrics\nAPI"
  participant datasetMetrics as "datasetmetrics\ntable"
  participant sushiParser as "SUSHI\nparser"
  participant citationParser as "citation\nparser"
end box

box "filesystem or S3" #DDD
  participant log as "activity-yyyy-mm.log"
  participant sushi as "sushi-yyyy-mm.json"
endbox

box "counter-processor" #DDD
  participant main.py
  database sqlite as "yyyy-mm.sqlite3"
end box

box "DataCite hub" #DDD
  participant reports as "reports\nendpoint"
  participant events as "events\nendpoint"
end box

box "Crossref" #DDD
  participant crossrefApi as "Crossref\nAPI"
end box

== Activity Logging ==
user --> datasetPage : GET /dataset.xhtml
datasetPage --> log : investigation/view
user --> downloadApi : GET /api/access/datafile
downloadApi --> log : request/download

== Log Processing ==

main.py --> log : read log nightly
main.py --> sqlite : write to SQLite
main.py --> reports : send SUSHI report to DataCite hub "reports" endpoint
main.py --> sushi : write SUSHI to filesystem or S3
sushiParser --> sushi : parse SUSHI file
sushiParser --> datasetMetrics : write metrics to datasetmetrics table

== Citation Retrieval ==

events --> crossrefApi : retrieve citations periodically 
citationParser --> events : retrieve citations nightly
citationParser --> datasetMetrics : write citations to datasetmetrics table

== Retrieving Usage Metrics ==

user --> metricsApi: request number of views, downloads, and citations
metricsApi --> datasetMetrics: get data from table
metricsApi --> user: return data

@enduml