A common use we see for Event Data is a publisher who would like to see all of the events associated with their articles and other works. Quite often their query will look like this:
http://0-api-eventdata-crossref-org.libus.csd.mu.edu/v1/events?from-occurred-date=2021-01-01&until-occurred-date=2021-01-31&obj-id=10.5555/1010001
repeated for all of their published DOIs and run every few weeks. But what if I told you that there is a much better way to stay more up-to-date using fewer queries? Here’s how.
First, use from-collected-date and until-collected-date. This means that you only need to query each time period once and store the results locally: if any events were missed and collected later on you will get them in a later query. Using occurred dates mean you will need keep going back periodically to see if anything changed, meaning more queries and less accurate data.
Second, search by prefix using obj-id.prefix. The results will include events for all of the DOIs across all associated DOIs. But what if the number of events is very large? To handle this, you can use a small timeframe, such as one day. Also, use cursors to page through results and make sure you pick up everything.
Third, be polite! Let us know your email address using “mailto=” then if we need to get in touch about your queries we can.
A much better query to make would be:
http://0-api-eventdata-crossref-org.libus.csd.mu.edu/v1/events?from-collected-date=2021-01-01&until-collected-date=2021-01-01&obj-id.prefix=10.5555&mailto=me@email.com
followed by, for example
http://0-api-eventdata-crossref-org.libus.csd.mu.edu/v1/events?from-collected-date=2021-01-01&until-collected-date=2021-01-01&obj-id.prefix=10.5555&mailto=me@email.com&cursor=ea4ede71-51e2-47fa-940e-22ab374e7a95
You can run this query once per day and be sure you are picking up every single event for the entire portfolio.
If you would like more advice on how to use the API, see our user guide.