It is mentionned in Cusor: that Cursor for forward pagination: tells the server where to start fetching the next set of results in a multi-page response. This applies only to queries without id. If an ID is specified, it supersedes any cursor/offset combinations. The cursor value specified here is from the pagination response field of a prior query.
I’m a bit confused since broadcaster_id is required parameter.
I still see some cursor values though when I query for a specific id.
What is the pagination for?
Is there a way to get the list of all current hype trains?
The endpoint returns Hype Train Events for a channel, each progression of a hype train is its own entry.
That line is referring to use of an actual id param, not broadcaster_id. If you specify id params it will limit results to those specific id’s. So if you request id=12345, you’ll only get results for that single event (which could be nothing if it doesn’t exist). There’s no pagination there because you can’t specify more than 100 id’s in a single request so can’t have more than 1 page of results anyway.
The pagination is so you can get the next page of results when not limiting the request to specific id’s. For example, by default the API returns the last 20 events, if you want to get more than that you would use the after param, with the value being the cursor, and that will get you the next 20. If you use first=100 then you’ll get 100 events per page. If you want to get all hype train events for a channel you would go page by page until there are no more results.