-
Notifications
You must be signed in to change notification settings - Fork 4
The seed parameter
0 generates a Random Seed
null disables Random Sorting (normally then sorting is done in alphabetically order if possible)
1-50 passing a seed
seed is used in pagination, i.e., when there are two or more pages of results, to keep the sorting across all pages. When retrieving a high number of items in a dataset it is desirable to have only a limited amount of results in each page.
It is possible to activate seed in two ways: in the dataset, choose a pagenumber (the number of the result page that will be shown first) or a pagesize (number of items in each page, we’ll use 15 in this example) and set seed to 0. At the beginning of query’s Response Body you will see something like:
{ "TotalResults": 10564, "TotalPages": 705, "CurrentPage": 1, "OnlineResults": -1, "Seed": "43", "Items": [ {
The remainder of the Response Body contains the first 15 sorted items. If you now want to retrieve page 2, page 56, or any other, use 43 as seed and write 2, 56, or the desired value as pagenumber.
If you do not enter the seed, you could find an item that was already shown before, because the API can not guarantee that the same sorting is used in different queries.
This wiki contains additional information about the Open Data Hub alongside the Open Data Hub - Official Documentation 🔗 .