5.16.35.16.3

DataSearch: Matching improvements and minor additions

We have made further improvements to the DataSearch integration for 5.16.3.

Automatic import criteria

The existing integration is very strict in the requirements of when a dataset will be automatically imported into Pure. The current requirement is that all related content (authors and organisations) must be matched to an already existing content in Pure. This meant that for the vast majority of datasets, only a very low percentage of these, would result in a direct import and instead become and import/update candidate. These requirements are still in place but are now configurable in the DataSearch configuration page. They will default to the most restrictive setting.

Matching

As a part of the paid API the DataSearch service provides enriched authors ids on the dataset. These are unstructured and therefore not directly linked to the author in the data we receive. We now use these ids to match to persons in Pure and then use these as a candidate list to do name matching. We will only match to a person if we have a unique match within the candidate subset of persons. 

Query limits

We have added another query limit, which can be used to specify the datasets you are interested in importing. The type limit will make it possible to limit to a certain amount of types as defined by DataSearch.

Matched count

In order to better highlight the effects of the query changes you might make, we have added a dataset count. This will show the amount of datasets found in DataSearch based on your current query. 

Full configuration page