We do not provide real-time search capability

Why?

Our Harmonized APIs are designed to return complete entities for business-to-business data and process integration.

As such, we do not provide features to allow you searching for fragments of text contents, for example on projects or publications references, titles, or abstracts.

This is a voluntary design limitation, justified by traffic, bandwidth, and latency considerations. The scope of the Harmonized APIs is to be serve our member's IT systems, not their end-users.

How to do it

If you wish to provide your users with very responsive search by fragments of texts, or even more, real-time matches while they are typing search fragments (a now traditional "autocompleter"), you will have to implement this feature at your site. This will generate an important stream of requests from every end-user session, this is the reason why we do not provide such services.

Our recommendation is that:

1. You setup a local search engine technology (e.g. ElasticSearch)
2. You schedule a periodic job to use our time-based queries to detect all changes and process only the returned entities
3. In every entity, you select only the fields of interest, for example: reference, title, abstract
4. You load these fields into your search engine