Clarification on RH API endpoints

[ubendre]

What is the difference between the RH API and the regular API.

[RichardSmithONS]

Custom endpoints were developed exclusively for Census 2001 components. The RH (Respondent Home) ones are no longer used and we have a ticket in the backlog to remove them. The EQ (Electronic Questionnaire) ones are still used by Social Survey applications so we are keeping these for now. The /eq endpoint does a bit more work that it really should in that it decides if the input should be forwarded to the /groupedpostcode , /bucket or /partial endpoints.

[ubendre]

Thank you for responding Richard. I see that the /addresses/groupedpostcode option does not work. I get a "Path not found". (I am using the docker compose version). However the addresses/postcode/ option works, and there is no /addresses/bucket in the Swagger.

So going forward should we be using the /addresses/eq itself, and let the API figure out whether it should forward to groupedpostcode, bucket or partial?

Will the EQ endpoint be permanent

[RichardSmithONS]

The bucket endpoint is an EQ only one whereas groupedpostcode is shared
https://github.com/ONSdigital/address-index-api/tree/develop/api-definitions/explore-the-api/eq/bucket
https://github.com/ONSdigital/address-index-api/tree/develop/api-definitions/explore-the-api/groupedpostcode

We're not doing anything with EQ at the moment, rather we're currently concentrating on getting the bulk matching part of the service to work inside the new IDS service, using a bulk manager API, BigQuery and PubSub (code not public).

[ubendre]

Thanks @RichardSmithONS . For all the API options that serve searching / parsing one address at a time, which endpoint is expected to be used in the long term? (If we are using the open source code, and a premium subscription to ONS data?)

For the bulk matching option, as you mentioned, if it is going to work inside the IDS service, and the code is not public, will this API endpoint be available for use here https://osdatahub.os.uk/docs ?