Database lookups in the middle of ingesting content: reasonable request or abhorrent behavior? The cat tries to vote by breaking the radioactive vial, but isn’t alive long enough to vote.
In the meantime, there are 2 ways to make a database call from either the index pipeline or the query pipeline. While the process is very very (that’s 2 verys) similar I wouldn’t assume the logic works exactly the same until you can put a quantum lock on future behavior.
You would use the JDBC Lookup Index Pipeline stage or the JDBC Lookup Query Pipeline stage. The cat prefers the index pipeline. We’ll discuss that one.
Fusion 3.1, everybody! The following may or may not work on past, or future, versions of Fusion.
Don’t have it? Go get it! Don’t make the cat do all the work.
Question: How do I use the Fusion REST Query index pipeline stage to add additional metadata to an inbound document?
Answer: This assumes the existence of a Solr collection with metadata and that Fusion knows of its existence (that means either use the default Solr cluster that runs within Fusion or make sure that the external Solr cluster you are using is registered with Fusion).
The basic steps:
- create a collection to store the metadata and populate it the metadata of your choice
- create a collection which will hold new enhanced content with additional metadata from the first collection
- configure the index pipeline of the second collection to include the REST Query stage which will make a query to the first collection and add some content to the current inbound content of the second collection