All 7 RedisSearch (Full Text Search) questions
More in RedisSearch (Full Text Search)
What is RedisSearch, and what capabilities does it add to a Redis deployment? How do you create a search index in RedisSearch, and what does defining a schema for that index actually do? How do you write a RedisSearch query that combines full text search with a numeric range filter? How do you sort and paginate search results returned by a RedisSearch query? How would you design a RedisSearch based product search feature that supports faceted filtering, such as showing available categories and their counts alongside search results? What are the memory and performance tradeoffs of adding RedisSearch indexes to a large existing dataset in Redis?
Advertisement
Back to results for “Commands”
Redis › RedisSearch (Full Text Search)

How do you update or delete a RedisSearch index without affecting the underlying data it was built from?

Intermediate

Answer

You use FT.DROPINDEX to remove an index definition, with the option to also delete the underlying documents if desired or leave them intact if you plan to recreate the index later, and to change an index's schema you typically need to drop and recreate it with the new field definitions, since RedisSearch does not support modifying an existing field's type directly.
FT.DROPINDEX product_idx
FT.CREATE product_idx ON HASH PREFIX 1 product: SCHEMA name TEXT price NUMERIC brand TAG
Real-world example A team needing to add a new searchable field to their product index drops the existing index definition and recreates it with the updated schema, while their actual underlying product data in Redis remains completely untouched throughout the process.

Common follow-ups: Does dropping an index affect the actual underlying data by default?;How long does it take to rebuild an index for a very large dataset?

Redis CLI & Basic Commands;Data Types
All 7 RedisSearch (Full Text Search) questions