Search

Indexes over hashes and JSON documents, searched by text, tag, number, geography and vector similarity, with aggregations and hybrid ranking (FT.*).

31 commands in KiviDB 1.1.2. Every command.

• FT._LIST: Returns the names of every index.
• FT.AGGREGATE: Runs a query and transforms its results: groups them, reduces each group, computes new fields, filters and sorts.
• FT.ALIASADD: Creates an alias for an index: a second name, usable wherever the index's name is.
• FT.ALIASDEL: Deletes an alias.
• FT.ALIASLIST: Returns the aliases that point to an index (Redis 8.10).
• FT.ALIASUPDATE: Points an alias at an index, creating it if needed, or moving it from another index.
• FT.ALTER: Adds fields to an index.
• FT.CONFIG: Reads and sets the search engine's settings, by RediSearch's names.
  • FT.CONFIG GET: Returns settings matching a name or pattern, * for all, as [name, value] pairs.
  • FT.CONFIG HELP: Lists the FT.CONFIG subcommands and what each does.
  • FT.CONFIG SET: Sets a setting.
• FT.CREATE: Creates a search index over the hashes, or JSON documents, whose keys start with the given prefixes.
• FT.CURSOR: Reads and closes the cursors FT.AGGREGATE ... WITHCURSOR returns.
  • FT.CURSOR DEL: Closes a cursor before it is read to the end.
  • FT.CURSOR READ: Returns the next rows of an aggregation started with WITHCURSOR, COUNT n at a time, and the cursor to read on with: 0 once there are no more, and the cursor is closed.
• FT.DEBUG: Maintenance of search indexes.
  • FT.DEBUG COMPACT: Starts a compaction of an HNSW vector field: a rebuild in the background that drops the tombstones of deleted vectors.
• FT.DICTADD: Adds words to a dictionary: a named list of words, which FT.SPELLCHECK can include or exclude.
• FT.DICTDEL: Removes words from a dictionary.
• FT.DICTDUMP: Returns the words of a dictionary, sorted.
• FT.DROPINDEX: Deletes an index.
• FT.EXPLAIN: Returns how a query is understood, as Redis writes a query plan: INTERSECT { ... }, UNION { ... }, NOT, OPTIONAL, TAG:@field { ... }, NUMERIC {lo <= @field <= hi}, VECTOR { ... }, a clause to a line.
• FT.EXPLAINCLI: FT.EXPLAIN, as an array of its lines.
• FT.HYBRID: Searches an index by text and by vector together, and fuses the two rankings into one.
• FT.INFO: Returns an index's definition and state: its key type and prefixes, its fields with their options, how many documents and terms it holds, and whether it is still indexing.
• FT.PROFILE: Runs a query, SEARCH, AGGREGATE or HYBRID, and returns its result with a profile: total, parsing and lock-wait times, warnings, and the tree of iterators and result processors the query used.
• FT.SEARCH: Searches an index and returns the matching documents.
• FT.SPELLCHECK: Suggests spellings for the words of a query that are not in the index.
• FT.SYNDUMP: Returns an index's synonym groups, as each word followed by the groups it is in.
• FT.SYNUPDATE: Creates or updates a synonym group of an index: words to be treated as the same.
• FT.TAGVALS: Returns the distinct values of a TAG field across the indexed documents, sorted.