Does the query optimizer coalesce identical calls to deterministic functions?
I’m thinking in particular of a statement like:
CREATE INDEX byFoo ON jtable (json_extract(body, ‘$.foo’)) WHERE
json_extract(body, ‘$.foo’) IS NOT NULL
where it would improve performance if the JSON were not parsed twice.
(Or if there’s any way to express this statement without writing the
json_extract call twice, please let me know. Assume that a substantial number
of rows do not contain a ‘foo’ property in their JSON, making it valuable to
use a partial index.)
—Jens
_______________________________________________
sqlite-users mailing list
[email protected]
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
