Hi folks, thanks for creating Typesense! I have a ...
# community-help
c
Hi folks, thanks for creating Typesense! I have a qq (more details in 🧵)
I'm running the following query:
Copy code
{'q': 'quick todos', 'query_by': 'content', 'per_page': 50, 'limit_hits': 50, 'num_typos': 2, 'prefix': 'false'}
One doc indexed into Typesense has the following content:
Fast TODOs...
(there are also 1-2 other docs that somewhat match this query, but for the sake of this example, I'm focused on the Fast TODOs doc). This doc is not returned as a match. Is this expected behavior? If so, is there any way to make these partial matches get returned in the case of no better match being found? NOTE: with
{'q': 'todos', ...}
the doc is returned as expected
j
Could you check if the solution in this thread helps with your use-case: https://typesense-community.slack.com/archives/C01P749MET0/p1687882851838579
c
This fixes it, thanks so much!
🙌 1
Do you know off the top of your head how this might effect performance on large corpuses?
j
It is definitely more CPU intensive when you turn on exhaustive_search… Hard to give exact numbers because of the number of variables involved. But for let’s say under 500K records the difference shouldn’t be significant.
👌 1
c
Got it, got it, thanks! Another qq: what's the reasoning for using both
exhaustive_search
and
drop_tokens_threshold
? The docs mention that
drop_tokens_threshold
is ignored if
exhaustive_search
is turned on
j
Oh you’re right! I forgot that little detail. So just
exhaustive_search: true
should suffice
❤️ 1