Split the attribute
ranking rule into two sub-ranking rules: field
and position
.
#653
loiclec
started this conversation in
Feedback & Feature Proposal
Replies: 2 comments 1 reply
-
Hi I am in favor of this as we have in our index data like that
When querying for "Smith", we'd like
So totally regardless of the position cos the position there really does not matter. |
Beta Was this translation helpful? Give feedback.
0 replies
-
Looks like new 1.6 version with it's proximity customization could solve this, not sure tho as my problem was on "implicit" |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Following meilisearch/meilisearch#3222 , I think it would be a good idea to split the
attribute
ranking rule into two sub-ranking rules.field
would prioritise documents containing query words in important fields firstposition
would prioritise documents containing query words appearing at the beginning of the fieldsNote that
field
interacts with the settingsearchableAttributes
and its functionality overlaps with another proposed ranking rule:words in [attribute set]
.The names
field
andposition
probably aren't the best. Other suggestions wereattribute:no-position
andattribute:position
.Note that technically, within the search engine, these two ranking rules are already implemented but not exposed to the user. In fact, the current rule
attribute
corresponds to the rulefield
immediately followed byposition
.Beta Was this translation helpful? Give feedback.
All reactions