Many the fish
Many the fish
## The algorithm The algorithm based on **Set** iterate over intervals of position giving **sets of docids** (`word_level_position_docids` database), and choose between the different **Query derivations** the best interval giving...
The goal of this issue is to use the work made in meilisearch/milli#660 at search time to reduce the list of languages that can be detected by Charabia. This would...
The current behavior of the Attribute ranking rule is to iterate over all possible word positions of each Attribute in order to sort documents. ## A criticizable relevancy This strategy...
# Summary A search with a phrase containing only stop words was returning an HTTP error 500, this PR filters the phrase containing only stop words dropping them before the...
# Pull Request > ⚠️ this PR is not meant to be merged. This PR deactivates Chinese tokenization forcing Japanese tokenization to be used by Meilisearch. It's a hotfix provided...
This PR changes the prefix database creation thresholds to reduce their impact on the indexing time. ## Explanations There are two thresholds that we can change in the prefix database:...
# Summary This PR aims to reduce the whole indexing time by reducing the indexing time of the word pair proximity database computing ## Why focus on the word pair...
# Pull Request The match positions object returned in `_matchesPosition` is wrong, the `start` field is in bytes length but the `len` field is in character length, both should be...
Having a typo tolerance penalty on `ngrams` and `split words` is not relevant for any word that has been split in another way than relying on a separator like dictionary-based...
The feature **Define searchable fields at runtime** introduced changes in the search database cache that could be rewritten in a better way, related comment: https://github.com/meilisearch/meilisearch/pull/3834#pullrequestreview-1488380472