1174: Limit query words number r=MarinPostma a=MarinPostma
This pr adds a limit to the number of words taken into account in a search query. Using query string that are too long leads to huge performance hits and ressources consumtion, that occasionally crashes the machine. The limit has been hard set to 10, and tests have been added to make sure that it is taken into account.
close#941
Co-authored-by: mpostma <postma.marin@protonmail.com>
1207: fix homebrew name r=MarinPostma a=fharper
brew is the command, the package manager name is homebrew
Co-authored-by: Frédéric Harper <hi@fred.dev>
1185: fix cors issue r=MarinPostma a=MarinPostma
This PR fixes a bug where foreign origin were not accepted.
This was due to an update to actix-cors
It also fixes the cors bug when authentication failed, with the caveat that request that are denied for permissions reason are not logged.
it introduces a bug described in #1186
Co-authored-by: mpostma <postma.marin@protonmail.com>
1167: Update dumps ci r=LegendreM a=MarinPostma
Now that the dump test are re-entrant, they can be run from a multithreaded context, whereas they used to be ran from a single threaded context, in a separate CI task.
Co-authored-by: mpostma <postma.marin@protonmail.com>
1091: New tokenizer r=LegendreM a=MarinPostma
Integration of the new tokenizer to meilisearch.
- Tokenize and normalizes the query string for better search results
- Language sensitive tokenization and normalization during indexation
- better support for Chinese thanks to jieba (when Chinese characters are detected)
To do in a later PR:
- Use a common tokenization instance
- use tokenization for synonyms
close#624
Co-authored-by: mpostma <postma.marin@protonmail.com>
Co-authored-by: many <maxime@meilisearch.com>
