914: lazily create an index on documents push r=LegendreM a=qdequele
Create an index if it's possible when a user trying to send data to a non-existing index. https://github.com/meilisearch/MeiliSearch/issues/918
Co-authored-by: qdequele <quentin@meilisearch.com>
Co-authored-by: qdequele <quentin@dequelen.me>
984: Add test search r=LegendreM a=LegendreM
- Get an error if the index does not exist
- Get an error if a parameter is not expected (e.g.: "lol")
- Check a basic search with no parameter
- Check a basic search with only a q parameter
isssue #814
Co-authored-by: many <maxime@meilisearch.com>
946: Sort displayedAttributes field r=MarinPostma a=gorogoroumaru
Fix#943
displayedAttributes use the HashSet struct which is an unsorted structure, so I changed the implementation from HashSet into BTreeSet.
Co-authored-by: gorogoroumaru <zokutyou2@gmail.com>
976: Revert 944 r=MarinPostma a=MarinPostma
revert #944
@bidoubiwa @curquiza @eskombro, this was a misunderstanding from our side. Doing this would in fact be an error, and would prevent us to do this: https://github.com/meilisearch/MeiliSearch/issues/945#issuecomment-685526678, which is what we are really after. We are resetting this to its default behaviour before it goes in prodution. Sorry for the confusion.
Co-authored-by: mpostma <postma.marin@protonmail.com>
* trigger backup importation via http route
* follow backup advancement with status route
* import backup via a command line
* let user choose batch size of documents to import (command lines)
closes#884closes#840
846: Change settings behavior r=LegendreM a=MarinPostma
partially implements #824.
Returning the field distribution for all know fields is more complicated that anticipated, see https://github.com/meilisearch/MeiliSearch/issues/824#issuecomment-657656561
If we decide to to it anyway, and find a reasonable solution, I will make another PR.
fix#853 by resetting displayed and searchable attributes to wildcard when attributes are set to `[]` in the all settings route. @curquiza @bidoubiwa can you confirm me that this is the expected behavior?
Co-authored-by: mpostma <postma.marin@protonmail.com>
