Open curquiza opened 2 months ago
I would like to work on this. But the only thing I had was, when calling the api from my local.
{method=POST host="localhost:7700" route=/indexes/movies/similar query_parameters= user_agent=curl/8.1.1 status_code=400 error=Using the similar API requires enabling the
vector storeexperimental feature. See https://github.com/meilisearch/product/discussions/677}
py
Is this available with 1.10?
It is available in 1.9, but you need to enable vector search. In the tests this fixture enables it.
Following this central issue
This is related to a newly introduced feature in Meilisearch v1.9.0: get similar documents.
Refer to docs for more information.
searchSimilarDocuments()
(adapt the case to the language) having 1 parameter. This parameter is an object corresponding to the body you should pass to thePOST /similar
route. Do not manage error, Meilisearch will handle them if a field is missing in the body. Do NOT implement the route forGET /indexes/{index_uid}/similar
, we will add it in a second iteration if requested..code-samples.meilisearch.yaml
get_similar_post_1
key and “translate” the following curl example by using the newly added methods: https://github.com/meilisearch/documentation/blob/04f26a35771305378ed0a506f4d60ef06d07bd7e/.code-samples.meilisearch.yaml#L1333-L1340