Louis Dureuil

After discussing with @Kerollmops, here are some points that are still to be done after #3331 : - [ ] Check that the lock cannot become poisoned due to an...

Indexing performance concern: As we have to wait for the indexing task to fail when indexing, if the indexed data is big, we might resize the index multiple times, and...

A more recent prototype is available, [see my other message below](https://github.com/meilisearch/meilisearch/issues/3382#issuecomment-1436523578) --- A prototype that removes the restriction on the number or size of indexes is available! ### How to...

A second prototype that removes the restriction on the number or size of indexes is available! Compared with the previous prototype, this one removes the performance concern wrt index resize...

Good news, we reached a good first level in preparation for v1.1 RC0 :tada:. However, I'm reopening this issue because we have some follow-ups that should be addressed before closing:...

So I'm not sure how to best express this and *where in the specification this should appear*, but following the changes made for v1.1, the current status, relevant to the...

Discussed with gmourier and opened the following specification PR: https://github.com/meilisearch/specifications/pull/226 Issue description and labels are up to date

Hello @qdequele 👋 thanks for the feedback on my PR! > there is no way to know which hits if for which index Not from the response alone, indeed. The...

Done. Click to see request/response ```json ~/datasets ❯ curl \ -X POST 'http://localhost:7700/search' \ -H 'Content-Type: application/json' \ --data-binary '[{ "indexUid": "index-0", "q": "toto", "limit": 1 }, {"indexUid": "index-1", "q":...

Added some tests. If any of the search requests is faulty, the response does not specify which one. This is an ergonomic issue, do you think we should address this?