Many the fish

This issue has been partially fixed through different improvements: - the [computed proximity between words has been reduced from 7 to 3](https://github.com/meilisearch/meilisearch/pull/4131) - The word prefix pair docids database no...

Hello @cloudrac3r, I understand that is counterintuitive that data stored in a database takes more space on the disk than the same data serialized in JSON, but Meilisearch is not...

Yes, 😄 Meilisearch is taking more space on the disk because it goes further than the other databases regarding pre-computed data structures. On the top of the inverted index, we...

Closing this issue because the initial question was answered, see you

Hello @northes and @LinHeLurking, Thank you for your reports, what we could try is to keep both versions in a word: `理性` would be normalized `lǐ理xìng性`. It looks weird, but...

Hmmm understood, 🤔 so we have to refactor the tokenizer to give us all the alternate versions of each token. I'm wondering if it's possible to retrieve all the possible...

Hello @yale8848, Meilisearch v1.6.0 doesn't bring any change in Chinese Language support. The work needed to fix this issue properly is too big to be done quickly without impacting the...

Hello @surenderaphro, The issue behind what you are trying is that you will deactivate the comma everywhere in the index, so if you have several searchable attributes, they will all...

Hello @surenderaphro, Could you please send your settings with a sample of documents (around 5)? this way, I can test your use case and see if there is any bug...

Hello @surenderaphro, After investigating your dataset, I don't see how your example showcases the issue. I may have misunderstood your issue. Could you explain what you expect with your example?...