Changes related to Meilisearch v1.8.0

[curquiza]

This issue gathers the changes related to the v1.8.0 of Meilisearch that will impact the integrations scope.

📅 Release date: 2024-05-06

Timelines & steps

Pre-release

• With the help of the Product team and this CI, define which integrations should be updated and how (New feature? Update README? Update tests?) -> Fill in the "What to implement?" section below in this issue 👇.
  Minial implementation: PHP, JS, Instant-meilisearch.
• Create a branch by running Octopus script: only open branches for the integrations we choose to update (defined in the previous step) + Kubernetes repository + Cloud provider repository (changing the version)
• Update integrations according to the decisions (cf "What to implement?" section below in this issue 👇)
  ⚠️ If possible, this step is done before pre-release, once the feature is ready thanks to the prototype released by the engine team
  • JS - implementations of search cutoff + AI changes
  • PHP - implementations of search cutoff + AI changes
  • Python - update tests related to AI changes
  • Instant-meilisearch - update meilisearch-js version
• Add code samples for the chosen up-to-date integrations with the new version of Meilisearch
• Update the library version of the related integrations and prepare the changelogs

Release day

• Release the integrations
• Merge the related PR in K8s repository
• Publish DevOps tools:
  • create the git tag
  • publish images (steps are in CONTRIBUTING.md)
• Open issues in the repositories that are not up-to-date with the latest version of Meilisearch (including code samples)
  • Update the Dart issue related to vector search about v1.8 changes
  • Add missing issues for search cutoff setting

What to implement?

Search cutoff - new index settings searchCutoffMs

Related issue in the engine: meilisearch/meilisearch#4488

Introduction of the new index setting named searchCutoffMs. Available via /indexes/:uid:/settings (GET, PATCH, DELETE) and /indexes/:uid:/settings/search-cutoff-ms (GET, PUT, DELETE)
Default value is null. Expects an int as value.

TODO:

• JS: [v1.8] New searchCutoffMs index settings meilisearch-js#1639
• PHP: [v1.8] New searchCutoffMs index settings meilisearch-php#630

The following integrations will not have the feature available for the release day, except if the community implements it: Dotnet, Dart, Go, Java, Python, Ruby, Rust, Swift

Hybrid/semantic search changes

Related issue in the engine: meilisearch/meilisearch#4481
Usage: https://meilisearch.notion.site/v1-8-AI-search-API-usage-135552d6e85a4a52bc7109be82aeca42?pvs=4

Changes:

• Fix tests (and code base?) to handle breaking changes about search
  • vector is not longer displayed in the search response
  • _semanticScore is no longer returned in the search response
  • When adding "showRankingScoreDetails": true to a semantic search query, the vector and its value are no longer displayed
• Add new models
  • REST embedder
  • Ollama
• Addition of distribution field in settings
• New semanticHitCount field in search response
• Now Meilisearch hides API key -> fix PHP & JS tests

Current state before implementation:
Vector search feature fully available in: PHP, JS, Python, Go
Vector search feature partially available (changes of v1.6 not implemented), and not impacted by breaking changes: Dart, Ruby
Feature not implemented at all: Dotnet, Rust, Swift, Java.

TODO:

• JS: [v1.8] Hybrid search improvements meilisearch-js#1641
• PHP: [v1.8] Hybrid search improvements meilisearch-php#632
• Python: fix python tests

The following integrations will not have the feature available for the release day: Python (but no issue because of breaking changes), Go (but no issue because of breaking changes), Dart, Dotnet, Rust, Swift and Java
Available by default in Ruby thanks to the no-strongly type