meilisearch v1.32.0 版本更新介绍
发布日期: 2026-01-12
版本号: v1.32.0
本次更新主要包含性能提升、问题修复和维护优化三方面内容。在性能方面,为日志搜索添加了详细的进度追踪与耗时记录,便于性能分析;并行处理文档操作显著提升了索引速度,插入四百万文档时实现七倍加速,性能随CPU数量线性增长。问题修复包括:解决向量排序时相似度相同文档的分桶问题,确保后续排序规则正确应用;修复修改可搜索属性后可能导致的数据库残留数据问题;在无转储升级时重建图链接以修复召回率问题。维护方面,将JavaScript SDK测试的包管理器从Yarn切换为pnpm以适配当前工具链,并升级了LRU依赖以修复内存安全问题。
更新内容 (中文)
🌈 改进
日志搜索性能追踪
为 Meilisearch 中的搜索操作引入了全面的进度追踪和日志记录功能。为搜索过程的每个步骤添加了详细计时信息,以提升可观测性和性能分析能力。
由 @ManyTheFish 提交,详见 https://github.com/meilisearch/meilisearch/pull/6072
并行提取有效负载中的文档操作
通过批量处理大量任务或并行处理大量记录来加速文档索引。我们通过并行提取各项变更并分配内部 ID 来加快有效负载的准备速度。使用四个 CPU 处理四百万文档插入时实现了 7 倍加速,且性能随 CPU 数量线性提升。
使用 skipCreation 的任务中的 indexedDocuments 字段不再精确反映执行的文档操作数量(特别是 POST 和 PUT 操作)。该计数可能高于实际操作数,但不影响计算过程,仅影响报告的计数值。在此我们优先考虑速度而非完美准确性,文档仍能像以前一样被正确索引。
由 @Kerollmops 提交,详见 https://github.com/meilisearch/meilisearch/pull/6080
🐛 错误修复
向量排序:桶分组相似度相同的文档
修复了向量排序的桶分组逻辑,确保相似度得分相同的文档被正确分组,从而保证后续排序规则能准确应用。
由 @dureuill 提交,详见 https://github.com/meilisearch/meilisearch/pull/6081
正确删除基于 FID 数据库中的文档
修复了当 searchableAttributes 从 ["*"] 更改为字段子集时,基于 FID 的数据库中会残留孤立数据的问题,该问题会导致搜索时出现数据损坏和警告。
由 @ManyTheFish 提交,详见 https://github.com/meilisearch/meilisearch/pull/6076
无转储升级时重建图链接
升级 hannoy 至 v0.1.3-nested-rtxns 版本,该版本修复了图相关的召回问题,并新增了重建图链接的方法以恢复先前格式错误的图结构。同时修复了无转储升级流程中的一个小问题:升级描述未正确显示,且相关操作未与升级正确关联。
由 @Kerollmops 提交,详见 https://github.com/meilisearch/meilisearch/pull/6055
🛠️ 维护与杂项
更新 JS SDK 测试使用 pnpm 替代 yarn
在 CI 工作流中将 JavaScript SDK 测试的包管理器从 yarn 更换为 pnpm,确保 SDK 测试套件能与当前工具链正确且一致地运行。
由 @Strift 提交,详见 https://github.com/meilisearch/meilisearch/pull/6075
适配 SDK 测试中的 JS 测试
更新了 JavaScript SDK 的 SDK 测试 CI 工作流。
由 @curquiza 提交,详见 https://github.com/meilisearch/meilisearch/pull/6050
将 lru 从 0.16.2 升级至 0.16.3
修复了 IterMut 中的 Stacked Borrows 违规问题。
由 @dependabot[bot] 提交,详见 https://github.com/meilisearch/meilisearch/pull/6087
完整更新日志:https://github.com/meilisearch/meilisearch/compare/v1.31.0…v1.32.0
更新内容 (原始)
🌈 Improvements
Log Search performance trace
Introduces comprehensive progress tracking and logging for search operations in Meilisearch. It adds detailed timing information for each step of the search process, enabling better observability and performance analysis.
by @ManyTheFish in https://github.com/meilisearch/meilisearch/pull/6072
Extract document operations from payloads in parallel
We accelerate document indexing by processing a large number of tasks in batches or a large number of records in parallel. We expedited the preparation of the payloads by extracting the various changes and assigning internal IDs in parallel. We achieved a 7x speedup on a four-million-document insertion using four CPUs, and the performance scales with the number of CPUs.
The indexedDocuments field in tasks using skipCreation no longer precisely reflects the number of document operations performed, specifically for POST and PUT operations. This count may be higher than the actual number of operations, but it doesn’t affect the computation; only the reported count is impacted. We prioritize speed over perfect accuracy here, and the documents are still correctly indexed as before.
by @Kerollmops in https://github.com/meilisearch/meilisearch/pull/6080
🐛 Bug fixes
Vector sort: Bucket documents with same similarity
Fixed vector sort bucketing so documents with identical similarity scores are grouped together, ensuring subsequent ranking rules are applied correctly.
by @dureuill in https://github.com/meilisearch/meilisearch/pull/6081
Properly Delete Documents from FID-Based Databases
Fixes a bug where changing searchableAttributes from ["*"] to a subset of fields left orphaned data in fid-based databases, causing corruption and warnings during search.
by @ManyTheFish in https://github.com/meilisearch/meilisearch/pull/6076
Rebuild the graph links when dumpless-upgrading
Bumps hannoy to v0.1.3-nested-rtxns, which fixes graph-related recall issues and adds a method to rebuild graph links to recover previously malformed graphs. Also fixes a minor issue in the dumpless upgrade flow where the upgrade description was not displayed correctly and related operations were not properly associated with the upgrade.
by @Kerollmops in https://github.com/meilisearch/meilisearch/pull/6055
🛠️ Maintenance and Misc.
Update JS SDKs tests to use pnpm instead of yarn
Updated the JavaScript SDK tests to use pnpm instead of yarn in CI workflows, switching the package manager across test configurations to ensure the SDK test suite runs correctly and consistently with the current tooling.
by @Strift in https://github.com/meilisearch/meilisearch/pull/6075
Adapt JS tests in SDK tests CI
Updated the SDK tests CI workflow for the JavaScript SDKs
by @curquiza in https://github.com/meilisearch/meilisearch/pull/6050
Bump lru from 0.16.2 to 0.16.3
Fix Stacked Borrows violation in IterMut.
by @dependabot[bot] in https://github.com/meilisearch/meilisearch/pull/6087
Full Changelog: https://github.com/meilisearch/meilisearch/compare/v1.31.0...v1.32.0
下载链接
- meilisearch-enterprise-linux-aarch64
- meilisearch-enterprise-linux-amd64
- meilisearch-enterprise-macos-amd64
- meilisearch-enterprise-macos-apple-silicon
- meilisearch-enterprise-windows-amd64.exe
- meilisearch-linux-aarch64
- meilisearch-linux-amd64
- meilisearch-macos-amd64
- meilisearch-macos-apple-silicon
- meilisearch-openapi.json
- meilisearch-windows-amd64.exe
- meilisearch.deb