Milli v0.3.1

Changes

• Fix the benchmarks script and names (#213) @Kerollmops
• Add --locked in CI tests (#214) @curquiza
• Enable optimization for build.rs and macro (#218) @irevoire
• Use the index primary key when deleting documents (#221) @Kerollmops
• Save the criteria field name in the fields ids map (#219) @Kerollmops
• Make hard separators split phrase query (#220) @ManyTheFish
• Move the UpdateStore into the http-ui crate (#222) @Kerollmops
• Introduce integration test on criteria (#212) @ManyTheFish
• Enable optimization in every profile (#224) @irevoire

Thanks again to @Kerollmops, @ManyTheFish, @curquiza and @irevoire! 🎉

Milli v0.3.0

Changes

• Exactness criterion (#188) @ManyTheFish
• Make bucket candidates optionals (#190) @ManyTheFish
• Fix plane sweep algorithm (#200) @ManyTheFish
• Optimize roaring codec (#169) @MarinPostma
• Benchmarks (#207) @irevoire
• Check the benchmarks in the CI (#210) @Kerollmops
• Fix primary key behavior (#193) @MarinPostma

Breaking changes ⚠️

• Transfer numbers and strings facets into the appropriate facet databases (#184) @Kerollmops
• Make the MatchingWords return the number of matching bytes (#203) @ManyTheFish
• Add field id word count docids database (#202) @ManyTheFish
• Decorrelate Distinct, Asc/Desc, Filterable fields from the faceted fields (#204) @Kerollmops

Thanks again to @Kerollmops, @ManyTheFish, @MarinPostma, @curquiza and @irevoire! 🎉

Milli v0.2.1

Changes

• Remove tests on main (#183) @MarinPostma
• Provide an iterator over all the documents in a milli index (#185) @irevoire
• Fix fields distribution after documents merge (#187) @shekhirin

Thanks again to @MarinPostma, @curquiza, @irevoire and @shekhirin! 🎉

Milli v0.2.0

Changes

• Support query words limit (#160) @shekhirin
• Make clippy happy (#164) @irevoire
• Introduce the Attribute criterion (#122) @Kerollmops
• Optimize proximity criterion (#172) @LegendreM
• Do not use echo that espaces newline (#176) @yanns
• Remove excluded document in criteria iterations (#173) @LegendreM
• Make document addition number visible (#178) @MarinPostma
• Disable autogenerate_docids by default (#180) @shekhirin
• Upgrade Tokenizer version (v0.2.2) (#181) @curquiza

Breaking changes ⚠️

• Facet float-integer becomes facet number (#146) @Kerollmops

Thanks again to @Kerollmops, @LegendreM, @MarinPostma, @curquiza, @irevoire, @shekhirin and @yanns! 🎉

Milli v0.1.1

New features

• Add Stop words (#156) @irevoire
• Support synonyms (#158) @shekhirin

Other changes

• Introduce setting enum (refactor) (#148) @shekhirin
• Fix and refactor settings assert_(ser|de)_tokens (refactor) (#154) @shekhirin
• Update Tokenizer version to v0.2.1 (#159) @curquiza
• Implement distinct attribute (#125) @MarinPostma
• typo: wrong command in example (#162) @michaelchiche

Thanks again to @Kerollmops, @MarinPostma, @curquiza, @irevoire, @michaelchiche and @shekhirin! 🎉

First release ever 🎉

Merge pull request #141 from meilisearch/reorganize-criterion

reorganize criterion