- Add support for keyword
descriptionin$jsonSchema[Feature] - Integrate cargo tools to identify dependencies for pg_documentdb_gw [Feature] (#263)
- Add support for killSessions command [Feature] (#402)
- Support arbitrary database user and password [Feature] (#306)
- Fix crash in
bson_dollar_lookup_projectwhen matched documents array contains NULL elements from LEFT JOIN [Bugfix] (#465) - Fix crash in
$densifywithpartitionByFieldson shard key due to mismatched sort operators for INT8 partition expression [Bugfix] (#464) - Fix crash in
shard_collectionwhen implicit collection creation fails and returns NULL [Bugfix] (#462) - Fix crash in
$letwheninexpression is a constant [Bugfix] (#463) - Support for TTL cron job to repeat deletes in batches until the one minute budget is exhausted, instead of deleting one batch per index per minute.[Perf]
- Crash fix when zero rows reach $first/$last/$firstN/$lastN accumulators in $group stage with no $sort [Bugfix]. (#466)
- Support collation with find positional queries [Feature]
- Short-circuit in
$condruntime evaluation [Perf] - Support operator variables(eg: $map.as alias) in let variable spec [Bugfix]
- Support for
killOpadministrative command [Feature] - Fix
$addToSetbehavior and skip the top-level field rewrite because it's already done in the operator [Bugfix] - Performance improvements for $addToSet update operator up to ~70x for large existing and update arrays. [Perf]
- Removed feature flags
documentdb.enableCompact,documentdb.enableBucketAutoStageanddocumentdb.enableIndexHintSupport - Fix use-after-free segmentation fault in
$let[Bugfix] - Optimize
$makeArrayon constant expressions.[Perf] - Short-circuit in
$switchat parse time [Perf] - Enable ordered indexes by default. Can be turned off by specifying "storageEngine": {"enableOrderedIndex": false} for a single index or by turning off the
documentdb.defaultUseCompositeOpClassGUC. - Fix NULL document crash from
$in: []optimization on sharded collections.
- Top-level
letvariables and$$NOWsupported by default. - Fix collation support on find and aggregation when variableSpec is not available [Bugfix].
- Support
dropRolecommand [Feature] - Support
rolesInfocommand [Feature] - Fix concurrent upsert behavior, update the documents in case of conflicts during insert [Bugfix] (#295).
- Support collation with
$sortArrayaggregation operator [Feature] - Add support for keyword
requiredin$jsonSchema - Fix a segmentation fault when using ordered aggregate such as
$lastwith$setWindowFieldsaggregation stage. [Bugfix] - Fix crash when building lookup pipeline queries from nested pipelines and $group aggregates [Bugfix]
- Add basic support for compiling with pg18 [Feature]
- Drop unused environment variable
ENFORCE_SSLin dockerfile [Bugfix] (#313) - Remove the explicit dependency on the RUM extension (it's now implicit on the .so file). Flip to documentdb_extended_rum for PG18+ [Feature]
- Use the appropriate GUC for the user_crud_commands.sql [Bugfix] (#319)
- Provide Rust dev environment in devcontainer [Feature]
- Add extension that adds a gateway host that's run as a postgres background worker [Feature]
- Support sort by _id against the _id index using the enableIndexOrderbyPushdown flag [Feature].
- Improvements to explain for various scan types [Feature].
- Support schema enforcement with CSFLE integration [Preview]
- Validate $jsonSchema syntax during rule creation or modification(schema validation) [Preview]
- Add internal extension that provides extensions to the
rumindex. [Feature] - Enable let support for update queries [Feature]. Requires
EnableVariablesSupportForWriteCommandsto beon. - Enable let support for findAndModify queries [Feature]. Requires
EnableVariablesSupportForWriteCommandsto beon. - Add internal extension that provides extensions to the
rumindex. [Feature] - Optimized query for
usersInfocommand. - Support collation with
delete[Feature]. RequiresEnableCollationto beon. - Support for index hints for find/aggregate/count/distinct [Feature]
- Support
createRolecommand [Feature] - Add schema changes for Role CRUD APIs [Feature]
- Add support for using EntraId tokens via Plain Auth
- Support
$bucketAutoaggregation stage, with granularity types:POWERSOF2,1-2-5,R5,R10,R20,R40,R80,E6,E12,E24,E48,E96,E192[Feature] - Support
conectionStatuscommand [Feature].
- Add string case support for
$toDateoperator - Support
sortwith collation in runtime*[Feature]* - Support collation with
$indexOfArrayaggregation operator. [Feature] - Support collation with arrays and objects comparisons [Feature]
- Support background index builds [Bugfix] (#36)
- Enable user CRUD by default [Feature]
- Enable let support for delete queries [Feature]. Requires
EnableVariablesSupportForWriteCommandsto beon. - Enable rum_enable_index_scan as default on [Perf]
- Add public
documentdb-localDocker image with gateway to GHCR - Support
compactcommand [Feature]. Requiresdocumentdb.enablecompactGUC to beon. - Enable role privileges for
usersInfocommand [Feature]
- Support collation with aggregation and find on sharded collections [Feature]
- Support
$convertonbinDatatobinData,stringtobinDataandbinDatatostring(except withformat: auto) [Feature] - Fix list_databases for databases with size > 2 GB [Bugfix] (#119)
- Support half-precision vector indexing, vectors can have up to 4,000 dimensions [Feature]
- Support ARM64 architecture when building docker container [Preview]
- Support collation with
$documentsand$replaceWithstage of the aggregation pipeline [Feature] - Push pg_documentdb_gw for documentdb connections [Feature]
- Support index pushdown for vector search queries [Bugfix]
- Support exact search for vector search queries [Feature]
- Inline $match with let in $lookup pipelines as JOIN Filter [Perf]
- Support TTL indexes [Bugfix] (#34)
- Support joining between postgres and documentdb tables [Feature] (#61)
- Support current_op command [Feature] (#59)
- Support for list_databases command [Feature] (#45)
- Disable analyze statistics for unique index uuid columns which improves resource usage [Perf]
- Support collation with
$expr,$in,$cmp,$eq,$ne,$lt,$lte,$gt,$gtecomparison operators (Opt-in) [Feature] - Support collation in
find, aggregation$project,$redact,$set,$addFields,$replaceRootstages (Opt-in) [Feature] - Support collation with
$setEquals,$setUnion,$setIntersection,$setDifference,$setIsSubsetin the aggregation pipeline (Opt-in) [Feature] - Support unique index truncation by default with new operator class [Feature]
- Top level aggregate command
letvariables support for$geoNearstage [Feature] - Enable Backend Command support for Statement Timeout [Feature]
- Support type aggregation operator
$toUUID. [Feature] - Support Partial filter pushdown for
$inpredicates [Perf] - Support the $dateFromString operator with full functionality [Feature]
- Support extended syntax for
$getFieldaggregation operator. Now the value of 'field' could be an expression that resolves to a string. [Feature]
- Push $graphlookup recursive CTE JOIN filters to index [Perf]
- Build pg_documentdb for PostgreSQL 17 [Infra] (#13)
- Enable support of currentOp aggregation stage, along with collstats, dbstats, and indexStats [Commands] (#52)
- Allow inlining $unwind with $lookup with
preserveNullAndEmptyArrays[Perf] - Skip loading documents if group expression is constant [Perf]
- Fix Merge stage not outputing to target collection [Bugfix] (#20)
Initial Release