All pull requests

Pull requests

• Ensure that nested objects are deserialized in the order of serialization

  bug fix
  #4874·robert-brunel opened Nov 10, 2026Contributor·

  ·
  1

  

• Cover indexes on unnested record types with several unnestings and other definitions in yaml-tests

  testing improvement
  #4870·g31pranjal opened Nov 10, 2026Member·

  ·

• Support referencing nested fields in struct prepared parameters

  bug fix
  #4868·hazefully opened Nov 7, 2026Contributor·

  ·
  4

• Fix test for verifying partition

  bug fix
  #4867·ohadzeliger opened Nov 6, 2026Contributor·

  ·
  

• Make PartitionBinarySelectRule reorient predicates towards the quantifier they are pushed to

  enhancement
  #4865·robert-brunel opened Nov 6, 2026Contributor·

  ·
  1

  34

  

• Update statThreshold value for HNSW operations test

  testing improvement
  #4863·shivanshrustagi opened Nov 6, 2026Contributor·

  ·
  1

• Draft fix for Issue #4170

  bug fix
  #4853·robert-brunel opened Nov 5, 2026Contributor·

  ·

• Support the SQL RANK() window function over a rank index

  enhancement
  #4845·hatyo opened Nov 5, 2026Contributor·

  ·
  1

• Automatically mark READABLE_UNIQUE_PENDING indexes READABLE once the last uniqueness violation is resolved

  enhancement
  #4828·jjezra opened Nov 2, 2026Contributor·

  ·
  20

• Fix DirectoryLayer-allocator races in core tests

  testing improvement
  #4827·ScottDugas opened Nov 2, 2026Collaborator·

  ·
  1

  3

• synthetic joined record type planning

  enhancement
  #4817·g31pranjal opened Nov 2, 2026Member·

  ·
  1

• Replace ExplodeExpression with a table function over an ExplodeValue

  DO NOT MERGEenhancement
  #4816·g31pranjal opened Nov 1, 2026Member·

  ·
  1

• Reduce log volume of FDBRecordStore.checkVersion

  cleanup
  #4810·alpaker opened Oct 30, 2026Contributor·

  ·
  1

  34

• Use APPEND_IF_FITS when saving compact nodes in HNSW

  performance
  #4809·alecgrieser opened Oct 30, 2026Collaborator·

  ·
  1

• Improve the planner’s reasoning about the ordering of at-most-one-record results

  enhancement
  #4801·robert-brunel opened Oct 29, 2026Contributor·

  ·
  1

  

• Add an index maintainer scenario testing framework

  testing improvement
  #4761·ScottDugas opened Oct 19, 2026Collaborator·

  ·

• Prefer COVERING index scans without a fetch over primary scans in the planning cost model

  enhancement
  #4752·robert-brunel opened Oct 15, 2026Contributor·3/3·

  ·
  1

  1

  

• Enable COVERING index scans to read computed index columns out of the index entry

  enhancement
  #4749·robert-brunel opened Oct 14, 2026Contributor·2/3·

  ·
  1

  2

  

• Introduce IndexEntryColumnValue

  enhancement
  #4748·robert-brunel opened Oct 14, 2026Contributor·1/3·

  ·
  1

  

• Support scalar subqueries in value positions

  enhancement
  #4745·robert-brunel opened Oct 14, 2026Contributor·7/7·

  ·
  2

  

• Introduce a RecordQueryMaxOneRecordPlan plan node

  enhancement
  #4742·robert-brunel opened Oct 13, 2026Contributor·6/7·

  ·
  

• Cost cached plans with a dedicated cost model instead of their plan hash

  bug fixRun mixed-mode
  #4739·hatyo opened Oct 13, 2026Contributor·

  ·
  1

  12

• Make MetaDataProtoEditor always rename the canonically named union field

  bug fix
  #4731·robert-brunel opened Oct 9, 2026Contributor·5/5·

  ·
  1

  

• Prepare ARRAY_AGG() for an in-call ORDER BY clause

  enhancement
  #4729·robert-brunel opened Oct 9, 2026Contributor·1/3·

  ·
  

• Make MetaDataProtoEditor reject renaming record types with indexes declared via the (field).index extension

  breaking changeenhancement
  #4715·robert-brunel opened Oct 6, 2026Contributor·4/5·

  ·
  1

  4