RedisGraph 1.2.2 (May 2019)

Update urgency: Medium This is a maintenance release for version 1.2.

This release improves overall stability and provides fixes for issues found after the previous release.

RedisGraph 1.2.0 (April 2019)

This is a major release for RedisGraph. Compared to previous release 1.0.15:

  • Added functionality
    • #452 Support multiple relationships of the same type R connecting two nodes
    • #392 Lexer elides escape characters in string creation: support for property values with ' or "
  • Performance improvements
    • #442 Seek graph entity by id: MATCH (n) WHERE ID(n) = 5 RETURN n
    • #422 MATCH (n) RETURN COUNT(n) is now O(1)
    • #421 and #459 Lazy loading of matrices on subsequent writes of relationships and nodes
    • #399 Single transpose of matrices
  • Minor Bugfixes:
    • #456 Memory leaks
    • #439 When label doesn’t exists, the node count should return 0 #439
    • #435 Referencing non existing properties #435

Compared to 1.0:

  • Added functionality
    • #390 GraphBLAS 2.3.0 release notes
    • #452 Support multiple relationships of the same type R connecting two nodes
    • #392 Lexer elides escape characters in string creation: support for property values with ' or "
    • Enhanced cypher
      • Initial support for WITH clause: Allows query parts to be chained together, piping the results from one to be used as starting points or criteria of the next. ref p.78
      • #247 Initial support for UNWIND
      • #244 RETURN * (give me everything bro)
      • #236 TYPE function (returns the type of a relationship)
      • #288 Querying patterns where a relationship between two nodes can be of different types. ()-[:A|:B]->()
      • #252 Multiple MATCH clauses
      • #327 Multiple CREATE clauses
      • #305 MERGE + SET
      • #348 Smaller memory footprint - encode properties
    • Faster and better bulk loading
  • Performance improvements
    • #442 Seek graph entity by id: MATCH (n) WHERE ID(n) = 5 RETURN n
    • #422 MATCH (n) RETURN COUNT(n) is now O(1)
    • #421 and #459 Lazy loading of matrices on subsequent writes of relationships and nodes
    • #399 Single transpose of matrices
    • #393 Discard distinct when performing aggregation
    • #289 Index utilization when performing Cartesian product
    • #308 Granular writer locking
  • Most noticeable bugfixes
    • #456 Memory leaks
    • #439 When label doesn’t exists, the node count should return 0
    • #435 Referencing non existing properties
    • #359 Adding ORDER BY changes the number of returned hits when used in combination with LIMIT
    • #363 Remove graph entity property when it is set to null
    • #386 Return updated values on queries that modify data
    • #249 reset operation within execution plan should propagate upwards
    • #259 Replace operations appropriately when rewriting execution plan
    • #262 Entity returned from datablock should have its internals cleared
    • #264 Loaded triemap strings are not guaranteed space for a null terminator