Permalink
Commits on Jun 2, 2018
  1. Add project strategy to edge induced subgraph

    s1ck committed Jun 2, 2018
    * also adds validation strategy for regular subgraph
    * fixes #796
Commits on Mar 5, 2018
  1. [#744] Implement ApplyTransformation on Tx Collection Layout

    s1ck committed Mar 5, 2018
    * Move ApplyAggregation tests into own test class
    * Automatically run ApplyAggregationTest on GVE and Tx Layout
    * Automatically run ApplyTransformationTest on GVE and Tx Layout
    * Enforce implementation for execution on GVE and Tx layout in Apply
    * Implement ApplyTransformation for Transaction Layout
    * Fixes #744
Commits on Mar 4, 2018
  1. [#752] Add minor changes to Gradoop examples

    s1ck committed Mar 4, 2018
    * Directory loading instead of explicit file loading in JSON Example
    * Update communities examples
    * Add method for DOT to PNG conversion using a system process calling `dot`
    * fixes #752
Commits on Feb 27, 2018
  1. [#742] Update version to 0.3.2-SNAPSHOT

    s1ck committed Feb 27, 2018
    * fixes #742
  2. Update README.md

    s1ck committed Feb 27, 2018
Commits on Dec 21, 2017
  1. Merge pull request #715 from dbs-leipzig/graph_construction_poc

    s1ck committed Dec 21, 2017
    [#703] Add basic graph construction via pattern
  2. [#703] Refactor graph construction code and update docs

    s1ck committed Dec 19, 2017
    * simplified code in CypherPatternMatching
    * moved helper parts to QueryHandler
  3. [#703] Add basic graph construction via pattern

    s1ck committed Nov 21, 2017
    [#703] Add basic graph construction via pattern
    
    filter embeddings by return pattern variables
    
    debug outputs
    
    implement add embedding functions based on a return pattern
    
    bugfix(embedding): add got accidentally called twice in map function
    
    remove debug outputs and enable outer right join in post processing
    
    implement label mapping for epgm elements that were created for the return pattern
    
    code cleanup
    
    clean code and write documentation
    
    refactor(CypherPatternMatching): label mapping contains newly created edges + vertices only
    
    fix(tests): avoid nullpointer exception by initializing empty map
    
    reformat code
    
    add missing documentation
Commits on Dec 19, 2017
Commits on Oct 4, 2017
Commits on Sep 13, 2017
  1. [docs] Update readme

    s1ck committed Sep 13, 2017
    * added link to Flink Forward presentation
Commits on Sep 4, 2017
Commits on Sep 3, 2017
Commits on Sep 1, 2017
  1. [#679] Add null checks around property access

    s1ck committed Sep 1, 2017
    * added @nullable annotation to Element.getProperties to give compiler warnings
    * fixes #679
Commits on Aug 21, 2017
  1. [#668] Add Indexed CSV data source

    s1ck committed Aug 21, 2017
    * adds a CSV-based datasource that reads vertex/edge files partitioned by label
    * fixes #668
  2. [#671] Push label predicates down to layout

    s1ck committed Aug 21, 2017
    * if a vertex/edge has a label predicate, the corresponding graph layout method will be called instead of loading all vertices/edges and applying a filter transformation
    * ff used in combination with IndexedGVELayout, this avoids duplicating objects in a Flink dataflow caused by the filter transformation
    * fixes #671
Commits on Aug 6, 2017
  1. [#636] Remove redundant classes (#664)

    s1ck committed Aug 6, 2017
    * fixes #636
  2. [#533] Add IndexedGVELayout

    s1ck committed Aug 6, 2017
    * provides access to label-partitioned data sets
    * required for performance optimizations in Cypher engine
    * fixes #533
Commits on Aug 4, 2017
  1. [#658] Decouple graph representation from Graph and Collection

    s1ck committed Aug 4, 2017
    * extract graph / collection operations to interfaces
    * introduced factories for LogicalGraph and GraphCollection
    * de-coupled graph layout and  graph operations
    * removed GraphTransactions
    * added factories for graph layouts
    * added GVELayout and TxCollectionLayout
    * fixes #658
Commits on Jun 12, 2017
  1. [doc] update readme

    s1ck committed Jun 12, 2017
    * added gradoop-hbase module description
Commits on Apr 18, 2017
  1. [doc] updates articles

    s1ck committed Apr 18, 2017
Commits on Mar 30, 2017
  1. [docs] Update Readme

    s1ck committed Mar 30, 2017
    * added DIMSpan paper
Commits on Mar 29, 2017
  1. Merge pull request #584 from dbs-leipzig/#380

    s1ck committed Mar 29, 2017
    * adds Cypher Pattern Matching
    * fixes #380
  2. [#579] Integrated Cypher pattern matching in Gradoop API

    s1ck committed Mar 29, 2017
    * added UnaryGraphToCollection operator
    * convenience methods for logical graph
    * added post processing
    * integrated Cypher engine with existing pattern matching tests
    * added tests for Cypher Isomorphism/Homomorphism and variable length paths
    * added debug printer for embedding
    * fixed loop and path evaluation
    * fixes #579
Commits on Mar 22, 2017
  1. [#567] Grouping examples for BTW demo

    s1ck committed Mar 22, 2017
    * added few example programs
    * added RandomNodeSamplingRunner
    * updated PokecExample
    * updated checkstyle config to ignore empty catch blocks
    * fixes #567
Commits on Mar 6, 2017
  1. [#565] Fixed PlanTableEntry.toString()

    s1ck committed Mar 6, 2017
    * toString triggered estimator to run
    * fixes #565
Commits on Mar 5, 2017
  1. [hotfix] fixed build failure after rebase

    s1ck committed Mar 5, 2017
    * caused by method renaming
  2. [#545] Improve MergeEmbeddings

    s1ck committed Feb 24, 2017
    * pre-compute non-join columns
    * perform distinctiveness check only when necessary
    * replace list iterations with array iterations
    * exposed JoinHint in JoinEmbeddingsNode
    * fixes #545
  3. [#533] Add triple support to QueryHandler

    s1ck committed Feb 24, 2017
    * fixes #533
  4. [#469] Add basic greedy planner

    s1ck committed Feb 17, 2017
    * Implemented first version of a greedy query planner considering different query plans based on estimated cardinality
    * Updated QueryHandler to support access to auto-generated query variables
    * Changed AND semantics of CNF (removes duplicates)
    * Planner considers variable length paths
    * Fixed CNF generation for NOT expression
    * Added simple test cases to test correct estimated and actual result cardinality
    * fixes #469
  5. [#517] Add basic query plan estimator

    s1ck committed Jan 28, 2017
    * added join estimation for 1-hop edges and paths
    * added filter estimator (based on label)
    * introduced marker interfaces for join, filter and projection nodes
    * additional tests
    * fixes #517
  6. [#508] Moved query plan to separate package

    s1ck committed Jan 25, 2017
    * fixes #508