v0.1.11

"In silico"

Notable changes:

• bump Dataflow Java SDK to 1.6.1 #209
• refactor subtractByKey and intersectByKey #206
• refactor HDFS #212
• fix ObjectFileIO in test #211
• allow usage of different ScalaTest specs #208
• bump ScalaTest and ScalaCheck versions
• various JobTest fixes and enhancements

v0.1.10

"Memento vivere"

Notable changes:

• mitigate local-mode BigQuery bug GoogleCloudPlatform/DataflowJavaSDK#331 (#193)
• detect BigQuery project from default gcloud config (#195) (active config coming via GoogleCloudPlatform/DataflowJavaSDK#348)
• bump Bigtable version to 0.9.1 (#199)
• add intersectByKey method (#202)
• include Class-Path jars from jar manifest in pipeline artifacts (#203)
• add ability to specify compression for text file (#205)

v0.1.9

"Lucem sequimur"

Notable changes:

• upgrade bigtable dependency to 0.9.0 (#187)
• add function to update the number of bigtable nodes (#188)
• Add ProtoBuf IO and improve serialization #182, #186
• use bytes in Avro for object file IO #178 (#180)
• [BEAM-371] Backport HDFS IO enhancements (#192)
• Fix temporary directory for local runner (#190)
• Fix parsing of timestamp strings with missing fractional seconds; add unit test (#185)
• use Semigroup[Array[T]] in DoubleSCollectionFunctions#histogram (#177)
• make KryoAtomicCoder thread safe 6974ea6
• Update scalacheck to 1.13.1 (#189)
• Bump algebird to 0.12.1 (#181)
• Lots of tweaks to decouple from Dataflow SDK and prepare for Apache Beam #114 #139, #175, #184, #191

Breaking changes:

• Object fie IO now uses byte array in Avro and is not compatible with earlier versions #180
• Bigtable IO now uses --bigtableInstanceId instead of --bigtableClusterId and --bigtableZoneId #187

v0.1.8

"Nemo saltat sobrius"

• Bump Dataflow Java SDK to 1.6.0 #168
• Fix app name detection #157
• Fix empty SCollection writes to HDFS Sink #158
• Support for multiple paths for Hadoop dist cache #144
• Remove getFloat and getInt in RichTableRow #162
• REPL build enhancements # 165 #167
• Fix local file output directory #172
• Decouple DataflowPipelineOptions #139

v0.1.7

"Spem gregis"

• Add skewed join with CountMinSketch #142
• Improve Bigtable saveAsMultipleTable #145 #152
• Remove V2 Bigtable API #149
• Add Bigtable netty dependency #141 #151
• Decouple Dataflow specific code #140
• Fix AvroCoder usage in scio-hdfs #154
• Tweak integration with rouzwawi/flo #147 #153

v0.1.6

"Sic infit"

• Improve toPrettyString in BigQueryType #24 #107 #103 #116
• Add support for hadoopDistCache #115
• Add scio-extra to scio-repl #110
• Fix MultiJoin #133
• Allow string formatting in BigQueryType.fromTable #134
• Bump bigtable dependency to 0.3.0 #102
• Fix HDFS authentication #124

v0.1.5

"Ad astra"

• Bump SDK to 1.5.1, add row filter support for Bigtable V2 API #93
• Allow flattenResults and destinationTable in BigQuery #88
• Add compression for HDFS output #85
• Add SCollection#closeAndCollect() in REPL #83
• Add BigQuery timeout #81
• REPL I/O commands #72
• Fix race condition in DistCache #69

v0.1.4

"Ad arbitrium"

• V2 Bigtable API
• SimpleAuth for HDFS source
• Windowing API improvements
• API refinement
• Lambda and closure serialization improvements

v0.1.3

"Ut cognoscant te"

• Add support for writing to multiple BigTable tables
• Enhance testing framework
• Bump DataflowJavaSDK dependency to 1.5.0
• Discover BigQuery project on REPL startup

v0.1.2

"Sapere aude"

• REPL enhancements
• GitHub release integration
• Bug fixes