com.ibm.streamsx.parquet 1.1.0

• Supports writing files in parquet format to file system, #7

Parquet Toolkit 1.0.3 for InfoSphere Streams 4.x.x

NULL support for rstring, timestamp and floats(32/64).

Parquet Toolkit 1.0.2 for InfoSphere Streams 4.x.x

This release contains:

• List/Map of basic types support added
• Added the following types support: INT8,UINT8, INT16, UINT16, UINT32. All mapped on int32 type in parquet schema
• Logger: add trace enablement verification
• High-availability policy changed to better support datanode temporarily unavailability
• Tuple-based partition generation
• Optional output port containing dynamic partition meta-data. Can be used for automatic partition addition.
• Close on punct support
• Parameter "autoCreate" controls if empty files should be created when no data arrived
• Migration to the parquet 1.6.0 and hadoop client 2.7.2

Big Thanks to Michael Bhola for very valuable contribution to this release!

Parquet Toolkit 1.0.1 for InfoSphere Streams 4.0.1

This release contains:

• bug fixes related to flatten (no partitioning) files generation.
• Streams 4.0.1 support
• new samples including BigSQL ddl's, scripts for samples build and run from command line

The tookit was tested with BI and BigSQL 4.0.1 and 4.1.