What's changed changed in version 2.0.1?

Specialized toolkits - release 4.3.1.0-prod20190605 > com.ibm.streams.teda 2.2.1 > What's changed? > What's changed changed in version 2.0.1?

  • Application Framework
    • The parameter releaseOnPunct of the <namespace>.lookup::LookupCache composite has "true" as new default.
    • In <namespace>.streams.custom::TypesCustom.spl the comments for the Table1 and Table1StreamType types have been updated.
    • The composite <namespace>.chainprocessor.reader.custom::CustomParserTemplate is renamed to <namespace>.chainprocessor.reader.custom::CustomReaderTemplate.
    • The parameter "compression" and "encoding" comments have been added to <namespace>.chainprocessor.reader.custom::FileReaderCustom.
    • The daysToKeep parameters are removed from configuration. Use the timeToKeep configuration parameters instead. The list of removed parameters:
      • ite.businessLogic.group.custom.daysToKeep
      • ite.businessLogic.group.deduplication.daysToKeep
      • ite.ingest.deduplication.daysToKeep
    • The ContextCheckpointStreamType type is moved from <namespace>.context.custom::ContextDataProcessor.spl to <namespace>.streams.custom::TypesCustom.spl.
    • The LookupCache operator has the new optional lookupFilter parameter to specify a filter expression that must be matched to perform the lookup.
    • The <namespace>.functions::stripPathFromFilename, <namespace>.functions::basename, and the internal com.ibm.streams.teda.internal.fileutils::getPathFromFilename functions do not exist anymore.
    • The new application templates provide the 2.0.0 version. The product version dependencies are updated:
      • IBM Streams: 4.2.0.0
      • com.ibm.streams.teda toolkit: 2.0.0
      • com.ibm.streams.db toolkit (Lookup Manager only): 2.0.0

      A code migration is required for existing applications as described in the Knowledge Center: Reference > Toolkits > SPL standard and specialized toolkits > com.ibm.streams.teda > Developing applications > Upgrading existing applications to the newest application framework

      After migration, use the new com.ibm.streams.teda.file.path::filename(rstring) instead of the removed stripPathFromFilename and basename functions, and the com.ibm.streams.teda.file.path::dirname(rstring) function as replacement for the getPathFromFilename function.

      The following customizable files are affected:

      • <namespace>.chainsink.custom::FileWriterCustom.spl
      • <namespace>.chainprocessor.reader.custom::CustomReaderTemplate.spl
      • copies of <namespace>.chainprocessor.reader.custom::CustomReaderTemplate.spl
  • Lookup Manager customizing
    • The structure of the customizing XML file changed
      • New attribute naming.
      • Command mapping is removed. Each application can use the init, update and delete command.
      • The IndexInInputFile attribute is optional now and you can remove it from the configuration that uses databases as data source
      • The new optional attribute DbAccessSpecification defines the reference to access specification that you declare in the 'connections.xml' file. You can use your own name for the data source and use a reference to the access specification name.
      • The new optional attribute DbColumnName defines the reference to the name of the database table column. You can define your own data source attribute names and use the references to the database table columns.
      • You can use the database queries to remove the repository entries. You can use the CSV and the DB queries in parallel file to remove of the repository entries.

Deprecated functionality

Functionality is deprecated when it is supported but no longer recommended and might become obsolete. In some cases, it might be advisable for you to plan to discontinue the use of deprecated functionality.

  • The 'Monitoring GUI' is deprecated because the Streams Console provides functionality that replaces the toolkit proprietary implementation.