Release notes for version 2.0.6 on June 20, 2011
The WRITE Adapter property AutoCommitAfter
allows you to control how FlowHeater builds import/update transactions when writing to a database.
- 0 = a single transaction is used for the complete import/update process (default)
- X = (a number) after each X rows are written, an AutoCommit is executed
- -1 (new) no database transactions are used
The new (-1) option disables database transactions (each row is effectively committed as soon as it is written) and this is useful when used together with the ODBC Adapter or OleDB Adapter when the underlying ODBC/OleDB database or driver cannot cope with database transactions.
Release notes for version 2.0.5 on June 6, 2011
Replace Heater with extended settings
The Replace Heater is enhanced with some powerful capabilities. You can now replace tabs, line breaks and define complete lists of replacements. Replacement lists can consist of two alternatives:
Release notes for version 2.0.4 on May 16, 2011
The MySQL Adapter now permits space characters embedded in field names.
In READ (import) mode, the File Heater can also cope with an HTTP URL as file location.
Release notes for version 2.0.0 on April 2, 2011
Lots of improvements and new features were integrated into FlowHeater for version 2 . New feature highlights include: a new Adapter for OleDB data sources, a new and powerful Heater for importing/exporting collections of external files , a new RAW data type. The latter permits almost all previously unrecognized SQL data types to be processed. See below for further details and a complete list of all the new features.
There are no problems in executing Definition files from version 1.x with the latest FlowHeater and there are attractively priced upgrade options for holders of version 1.x licenses.
Release notes for version 1.3.2 on March 5, 2011
Bug Fix: GroupBy Heater
When the GroupBy Heater was used together with the Filter Heater in a Definition and the first record / data row was filtered out by the Filter Heater, processing of the first grouping failed to process properly. This problem only occurred in Memory Mode, whereas Mass Data Mode was not affected. The problem is resolved with this release.