|Table of Contents|
Problems starting spoon
Q: When I start Spoon.bat in a Windows environment nothing happens. How can I analyze the problem?
Technically, most of the steps use optimization techniques which map column names into field numbers (e.g. The column with name "sid" is field 4). Having different row structures would cause these steps to break.
Q: Is it possible to add/mix different meta-data into one Dummy Step?
A: You can do this. But be aware that it will only use the first row meta-data to display all rows. Even with the "Safe mode" on, which is used to find issues with different data types, does not check for different meta-data. The reason is, that PDI keeps internally all the available precision and changes the format only when viewing (preview) or saving into a file for example. To solve, this issue, all meta-data in the incoming streams have to be the same.
About duplicate fieldnames in a transformation
A: One of the basic design principles in PDI is that all of the steps in a transformation are executed in parallel. This would require architectural changes to PDI and sequential processing would also result in very slow processing.
How to make a DB connection different for different Kettle installations (e.g. sandbox/staging/production)?