Integer Leading Spaces

We have recently started experimenting with Pentaho Data Integration; so far we have had excellent luck, it really is a fine piece of software. So far I have managed to tackle every problem I ran into thanks to all the previous forum posts and tracker info.

However, this one is just beyond me. The problem is getting really annoying:

- A REST Client retrieves JSON data
- The JSON data is processed by the JSON input step
- Which is in turn passed to a Table output step

Everything works fine, except that for the table output I get the following error:
Data truncation: Out of range value for column 'someinteger' at row 1

So I replaced table output with a text output step. I found that all integers got leading spaces for some reason:
I managed to replicate this issue with an XML Output as well.

Now, if the data is handled the following way by the stream:
- JSON Input interprets everything as Strings
- And everything is output as strings
==> Then things work. Except that I can't pass a string to a database's int column.

However, if I convert a string to an integer, no matter how, I get a leading space
- if JSON interprets it as an integer, I get the leading space
- if JSON interprets it as a string, and I convert it by select value, I get the leading space
- if JSON interprets it as a string and I convert it in the text output, I get the leading space

I have a wild guess this leading space causes the table output issue in the first place. Is there a way that I can sanitize these integers?
I am using Kettle stable release 4.4.0