split a 1000K records text file using camel splitter

asinghal223 Nov 29, 2011 10:40 AM

Hi,

I am new to fuse esb and apache camel (usinf fuse esb 4.4.1). I have a fixed length text file of 1million records and I want to split it based on the record count. lets say 10 files each with 100K records. IS there any way to do that using camel splitter. I saw in one of the example that based on \n, the file was being split line wise but i wud want to do it based on row count.

any pointers wud be highly appreciated.

Thx,

Aashish

1. Re: split a 1000K records text file using camel splitter

davsclaus Nov 30, 2011 3:03 PM (in response to asinghal223)

You can implement your own Camel Expression and use that with the splitter. Then you can read the CXF data in chunks, such as X rows.

This should be implemented as an Iterator, which the Camel Splitter, then will use.
You would also need to enable streaming mode on the Splitter.

For example see the source code, such as the token pair iterator
https://svn.apache.org/repos/asf/camel/trunk/camel-core/src/main/java/org/apache/camel/support/TokenPairExpressionIterator.java

The API has changed slightly in Camel 2.9, so you would need to adjust accordingly in older Camel releases.
Actions