XML file split based on number of records

[venkatagiri_gongadi]

Hi,

I have a requirement as follows,

PI needs to pick up xml file from source directory and split  file for every 10K records and generate a xml files then place it in target directory.

I have gone through some forums but did not help me,can you please provide me some inputs on the same.

Please also let me if there is a way to split xml file based on size and place xml files in the target directory.

Cheers,

Giri

[anupam_ghosh2]

Hi Giri,

            Where do you want to write the files after split? to NFS or FTP or SFTP location.

Regards

Anupam

[venkatagiri_gongadi]

Hi,

Needd to place files in NFS.

Cheers,

Giri

[former_member190293]

Hi Venkatagiri!

If your source structure isn't much complicated you could use approach with pure graphical mapping described in this thread:

As other option I would consider using java mapping with SAX parsing as it works faster than DOM and consumes less of resources.

Regards, Evgeniy.

[former_member186851]

Hello Venkat.

You can try the split approach mentioned by Venkat using the UDF in the below link

Split Mapping using UDF - Process Integration - SCN Wiki

[manoj_khavatkopp]

Try with multi-mapping and an udf to split records after 10k records it should work.

and also splitting on size is possible but it wont work in your case as this is xml file and will result in invalid xml by splitting based on filesize.

br,

Manoj