Hi,

I am publishing data from an integrationin webMethods 6.x to a target integration in webmethods6.x.

In my integration I have fields whose column type in database is Number and output field type is java.lang.string in the standard select statement,and in canonical it is a string

1)When there are no values in the database for those fields,the publish integration in wm6.x is publishing data and it is published …which when saved to pieline at the target integration shows as below

123

The fields CREDIT_LIMIT and SET_OF_BOOKS_ID are showing as above

This is throwing an error at the target subscribe integration which is using transformer String to BigDecimal for these fileds

The error is ----ERROR TYPE : com.wm.app.b2b.server.ServiceException ERROR DUMP : java.lang.NullPointerException

2)But when the same data when it is published fro thesame integration developed with webmethods4.x it succesfully processingthe data.

When save to pipe line is used for wm4.x for the same it is showing

Can any one clear me the following

1)Why is the difference between the tags of wm6.x and wm 4.x for fields in save to pipe line
example: webmethods6.x—
webmethods4.x—

2)How will wm4.x will take a field whose column type in database is Number and output field type is java.lang.string in the standard select statement,and in canonical it is a string…If there is no value in the database for that field…

Will it take as blank unlike wm6.x which is taking as null

Regards
Surendra

Surendra,

I think the 6.x approach for dealing with empty tags is more correct than the 4.6 approach. This may require you however, to check for nulls before invoking your transformer or to build a more forgiving transformer that implemented a desired default behavior when attempting to map a null String to a BigDecimal.

Mark