March 22, 2006 at 7:53 am
Hi All
Is there any way to set dynamic mappings in 'Data Flow Task'
That is, I am setting 'source' and 'destination' tables (OLDB Tasks) from Package Variables...
Since the structure of these tables are varying, I have to set the mappings dynamically..
Any way to achieve this through 'Script Task' or Any other ways?
Regards,
Agson Chellakudam
March 23, 2006 at 9:39 am
You cannot change the metadata (i.e. column info) of a data-flow at runtime.
-Jamie
Jamie Thomson
http://sqlblog.com/blogs/jamie_thomson
March 23, 2006 at 9:47 am
Hi All
After going through some links and trying myself I understood that, ther is no easy and direct way to handle this situation..
But I noticed a solution for a particular situation..
ie for me column names and number of columns are same for all the tables...
Only the varying factor is, its data types..
I will explain with an Example
Consider two tables, Source_Table(Col1 varchar(10)) and Dest_Table(col1 varchar(10))
I set these table names to two variables 'Source_Variable' and 'Dest_Variable'..
OLEDB Source and Destination Tasks are picking the respective table name from these variables
Package Executed Successfully...
Now Consider other two tables, Source_Table_Sec(col1 varchar(20)) and Dest_Table_Sec(col1 varchar(20))
Same variables are initialized with these table names...
While executing the package, I am getting a truncation error, because of Source_Table_Sec.col1 data length is > 10 (it is remembering previous metadata)
To handle this situation, I redirected all the raws(using Error Output of Source OLEDB) to another OLEDB Destination which will pick the table name form same variable (Dest_Variable)
So I am able to insert all the data to destination table woth out failing the component...
Regards
Agson Chellakudam
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply