Link to home
Start Free TrialLog in
Avatar of conardb
conardbFlag for United States of America

asked on

What task or combo of tasks can I use in ssis 2015 to load different file names into different tables into sql server?

I have a for each look container using folder and filename variables.  The initial dataflow task in the for each loop container does a conditional split on the file name (2 different file names) and inserts into a seperate initial staging table for each file.  The inital data flow has 2 precedence constraints with expression and constraint for the filename and then processes to seperate subsequent dataflows for each file.   Dataflow for file1 is reprocessing 1 extra file.  screenprint attached.
packageflow.docx
Avatar of Jim Horn
Jim Horn
Flag of United States of America image

What exactly is your question here?
Avatar of conardb

ASKER

Is using a dataflow task with conditional split on filename(s) to process different files with different layouts a proper approach to use ssis tasks?    What is your expert opinion?
If it works and is fast enough, use it. If you think you might need to support additional filenames in the future, you might want to consider what would facilitate doing that.

You mention both data flow tasks (shown in the attachment) and conditional splits, which are transformations within a data flow task. What is the purpose of testing the filename at both levels?
ASKER CERTIFIED SOLUTION
Avatar of Jim Horn
Jim Horn
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of conardb

ASKER

ok, thanks... Presently, I have 1 dataflow that is using conditional split logic to select by file name from the same directory each of 2 different files / formats and write to a different staging table for each.  Then seperate dataflows for each of those types derive the fixed length non-delimted row into  seperate sql server tables.  It's working but I suspected not best practice as the filter by name is redundantly used in the initial dataflow then afterward as a precedence to direct processing to derive to seperate tables .  Jim, could you provide an example or link to an example of your approach?  Thanks again.
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of conardb

ASKER

sorry I took so long, thanks again