Dear All Finance department has come up with the requirement to load a delimiter file s but mostly they will date (YYYYMMDD) appended to file so that they can keep the track and want to load it without changing name. E.g. For Jan 2011 US Local data file on 1-Jan-2011 will be: Us_local_20110101.csv UK Local data file on 1-Jan-2011 will be: Uk_local_20110101.csv For Feb 2011 US Local data file on 1-Feb-2011 will be: Us_local_20110201.csv UK Local data file on 1-Feb-2011 will be: Uk_local_20110201.csv And so on What I was looking for is a technique where I can load the file by specifying the names like ‘Us_local_*.csv’ and ‘Uk_local_*.csv’ Additionally if I can read the latest file for each category then I can keep the previous file in the directory rather than deleting it from the folder to have the latest files only. Please adviseDigitalFM

Doesn't make sense for me clearly. do you want to load these data into same table? or different table per each country and date? http://www.rad.pasfu.com

Do you intend to pick all the files for US or UK? or do you want to pick the latest file for any specific country? and yes then load the data to one table?

I want to load only the latest files each of UK and US and load in the separate tables i.e different table for each country data DigitalFM

what about structure of data? I mean is the structure of file and table differs for 'us' and 'uk' ?http://www.rad.pasfu.com

Have a look at my blog for some ideas.

When loading the files in SSIS, you can have two data flow tasks which have a different file-iterating Foreach loop in them. The first foreach loop would use the filter Us_local_*.csv and the second would use Uk_local_*.csv See http://www.sqlis.com/post/Looping-over-files-with-the-Foreach-Loop.aspx for screenshots and examples

this method will find last created file in the special directory, you don't event need to use foreach loop, just use script task. http://www.rad.pasfu.com/index.php?/archives/30-Find-Last-Created-File-in-Special-Directory-SSIS.htmlhttp://www.rad.pasfu.com

Yes structure is differnent beccase of regional requirements (European Union , USA)DigitalFM

Yes structure is differnent beccase of regional requirements (European Union , USA) DigitalFM In that case you cannot use any of the suggested methods. The data flow task is not designed for dynamic metadata.

Yes structure is differnent beccase of regional requirements (European Union , USA) DigitalFM In that case you cannot use any of the suggested methods. The data flow task is not designed for dynamic metadata.

you can use two separate dataflows with different mappings from separate directories or with different filters as mentioned above

you can use two separate data flows, as I think you are doing this already. but for finding last file of each data flow source go with my previous posted link. let us know if you have any problem.http://www.rad.pasfu.com

Hi Raza Yes I am using two separate data flows. I have see your link and I hope it will do what i am looking for but i am not sure where i should put that c~ code becase i have not work on this part of SSIS. Please advise DigitalFM

http://chanmingman.wordpress.com/2009/05/06/write-a-%E2%80%9Chello-world%E2%80%9D-in-ssis-sql-server-integration-services-using-c/DigitalFM