Posted: 12/19/2011
Hello All,
I have a package which extracts the file from specified location and loads into a table. Rightnow I am doing it as a daily process. Everyday I will be having a new file concatenated with the date(data_mart_account_2011-12-19) and I should import the file into a table. I have scheduled in a sql agent to run the job for importing a new file into a table everyday. Rightnow it is working fine. But the issue is I need to load some history files from dec 1st to dec 12th and those files have less number of columns than the current file. I mean data_mart_account_2011-12-01 to data_mart_Account_2011-12-12 has 10 columns and data_mart_Account_2011-12-18 onwards has 14 columns. I need to load the files of dec 1st to dec 12th using the same package as I am doing for the current file. Since the mappings differ in the source files I am unable to use the same package to load the history files.
Is there a way to achieve this. Please assist.
Thanks
Posted: 12/20/2011
Hi sqlkala.
while I was reading your post, I was thinking in suggest you to take package with 14 columns and remove those 4 colums in order you to load files from Dec 1st to 12.
But since you want to load files with the same package with 14 columns, I recommend you to open the older files from Dec 1st to 12, add 4 columns and format them equally as the newer files. Is it a hard work to do, or is it feasible?
Unfortunately there is no way to add the columns for older files. We should load the files as it is. I am just creating an other package to load the older files.