I have several Excel files that require manipulation before they can be loading into a database table via SSIS. I was looking into using OpenXML 2.0 SDK to do this, but it looks like for some of my manipulations this won't be too feasible. Some of the Excel files have as many as 10-15 "header" rows that are deleted, and then some have columns being inserted that hold parsed data from another column.
I then figured that it would probably be a better solution to use OpenXML to parse the Excel file into a new CSV file. Upon some research for that I was finding that if cells are blank it skips them which could cause a varying number of columns in each row.
Is there a good solution to approach this that would have examples? I'm trying to automate the load process on our server and trying to avoid using Excel itself since that isn't meant to be run from a server due to all the messageboxes it throws.
Your input is appreciated. Thanks