That's definitely one of the things that I got out of Phil's article. Contrary to popular belief, CSV is still used a whole lot especially in the shop I work in. Microsoft is seriously remiss in thinking they don't need to spend some time on it via T-SQL. The normal answer I get from MS or anyone closely associated with MS (some damned gold partner, whatever) is that we should be using SSIS. My thought on that suggestion is, NNNNNOOOOOOOOOOO!!!!!! We just converted our stuff to all T-SQL and final got rid of the train wreck known as SSIS last year. In previous companies, one of my major tasks was to write stored procedures that did imports because SSIS couldn't actually handle them quick enough.
One example was that an SSIS package was taking 45 minutes just to get one file ready for import to a table. We needed to process hundreds of such files per day and so 45 minutes of file prep time just wasn't cutting it. When I got done with it, we were importing 8 such files, validating their content, and merging the data with the final tables every 2 minutes.
My only regret was the amount of time I had to spend building the tools (and the 8K per row limit in SQL Server 2000 back then). I'd have loved to have some of the tools that MySQL and other database engines have available in them.
To Phil's point, "C'mon Microsoft"! You regularly spend time building features that very few people actually use... Can you catch up on the practical world, please???"
Sorry for the rant but Phil's article reminded me of the huge disappoint I have with Microsoft's poor file handling capabilities when that should be one of the most important and, no, SSIS doesn't cut it for the stuff that I do and neither does (gaaaahhh) XML or JSON bloatware.