Jeff Moden (8/3/2015)
Heh... ironically, I absolutely agree with the idea of writing SQLCLR for file handling, etc. Equally ironic is the amount of FUD concerning such CLR endeavors and I've actually found (read that as "feels like") more people would accept the use of xp_CmdShell than those who would accept the use of SQLCLR. Considering that I had some pretty idiotic SQLCLR code submitted to me for review in the past, I can't say I blame people much. The most idiotic CLR I ever had to turn down was one that calculated MODULUS because the "developer" didn't know (or even try to find out) that SQL Server has a modulus operator built in.
Rather strangely (because I think it would be better to extend SQL to do some of the things I currently would have to use SQLCLR for) I think SQLCLR is an extremely good thing; but, in the hands of brain-dead idiots (or which there are far too many screwing up databases, whether in CLR or in SQL) it rather obviously could be a bit of a pain, just as hopelessly bad as incompetently written XPs used to be.
Tom