After further discussion, I'm planning to close this as notabug.
Does anyone else on the list have any objection to requiring the user to manage their own
pool naming schemes without us holding their hands through it (see below for more info)?
--
Kenneth Keiter
Sr. Software Engineer, Cloud BU
kenkeiter(a)redhat.com
http://twitter.com/kenkeiter
IRC: tigger or tiggerer
On Feb 20, 2012, at 2:40 PM, Matt Wagner wrote:
On Mon, Feb 20, 2012 at 02:27:21PM -0800, Kenneth Keiter wrote:
> Interesting.. I wonder why it didn't fail testing for me?
I wonder if you're using a different database. Postgres is
case-sensitive by default, but MySQL is not. I don't know about sqlite.
> Are we (canonically?) supporting unique names as case-sensitive, or insensitive? It
might just be best to leave this issue alone and allow the user to deal with it, rather
than creating too much trouble for ourselves. Thanks for testing it.
I was actually kind of secretly favoring that option, but decided to not
press the issue. I'm glad you did.
I don't really believe that the root BZ was an actual bug. It could be
argued that we should be case-insensitive, but I think you're spot-on in
suggesting that this would introduce an inconsistency -- we should
either be case-insensitive across the board, or we should be
case-sensitive across the board. Making a special case here because of
the bug would likely just make things worse in the long-term.
Combined with the fact that becoming case-insensitive now would preclude
us from using indexes on those fields, I'm strongly in favor of what you
propose -- leaving this issue alone. (And perhaps closing the BZ as
notabug, but I don't want to put words in your mouth.)
-- Matt