Andrew Dunstan <[EMAIL PROTECTED]> writes: > We wouldn't be having this discussion at all if we had not had a > horribly long period beween feature freeze and beta.
I'm not sure about that. The bottom line to me is that we are doing a favor to the Slony and Skytools projects, who figured out *after* our feature freeze that they could use some common code but it would be too late if it wasn't available in 8.3. Even had the freeze period been only a couple weeks, that could have happened --- it was just the sort of unfortunate timing that the universe likes to play on humans ;-) There was some complaining in this thread that we were showing favoritism to the txid patch. I'll plead guilty instead to favoritism to the Slony and Skytools projects --- the patch would certainly have gotten bounced without the support of those projects. I don't have a good grasp on the importance of Skytools, but I do know about Slony, and I don't have a problem with cutting them a break on a small, low-risk patch, even if it is a feature addition. The patches that got passed over for 8.3 were neither small nor low-risk; in fact, the reason the process dragged out so long was exactly that we busted our butts to get in everything that had any chance at all of getting in. regards, tom lane ---------------------------(end of broadcast)--------------------------- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match