Closed (fixed)
Project:
Image
Version:
6.x-1.x-dev
Component:
image.module
Priority:
Normal
Category:
Bug report
Assigned:
Unassigned
Reporter:
Created:
17 Feb 2008 at 20:12 UTC
Updated:
28 Apr 2008 at 00:06 UTC
Jump to comment: Most recent file
Comments
Comment #1
drewish commentedComment #2
HorsePunchKid commentedThe patch does fix the error in Postgres. But I think this just reintroduces the possibility of having duplicates, since one could (and almost certainly would) have the same node ID but a distinct random number. In other words, it's pretty much the same as not having the
DISTINCTthere in the first place. Unless I don't understand how that keyword works--but I'm pretty sure it applies to the entireSELECTlist.I don't know of any good solution for Postgres. However, the error is pretty bad, and an unlikely duplicate is better than a guaranteed error. We could fork the queries into MySQL and Postgres versions, I suppose.
Comment #3
drewish commentedthat's kind of a mixed review... should this really be committed as it is?
Comment #4
drewish commentedcouldn't it be DISTINCT(n.nid), RAND() ? that would make sure the nid is distinct...
Comment #5
HorsePunchKid commented#4 seems to work fine in Postgres. Not sure why that didn't occur to me.
:)Comment #6
drewish commentedgreat, committed to DRUPAL-5, DRUPAL-5--2, and HEAD.
Comment #7
Anonymous (not verified) commentedAutomatically closed -- issue fixed for two weeks with no activity.