Tue Jul 18 23:53:10 PDT 2006
- Previous message: [Slony1-general] Standards conforming strings
- Next message: [Slony1-general] 1.2 RC1 RSN
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On 7/18/2006 2:21 PM, Christopher Browne wrote: > Christopher Browne wrote: >> It sounds like "standards_conforming_strings" is something we need to >> talk about further. Seems like something where our timing is very right >> to fix it now, and thus avoid forcing 8.2 users to do any "instant >> upgrades." (In some ways, that's a lot to hope for, since 8.2 doesn't >> really exist, and probably won't be released 'til ~November, but we >> might as well avoid any anvils that we see hanging over our heads :-).) >> _______________________________________________ >> > After some discussion offline, it looks like we have an easy way to > address this; on 8.2, we detect the availability of this, and set all > connections to use "standards_conforming_strings = off" > > The change is that in 8.2, you can specify that the server prefers an > SQL standard scheme for representing backquotes (\) and such. > > Slony-I at present *needs* to NOT be "standards-conforming" because > versions 7.4, 8.0, 8.1 do not accept that kind of configuration. > > In effect, Slony-I will need to be "non-standards-conforming" on this > aspect of string handling until such time as we drop support for all > versions < 8.2... > > The open question, at this time, is whether or not to directly support > 8.2 with "scs=on"... It looks as though it should be pretty easy, but > it's debatable whether we want to force this to 1.2.0... Even if we did it would have to be configurable if Slony should actually use this on 8.2. I would expect it to be rather problematic to replicate between 8.1 and 8.2 otherwise. With that said, what would Slony actually gain from using scs=on? To my knowledge, scs=off just means that it's the application (our log triggers and slon) that has to be carefull about escaping and that there is less guarding on the side of the backend against injection attacks. As long as we can provide escaping for literals that is injection safe, I don't see what we win by using scs. Jan -- #======================================================================# # It's easier to get forgiveness for being wrong than for being right. # # Let's break this rule - forgive me. # #================================================== JanWieck at Yahoo.com #
- Previous message: [Slony1-general] Standards conforming strings
- Next message: [Slony1-general] 1.2 RC1 RSN
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Slony1-general mailing list