Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.databases.ms-sqlserver > #1851
| From | Jim <jgeissman@socal.rr.com> |
|---|---|
| Newsgroups | comp.databases.ms-sqlserver |
| Subject | Re: CASE issue |
| Date | 2014-11-28 16:07 -0800 |
| Message-ID | <MPG.2ee2ae00cb38741598968b@news.powerusenet.com> (permalink) |
| References | <547306f8$0$41749$c3e8da3$5d8fb80f@news.astraweb.com> <XnsA3EF7E8B63B6CYazorman@127.0.0.1> <m4vlom$q0c$1@dont-email.me> <XnsA3EFF04705958Yazorman@127.0.0.1> <m513rc$8vr$1@dont-email.me> |
In article <m513rc$8vr$1@dont-email.me>, erik.lennart.jonsson@gmail.com says... > > On 11/24/2014 11:37 PM, Erland Sommarskog wrote: > > Lennart Jonsson (erik.lennart.jonsson@gmail.com) writes: > >> I think (don't have time to look it up now) the SQL-standard says that > >> the result of the CASE statement is the first condition that evaluates > >> to true. From your answer I get the impression that this is not true for > >> SQL-server, correct? > > > > That is indeed true. The issue is that the expression on the left-hand side > > is reevaluated for every branch in the CASE. > > > > I see, thanks > > > /Lennart I believe RAND() is evaluated once at the beginning, while NEWID() is evaluated each time. So one solution would be to evaluate the NEWID() version and park the result in a variable and use the variable, or use RAND() but base it on a seed that doesn't involve NEWID(), or at least if it does, determine it outside of the CASE statement.
Back to comp.databases.ms-sqlserver | Previous | Next — Previous in thread | Next in thread | Find similar
CASE issue "twenty-six@b-mint.net" <twenty-six@b-mint.net> - 2014-11-24 10:22 +0000
Re: CASE issue Erland Sommarskog <esquel@sommarskog.se> - 2014-11-24 11:26 +0000
Re: CASE issue bradbury9 <ray.bradbury9@gmail.com> - 2014-11-24 04:16 -0800
Re: CASE issue Erland Sommarskog <esquel@sommarskog.se> - 2014-11-24 14:28 +0000
Re: CASE issue Erland Sommarskog <esquel@sommarskog.se> - 2014-11-28 22:47 +0100
Re: CASE issue Jim <jgeissman@socal.rr.com> - 2014-11-28 17:19 -0800
Re: CASE issue "twenty-six@b-mint.net" <twenty-six@b-mint.net> - 2014-11-24 13:32 +0000
Re: CASE issue Erland Sommarskog <esquel@sommarskog.se> - 2014-11-24 14:25 +0000
Re: CASE issue "twenty-six@b-mint.net" <twenty-six@b-mint.net> - 2014-11-24 16:06 +0000
Re: CASE issue Lennart Jonsson <erik.lennart.jonsson@gmail.com> - 2014-11-24 17:18 +0100
Re: CASE issue Erland Sommarskog <esquel@sommarskog.se> - 2014-11-24 23:37 +0100
Re: CASE issue Lennart Jonsson <erik.lennart.jonsson@gmail.com> - 2014-11-25 06:25 +0100
Re: CASE issue Jim <jgeissman@socal.rr.com> - 2014-11-28 16:07 -0800
Re: CASE issue Erland Sommarskog <esquel@sommarskog.se> - 2014-11-29 10:43 +0100
csiph-web