home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ogicse!uwm.edu!linac!uchinews!gsbacd.uchicago.edu!cs_mj
- From: cs_mj@gsbacd.uchicago.edu (Mark Jaeger)
- Newsgroups: comp.databases.ingres
- Subject: Re: HELP! SQL (Ingres) problem
- Message-ID: <1992Dec13.111807.1@gsbacd.uchicago.edu>
- Date: 13 Dec 92 17:18:07 GMT
- Article-I.D.: gsbacd.1992Dec13.111807.1
- References: <1992Dec9.123448.10686@vax.oxford.ac.uk>
- Sender: news@uchinews.uchicago.edu (News System)
- Lines: 26
-
- In article <1992Dec9.123448.10686@vax.oxford.ac.uk>, mjl@vax.oxford.ac.uk
- writes:
- > As far as I can see SQL is not following simple logic here. The first
- > query is , two clauses A and B, which each evaluate to TRUE or FALSE.
- > The number of rows returned by the query 'A OR B' should be AT LEAST as
- > many as the sum of 'those returned by clause A run separately and the
- > number of rows returned by clause B run separately'. This is dictated
- > by standard boolean logic.
- >
- > As the second and third queries show, this is not the case. The sum of clause
- > A and clause B separately is 3930, but the query A OR B is only 1273.
- >
- > Is Ingres giving me incorrect unlogical results or am I missing some
- > Ingres feature.
-
- I may be barking up the wrong tree, but whenever I see weird behavior
- like this, I suspect nulls.
-
- Can you supply us with full table definitions ("create table" statements
- from the "copydb" utility) and some sample data?
-
- --Mark Jaeger internet: cs_mj@gsbvax.uchicago.edu
- Graduate School of Business yellnet: (312) 702-0328
- University of Chicago faxnet: (312) 702-0233
- Disclaimer: My opinions are my own and not those of my employer.
- Ich bin ein Virus. Mach' mit und kopiere mich in Deine .signature.
-