Re: regexp_like weirdness in 10G?

From: Vladimir M. Zakharychev <vladimir.zakharychev_at_gmail.com>
Date: Sun, 14 Feb 2010 13:46:46 -0800 (PST)
Message-ID: <e73fe9e8-7af8-4940-a852-cc4f6a988dbb_at_k41g2000yqm.googlegroups.com>

On Feb 14, 11:55 pm, Ken Quirici <kquir..._at_yahoo.com> wrote:
> I'm getting this column contents (the column is a varchar2(200)):
>
> 'K+CS2(26)/DP'
>
> matched by this:
>
> regexp_like(<col-name>,'^(\d)+\(B|\d\).*DP.*')
>
> Just in case, I also returned the column length and got 12, which is
> exactly the number of
> characters between the single quotes.
>
> I thought '^' meant start-of-string, in this case, start-of-database-
> column-string, so I
> should not be getting a match with a string starting with 'K+CS'>
>
> When I remove the '^' I get exactly the same results.
>
> I'm running this test in TOAD.

Unless the booze really got hold of me, your RE says: either "^(\d)+\ (B" or "\d\).*DP.*". "\d\).*DP.*" (a digit followed by literal ")", then at least one character, then literal "DP", then at least one character - note that end-of-line counts as a character here) perfectly matches the input.

RE is not for the faint-hearted. In fact, you need to be very precise in what you're after...

Regards,

   Vladimir M. Zakharychev
   N-Networks, makers of Dynamic PSP(tm)    http://www.dynamicpsp.com Received on Sun Feb 14 2010 - 15:46:46 CST