Re: how do i split a string
From: Bob Jones <email_at_me.not>
Date: Sat, 18 Oct 2008 13:33:58 -0500
Message-ID: <g_pKk.922$%11.556@flpi144.ffdc.sbc.com>
> If you look again on original topic, you may probably notice, that the
> question was to find 1st and 2nd occurence, and "3263" was just an example
> of required output. I took the freedom to provide an alternative solution
> to question and disregard the example just because (again, in my opinion)
> that example wasn't very suitable - if either length or position is
> greater than 9, output becomes ambiguous.
> On the other side, there are a lot of well known string aggregation
> techniques in oracle, which you probably are aware of, so, if OP (or
> someone else) insist on that exact output format - it is matter of some
> (few) additional lines of code to get it
> Now, i hope answered your questions in context of this thread, and as we
> start to be more and more offtopic, i assume, we can safely finish this
> discussion.
> Best regards
>
> Maxim
Date: Sat, 18 Oct 2008 13:33:58 -0500
Message-ID: <g_pKk.922$%11.556@flpi144.ffdc.sbc.com>
"Maxim Demenko" <mdemenko_at_gmail.com> wrote in message
news:48FA24B4.2060103_at_gmail.com...
> Bob Jones schrieb:
>> "Maxim Demenko" <mdemenko_at_gmail.com> wrote in message >> news:48FA0A98.7050201_at_gmail.com... >>> Bob Jones schrieb: >>>> Ok, I just wasn't sure why you were providing an alternative solution >>>> to a different problem. >>> You see - to find *all* occurences of repeated characters within given >>> string (meant - to find corresponding position and length) in my opinion >>> is not much different, it is just a more general case of what was >>> asked - to find *first two* occurences. Do you think, it isn't worth to >>> look for alternative solutions for more general classes of problems, if >>> a solution for a particular case is known? >>> >> >> I am all for alternative or better solutions, but you can't call it an >> alternative if it returns a different result. >> >> This result is not the same as "3263" that OP was looking for. >> >> 3 2 >> 6 3 >> 12 6 >> 20 2 >> >> Now you can probably write something does a lot more, but whether you >> need it is the question. >
> If you look again on original topic, you may probably notice, that the
> question was to find 1st and 2nd occurence, and "3263" was just an example
> of required output. I took the freedom to provide an alternative solution
> to question and disregard the example just because (again, in my opinion)
> that example wasn't very suitable - if either length or position is
> greater than 9, output becomes ambiguous.
> On the other side, there are a lot of well known string aggregation
> techniques in oracle, which you probably are aware of, so, if OP (or
> someone else) insist on that exact output format - it is matter of some
> (few) additional lines of code to get it
> Now, i hope answered your questions in context of this thread, and as we
> start to be more and more offtopic, i assume, we can safely finish this
> discussion.
>
> Best regards
>
> Maxim
I don't really have issues with alternative solutions. I just disagree with that all occurrences on multiple lines is the same as the first 2 occurrences on the same line. Received on Sat Oct 18 2008 - 13:33:58 CDT