Hi,
since IDs may not be changed in your case (I suppose so) you can
simply use existing ones for distribution algorithm.

TF0806500123;3 means that this deal was created at the year 2008, day
number 065.


On Sep 20, 11:01 pm, Jim Idle <[email protected]> wrote:
> On 09/20/2009 05:04 AM, [email protected] wrote:
>
> > The problem of using any data from the record is what happens when you
> > retrieve the record from the file. Let's say you want to se the record
> > with ID TF0806500123;3. T algorithm will not receive the record(
> > that's precisely what you are trying to obtain). The algorithm to
> > calculate the partition is used in all access to the record, so it
> > shouldn't use the record, since in a read it will not have that record
> > info.
>
> I meant that you can use the data in the record to re-write the record
> ID as the user program sees it, not calculate the ID the other way
> around. If the ID requested is xxxx-2008, then the file is in 2008, but
> you can store it without the 2008 and that might be a better hash
> pattern. The ID returned for a read can be changed on the fly to append
> the -2008 -2009 and so on. BUt you need to be able to read the record
> with just the ID given.
>
> Jim

--~--~---------~--~----~------------~-------~--~----~
Please read the posting guidelines at: 
http://groups.google.com/group/jBASE/web/Posting%20Guidelines

IMPORTANT: Type T24: at the start of the subject line for questions specific to 
Globus/T24

To post, send email to [email protected]
To unsubscribe, send email to [email protected]
For more options, visit this group at http://groups.google.com/group/jBASE?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to