Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.misc -> Re: Year 2000 Issue

Re: Year 2000 Issue

From: Authorised User <bp_at_bp.com>
Date: 1998/02/18
Message-ID: <01bd3cc5$b4623200$826964a1@azmelw1358.mel.az.bp.com>#1/1

issues:

  1. trigger should be on update too - if you read a date, say 2003, into a Forms3 dd-mon-yy field, it can get switched to the wrong century, 1903, without you even noticing - I tested all this 18 months back, but can't remember specifics. Behaviour varies according to whether you have a format mask or not.
  2. not sure about performance of triggers under 7.1.6. try it and see. you'd be much better off on 7.3.x where triggers are precompiled.
  3. say you enter 01-jan-03 on screen. Forms has 01-jan-1903 internally. database switches that to 01-jan-2003. Now, you commit. Then you go to reedit the record. Forms checks the database to make sure that the record looks the same as it does in memory. Question is, will Forms think that 1903 internally is different to 2003 in the database? If it is, you'll need to requery before being able to edit. This should be simple to test - I just don't have the facilities.
  4. Certain dates should not use this - eg., birth dates.They need to be YYYY for entry and display, and no trigger.

Assuming all the above are non-issues, go with it.

Andrew

NiceGirl <nicegirl_at_nicegirl.com> wrote in article <34e256db.174177904_at_news.globalserve.net>...
> Our company is using Oracle 7.1.6 running forms 3.0. We have around 500
 tables
> containing date fields. The problem that we have is forms 3.0 does not
 work with
> 'DD-MON-RR' date mask.
>
> One of our developer comes up with a bright idea to solve our Y2k
 problem.
>
> Create a database trigger for all tables.
>
> create trigger y2000
> before insert on <tablename>
> for each row
> begin
> :new.date1 := to_date(to_char(:old.date1, 'DD-MON-YY'), 'DD-MON-RR');
> end;
>
> This looks like a very good and cheap solution and we can always enhance
 this
> trigger to make it foolproof.
>
> However, my major concern is what is the implication? Where is the
 pitfall? How
> does it affect the performance? Is there something that I have oversee?
>
> Please post your comments on this news group or email me. Any
 ideas/comments
> would be greatly appreciated.
>
Received on Wed Feb 18 1998 - 00:00:00 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US