So you would recommend the OP modify their entire system of three thousand 
programs
Instead of accept a 98% solution?

Very nice.



-----Original Message-----
From: Robert Porter <[email protected]>
To: u2-users <[email protected]>
Sent: Fri, Mar 23, 2012 11:44 am
Subject: Re: [U2] Record lock


I whole-heartedly disagree... 
) I have numerous processes that lock/update/release the same record repeatedly 
unning 24x7 - on the sale process id. 

) I don't want code that works 98% of the time - that's unacceptable.  I don't 
ant calls in the middle of the night because it appears processes are locked up 
hen in fact they are running perfectly fine because some program fails 2% of 
he time.  

o this end, ALL READU's are wrapped inside of a subroutine that tracks 
locking, time held, etc.  
xceptions are a part of programming... Not dealing with the exceptions is bad 
rogramming practice. Would you not validate input because the user is expected 
o enter the right result 98% of the time, and the other 2% is ok?  

ears ago, we changed from a 5 digit numeric specimen id to 5 alpha/numeric... I 
at in the pre-live meeting, and brought up the fact that words could be 
pelled. Words we might not want patients to see... With over 60 million 
ossible permutations, the likelyhood was well under 2%. Would that have been an 
cceptable error rate to you?  It's not to us. We excluded the vowels so words 
ouldn't be spelled... That's programming for the exception.
 

_______________________________________________
2-Users mailing list
[email protected]
ttp://listserver.u2ug.org/mailman/listinfo/u2-users

_______________________________________________
U2-Users mailing list
[email protected]
http://listserver.u2ug.org/mailman/listinfo/u2-users

Reply via email to