Subject | Re: [ib-support] Re: string concatination, triggers |
---|---|
Author | Ann W. Harrison |
Post date | 2001-04-10T13:45:14Z |
Diane Brown wrote
source of the confusion having at various times said something like:
"Anything and null is null, and null evaluates to false." Sounds just
like me. What I meant was that
if (<null value> = <value>) is false
and if (<null value> <> <value>) is false
and if (NOT (<null value> == <value>) is false
The nullness just comes through, like ink in a shirt pocket when the
pen cap is on the desk.
In a more reasonable world, the syntax might be:
if <condition>
true <action>
false <action>
null <action>
but a more reasonable world would have a solvent/neutralizer that
would remove the ink, and that, in my opinion, would be of more value
to the world than three-valued ifs.
Regards,
Ann
>Either Value + NULL evaluates to NULL/unknown, or Value + NULLDiane is, of course, correct. I must acknowledge that I am the likely
>evaluates to FALSE ... even in the SQL standard, you can't have it
>both ways.
>
>I can see that you might say that Value + NULL is equivalent to FALSE
>in an IF .. THEN ... ELSE ... construct, but even there I'd quibble.
>That thinking seems to cause people to get messed up when they write
>IF *NOT* ... THEN ... constructs, at least in my experience.
source of the confusion having at various times said something like:
"Anything and null is null, and null evaluates to false." Sounds just
like me. What I meant was that
if (<null value> = <value>) is false
and if (<null value> <> <value>) is false
and if (NOT (<null value> == <value>) is false
The nullness just comes through, like ink in a shirt pocket when the
pen cap is on the desk.
In a more reasonable world, the syntax might be:
if <condition>
true <action>
false <action>
null <action>
but a more reasonable world would have a solvent/neutralizer that
would remove the ink, and that, in my opinion, would be of more value
to the world than three-valued ifs.
Regards,
Ann