Martin Stjernholm,  Roxen IS @ Pike  developers forum wrote:
>> Well, a NULL value basically has any type (in SQL).  I'm not sure what
>> would be the most natural way to extend this into Pike, but I figured
>> that the closest we could get, was something like the code above.
>> Since the NULL value has "any" type, it's not lying as such.

>I think one should view this more from the Pike angle. If e.g.
>arrayp(x) is true, a pike programmer typically expects that all
>operators and calls for arrays would work for x. It's not possible to
>achieve that when x is an object (most builtin functions expecting
>arrays would still balk, for example). So even if you added a bunch of
>backtick functions it'd still not work 100%.

Ok, I see your point.  I'll try and pickup the slack as soon as 7.9 sees
the light and help to get this going.
-- 
Sincerely,
           Stephen R. van den Berg.

"Good moaning!"
  • NULL result... Stephen R. van den Berg
    • Re: NU... Martin Bähr
      • Re... Stephen R. van den Berg
    • NULL r... Martin Nilsson (Opera Mini - AFK!) @ Pike (-) developers forum
      • Re... Stephen R. van den Berg
    • NULL r... Martin Stjernholm, Roxen IS @ Pike developers forum
      • Re... Stephen R. van den Berg
        • ... Martin Stjernholm, Roxen IS @ Pike developers forum
          • ... Stephen R. van den Berg
            • ... Martin Nilsson (Opera Mini - AFK!) @ Pike (-) developers forum
              • ... Stephen R. van den Berg

Reply via email to