[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bug-apl] Lambda valence error

From: Juergen Sauermann
Subject: Re: [Bug-apl] Lambda valence error
Date: Sun, 05 Jun 2016 16:54:31 +0200
User-agent: Mozilla/5.0 (X11; Linux i686; rv:31.0) Gecko/20100101 Thunderbird/31.4.0


it kind of worked earlier, but not really.

I changed the behavior in SVN 622 in order to fix a bug reported earlier.
If I remember correctly then:

- )LOAD of a workspace with a monadic function called dyadically on the )SI stack
 would crash due to a null pointer in a value token, and
- IBM APL2 would also raise a valence error when a monadic function is called dyadically

/// Jürgen

On 06/04/2016 08:58 AM, Daniel H. Leidisch wrote:

On r734:

      1 {⍵} 2
      1 λ1 2
      ^ ^

Unless I'm mistaken, this used to work. I assumed it to be a regression,
but reading the info section about lambdas, it actually seems to be
intended behavior. ("If only ⍵ is present then it is monadic, and if
neither ⍺ nor ⍵ is present then the lambda is niladic.")

I can get around it using something like {⍺ ⊢ ⍵}, but it feels clumsy
and isn't needed in Dyalog and NGN. Is it intended behavior, and if yes,
what was the rationale behind it?



reply via email to

[Prev in Thread] Current Thread [Next in Thread]