[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bug-apl] Locked Functions
From: |
David B. Lamkins |
Subject: |
Re: [Bug-apl] Locked Functions |
Date: |
Tue, 8 Sep 2015 21:05:10 -0700 |
User-agent: |
Mutt/1.5.23 (2014-03-12) |
This was covered quite a while back, Mike; you probably missed it.
Not hiding the content of a locked function is intentional as it goes against
the spirit of open software.
Take a look at quad-FX: the optional left argument can be used to prevent
suspension. I use this in the iso-apl-cf package.
On Tue, Sep 08, 2015 at 06:33:08PM -0700, Mike Duvos wrote:
> The workings of locked functions are supposed to be completely
> concealed from the user, and if an error appears in one, it is
> considered to have occurred in the line invoking the locked function,
> as if it were a primitive.
> We can see this in APL2.
> ∇FOO
> [1] BAR⍫
> ∇BAR
> [1] →1∇
>
> FOO
> INTERRUPT
> FOO
> ^
>
> In GNU APL, however, we can get suspensions in locked functions and the
> things they call on our state indicator.
> FOO
> ^CATTENTION
> BAR[1] →1
> ^
> )si
> BAR[1]
> FOO[1]
> ⋆
> I was wondering what the standard says about this, as it is definitely
> contrary to the APL2 documentation.