[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: indirect function call issues (2)
From: |
arnold |
Subject: |
Re: indirect function call issues (2) |
Date: |
Thu, 16 Jan 2025 00:47:34 -0700 |
User-agent: |
Heirloom mailx 12.5 7/5/10 |
Denis Shirokov <cosmogen@gmail.com> wrote:
> Another question on a completely different topic — look:
>
> Example:
>
> BEGIN {
> length(A[1])
> print "A[1] : " typeof(A[1])
>
> length(B)
> print "B : " typeof(B)
> }
>
> Outputs:
>
>
> A[1] : untyped B : unassigned
>
> In the first case, length() does not make A[1] a string or an array,
> leaving A[1] in the *untyped* state, whereas in the second case, for some
> reason, it makes the neutral B *unassigned*. Is this correct in light of
> the new (but long-awaited!) gawk feature allowing uninitialized subscripts
> in arrays without automatically making them *unassigned*?
>
> Is such behavior correct?
This is correct. It's an issue of historical compatibility, where
length() has always converted an untyped variable into a scalar.
Arnold
- indirect function call issues (2), Denis Shirokov, 2025/01/14
- Re: indirect function call issues (2), Eli Zaretskii, 2025/01/15
- Re: indirect function call issues (2), arnold, 2025/01/15
- Re: indirect function call issues (2), Eli Zaretskii, 2025/01/15
- Re: indirect function call issues (2), Denis Shirokov, 2025/01/15
- Re: indirect function call issues (2), Eli Zaretskii, 2025/01/15
- Re: indirect function call issues (2), Denis Shirokov, 2025/01/15
- Re: indirect function call issues (2), Eli Zaretskii, 2025/01/15
- Re: indirect function call issues (2),
arnold <=
- Re: indirect function call issues (2), mekanofox, 2025/01/15
- Re: indirect function call issues (2), arnold, 2025/01/16
- Re: indirect function call issues (2), mekanofox, 2025/01/16
- Re: indirect function call issues (2), mekanofox, 2025/01/16
Re: indirect function call issues (2), Denis Shirokov, 2025/01/15