Re: [bug-gawk] posix flag bug?

[Ed Morton]

thanks for the quick action. I see do_lint and do_traditional being tested in 
the fix - does that cover the presence of the "--posix" argument too or is there 
a missing do_posix variable?

       Ed.

On 9/26/2015 1:29 PM, Aharon Robbins wrote:
> Greetings. Re this:
>
> > To: address@hidden
> > From: Ed Morton <address@hidden>
> > Date: Sat, 26 Sep 2015 08:45:10 -0500
> > Subject: [bug-gawk] posix flag bug?
> >
> > Since multi-dimensional arrays are gawk extensions I expected this:
> >
> >      $ echo 'x' | awk --posix '{arr[1][2]=$0; for (i in arr) for (j in arr[i])
> >      print arr[i][j]}'
> >      x
> >
> > to fail, but clearly it didn't. It didn't fail with the `--traditional` flag
> > either. Shouldn't it?
> >
> >       Ed.
> >
> > P.S. I'm using:
> >
> >      $ awk --version
> >      GNU Awk 4.1.3, API: 1.1 (GNU MPFR 3.1.3, GNU MP 6.0.0)
> >
> > in bash on cygwin.
> Here's the fix. It'll get pushed eventually.
>
> Arnold
> -----------------------------------------------
> diff --git a/awkgram.y b/awkgram.y
> index f311f29..db61aa6 100644
> --- a/awkgram.y
> +++ b/awkgram.y
> @@ -3224,6 +3227,10 @@ retry:
>                 c = nextc(true);
>                 pushback();
>                 if (c == '[') {
> +                       if (do_traditional)
> +                               fatal(_("multidimensional arrays are a gawk 
> extension"));
> +                       if (do_lint)
> +                               lintwarn(_("multidimensional arrays are a gawk 
> extension"));
>                         yylval = GET_INSTRUCTION(Op_sub_array);
>                         lasttok = ']';
>                 } else {