Re: [svn:parrot] r45779 - branches/immutable_strings_part2/runtime/parrot/library/Test

[chromatic]

On Sunday 18 April 2010 at 04:19, bacek wrote:

> Log:
> Update Test::More to not use removed ops.
>
> Modified:
> branches/immutable_strings_part2/runtime/parrot/library/Test/More.pir
>
> --- branches/immutable_strings_part2/runtime/parrot/library/Test/More.pir
Sun Apr 18 11:18:54 2010 (r45778)
> +++ branches/immutable_strings_part2/runtime/parrot/library/Test/More.pir
Sun Apr 18 11:19:06 2010 (r45779)
> @@ -576,7 +576,7 @@
> if does_flag goto compare_hash
>
> diagnosis = typeof left
> - diagnosis .= ' is not a nested data structure'
> + diagnosis = diagnosis . ' is not a nested data structure'
> result = 0
> goto report_result

That's going to lead to annoying code. We should instead consider patching
IMCC to rewrite the infix op to the concat_s_s_s form.

-- c
_______________________________________________
http://lists.parrot.org/mailman/listinfo/parrot-dev

--
You received this message because you are subscribed to the Google Groups "parrot.dev" group.
To post to this group, send email to parro...@googlegroups.com.
To unsubscribe from this group, send email to parrot-dev+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/parrot-dev?hl=en.

[Vasily Chekalkin]

Sorry. Hit wrong button.


---------- Forwarded message ----------
From: Vasily Chekalkin <ba...@bacek.com>
Date: Mon, Apr 19, 2010 at 7:25 PM
Subject: Re: [svn:parrot] r45779 -
branches/immutable_strings_part2/runtime/parrot/library/Test
To: chromatic <chro...@wgz.org>


Hello.

On Mon, Apr 19, 2010 at 5:10 PM, chromatic <chro...@wgz.org> wrote:
>> -    diagnosis .= ' is not a nested data structure'
>> +    diagnosis  = diagnosis . ' is not a nested data structure'
>>      result     = 0
>>      goto report_result
>
> That's going to lead to annoying code.  We should instead consider patching
> IMCC to rewrite the infix op to the concat_s_s_s form.

I tried it. But there is no such functionality in IMCC. It just emit
"concat $S0, $S1" when it see "$S0 .= $S1".

I actually considering to put clear statement somewhere in docs stated
"'in-place' concat op is syntax sugar for 3-args ops. E.g. C<concat
foo, bar> is actually C<foo = concat foo, bar>" and keep 2-args form.

--
Bacek.