Re: [Chicken-hackers] [PATCH] Fix for #905

[Mario Domenech Goulart]

Hi,

On Tue, 28 Aug 2012 21:21:39 +0200 (CEST) Felix <address@hidden> wrote:

> From: Peter Bex <address@hidden>
> Subject: [Chicken-hackers] [PATCH] Fix for #905
> Date: Tue, 28 Aug 2012 21:04:37 +0200
> 
>> As Ivan found out the hard way, I forgot to modify hash-table-copy
>> when introducing the new randomization factor.  Funny enough, the
>> tests did check whether the copied hash table has the same contents,
>> but it doesn't reference any values in the hash table, which means
>> the tests for copying passed with flying colours even though the
>> hash table would effectively be unusable.
>> 
>> I was a little tired yesterday, but meanwhile our tireless friend Mario
>> found the bug and proposed a patch.  Here it is as a git changeset, with
>> a small tweak to the regression tests to catch this situation.
>
> Signed off and pushed. IIRC, this closes the last open tickets.  If
> Ivan can confirm that nemo works with this patch, and salmonella
> doesn't report any new errors for master, I'll update the manual
> (#908) and we can tag a new release candidate.

Alright.  Ivan confirms the fix works for his case.

Salmonella doesn't report any breakage:
http://tests.call-cc.org/master/linux/x86/2012/08/29/yesterday-diff/

Actually the tests for sets now pass.  They had been failing since
2012-01-10
(http://tests.call-cc.org/master/linux/x86/2012/01/10/salmonella-report/tests/sets.html),
but went unnoticed.  The code for sets uses hash-tables.

So it seems that we are ready for rc3.


Best wishes.
Mario
-- 
http://parenteses.org/mario