Re: [PATCH] Fix for Hashtable contains blowing up the stack

[Mark Wielaard]

Hi,

On Fri, 2003-11-28 at 17:39, Dalibor Topic wrote:
> Thanks for pointing that mistake out. I've attached an updated version 
> of the patch.
> 
> 2003-11-28  Dalibor Topic <address@hidden>
> 
>          Reported by: Jim Pick <address@hidden>
> 
>          * libraries/javalib/java/util/Hashtable.java
>          (internalcontainsValue): New method.
>          (contains) Delegate to internalContainsValue.
> 
>          Reported by: Mark Wielaard  <address@hidden>
> 
>          * libraries/javalib/java/util/Hashtable.java
>          (contains): Improved comment.
> 
>          Reported by: Jeroen Frijters  <address@hidden>
> 
>          * libraries/javalib/java/util/Hashtable.java
>          (containsValue): Delegate to contains(Object) to make sure older
>          code overwriting it continues to work.
> 
> I'm not sure how to deal with mutliple people's bug reports being fixed 
> in this patch in the ChangeLog entry, so I'd appreciate a hint from the 
> ChangeLog police.

Looks fine to me. (I would loose the extra empty line between Reported
by: and the actual change, but feel free to ignore me).

Also you could now pull the complete internalContainsValue() into the
containsValue() method, but that would make the patch bigger, so don't
bother (compiler/VM will inline the private method anyway).

If this solves both Mauve Hashtable tests, please check it in.

> Yes, that makes the most sense to me, too. Stuart Ballard started 
> working on a 'override compatibility' test suite for collection classes. 
>   See [1] for details.
>
> [1] http://www.kaffe.org/pipermail/kaffe/2003-May/042181.html

Interesting. Might be a candidate for our extended Mauve tests.

Cheers,

Mark

signature.asc
Description: This is a digitally signed message part