On Thursday, March 7, 2002, at 10:09 PM, Alexander Malmberg wrote:

> Hi,
>
> NSMapTable doesn't properly handle the case where you call NSMapInsert()
> on a key that already has a value. Specifically:

<snip>

> I've attached a patch that fixes this. (It could be optimized to modify
> the value in place with additional complexity.)

Actually, it has to do that to conform to the spec. - so I wrote that
change instead.  I also fixed a similar bug in the NSHashTable code.

I guess on my recent rewrite of these functions to use the GSI macros,
I forgot that the macros did not have these behaviors but were
simpler primitives.


_______________________________________________
Bug-gnustep mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-gnustep

Reply via email to