Re: [bug #38393] NSBezierPath does not have a complete implementation of keyed archive coding

[Christopher Armstrong]

Hi Fred

That's definitely true, I was actually just completing the code in the 
initWithCoder: method. From what I could see, storing any points for the close 
path primitive was pointless anyway.

I'm in no rush to get the code in, it was just to fix unit tests in EtoileUI.

Chris

Sent from my iPad

On 24/02/2013, at 23:24, Fred Kiefer <INVALID.NOREPLY@gnu.org> wrote:

> Follow-up Comment #1, bug #38393 (project gnustep):
> 
> Thank you for this great patch. I would really love to add that to the
> upcoming gui release, even though we are in a code freeze.
> 
> There is one interesting aspect to your patch. For "close" elements you are
> using two points returned by elementAtIndex:associatedPoints: Now that method
> in GNUstep doesn't return any points for "close" elements and according to the
> Apple documentation it should return one.
> 
> This means that we have an important issue here in the current GNUstep code
> and that your code has an issue as well. Your issue could be easily fixed by
> using the same point twice. But where should that point come from? I think the
> best way to fix it now is to moce some code from currentPoint into the
> elementAtIndex:associatedPoints: method and later one rethink the whole code
> once more.
> 
>    _______________________________________________________
> 
> Reply to this item at:
> 
>  <http://savannah.gnu.org/bugs/?38393>
> 
> _______________________________________________
>  Message sent via/by Savannah
>  http://savannah.gnu.org/
>