gnash-commit
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnash-commit] gnash ChangeLog backend/render_handler.h libgeo...


From: strk
Subject: Re: [Gnash-commit] gnash ChangeLog backend/render_handler.h libgeo...
Date: Mon, 5 Mar 2007 13:32:06 +0100

On Fri, Mar 02, 2007 at 03:38:45PM +0100, Udo Giacomozzi wrote:
> Hello strk,
> 
> Thursday, March 1, 2007, 2:24:16 PM, you wrote:
> s> This was the reason to provide a 'visitor' model for SnappingRanges2d, 
> hiding
> s> the underlying container type. This use uses a random access iterator when
> s> it doesn't even need to. For a source-level compatibility with future 
> container
> s> switch we could expose the iterator interface, or provide the visitor 
> pattern.
> 
> Ok. But I ask you to do this as you are probably more experienced with
> it.

I've committed the visit() method, but it is still untested.
Do you want to try it out ? If you do examples of use are
in sprite_instance.cpp. One example prematurely ending the loop is
MouseEntityFinder class.

        bool operator() (character* ch)

Should be changes with

        bool operator() (Range2d<xx>& ch)

You can have multiple overloaded operator() to accept different
kind of Range2d<> classes, or even make the visitor class itself
a template.

--strk;




reply via email to

[Prev in Thread] Current Thread [Next in Thread]