I am not sure I quite understand you. If you don't need a map, why do you try to fetch it? I don't remember details right now, but wouldn't expect anything better than NULL or some garbage when reading uninitialized internal field.
If you only need to call a callback, maybe you could omit all the code which fetched the map? yours, anton. On Wed, May 26, 2010 at 4:33 PM, mgen <[email protected]> wrote: > I actually am not trying to wrap a native pointer, but want to fire a > callback (JS one not C++) on any access to any property of an object. > Although does this mean it would be easier to create a map myself to > store it on the object rather than trying to access the map created > for properties by default? > > On May 26, 6:08 am, Anton Muhin <[email protected]> wrote: >> I don't see where you set the internal field. And to store native >> pointers (esp. aligned), {Get,Set}Pointer{From,To}InternalField is a >> simpler and preferred API. >> >> yours, >> anton. >> >> >> >> On Wed, May 26, 2010 at 8:38 AM, mgen <[email protected]> wrote: >> > When using a NamedPropertyHandler as an interceptor, I was wondering >> > how to access the normal map for an object's properties. Copy and >> > pasting from the process.cc example's use of UnwrapMap with some >> > massaging only led to seg faults so far. If someone could explain the >> > proper way to access these properties (something akin to ForceSet but >> > with a Get?). >> >> > Paste of code is athttp://gist.github.com/414012 >> >> > -- >> > v8-users mailing list >> > [email protected] >> >http://groups.google.com/group/v8-users > > -- > v8-users mailing list > [email protected] > http://groups.google.com/group/v8-users > -- v8-users mailing list [email protected] http://groups.google.com/group/v8-users
