Re: linking tables to C++ objects?

[Graham Wakefield <lists@...>]

Ah, now I read Rici's note.

So in pseudocode again:

// creation (synchronized) of table, userdata & c++ instance:

ptr = [C++ instance]
udata = userdata[ptr]	// boxed pointer
table = lua_newtable
setfenv(udata, table)
REGISTRY[weakvaluedtable][ptr] = udata
REGISTRY[weakvaluedtable][table] = ptr

// from C++ ptr to table:

getfenv(REGISTRY[weakvaluedtable][ptr])

// from table to C++ ptr:

REGISTRY[weakvaluedtable][table]

// garbage collection:

udata.__gc { // remove ptr & table entries from REGISTRY 
[weakvaluedtable]) }




Surely there is a more efficient way to do this!?


On Feb 26, 2007, at 12:44 PM, Graham Wakefield wrote:

> This works for getting from a userdata (on the stack) to a table.
>
> This doesn't help in getting from a C++ pointer to the table, since  
> I can't get a userdata on the stack from a C++ pointer; or can I?
>
> On Feb 26, 2007, at 6:29 AM, David Burgess wrote:
>
> > Answering one of your questions:
> >
> > Use LUA_ENVIRONINDEX rather than the LUA_REGISTRYINDEX,
> > see liolib.c for an example of its usage.
> >
> > And I think this recent quote from Rici Lake applies:
> >
> > > The simplest way is to keep a weak-valued table whose keys are
> > > lightuserdata (the pointer to the c structure) and whose values  
> > > are the
> > > equivalent userdata object (containing the boxed pointer).
> >
> > db