Re: Strange userdata metatable behavior in C API

[Peter Odding <xolox@...>]

dormando wrote:
> dormando wrote:
> > Hey,
> >
> > I've been fighting with a nasty bug in the C API for a 
> > soon-to-be-GPL'ed project o' mine... for several hours, and I'm wholly 
> > stuck. The problem is pretty easy to illustrate:
>
> As usual, a few hours later I'll make a dick of myself.
>
> The issue is that I'm using a lightuserdata instead of a full userdata.
>
> Setting up a **pointer and shoveling that inside the full userdata has 
> completely ameliorated the weirdness.
>
> I do have one real question however; I have the 2nd ed PiL book, the 5.1 
> ref man, and the online manual. All of which devote a single sentence 
> saying that lightuserdata does not have a metatable. However, this 
> appears to not be the case :P It has a single metatable for all types 
> (like numbers, strings, etc). So when I set another metatable for the 
> second lightuserdata object, it overwrote it globally.
>
> Am I reading the docs wrong and this is intended behavior?
>
> Thanks,
> -Dormando

You made me curious so I checked the manual. Here's the entry for 
lua_pushlightuserdata (*emphasis* mine):

--------------------

void lua_pushlightuserdata (lua_State *L, void *p);

Pushes a light userdata onto the stack.

Userdata represent C values in Lua. A light userdata represents a pointer. It is 
a value (like a number): you do not create it, *it has no individual metatable*, 
and it is not collected (as it was never created). A light userdata is equal to 
"any" light userdata with the same C address.

--------------------

So this is in fact documented, though I must admit I wasn't aware of it either.

 - Peter Odding