Re: [PATCH] make light userdata a little bit heavier

[Ross Bencina <rossb-lists@...>]

On 22/04/2014 4:44 PM, steve donovan wrote:
> Man, it took me a while in the beginning to understand the difference
> between light and full, but it's easy enough: light is just a pointer,
> and all light userdata share the same metatable (like strings, etc);
> full metadata can have individual metatables.  Now imagine writing the
> manual to explain the meaning of 'shallow' ;)

[snip]

> But conceptually, three is a much larger number than two...

Conceptually, there are still two types of userdata: light and full. The 
proposal extends the concept of light userdata by allowing multiple 
distinct classes (tagged subtypes) of light userdata.

C language embedders may introduce additional classes of light userdata 
(class-0, class-1, class-2, etc.). Each class has a distinct metatable, 
but otherwise behaves as light userdata does currently. Existing code 
gets class-0 light userdata, and so works exactly as it does now.

But still, conceptually, two is a much larger number than one.

Ross.