On 18/11/2010, at 9:25 PM, john skaller wrote:
> well this is weird, I have done some transforming of code that I thought
> was "transparent" but now:
The serious one is this:
Client Error binding expression new ((darray_ctl ((_ctor_varray[T] 20lu),
dflt_resize)))
CLIENT ERROR
[lookup_name_with_sig] Can't find darray_ctl of varray[<T10064>] * (ulong^2 ->
ulong)
In build/release-optimized/lib/std/darray.flx: line 34, cols 33 to 42
33: // make an empty darray, give it 20 slots for no particular reason
34: gen _ctor_darray[T] () => new darray_ctl( _ctor_varray[T] 20ul ,
dflt_resize);
**********
35:
This doesn't involve any cfuns .. but it does involve wrappers: we have a
struct constructor thing here. Maybe it is because there are
two ticaches in use .. a global one syms.ticache and a separate one
in the lookup state object..
This uses the global one, in "syms":
src/compiler/flx_bind/flx_bbind.ml: if not (Hashtbl.mem state.syms.ticache
symbol_index) then begin
src/compiler/flx_bind/flx_bbind.ml: Hashtbl.add state.syms.ticache
symbol_index t
but this:
src/compiler/flx_bind/flx_lookup.ml: try Hashtbl.find state.ticache index with
Not_found ->
uses the one in the lookup state. That thing was just supposed to be a cache,
I guess I'm hacking it somewhere I didn't expect :)
--
john skaller
[email protected]
------------------------------------------------------------------------------
Beautiful is writing same markup. Internet Explorer 9 supports
standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 & L3.
Spend less time writing and rewriting code and more time creating great
experiences on the web. Be a part of the beta today
http://p.sf.net/sfu/msIE9-sfdev2dev
_______________________________________________
Felix-language mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/felix-language
