[Date Prev][Date Next][Thread Prev][Thread Next]
[Date Index]
[Thread Index]
- Subject: Lua research and static type checking
- From: Kornel Kisielewicz <epyon@...>
- Date: Tue, 5 Feb 2013 22:10:04 +0100
Hello,
My name is Kornel Kisielewicz, and I'm a CS PhD student at the Wroclaw
University in Poland. Lua has recently become my main research
interest. I am quite surprised at the lack of research materials about
Lua - while dynamic languages are usually unpopular for research,
Lua's extremely compact size and tight design makes it ideal for a
number of research topics.
Hence, I have a couple questions related to the research side -- the
main one being - is there a formal specification (or an attempt of) of
the virtual machine and the language? If not, then would one be
considered valuable for the community?
Secondly, my main spark of interest was the idea of a limited form of
typechecking, or rather an annotation mechanism for static type
checking during compilation. The idea would be to build a extended
compiler, that would be fully compatible with the existing one in
normal operation (ignoring any type information), but could also run
in static typechecking mode (which would generate compiler errors and
warnings where types would be deduced to conflict) and a dynamic
typechecking mode, where the type annotations would additionally
generate run-time type checking code (a application debug version). As
I understand such a tool would be invaluable in real-life Lua
applications.
I saw a few projects that had a similar aim, but most of them were
usually a set of hacks. Were there any more serious attempts at that?
I did go through all things linked at
http://lua-users.org/wiki/LuaTypeChecking including the MetaLua
solution (which seems to be abandoned and compile time only).
In particular - did anyone attempt to introduce type annotation
on compiler level?
Finally, I'm looking for as much sources on Lua in CS research as
possible -- unfortunately except the papers linked from the Lua site -
most papers are behind a paywall (which our university unfortunately
doesn't cover), and for any potential active research topics in it's
field. Any help on list or off list would be greatly appreciated!
--
regards,
Kornel Kisielewicz