RE: need for LuaLint? (Re: Experience with Large Applications in Lua?)

["Andre Carregal" <carregal@...>]

Hi,

I don't know if it would fit your requirements, but the Kepler team is
working on a suite of Lua dev tools. Please keep in mind that we will not
even try to solve the "static typing" problem per se, but simply offer a
collection of tools to ease the development of bigger Lua modules.

The tools in the suite are LuaDoc, LuaProfiler, LuaUnit, a Lua lint and a
wiki to generate the web reports using Xavante and CGILua. Most of the tools
are already running, but we still need to work on LuaDoc, the lint and the
wiki.

Basically LuaDoc makes it easier to write about your functions inside Lua
comments, just like JavaDoc does. LuaDoc can extract the descriptor data
from Lua sources, allowing the integration of a Lua lint that could check
some of the restrictions imposed. We will be working in tandem with the
LuaEclipse team to make both tools work nicely together, but you can also
use LuaDoc as a command line tool.

It would be nice to know what people would expect from such a system, and
how we could better structure the modules so one could use just part of them
if needed.

Thanks

Andre Carregal

> -----Original Message-----
> From: lua-bounces@bazar2.conectiva.com.br
> [mailto:lua-bounces@bazar2.conectiva.com.br]On Behalf Of Asko Kauppi
> Sent: Wednesday, July 07, 2004 9:23 AM
> To: Lua list
> Subject: need for LuaLint? (Re: Experience with Large Applications in
> Lua?)
>
>
>
> How about a tool that would collect the suggested function
> 'descriptors' and apply them to all invocations of the functions?
>
> I'm certainly _not_ going to do such a tool, but .. if
> someone will.. :)
>
> That would find often experienced bugs s.a. wrong params to
> 'string.find' (I tend to forget the string itself! :)  and such.
> Basically, a Lint
> (http://computing-dictionary.thefreedictionary.com/lint) for Lua.
>
> -ak
>
>
> 7.7.2004 kello 13:53, Jamie Webb kirjoitti:
>
>   On Wed, Jul 07, 2004 at 10:50:21AM +0200, Steve Donovan wrote:
> >> In a nutshell: we need strong testing, not strong typing.
> >
> > The problem I have with this argument is that a static type checker:
> > a) is more reliable
> > b) will tell you exactly where the problem is, rather than
> throwing up
> > an error at some point later when you try and misuse a value.
> >
>
>