Alien is a binding to the libffi library; http://sourceware.org/libffi/
So you need libffi to be able to build the binding.
Van: jhernanca...@gmail.com [mailto:jhernanca...@gmail.com] Namens Hernan Cano
Verzonden: Saturday, 19 March, 2016 18:41
Aan: luarocks-developers@lists.sourceforge.net
Onderwerp: Re: [Luarocks-developers] "Quick Start" for Windows?
Hi, friends.
I have some problem related with the comments in this thread.
Since LuaRocks I want to install Alien. I write:
luarocks install alien
But I receive a message that informs that it lacks something as ffi.lib,
ffi.dll, libffi.dll .....
If I write one of the following
luarocks install ffi
luarocks install luaffi
luarocks install luaffifb
luarocks install libffi
it says that it cannot be found.
If Alien is included in LuaRocks (and LuaFFI too), how can I install Alien
using LuaRocks?
Thanks, friends.
2016-03-18 18:59 GMT-05:00 Stefano:
On 18 Mar 2016 23:08, "Thijs Schreijer" wrote:
>
> > >
> > > I don't think that the LuaRocks installation is the issue here of
> > > holding back Windows users to use it. The problem lies in the harmony
> > > required between the Lua version (actually the underlying runtime lib
> > > used), compiler toolchain and LuaRocks
> > >
> > > Note: chocolatey only does binary distribution
> > >
> > > And within the Lua ecosystem we're leaning towards the gcc toolchain,
> > probably mainly because most Lua code is unix based and gcc/MinGW is
> > closest to that.
> > >
> > > It's a different angle, but this is mainly where I got the inspiration
> > for my luawinmake and luawinmulti projects. I might add a similar
> > instruction to the LuaWinMulti project to make it even easier. Though the
> > culprit for the average Windows user will always be the need for a
> > compiler toolchain, which is very non-Windows as Windows is basically a
> > binary distribution platform.
> >
> > I see. I wonder if we could ever do something about that, but I believe we
> > just don't have the resources to provide binary packages. I don't know if
> > there is any way we could automate this significantly and work around the
> > issue of external dependencies in a good way.
> >
> > -- Hisham
>
> It's the external dependencies what's causing the trouble. If it's just a
> plain Lua library in C, that can be automated by using LuaRocks to generate
> and upload the results I think. But once it gets to dependencies, binding to
> external libraries, that's where it gets really hairy.
>
> It would be interesting to see whether we could automate building LuaSocket
> and LuaFileSystem for example for the different flavours (matrix of 6 or 7
> runtimes, against 32 and 64 bit versions).
> We might be able to utilize build artifacts of travis ci for example.
Or just use http://ulua.io that does right that? ;)
Btw I had plans to create some GitHub issues (for my distro) on how to move
forward on this exact same issue of external dependencies.
Basically supporting automated builds for git repos which have tagged revisions
via simple scripts seems achievable.
I'd work on this only if someone is interested in joining efforts.
Stefano
>
> LuaRocks would also need to make that distinction in the binary packages by
> the way...
>
> Thijs
>
>
> ------------------------------------------------------------------------------
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785231&iu=/4140
> _______________________________________________
> Luarocks-developers mailing list
> https://lists.sourceforge.net/lists/listinfo/luarocks-developers
------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785231&iu=/4140
_______________________________________________
------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785231&iu=/4140
_______________________________________________
Luarocks-developers mailing list
Luarocks-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/luarocks-developers
