[Date Prev][Date Next][Thread Prev][Thread Next]
- Subject: Re: Makefile vs LUA_PATH inconsistency
- From: Philipp Janda <siffiejoe@...>
- Date: Tue, 20 May 2014 11:07:28 +0200
Am 20.05.2014 09:59 schröbte Thijs Schreijer:
From: email@example.com [mailto:firstname.lastname@example.org] On
Behalf Of Ulrich Schmidt
Sent: dinsdag 20 mei 2014 8:24
Subject: Re: Makefile vs LUA_PATH inconsistency
Am 20.05.2014 01:56, schrieb Luiz Henrique de Figueiredo:
But it's the inconsistency between luaconf.h and the makefile (default
behaviors) that I intended to address.
What exactly do you propose to change in the top-level Makefile?
Changing INSTALL_LMOD and INSTALL_CMOD dos not really work because the
paths in luaconf.h include the location of lua.exe, which is not known
at build time and changes dynamically at each invocation.
We need to distinguish between windows- and linux installation 1st.
How is MinGW classified? Is it windows or unix or both?
MinGW is just a C compiler, so it is neither Unix nor Windows. The only
difference to other C compilers on Windows is that it is similar enough
to Unix C compilers that it can be handled in the same Makefile. You
probably meant MSYS or even Cygwin.
So, the executables produced via MinGW should follow the normal Windows
Can't we use the currently created structure as the "widely accepted
subfolder tree"? There is no technical reason that it won't work.
It's just not very windowish, and it limits platform differences.
The way I see it is: There is no default directory structure for the
standalone interpreter on Windows. The mingw target is included in the
Unix Makefile as a convenience, and that Unix Makefile uses the usual
Unix installation locations. No reason to read something into it for