Thanks for the feedback.

For those that are interested in the "base/1/play" and "base/1/support"
layout, where (GNU/Linux) binaries are laid out for "./play-lua-tec"'s
benefit, here's a snippet of the Lua code that names the binaries that
are copied when "gather" is executed:

---- Start Lua script excerpts ----

local im = {
        Name = "im",
-- [...]
        RunFiles = {
                "libimlua51.so",
                "../libim.so",
        }
}

local cd = {
        Name = "cd",
-- [...]
        RunFiles = {
                "libcdlua51.so",
                "libcdluapdf51.so",
                "../libcd.so",
                "../libcdpdf.so",
                "../libpdflib.so",
        },
-- [...]
}

local iup = {
        Name = "iup",
-- [...]
        RunFiles = {
                "libiuplua51.so",
                "../libiup.so",
        }
}

---- End Lua script excerpts ----

The "RunFiles" array names the build products to be gathered in the
"base/1/support/" directory.  Another script, "platform", is a helper
script to find where the binaries are on the local machine; it is
run using io.popen:

        $ cat base/build/platform
        #!/bin/bash

        ### Platform -- Use tec_uname to report the OS/Kernel/Arch used by the
        # build scripts.  The caller (a Lua script) uses this path to gather
        # all needed files into a private folder, without requiring system
        # privileges (maybe; some dynamic linking avoids
        # non-system-sanctioned paths as these can be a security risk).

        . tec_uname

        ComputeTecUname
        ComputeLuaVersion

        echo 
"lib/$TEC_SYSNAME$TEC_SYSVERSION${TEC_SYSMINOR}_$TEC_SYSARCH/Lua$LUA_SFX"

NO WARRANTY is given as to the correctness/completeness of the above
excerpt/script pieces.

This lack of warranty is especially true, given that I don't have a
comprehensive test rig to check the correctness/completeness of my
work.  A regression test suite, one for each of IM, CD and IUP, would
be highly desirable, but is non-trivial:  For example, in IUP, the
test would need to be able to automate inputs from various sources,
including keyboard and mouse, but perhaps also including events such
as refresh/redraw delivered by the windowing system, and have a way
of serialising the results in a way that captures all elements that
matter, so the results can be checked in an automated fashion.

cheers,

sur-behoffski
programmer, Grouse Software


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Iup-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/iup-users

Reply via email to