How should the win16 test suite be arranged?

[Dan Kegel]

So we're collecting the existing win16 tests,
updating and fixing them, and will post them
to wine-patches sometime soon.
At first, the test harness is going to be
trivial: a top-level batch file to build all the
tests with openwatcom c compiler (it's free,
but not quite free enough to make it into debian),
and another top-level batch file to run all the tests.
(We'll also provide a batch file to install open watcom.)

Given that not many developers will bother installing
openwatcom just to be able to compile the win16
tests, I suspect that might be enough of a test
harness for now.  Objections?
- Dan

[Alexandre Julliard]

"Dan Kegel"  writes:
> Given that not many developers will bother installing
> openwatcom just to be able to compile the win16
> tests, I suspect that might be enough of a test
> harness for now.  Objections?

I don't think there's any point in putting into the tree test cases
what won't be built or executed. They will just bitrot. We already
have a hard time keeping the normal test suite in a usable state.

[Dan Kegel]

On 7/16/07, Alexandre Julliard  wrote:
> I don't think there's any point in putting into the tree test cases
> what won't be built or executed. They will just bitrot. We already
> have a hard time keeping the normal test suite in a usable state.

So if I want these in the tree, they should be built and run
by default.  Got it.  Next question: is it acceptable to
require OpenWatcom?  (That's the only free compiler I
know of that targets win16.  Turbo C++ is also a free
download, but it's not open source.)
Since it's not available via apt-get install,
the easiest way to use it might be to have
'make test' install the windows version, and then use
it to compile the test cases.  The download and install for the
compiler could be cached, so that part would only be slow
the first run.

Alternately, I could recode the win16 test cases in
assembly.  Since the usual tools can't link win16
executables, the assembly would have to also encode
the executable headers.  This would make sure the
test cases got run every time, but would make
adding new test cases hard.  (Doesn't sound like
a good idea to me.)

Or would you prefer the win16 tests stay out of tree
forever?  That's an option, too; people who want
to run them could download them from kegel.com or
something, and it wouldn't stop Codeweavers from
running them in cxtest.
- Dan

[Alexandre Julliard]

"Dan Kegel"  writes:
> So if I want these in the tree, they should be built and run
> by default.  Got it.  Next question: is it acceptable to
> require OpenWatcom?  (That's the only free compiler I
> know of that targets win16.  Turbo C++ is also a free
> download, but it's not open source.)
> Since it's not available via apt-get install,
> the easiest way to use it might be to have
> 'make test' install the windows version, and then use
> it to compile the test cases.  The download and install for the
> compiler could be cached, so that part would only be slow
> the first run.

I don't think we want to do it that way, we should use a Unix version
of OpenWatcom, and detect it in configure.
> Or would you prefer the win16 tests stay out of tree
> forever?  That's an option, too; people who want
> to run them could download them from kegel.com or
> something, and it wouldn't stop Codeweavers from
> running them in cxtest.

It depends how many tests there are, but if it's just a few that's
probably the best solution, we wouldn't want to have to maintain a lot
of makefile infrastructure just for a couple of tests. Somehow I doubt
that we'll see a lot of development on win16 tests...

[Dan Kegel]

On 7/16/07, Alexandre Julliard  wrote:
> I don't think we want to do it that way, we should use a Unix version
> of OpenWatcom, and detect it in configure.

The OpenWatcom guys haven't gotten around
to making Linux packages available from their main
download site.  It should happen eventually.
> > Or would you prefer the win16 tests stay out of tree
> > forever?  That's an option, too; people who want
> > to run them could download them from kegel.com or
> > something, and it wouldn't stop Codeweavers from
> > running them in cxtest.
>
> It depends how many tests there are, but if it's just a few that's
> probably the best solution, we wouldn't want to have to maintain a lot
> of makefile infrastructure just for a couple of tests. Somehow I doubt
> that we'll see a lot of development on win16 tests...

OK, separate download for now.
- Dan