From shlomif@... Thu Jul 31 03:38:56 2008
Return-Path: <shlomif@...>
X-Sender: shlomif@...
X-Apparently-To: fc-solve-discuss@yahoogroups.com
X-Received: (qmail 46003 invoked from network); 31 Jul 2008 10:38:54 -0000
X-Received: from unknown (66.218.67.97)
by m52.grp.scd.yahoo.com with QMQP; 31 Jul 2008 10:38:54 -0000
X-Received: from unknown (HELO mxout4.netvision.net.il) (194.90.9.27)
by mta18.grp.scd.yahoo.com with SMTP; 31 Jul 2008 10:38:54 -0000
X-Received: from khoresh.eonspace.net ([212.143.218.31])
by mxout4.netvision.net.il
(Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007))
with ESMTP id <0K4V009989KTHTC0@...> for
fc-solve-discuss@yahoogroups.com; Thu, 31 Jul 2008 13:38:53 +0300 (IDT)
X-Received: by khoresh.eonspace.net (Postfix, from userid 8) id D4B957B540; Thu,
31 Jul 2008 13:27:11 +0300 (IDT)
X-Received: from [192.168.1.105]
(bzq-219-139-216.static.bezeqint.net [62.219.139.216])
(Authenticated sender: shlomif.shlomifish) by khoresh.eonspace.net (Postfix)
with ESMTP id 772E57B52E; Thu, 31 Jul 2008 13:27:08 +0300 (IDT)
Date: Thu, 31 Jul 2008 13:27:04 +0300
To: Freecell Solving Discussions <fc-solve-discuss@yahoogroups.com>
Cc: Hackers-IL <hackers-il@yahoogroups.com>, Perl in Israel <perl@...>
Message-id: <200807311327.05397.shlomif@...>
MIME-version: 1.0
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7BIT
Content-disposition: inline
X-Spam-Result: No, tests=bogofilter, spamicity=0.000000, version=1.1.5
User-Agent: KMail/1.9.9
X-Originating-IP: 194.90.9.27
X-eGroups-Msg-Info: 1:12:0:0:0
From: Shlomi Fish <shlomif@...>
Subject: Freecell Solver's Test Suite + Conversion to CMake
X-Yahoo-Group-Post: member; u=170008011; y=zCfrbyOZu9WGZuMyKxKSkmFk-xWwyym5DrZuNuwOWjmJwR4
X-Yahoo-Profile: shlomif2
Hi all!
(All the lists should have a reply-to, so please only reply-to one list).
I recently resumed work on Freecell Solver (FCS) somewhat more intensively.
Having gotten tired of Autoconf/Automake/Libtool (a.k.a Autohell), I decided
to convert to CMake ( http://www.cmake.org/ ) for its configuration stage.
However, since FCS lacked automated tests, I was afraid that by the
conversion I would break something. (It's kinda paranoid, but still).
As a result, I started working on a test suite. The first tests I wrote were C
files generated by Perl and Template toolkit to test the card input/output.
It was a start but not enough.
Then I started working on http://fc-solve.berlios.de/verify-code/ , which is a
Perl CPAN module to verify solutions of Freecell-like games. At first, it
could only do Freecell, but was eventually made more generic and can now
solve many other variants. It is primarily intended for Freecell Solver, but
is generic enough to be of use for other solvers.
Using it, I added tests that verify the entire solutions of about 10 deals and
configurations of various games. Afterwards, I also made sure that the exact
output of the tests remain the same (an output regression) by keeping track
of the SHA-256 hash of the solution.
Then I came to what I wanted - CMake. I began converting FCS to CMake here:
http://svn.berlios.de/svnroot/repos/fc-solve/branches/conversion-to-cmake/
It's naturally a long process, because CMake is not backwards compatible with
the GNU Autotools (and for a good reason, because they are very quirky), but
I've been making progress. Hopefully, I can make use of my experience with
converting FCS to CMake in working on http://thewml.org/ .
I should also note that someone I met on IRC expressed interest in
contributing to FCS. However, he hasn't even played Freecell yet, so it may
take some time, until he actually contributes.
Regards,
Shlomi Fish
-----------------------------------------------------------------
Shlomi Fish http://www.shlomifish.org/
Interview with Ben Collins-Sussman - http://xrl.us/bjn8s
I met a guy in the bar, talked to her and she gave me her phone number.
|
Thu Jul 31, 2008 10:27 am
Shlomi Fish <shlomif@...>
shlomif2
Offline Send Email
|