class Yamcha < Formula desc "NLP text chunker using Support Vector Machines" homepage "http://chasen.org/~taku/software/yamcha/" url "http://chasen.org/~taku/software/yamcha/src/yamcha-0.33.tar.gz" sha256 "413d4fc0a4c13895f5eb1468e15c9d2828151882f27aea4daf2399c876be27d5" bottle do cellar :any rebuild 1 sha256 "b9f2e9521d25dafc70617857f32b1742b8bb29046b3ea930eafb3261a0727e36" => :sierra sha256 "b65fade9c6ddcced1d3c3fc6700f18ed2ddd16b62437fc71f9a85a3568851520" => :el_capitan sha256 "b038ddce247b7f56041c4325fd01c0ab0b32399d1ca602df37b65739a09b74e0" => :yosemite sha256 "6b43b01d7d2385706e3259a49e8944368edef3a20755ea4093f4e3ebbf56eb27" => :mavericks end depends_on "tinysvm" def install system "./configure", "--disable-debug", "--disable-dependency-tracking", "--prefix=#{prefix}", "--mandir=#{man}" system "make", "install" end test do libexecdir = shell_output("#{bin}/yamcha-config --libexecdir").chomp assert_equal libexecdir, "#{libexec}/yamcha" (testpath/"train.data").write <<-EOS.undent He PRP B-NP reckons VBZ B-VP the DT B-NP current JJ I-NP account NN I-NP deficit NN I-NP will MD B-VP narrow VB I-VP to TO B-PP only RB B-NP # # I-NP 1.8 CD I-NP billion CD I-NP in IN B-PP September NNP B-NP . . O He PRP B-NP reckons VBZ B-VP .. EOS system "make", "-j", "1", "-f", "#{libexecdir}/Makefile", "CORPUS=train.data", "MODEL=case_study", "train" %w[log model se svmdata txtmodel.gz].each do |ext| assert File.exist? testpath/"case_study.#{ext}" end end end