class Hive < Formula desc "Hadoop-based data summarization, query, and analysis" homepage 'https://hive.apache.org' url 'https://www.apache.org/dyn/closer.cgi?path=hive/hive-1.1.1/apache-hive-1.1.1-bin.tar.gz' sha256 "71cb92f87aaea1af69ff27f95878f3190e9b184cdff84b8f8740af4cc99d81c3" depends_on 'hadoop' depends_on :java def install rm_f Dir["bin/ext/*.cmd", "bin/ext/util/*.cmd"] libexec.install %w[bin conf examples hcatalog lib scripts] bin.write_exec_script Dir["#{libexec}/bin/*"] end def caveats; <<-EOS.undent Hadoop must be in your path for hive executable to work. After installation, set $HIVE_HOME in your profile: export HIVE_HOME=#{libexec} If you want to use HCatalog with Pig, set $HCAT_HOME in your profile: export HCAT_HOME=#{libexec}/hcatalog You may need to set JAVA_HOME: export JAVA_HOME="$(/usr/libexec/java_home)" EOS end end