35 lines
1.1 KiB
Ruby
35 lines
1.1 KiB
Ruby
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-2.0.0/apache-hive-2.0.0-bin.tar.gz"
|
|
sha256 "e854798954f48bdbc40235b1367287218464b62376d29dfcd2a328a0e7a184e6"
|
|
|
|
bottle :unneeded
|
|
|
|
depends_on "hadoop"
|
|
depends_on :java
|
|
|
|
def install
|
|
rm_f Dir["bin/*.cmd", "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
|
|
|
|
test do
|
|
system "#{bin}/schematool", "-initSchema", "-dbType", "derby"
|
|
assert_match "Hive #{version}", shell_output("#{bin}/hive --version")
|
|
end
|
|
end
|