| <?xml version="1.0"?> |
| <project name="benchmark" default="default"> |
| |
| <description> |
| Lucene Benchmarking Contributions |
| </description> |
| |
| <import file="../contrib-build.xml"/> |
| <property name="working.dir" location="work"/> |
| |
| <!-- the tests have some parallel problems --> |
| <property name="tests.threadspercpu" value="0"/> |
| |
| <target name="check-files"> |
| <available file="temp/news20.tar.gz" property="news20.exists"/> |
| |
| <available file="${working.dir}/20_newsgroup" property="news20.expanded"/> |
| |
| <available file="temp/reuters21578.tar.gz" property="reuters.exists"/> |
| <available file="${working.dir}/reuters" property="reuters.expanded"/> |
| <available file="${working.dir}/reuters-out" property="reuters.extracted"/> |
| <available file="temp/20news-18828.tar.gz" property="20news-18828.exists"/> |
| <available file="${working.dir}/20news-18828" property="20news-18828.expanded"/> |
| <available file="${working.dir}/mini_newsgroups" property="mini.expanded"/> |
| |
| <available file="temp/enwiki-20070527-pages-articles.xml.bz2" property="enwiki.exists"/> |
| <available file="temp/enwiki-20070527-pages-articles.xml" property="enwiki.expanded"/> |
| <available file="${working.dir}/enwiki.txt" property="enwiki.extracted"/> |
| <available file="temp/${top.100k.words.archive.filename}" |
| property="top.100k.words.archive.present"/> |
| <available file="${working.dir}/top100k-out" |
| property="top.100k.word.files.expanded"/> |
| </target> |
| |
| <target name="enwiki-files" depends="check-files"> |
| <mkdir dir="temp"/> |
| <antcall target="get-enwiki"/> |
| <antcall target="expand-enwiki"/> |
| </target> |
| |
| <target name="get-enwiki" unless="enwiki.exists"> |
| <get src="http://people.apache.org/~gsingers/wikipedia/enwiki-20070527-pages-articles.xml.bz2" |
| dest="temp/enwiki-20070527-pages-articles.xml.bz2"/> |
| </target> |
| |
| <target name="expand-enwiki" unless="enwiki.expanded"> |
| <bunzip2 src="temp/enwiki-20070527-pages-articles.xml.bz2" dest="temp"/> |
| </target> |
| |
| <target name="get-news-20" unless="20news-18828.exists"> |
| <get src="http://www-2.cs.cmu.edu/afs/cs.cmu.edu/project/theo-20/www/data/news20.tar.gz" |
| dest="temp/news20.tar.gz"/> |
| |
| </target> |
| <target name="get-reuters" unless="reuters.exists"> |
| |
| <get src="http://www.daviddlewis.com/resources/testcollections/reuters21578/reuters21578.tar.gz" |
| dest="temp/reuters21578.tar.gz"/> |
| </target> |
| |
| <target name="expand-news-20" unless="news20.expanded"> |
| <gunzip src="temp/news20.tar.gz" dest="temp"/> |
| <untar src="temp/news20.tar" dest="${working.dir}"/> |
| </target> |
| <target name="expand-reuters" unless="reuters.expanded"> |
| <gunzip src="temp/reuters21578.tar.gz" dest="temp"/> |
| <mkdir dir="${working.dir}/reuters"/> |
| <untar src="temp/reuters21578.tar" dest="${working.dir}/reuters"/> |
| <delete > |
| <fileset dir="${working.dir}/reuters"> |
| <include name="*.txt"/> |
| </fileset> |
| </delete> |
| |
| </target> |
| <target name="extract-reuters" depends="check-files" unless="reuters.extracted"> |
| <java classname="org.apache.lucene.benchmark.utils.ExtractReuters" maxmemory="1024M" fork="true"> |
| <classpath refid="run.classpath"/> |
| <arg file="${working.dir}/reuters"/> |
| <arg file="${working.dir}/reuters-out"/> |
| </java> |
| </target> |
| <target name="get-20news-18828" unless="20news-18828.exists"> |
| <get src="http://people.csail.mit.edu/u/j/jrennie/public_html/20Newsgroups/20news-18828.tar.gz" |
| dest="temp/20news-18828.tar.gz"/> |
| |
| </target> |
| <target name="expand-20news-18828" unless="20news-18828.expanded"> |
| <gunzip src="temp/20news-18828.tar.gz" dest="temp"/> |
| <untar src="temp/20news-18828.tar" dest="${working.dir}"/> |
| </target> |
| <target name="get-mini-news" unless="mini.exists"> |
| <get src="http://kdd.ics.uci.edu/databases/20newsgroups/mini_newsgroups.tar.gz" |
| dest="temp/mini_newsgroups.tar.gz"/> |
| </target> |
| <target name="expand-mini-news" unless="mini.expanded"> |
| <gunzip src="temp/mini_newsgroups.tar.gz" dest="temp"/> |
| <untar src="temp/mini_newsgroups.tar" dest="${working.dir}"/> |
| </target> |
| |
| <property name="top.100k.words.archive.filename" |
| value="top.100k.words.de.en.fr.uk.wikipedia.2009-11.tar.bz2"/> |
| <property name="top.100k.words.archive.base.url" |
| value="http://people.apache.org/~rmuir/wikipedia"/> |
| <target name="get-top-100k-words-archive" unless="top.100k.words.archive.present"> |
| <mkdir dir="temp"/> |
| <get src="${top.100k.words.archive.base.url}/${top.100k.words.archive.filename}" |
| dest="temp/${top.100k.words.archive.filename}"/> |
| </target> |
| <target name="expand-top-100k-word-files" unless="top.100k.word.files.expanded"> |
| <mkdir dir="${working.dir}/top100k-out"/> |
| <untar src="temp/${top.100k.words.archive.filename}" |
| overwrite="true" compression="bzip2" dest="${working.dir}/top100k-out"/> |
| </target> |
| |
| <target name="top-100k-wiki-word-files" depends="check-files"> |
| <mkdir dir="${working.dir}"/> |
| <antcall target="get-top-100k-words-archive"/> |
| <antcall target="expand-top-100k-word-files"/> |
| </target> |
| |
| <target name="get-files" depends="check-files"> |
| <mkdir dir="temp"/> |
| <antcall target="get-reuters"/> |
| <antcall target="expand-reuters"/> |
| <antcall target="extract-reuters"/> |
| </target> |
| |
| <path id="classpath"> |
| <pathelement path="${memory.jar}"/> |
| <pathelement path="${highlighter.jar}"/> |
| <pathelement path="${analyzers-common.jar}"/> |
| <path refid="base.classpath"/> |
| <fileset dir="lib"> |
| <include name="**/*.jar"/> |
| </fileset> |
| </path> |
| <path id="run.classpath"> |
| <path refid="classpath"/> |
| <pathelement location="${build.dir}/classes/java"/> |
| <pathelement path="${benchmark.ext.classpath}"/> |
| </path> |
| |
| <property name="task.alg" location="conf/micro-standard.alg"/> |
| <property name="task.mem" value="140M"/> |
| |
| <target name="run-task" depends="compile,check-files,get-files" |
| description="Run compound penalty perf test (optional: -Dtask.alg=your-algorithm-file -Dtask.mem=java-max-mem)"> |
| <echo>Working Directory: ${working.dir}</echo> |
| <java classname="org.apache.lucene.benchmark.byTask.Benchmark" maxmemory="${task.mem}" fork="true"> |
| <classpath refid="run.classpath"/> |
| <arg file="${task.alg}"/> |
| </java> |
| </target> |
| |
| <target name="enwiki" depends="compile,check-files,enwiki-files"> |
| <echo>Working Directory: ${working.dir}</echo> |
| <java classname="org.apache.lucene.benchmark.byTask.Benchmark" maxmemory="1024M" fork="true"> |
| <assertions> |
| <enable/> |
| </assertions> |
| <classpath refid="run.classpath"/> |
| <arg file="conf/extractWikipedia.alg"/> |
| </java> |
| </target> |
| |
| <property name="collation.alg.file" location="conf/collation.alg"/> |
| <property name="collation.output.file" |
| value="${working.dir}/collation.benchmark.output.txt"/> |
| <property name="collation.jira.output.file" |
| value="${working.dir}/collation.bm2jira.output.txt"/> |
| |
| <path id="collation.runtime.classpath"> |
| <path refid="run.classpath"/> |
| <pathelement path="${icu.jar}"/> |
| <fileset dir="${common.dir}/contrib/icu/lib" includes="icu4j*.jar"/> |
| </path> |
| |
| <target name="collation" depends="compile,jar-icu,top-100k-wiki-word-files"> |
| <echo>Running contrib/benchmark with alg file: ${collation.alg.file}</echo> |
| <java fork="true" classname="org.apache.lucene.benchmark.byTask.Benchmark" |
| maxmemory="${task.mem}" output="${collation.output.file}"> |
| <classpath refid="collation.runtime.classpath"/> |
| <arg file="${collation.alg.file}"/> |
| </java> |
| <echo>Benchmark output is in file: ${collation.output.file}</echo> |
| <echo>Converting to JIRA table format...</echo> |
| <exec executable="perl" output="${collation.jira.output.file}" failonerror="true"> |
| <arg value="scripts/collation.bm2jira.pl"/> |
| <arg value="${collation.output.file}"/> |
| </exec> |
| <echo>Benchmark output in JIRA table format is in file: ${collation.jira.output.file}</echo> |
| </target> |
| |
| <property name="shingle.alg.file" location="conf/shingle.alg"/> |
| <property name="shingle.output.file" |
| value="${working.dir}/shingle.benchmark.output.txt"/> |
| <property name="shingle.jira.output.file" |
| value="${working.dir}/shingle.bm2jira.output.txt"/> |
| |
| <path id="shingle.runtime.classpath"> |
| <path refid="run.classpath"/> |
| </path> |
| |
| <target name="shingle" depends="compile,get-files"> |
| <echo>Running contrib/benchmark with alg file: ${shingle.alg.file}</echo> |
| <java fork="true" classname="org.apache.lucene.benchmark.byTask.Benchmark" |
| maxmemory="${task.mem}" output="${shingle.output.file}"> |
| <classpath refid="run.classpath"/> |
| <arg file="${shingle.alg.file}"/> |
| </java> |
| <echo>Benchmark output is in file: ${shingle.output.file}</echo> |
| <echo>Converting to JIRA table format...</echo> |
| <exec executable="perl" output="${shingle.jira.output.file}" failonerror="true"> |
| <arg value="scripts/shingle.bm2jira.pl"/> |
| <arg value="${shingle.output.file}"/> |
| </exec> |
| <echo>Benchmark output in JIRA table format is in file: ${shingle.jira.output.file}</echo> |
| </target> |
| |
| <target name="init" depends="contrib-build.init,jar-memory,jar-highlighter,jar-analyzers-common"/> |
| |
| <target name="clean-javacc"> |
| <fileset dir="src/java/org/apache/lucene/benchmark/byTask/feeds/demohtml" includes="*.java"> |
| <containsregexp expression="Generated.*By.*JavaCC"/> |
| </fileset> |
| </target> |
| |
| <target name="javacc" depends="init,javacc-check" if="javacc.present"> |
| <invoke-javacc target="src/java/org/apache/lucene/benchmark/byTask/feeds/demohtml/HTMLParser.jj" |
| outputDir="src/java/org/apache/lucene/benchmark/byTask/feeds/demohtml" |
| /> |
| </target> |
| |
| <target name="dist-maven" depends="contrib-build.dist-maven"> |
| <m2-deploy-with-pom-template pom.xml="lib/lucene-xercesImpl-pom.xml.template" |
| jar.file="lib/xercesImpl-2.9.1-patched-XERCESJ-1257.jar" /> |
| </target> |
| </project> |