uimafit-docbook/src/docbook/tools.uimafit.pipelines.xml

<!--
	Licensed to the Apache Software Foundation (ASF) under one
	or more contributor license agreements. See the NOTICE file
	distributed with this work for additional information
	regarding copyright ownership. The ASF licenses this file
	to you under the Apache License, Version 2.0 (the
	"License"); you may not use this file except in compliance
	with the License. You may obtain a copy of the License at

	http://www.apache.org/licenses/LICENSE-2.0

	Unless required by applicable law or agreed to in writing,
	software distributed under the License is distributed on an
	"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
	KIND, either express or implied. See the License for the
	specific language governing permissions and limitations
	under the License.
-->
<chapter id="ugr.tools.uimafit.pipelines">
  <title>Pipelines</title>
  <para>UIMA is a component-based architecture that allows composing various processing components
    into a complex processing pipeline. A pipeline typically involves a <emphasis>collection
      reader</emphasis> which ingests documents and <emphasis>analysis engines</emphasis> that do
    the actual processing.</para>
  <para>Normally, you would run a pipeline using a UIMA Collection Processing Engine or using UIMA
    AS. uimaFIT offers a third alternative that is much simpler to use and well suited for embedding
    UIMA pipelines into applications or for writing tests.</para>
  <para>As uimaFIT does not supply any readers or processing components, we just assume that we have
    written three components:</para>
  <itemizedlist>
    <listitem>
      <para><classname>TextReader</classname> - reads text files from a directory</para>
    </listitem>
    <listitem>
      <para><classname>Tokenizer</classname> - annotates tokens</para>
    </listitem>
    <listitem>
      <para><classname>TokenFrequencyWriter</classname> - writes a list of tokens and their
        frequency to a file</para>
    </listitem>
  </itemizedlist>
  <para>We create descriptors for all components and run them as a pipeline:</para>
  <programlisting>CollectionReaderDescription reader = 
  CollectionReaderFactory.createReaderDescription(
    TextReader.class, 
    TextReader.PARAM_INPUT, "/home/uimafit/documents");

AnalysisEngineDescription tokenizer = 
  AnalysisEngineFactory.createEngineDescription(
    Tokenizer.class);

AnalysisEngineDescription tokenFrequencyWriter = 
  AnalysisEngineFactory.createEngineDescription(
    TokenFrequencyWriter.class, 
    TokenFrequencyWriter.PARAM_OUTPUT, "counts.txt");

SimplePipeline.runPipeline(reader, tokenizer, writer);</programlisting>
  <para>Instead of running the full pipeline end-to-end, we can also process one document at a time
    and inspect the analysis results:</para>
  <programlisting>CollectionReaderDescription reader = 
  CollectionReaderFactory.createReaderDescription(
    TextReader.class, 
    TextReader.PARAM_INPUT, "/home/uimafit/documents");

AnalysisEngineDescription tokenizer = 
  AnalysisEngineFactory.createEngineDescription(
    Tokenizer.class);

for (JCas jcas : SimplePipeline.iteratePipeline(reader, tokenizer)) {
  System.out.printf("Found %d tokens%n", 
    JCasUtil.select(jcas, Token.class).size());
}</programlisting>
</chapter>