You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@uima.apache.org by mb...@apache.org on 2007/11/19 17:48:57 UTC
svn commit: r596361 -
/incubator/uima/sandbox/trunk/RegularExpressionAnnotator/docbook/RegexAnnotatorUserGuide/regexAnnotatorUserGuide.xml
Author: mbaessler
Date: Mon Nov 19 08:48:56 2007
New Revision: 596361
URL: http://svn.apache.org/viewvc?rev=596361&view=rev
Log:
UIMA-642
update documentation: add featurePath matching
https://issues.apache.org/jira/browse/UIMA-642
Modified:
incubator/uima/sandbox/trunk/RegularExpressionAnnotator/docbook/RegexAnnotatorUserGuide/regexAnnotatorUserGuide.xml
Modified: incubator/uima/sandbox/trunk/RegularExpressionAnnotator/docbook/RegexAnnotatorUserGuide/regexAnnotatorUserGuide.xml
URL: http://svn.apache.org/viewvc/incubator/uima/sandbox/trunk/RegularExpressionAnnotator/docbook/RegexAnnotatorUserGuide/regexAnnotatorUserGuide.xml?rev=596361&r1=596360&r2=596361&view=diff
==============================================================================
--- incubator/uima/sandbox/trunk/RegularExpressionAnnotator/docbook/RegexAnnotatorUserGuide/regexAnnotatorUserGuide.xml (original)
+++ incubator/uima/sandbox/trunk/RegularExpressionAnnotator/docbook/RegexAnnotatorUserGuide/regexAnnotatorUserGuide.xml Mon Nov 19 08:48:56 2007
@@ -277,12 +277,12 @@
</para>
<para>
The listing below shows an abstract rule definition with
- all the possible sub elements. Please refer to
+ all possible sub elements and attributes. Please refer to
the sub sections for details about the sub elements.
</para>
<para>
<programlisting><emphasis><![CDATA[<rule ruleId="ID1" regEx="TestRegex" matchStrategy="matchAll"
- matchType="uima.tcas.DocumentAnnotation" confidence="1.0">
+ matchType="uima.tcas.DocumentAnnotation" featurePath="my/feature/path" confidence="1.0">
<matchTypeFilter>
<feature name="language">en</feature>
@@ -337,7 +337,8 @@
- The annotation type that is used
to match the regular expression pattern.
As input text for the match, the annotation span
- is used.
+ is used, but only if no additional <code>featurePath</code>
+ feature is specified.
</para>
</listitem>
</itemizedlist>
@@ -369,6 +370,29 @@
<xref linkend="sandbox.regexAnnotator.conceptsFile.annotationCreation.features"/>).
</para>
</listitem>
+ <listitem>
+ <para>
+ <code>featurePath</code>
+ - Specifies the feature path that should be used to match the regular expression pattern.
+ If a feature path is specified, the feature path value is used to match against the
+ regular expression instead of the match type annotation span.
+ The defined feature path must be valid for the specified match type annotation type.
+ The feature path elements are separated by "/".
+ </para>
+ <para>
+ The listing below shows how to match a regular expression on the <code>normalizedText</code>
+ feature of a <code>uima.TokenAnnotation</code>. So in this case, not the covered text of the
+ <code>uima.TokenAnnotation</code> is used to match the regular expression but the
+ <code>normalizedText</code> feature value of the annotation. The <code>normalizedText</code>
+ feature must be defined in the UIMA type system as feature of type <code>uima.TokenAnnotation</code>.
+ </para>
+ <para>
+ <programlisting><emphasis><![CDATA[<rule regEx="TestRegex" matchStrategy="matchAll"
+ matchType="uima.TokenAnnotation" featurePath="normalizedText">
+</rule>
+]]></emphasis></programlisting>
+ </para>
+ </listitem>
</itemizedlist>
<section
@@ -684,6 +708,15 @@
</listitem>
</itemizedlist>
</para>
+ <note>
+ <para>
+ When the rule definition defines a <code>featurePath</code> for a <code>matchType</code>,
+ the annotation boundaries for the created annotation are automatically set to
+ the annotation boundaries of the match input annotation. This must be done since
+ the matching with a feature value of an annotation has no relation to the document text, so the only
+ relation is the annotation where the feature is defined.
+ </para>
+ </note>
</section>
<section id="sandbox.regexAnnotator.conceptsFile.annotationCreation.validation">
<title>Annotation Validation</title>
@@ -701,8 +734,7 @@
method called <code>validate(String coveredText, String ruleID)</code>. The validate method is called by the annotator
before the match is added as annotation to the CAS. Annotations are only added if the validate method
returns <code>true</code>, otherwise the match is skipped. The <code>coveredText</code> parameter contains
- the text span that will be covered by the annotation and was defined using the annotation boundaries (see
- <xref linkend="sandbox.regexAnnotator.conceptsFile.annotationCreation.boundaries"/> for details).
+ the text that matches the regular expression.
The <code>ruleID</code> parameter contains the ruldId of the rule that creates the match. This can also be null
if no ruleID was specified. The listing below shows a sample implementation of the validation interface.
</para>
@@ -1032,7 +1064,10 @@
</para>
<para>
<programlisting><![CDATA[<?xml version="1.0" encoding="UTF-8"?>
-<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema">
+<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"
+ targetNamespace="http://incubator.apache.org/uima/regex"
+ xmlns="http://incubator.apache.org/uima/regex"
+ elementFormDefault="qualified">
<!--
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
@@ -1104,6 +1139,7 @@
</xs:simpleType>
</xs:attribute>
<xs:attribute name="matchType" type="xs:string" use="required"/>
+ <xs:attribute name="featurePath" type="xs:string" use="optional" />
<xs:attribute name="ruleId" type="xs:string" use="optional"/>
<xs:attribute name="confidence" type="xs:decimal" use="optional"/>
</xs:complexType>