comparison software/eXist/mpdl-modules/src/org/exist/xquery/modules/mpdltext/GetLemma.java @ 0:408254cf2f1d

Erstellung
author Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
date Wed, 24 Nov 2010 17:24:23 +0100
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:408254cf2f1d
1 /*
2 * eXist Open Source Native XML Database: Extension module
3 * Copyright (C) 2008 Josef Willenborg
4 * jwillenborg@mpiwg-berlin.mpg.de
5 * http://www.mpiwg-berlin.mpg.de
6 *
7 * This program is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU Lesser General Public License
9 * as published by the Free Software Foundation; either version 2
10 * of the License, or (at your option) any later version.
11 *
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU Lesser General Public License for more details.
16 *
17 * You should have received a copy of the GNU Lesser General Public License
18 * along with this program; if not, write to the Free Software
19 * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
20 *
21 * $Id: TextModule.java $
22 */
23 package org.exist.xquery.modules.mpdltext;
24
25 import java.util.ArrayList;
26
27 import org.exist.dom.QName;
28 import org.exist.memtree.DocumentImpl;
29 import org.exist.memtree.MemTreeBuilder;
30 import org.exist.xquery.BasicFunction;
31 import org.exist.xquery.Cardinality;
32 import org.exist.xquery.FunctionSignature;
33 import org.exist.xquery.XPathException;
34 import org.exist.xquery.XQueryContext;
35 import org.exist.xquery.value.Sequence;
36 import org.exist.xquery.value.SequenceType;
37 import org.exist.xquery.value.Type;
38
39 import de.mpg.mpiwg.berlin.mpdl.exception.ApplicationException;
40 import de.mpg.mpiwg.berlin.mpdl.lt.morph.app.Form;
41 import de.mpg.mpiwg.berlin.mpdl.lt.morph.app.Lemma;
42 import de.mpg.mpiwg.berlin.mpdl.lt.morph.app.MorphologyCache;
43
44 /**
45 * @author Josef Willenborg (jwillenborg@mpiwg-berlin.mpg.de)
46 */
47 public class GetLemma extends BasicFunction {
48
49 public final static FunctionSignature signature =
50 new FunctionSignature(
51 new QName("get-lemma", MPDLTextModule.NAMESPACE_URI, MPDLTextModule.PREFIX),
52 "A function which delivers the lemma of a given lemma name and language " +
53 "by the MPDL language technology",
54 new SequenceType[] { new SequenceType(Type.STRING, Cardinality.ZERO_OR_MORE),
55 new SequenceType(Type.STRING, Cardinality.ZERO_OR_MORE) },
56 new SequenceType(Type.NODE, Cardinality.EXACTLY_ONE));
57
58 public GetLemma(XQueryContext context) {
59 super(context, signature);
60 }
61
62 public Sequence eval(Sequence[] args, Sequence contextSequence) throws XPathException {
63 Sequence langSeq = args[0];
64 Sequence formNameSeq = args[1];
65 String language = "";
66 String formName = "";
67 if (langSeq.isEmpty() || formNameSeq.isEmpty())
68 return Sequence.EMPTY_SEQUENCE;
69 language = langSeq.getStringValue();
70 formName = formNameSeq.getStringValue();
71 ArrayList<Lemma> lemmas = null;
72 try {
73 MorphologyCache morphologyCache = MorphologyCache.getInstance();
74 Lemma lemma = morphologyCache.getLemma(language, formName, true);
75 if (lemma == null) {
76 lemma = new Lemma("standard analyzer", language, formName); // delivers at least one lemma with one form (with the formName)
77 }
78 lemmas = new ArrayList<Lemma>();
79 lemmas.add(lemma);
80 } catch (ApplicationException e) {
81 throw new XPathException(e);
82 }
83 DocumentImpl doc = null;
84 if (lemmas != null) {
85 MemTreeBuilder builder = context.getDocumentBuilder();
86 builder.startElement("", "lemmas", "lemmas", null);
87 for (int i=0; i<lemmas.size(); i++) {
88 Lemma lemma = lemmas.get(i);
89 builder.startElement("", "lemma", "lemma", null);
90 builder.startElement("", "provider", "provider", null);
91 builder.characters(lemma.getProvider());
92 builder.endElement();
93 builder.startElement("", "language", "language", null);
94 builder.characters(lemma.getLanguage());
95 builder.endElement();
96 builder.startElement("", "lemma-name", "lemma-name", null);
97 builder.characters(lemma.getLemmaName());
98 builder.endElement();
99 ArrayList<Form> forms = lemma.getFormsList();
100 builder.startElement("", "forms-size", "forms-size", null);
101 builder.characters(String.valueOf(forms.size()));
102 builder.endElement();
103 builder.startElement("", "forms", "forms", null);
104 for (int j=0; j<forms.size(); j++) {
105 Form form = forms.get(j);
106 builder.startElement("", "form", "form", null);
107 builder.startElement("", "provider", "provider", null);
108 builder.characters(form.getProvider());
109 builder.endElement();
110 builder.startElement("", "language", "language", null);
111 builder.characters(form.getLanguage());
112 builder.endElement();
113 builder.startElement("", "form-name", "form-name", null);
114 builder.characters(form.getFormName());
115 builder.endElement();
116 builder.startElement("", "lemma-name", "lemma-name", null);
117 builder.characters(lemma.getLemmaName());
118 builder.endElement();
119 builder.endElement();
120 }
121 builder.endElement();
122 builder.endElement();
123 }
124 builder.endElement();
125 doc = ((DocumentImpl)builder.getDocument());
126 } else {
127 return Sequence.EMPTY_SEQUENCE;
128 }
129 return doc;
130 }
131 }