1
|
/**
|
2
|
* '$RCSfile$'
|
3
|
*
|
4
|
* Copyright: 2016 Regents of the University of California and the
|
5
|
* National Center for Ecological Analysis and Synthesis
|
6
|
* Authors: Jing Tao
|
7
|
*
|
8
|
*
|
9
|
* This program is free software; you can redistribute it and/or modify
|
10
|
* it under the terms of the GNU General Public License as published by
|
11
|
* the Free Software Foundation; either version 2 of the License, or
|
12
|
* (at your option) any later version.
|
13
|
*
|
14
|
* This program is distributed in the hope that it will be useful,
|
15
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
16
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
17
|
* GNU General Public License for more details.
|
18
|
*
|
19
|
* You should have received a copy of the GNU General Public License
|
20
|
* along with this program; if not, write to the Free Software
|
21
|
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
22
|
*/
|
23
|
package edu.ucsb.nceas.metacat.service;
|
24
|
|
25
|
import java.io.IOException;
|
26
|
import java.io.Reader;
|
27
|
|
28
|
import org.apache.log4j.Logger;
|
29
|
import org.xml.sax.Attributes;
|
30
|
import org.xml.sax.InputSource;
|
31
|
import org.xml.sax.SAXException;
|
32
|
import org.xml.sax.XMLReader;
|
33
|
import org.xml.sax.helpers.DefaultHandler;
|
34
|
import org.xml.sax.helpers.XMLReaderFactory;
|
35
|
|
36
|
import edu.ucsb.nceas.metacat.properties.PropertyService;
|
37
|
import edu.ucsb.nceas.utilities.PropertyNotFoundException;
|
38
|
|
39
|
|
40
|
/**
|
41
|
* This class will parse the root element to figure out the namespace of the root element(we also call
|
42
|
* it the namespace of the element). If it doesn't have a namespace, but it does have an attribute of noNamespaceSchemaLocation
|
43
|
* at the root element, it will get the value as well.
|
44
|
* @author tao
|
45
|
*
|
46
|
*/
|
47
|
public class XMLNamespaceParser extends DefaultHandler {
|
48
|
|
49
|
private Reader xml = null;
|
50
|
private XMLReader parser = null;
|
51
|
private boolean rootElement = true;
|
52
|
private static Logger logMetacat = Logger.getLogger(XMLNamespaceParser.class);
|
53
|
private String namespace = null;
|
54
|
private String noNamespaceSchemaLocation = null;
|
55
|
|
56
|
/**
|
57
|
* Constructor
|
58
|
* @param xml the xml object which will be parsed
|
59
|
*/
|
60
|
public XMLNamespaceParser(Reader xml) throws SAXException, PropertyNotFoundException {
|
61
|
this.xml = xml;
|
62
|
initParser();
|
63
|
}
|
64
|
|
65
|
/*
|
66
|
* Initialize sax parser
|
67
|
*/
|
68
|
private void initParser() throws SAXException, PropertyNotFoundException {
|
69
|
// Get an instance of the parser
|
70
|
String parserName = PropertyService.getProperty("xml.saxparser");
|
71
|
parser = XMLReaderFactory.createXMLReader(parserName);
|
72
|
parser.setContentHandler(this);
|
73
|
|
74
|
}
|
75
|
|
76
|
/**
|
77
|
* Parse the xml file
|
78
|
* @throws SAXException if some sax related exception happens
|
79
|
* @throws IOException if the schema content couldn't be found
|
80
|
*/
|
81
|
public void parse() throws SAXException, IOException {
|
82
|
try {
|
83
|
parser.parse(new InputSource(xml));
|
84
|
} catch (ParsingEndException e) {
|
85
|
logMetacat.debug("XMLNamespace.parse - The parsing process stopped.");
|
86
|
}
|
87
|
|
88
|
}
|
89
|
|
90
|
/** SAX Handler that is called at the start of each XML element */
|
91
|
@Override
|
92
|
public void startElement(String uri, String localName, String qName,
|
93
|
Attributes atts) throws SAXException{
|
94
|
logMetacat.debug("XMLNamespace.startElement - uri: "+uri);
|
95
|
logMetacat.debug("XMLNamespace.startElement - local name: "+localName);
|
96
|
logMetacat.debug("XMLNamespace.startElement - qualified name: "+qName);
|
97
|
if(!rootElement) {
|
98
|
throw new ParsingEndException("We only parse the root elment. We got there and the parsing stopped.");
|
99
|
} else {
|
100
|
rootElement = false;
|
101
|
if(uri != null && !uri.trim().equals("")) {
|
102
|
namespace = uri;
|
103
|
}
|
104
|
logMetacat.debug("XMLNamespace.startElement - the namespace is: "+namespace);
|
105
|
if(atts != null) {
|
106
|
for(int i=0; i<atts.getLength(); i++) {
|
107
|
if((atts.getURI(i) != null && atts.getURI(i).equals("http://www.w3.org/2001/XMLSchema-instance")) &&
|
108
|
(atts.getLocalName(i) != null && atts.getLocalName(i).equals("noNamespaceSchemaLocation"))) {
|
109
|
if(atts.getValue(i) != null && !atts.getValue(i).trim().equals("")) {
|
110
|
noNamespaceSchemaLocation = atts.getValue(i);
|
111
|
}
|
112
|
logMetacat.debug("XMLNamespace.startElement - we found the attribute of the noNamespaceSchemaLocation and its value is: "+noNamespaceSchemaLocation);
|
113
|
break;
|
114
|
}
|
115
|
}
|
116
|
}
|
117
|
|
118
|
}
|
119
|
}
|
120
|
|
121
|
|
122
|
/**
|
123
|
* Get the namespace of the document (root element). The parse() method should be called first
|
124
|
* @return the value of the namespace. A null will be returned if it can't be found.
|
125
|
*/
|
126
|
public String getNamespace() {
|
127
|
logMetacat.debug("XMLNamespace.getNamespace - the namespace is: "+namespace);
|
128
|
return namespace;
|
129
|
}
|
130
|
|
131
|
|
132
|
/**
|
133
|
* Get the value of noNamespaceSchemaLocation of the document (root element). The parse() method should be called first
|
134
|
* @return the value of the noNamespaceSchemaLocation. A null will be returned if it can't be found.
|
135
|
*/
|
136
|
public String getNoNamespaceSchemaLocation() {
|
137
|
logMetacat.debug("XMLNamespace.getNoNamespaceSchemaLocation - the NoNamespaceSchemaLocation is: "+noNamespaceSchemaLocation);
|
138
|
return noNamespaceSchemaLocation;
|
139
|
}
|
140
|
|
141
|
/**
|
142
|
* A class signals that the parsing process stop early.
|
143
|
* @author tao
|
144
|
*
|
145
|
*/
|
146
|
class ParsingEndException extends RuntimeException {
|
147
|
public ParsingEndException(String message) {
|
148
|
super(message);
|
149
|
}
|
150
|
}
|
151
|
}
|