我有一个4.2 GB的XML文件！显然解析整个DOM是不切实际的。我一直在寻找SAX和STAX来完成解析这个巨大的XML文件。然而，我见过的所有例子都很简单。我正在处理的XML文件嵌套在嵌套的嵌套上。有些领域有10个以上的水平。
我找到了这个教程，但不确定它是否是一个可行的解决方案。
http://www.javacodegeeks.com/2013/05/parsing-xml-using-dom-sax-and-stax-parser-in-java.html（使用STAX的示例）
我不确定如何处理嵌套对象。
我创建了Java对象来模仿XML的结构。这里有一些，太多无法显示。
Record.java
public class Record implements Serializable {
String uid;
StaticData staticData;
DynamicData dynamicData;
}
Summary.java
public class Summary {
EWUID ewuid;
PubInfo pubInfo;
Titles titles;
Names names;
DocTypes docTypes;
Publishers publishers;
}
EWUID.java
public class EWUID {
String collId;
String edition;
}
PubInfo.java
public class PubInfo {
String coverDate;
String hasAbstract;
String issue;
String pubMonth;
String pubType;
String pubYear;
String sortDate;
String volume;
}
这是我到目前为止提出的代码。
public class TRWOSParser {
XMLEventReader eventReader;
XMLInputFactory inputFactory;
InputStream inputStream;
public TRWOSParser(String file) throws FileNotFoundException, XMLStreamException {
inputFactory = XMLInputFactory.newInstance();
inputStream = new FileInputStream(file);
eventReader = inputFactory.createXMLEventReader(inputStream);
}
public void parse() throws XMLStreamException{
while (eventReader.hasNext()) {
XMLEvent event = eventReader.nextEvent();
if (event.isStartElement()) {
StartElement startElement = event.asStartElement();
if (startElement.getName().getLocalPart().equals("record")) {
Record record = new Record();
Iterator<Attribute> attributes = startElement.getAttributes();
while (attributes.hasNext()) {
Attribute attribute = attributes.next();
if (attribute.getName().toString().equals("UID")) {
System.out.println("UID: " + attribute.getValue());
}
}
}
}
}
}
}
更新
XML中的数据已获得许可，因此无法显示完整文件。这是一个非常小的片段，我在其中加扰了数据。
<?xml version="1.0" encoding="UTF-8"?>
<records>
<REC>
<UID>WOS:000310438600004</UID>
<static_data>
<summary>
<EWUID>
<WUID coll_id="WOS" />
<edition value="WOS.SCI" />
</EWUID>
<pub_info coverdate="NOV 2012" has_abstract="N" issue="5" pubmonth="NOV" pubtype="Journal" pubyear="2012" sortdate="2012-11-01" vol="188">
<page begin="1662" end="1663" page_count="2">1662-1663</page>
</pub_info>
<titles count="6">
<title type="source">JOURNAL OF UROLOGY</title>
<title type="source_abbrev">J UROLOGY</title>
<title type="abbrev_iso">J. Urol.</title>
<title type="abbrev_11">J UROL</title>
<title type="abbrev_29">J UROL</title>
<title type="item">Something something</title>
</titles>
<names count="1">
<name addr_no="1 2 3" reprint="Y" role="author" seq_no="1">
<display_name>John Doe</display_name>
<full_name>John Doe</full_name>
<wos_standard>Doe, John</wos_standard>
<first_name>John</first_name>
<last_name>Doe</last_name>
</name>
</names>
<doctypes count="1">
<doctype>Editorial Material</doctype>
</doctypes>
<publishers>
<publisher>
<address_spec addr_no="1">
<full_address>360 PARK AVE SOUTH, NEW YORK, NY 10010-1710 USA</full_address>
<city>NEW YORK</city>
</address_spec>
<names count="1">
<name addr_no="1" role="publisher" seq_no="1">
<display_name>ELSEVIER SCIENCE INC</display_name>
<full_name>ELSEVIER SCIENCE INC</full_name>
</name>
</names>
</publisher>
</publishers>
</summary>
</static_data>
</REC>
</records>
答案 0 :(得分：2)
lscoughlin答案的类似解决方案是使用DOM4J，它有机制来处理这种情况：http://dom4j.sourceforge.net/
在我的蛋白质中，它更直接，更容易理解。但它可能不支持名称空间。
答案 1 :(得分：1)
我做了两个假设1）早期重复，2）你可以用部分文件做一些有意义的事情。
假设您可以移动某种级别的嵌套，然后多次处理文档，每次处理＆＃34;处理＆＃34;删除工作级别的节点。该文件。这意味着在任何给定时间内只有一个工作子树将在内存中。
这是一个有效的代码段：
package bigparse;
import static javax.xml.stream.XMLStreamConstants.CHARACTERS;
import static javax.xml.stream.XMLStreamConstants.END_DOCUMENT;
import static javax.xml.stream.XMLStreamConstants.END_ELEMENT;
import static javax.xml.stream.XMLStreamConstants.START_DOCUMENT;
import static javax.xml.stream.XMLStreamConstants.START_ELEMENT;
import java.io.FileNotFoundException;
import java.io.FileReader;
import java.io.StringWriter;
import javax.xml.parsers.DocumentBuilder;
import javax.xml.parsers.DocumentBuilderFactory;
import javax.xml.parsers.ParserConfigurationException;
import javax.xml.stream.XMLInputFactory;
import javax.xml.stream.XMLStreamException;
import javax.xml.stream.XMLStreamReader;
import javax.xml.transform.OutputKeys;
import javax.xml.transform.Transformer;
import javax.xml.transform.TransformerFactory;
import javax.xml.transform.dom.DOMSource;
import javax.xml.transform.stream.StreamResult;
import org.w3c.dom.Document;
import org.w3c.dom.Element;
public class BigParse {
public static void main(String... args) {
XMLInputFactory factory = XMLInputFactory.newInstance();
DocumentBuilderFactory documentBuilderFactory = DocumentBuilderFactory.newInstance();
try {
XMLStreamReader streamReader = factory.createXMLStreamReader(new FileReader("src/main/resources/test.xml"));
DocumentBuilder documentBuilder = documentBuilderFactory.newDocumentBuilder();
Document document = documentBuilder.newDocument();
Element rootElement = null;
Element currentElement = null;
int branchLevel = 0;
int maxBranchLevel = 1;
while (streamReader.hasNext()) {
int event = streamReader.next();
switch (event) {
case START_DOCUMENT:
continue;
case START_ELEMENT:
if (branchLevel < maxBranchLevel) {
Element workingElement = readElementOnly(streamReader, document);
if (rootElement == null) {
document.appendChild(workingElement);
rootElement = document.getDocumentElement();
currentElement = rootElement;
} else {
currentElement.appendChild(workingElement);
currentElement = workingElement;
}
branchLevel++;
} else {
workingLoop(streamReader, document, currentElement);
}
continue;
case CHARACTERS:
currentElement.setTextContent(streamReader.getText());
continue;
case END_ELEMENT:
if (currentElement != rootElement) {
currentElement = (Element) currentElement.getParentNode();
branchLevel--;
}
continue;
case END_DOCUMENT:
break;
}
}
} catch (ParserConfigurationException
| FileNotFoundException
| XMLStreamException e) {
throw new RuntimeException(e);
}
}
private static Element readElementOnly(XMLStreamReader streamReader, Document document) {
Element workingElement = document.createElement(streamReader.getLocalName());
for (int attributeIndex = 0; attributeIndex < streamReader.getAttributeCount(); attributeIndex++) {
workingElement.setAttribute(
streamReader.getAttributeLocalName(attributeIndex),
streamReader.getAttributeValue(attributeIndex));
}
return workingElement;
}
private static void workingLoop(final XMLStreamReader streamReader, final Document document, final Element fragmentRoot)
throws XMLStreamException {
Element startElement = readElementOnly(streamReader, document);
fragmentRoot.appendChild(startElement);
Element currentElement = startElement;
while (streamReader.hasNext()) {
int event = streamReader.next();
switch (event) {
case START_DOCUMENT:
continue;
case START_ELEMENT:
Element workingElement = readElementOnly(streamReader, document);
currentElement.appendChild(workingElement);
currentElement = workingElement;
continue;
case CHARACTERS:
currentElement.setTextContent(streamReader.getText());
continue;
case END_ELEMENT:
if (currentElement != startElement) {
currentElement = (Element) currentElement.getParentNode();
continue;
} else {
handleDocument(document, startElement);
fragmentRoot.removeChild(startElement);
startElement = null;
return;
}
}
}
}
// THIS FUNCTION DOES SOMETHING MEANINFUL
private static void handleDocument(Document document, Element startElement) {
System.out.println(stringify(document));
}
private static String stringify(Document document) {
try {
Transformer transformer = TransformerFactory.newInstance().newTransformer();
transformer.setOutputProperty(OutputKeys.INDENT, "yes");
StreamResult result = new StreamResult(new StringWriter());
DOMSource source = new DOMSource(document);
transformer.transform(source, result);
String xmlString = result.getWriter().toString();
return xmlString;
} catch (Exception e) {
throw new RuntimeException(e);
}
}
}
编辑：我犯了一个非常愚蠢的错误。它现在已经解决了。它的工作但不完美 - 应足以引导您朝着有用的方向发展。
答案 2 :(得分：1)
考虑使用以下形式的XSLT 3.0流式转换：
<xsl:template name="main">
<xsl:stream href="bigInput.xml">
<xsl:for-each select="copy-of(/records/REC)">
<!-- process one record -->
</xsl:for-each>
</xsl:stream>
</xsl:template>
您可以使用Saxon-EE 9.6进行处理。
“进程一记录”逻辑可以使用Saxon SQL扩展，或者它可以调用扩展函数：上下文节点将是一个REC元素及其包含的树，可在子树中完全导航，但无法导航在当前正在处理的REC元素之外。