我面临的问题是,即使在XSLT转换之后,也要保留特殊字符。我的源文件XHTML文件包含几个特殊字符,如
、—
、’
;在XSLT转换时,这些字符将被忽略。
我尝试过各种不同的答案,比如这和这。
如果我手动将特殊字符的值更改为相应的Unicode表示,则这些字符将保留在输出中。
例如,将
转换为 
,它会在输出中产生一个空格。请参阅下面的一些样本文件:
源XHTML :
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xmlns:text="http://giraffe.wkle.com/text" xmlns:epub="http://www.idpf.org/2007/ops">
<body>
<div class="section" id="section_1">
<p id="para_1" class="para">Content of paragraph—1.</p>
<p id="para_2" class="para">Content of paragraph—2.</p>
</div>
</body>
</html>
XSL模板:
<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:xs="http://www.w3.org/2001/XMLSchema">
<xsl:output method="xml" indent="yes"/>
<xsl:template match="node()|@*" name="identity">
<xsl:copy>
<xsl:apply-templates select="node()|@*" />
</xsl:copy>
</xsl:template>
<xsl:template match="*[local-name()='p']/text()">
<xsl:copy-of select="."/>
</xsl:template>
</xsl:stylesheet>
预期输出:
<?xml version="1.0" encoding="UTF-8"?>
<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" xmlns:text="http://giraffe.wkle.com/text">
<body>
<div class="section" id="section_1">
<p class="para" id="para_1">Content of paragraph—1.</p>
<p class="para" id="para_2">Content of paragraph—2.</p>
</div>
</body>
</html>
实际输出:
<?xml version="1.0" encoding="UTF-8"?>
<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" xmlns:text="http://giraffe.wkle.com/text">
<body>
<div class="section" id="section_1">
<p class="para" id="para_1">Contentofparagraph1.</p>
<p class="para" id="para_2">Contentofparagraph2.</p>
</div>
</body>
</html>
限制:
请告诉我是否有任何方法可以使用Unicode值转换特殊字符并将它们保留在我的输出XML文档中。
更新:
我使用这段Java代码来调用转换:
public class XSLTUtil {
public static String processXHTML(String sourceFileName, String outputXhtml, String xslFilePath) throws ParserConfigurationException, SAXException, IOException, TransformerException {
DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
DocumentBuilder docbuilder = factory.newDocumentBuilder();
Document doc = docbuilder.parse(new FileInputStream(sourceFileName));
FileOutputStream fos = null;
FileInputStream fis = null;
try {
fos = new FileOutputStream(outputXhtml);
fis = new FileInputStream(xslFilePath);
TransformerFactory transformfactory = TransformerFactory.newInstance();
Templates xsl = transformfactory.newTemplates(new StreamSource(fis));
Transformer transformer = xsl.newTransformer();
transformer.transform(new DOMSource(doc.getDocumentElement()),new StreamResult(fos));
return outputXhtml;
} finally {
if(fos != null) {
fos.close();
}
if(fis != null) {
fis.close();
}
}
}
public static void main(String args[]){
String sourceFileName = "C:\\source.xhtml";
String outputXhtml = "C:\\output.xhtml";
String xslFilePath = "C:\\xslTemplate.xsl";
String result = "-1";
try {
result = processXHTML(sourceFileName, outputXhtml, xslFilePath);
} catch (ParserConfigurationException e) {
e.printStackTrace();
} catch (SAXException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
} catch (TransformerException e) {
e.printStackTrace();
}
System.out.println("Result : "+ result);
}
}
发布于 2017-09-14 14:57:51
使用Apache Lang3 Lang3 StringEscapUtils的解决方案。需要图书馆:
<dependency>
<groupId>org.apache.commons</groupId>
<artifactId>commons-lang3</artifactId>
<version>3.0</version>
</dependency>
首先阅读内容并将所有实体替换为真实文本。
public static String processXHTML(String sourceFileName, String outputXhtml,
String xslFilePath) throws ParserConfigurationException, SAXException, IOException,
TransformerException {
Charset charset = StandardCharsets.UTF_8;
Path path = Paths.get(sourceFileName);
String source = new String(Files.readAllBytes(path), charset);
source = source.replaceAll("\\&(amp|lt|gt|quot);", "\u0001$1;");
source = StringEscapeUtils.unescapeHtml4(source);
source = source.replace('\u0001', '&');
byte[] bytes = source.getBytes(charset);
ByteArrayInputStream bais = new ByteArrayInputStream(bytes);
DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
DocumentBuilder docbuilder = factory.newDocumentBuilder();
docbuilder.setEntityResolver(new EntityResolver() {
@Override
public InputSource resolveEntity(String publicId, String systemId)
throws SAXException, IOException {
System.out.printf("resolveEntity PUBLIC %s SYSTEM %s%n", publicId, systemId);
return new InputSource(new StringReader(""));
}
});
//Document doc = docbuilder.parse(new FileInputStream(sourceFileName));
Document doc = docbuilder.parse(bais);
我跳过了XML实体。
我用短线连接了EntityResolver,因为读取和处理实体(通过网络)需要很长时间。似乎您已经安装了类似的东西,因为实体没有被替换。
在resolveEntity中,返回null以查看哪些DTD是递归加载的。
或者,您可以安装一个HTML目录,这是HTML的本地缓存,其中有一些。
发布于 2017-09-13 16:20:44
对较早版本的问题的过时回答
考虑使用Andrew的Lexev实用程序。它基本上对XML进行预处理,将实体引用转换为将通过XML解析保留的内容,然后对转换结果进行后处理以将实体引用放回。
http://andrewjwelch.com/lexev/
发布于 2017-09-14 13:54:22
我会处理掉DOM代码。当您只想将DOM转换为其他东西时,创建DOM是笨拙和低效的。如果您提供一个StreamSource或SAXSource,并让它们决定自己的树表示形式,那么Saxon和Xalan都会运行得更快。与撒克逊,这可以是5-10倍的速度,将使用更少的内存。
我不知道DOM为什么会丢失实体引用。对于所使用的XSLT处理器,您提供了相互矛盾的信息,因此不容易进行研究,但是由于DOM数据模型不同于XSLT/XPath数据模型(特别是在处理实体扩展方面),因此应该解决这个问题。
https://stackoverflow.com/questions/46200610
复制相似问题