%@ page
session="false"
contentType="text/html; charset=UTF-8"
import="java.io.*"
import="java.util.*"
import="org.apache.nutch.searcher.*"
import="org.apache.nutch.parse.ParseData"
import="org.apache.nutch.metadata.Metadata"
import="org.apache.hadoop.conf.Configuration"
import="org.apache.nutch.util.NutchConfiguration"
%><%
Configuration nutchConf = NutchConfiguration.get(application);
NutchBean bean = NutchBean.get(application, nutchConf);
bean.LOG.info("cache request from " + request.getRemoteAddr());
Hit hit = new Hit(Integer.parseInt(request.getParameter("idx")),
Integer.parseInt(request.getParameter("id")));
HitDetails details = bean.getDetails(hit);
String id = "idx=" + hit.getIndexNo() + "&id=" + hit.getIndexDocNo();
String language =
ResourceBundle.getBundle("org.nutch.jsp.cached", request.getLocale())
.getLocale().getLanguage();
Metadata metaData = bean.getParseData(details).getContentMeta();
String content = null;
String contentType = (String) metaData.get(Metadata.CONTENT_TYPE);
if (contentType.startsWith("text/html")) {
// FIXME : it's better to emit the original 'byte' sequence
// with 'charset' set to the value of 'CharEncoding',
// but I don't know how to emit 'byte sequence' in JSP.
// out.getOutputStream().write(bean.getContent(details)) may work,
// but I'm not sure.
String encoding = (String) metaData.get("CharEncodingForConversion");
if (encoding != null) {
try {
content = new String(bean.getContent(details), encoding);
}
catch (UnsupportedEncodingException e) {
// fallback to windows-1252
content = new String(bean.getContent(details), "windows-1252");
}
}
else
content = new String(bean.getContent(details));
}
%>
<%
out.flush();
%>
<%@ taglib uri="http://jakarta.apache.org/taglibs/i18n" prefix="i18n" %>