com.day.crx.statistics.loader.AccessLogLoader Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of aem-sdk-api Show documentation
Show all versions of aem-sdk-api Show documentation
The Adobe Experience Manager SDK
/*************************************************************************
*
* ADOBE CONFIDENTIAL
* ___________________
*
* Copyright 1997 Adobe Systems Incorporated
* All Rights Reserved.
*
* NOTICE: All information contained herein is, and remains
* the property of Adobe Systems Incorporated and its suppliers,
* if any. The intellectual and technical concepts contained
* herein are proprietary to Adobe Systems Incorporated and its
* suppliers and are protected by trade secret or copyright law.
* Dissemination of this information or reproduction of this material
* is strictly forbidden unless prior written permission is obtained
* from Adobe Systems Incorporated.
**************************************************************************/
package com.day.crx.statistics.loader;
import com.day.crx.statistics.Entry;
import com.day.crx.statistics.query.Query;
import com.day.crx.statistics.result.ResultSelected;
import java.io.InputStream;
import java.io.BufferedReader;
import java.io.InputStreamReader;
import java.io.IOException;
import java.net.URI;
import java.net.URISyntaxException;
import java.util.StringTokenizer;
import java.util.Locale;
import java.text.ParseException;
import java.text.SimpleDateFormat;
/**
* AccessLogLoader reads a CQSE access.log.
*
* @author mreutegg
*/
public class AccessLogLoader {
/**
* Date format as used in the CQSE access.log
*/
private final SimpleDateFormat dateFormat = new SimpleDateFormat("dd/MMM/yyyy:HH:mm:ss Z", Locale.US);
/**
* Optional progress listener.
*/
private ProgressListener listener;
/**
* Query location. Where extracted queries from the access.log are written.
*/
private final String ql;
/**
* Result location. Where extracted results from the access.log are written.
*/
private final String rl;
/**
* Query path prefix. URLs with this path prefix are considered searches.
*/
private final String qpp;
/**
* Query parameter name. The name of the query parameter.
*/
private final String qpn;
/**
* Optional result path prefix. Only result URLs with this prefix are
* considered.
*/
private final String rpp;
/**
* Creates a new AccessLogLoader.
*
* @param ql the query location.
* @param rl the result location.
* @param qpp the query path prefix.
* @param qpn the query parameter name.
* @param rpp the result path prefix.
*/
public AccessLogLoader(String ql, String rl, String qpp,
String qpn, String rpp) {
this.ql = ql;
this.rl = rl;
this.qpp = qpp;
this.qpn = qpn;
this.rpp = rpp;
}
/**
* Processes an access.log and notifies an attached listeners about entries
* read.
*
* @param in the input stream where to read from the access.log.
* @return an integer array of length 2, where the first value indicates the
* number of queries and the second the number of results loaded.
* @throws IOException if an error occurs while reading from the
* access.log.
*/
public int[] processFile(InputStream in) throws IOException {
BufferedReader reader = new BufferedReader(new InputStreamReader(in, "ISO-8859-1"));
try {
String line;
int lineCount = 0;
int queryCount = 0;
int resultCount = 0;
while ((line = reader.readLine()) != null) {
Entry entry = null;
lineCount++;
try {
Access a = Access.fromString(line);
URI uri = new URI(a.requestURI);
URI referer = new URI(a.referer);
if (referer.getPath() != null && referer.getPath().startsWith(qpp)) {
// result selected
long time = dateFormat.parse(a.dateTime + " " + a.timeZone).getTime();
String path = uri.getPath();
// check result path prefix if any
if (rpp != null && path.startsWith(rpp)) {
int extIdx = path.lastIndexOf('.');
if (extIdx != -1) {
path = path.substring(0, extIdx).trim();
}
String query = getQuery(referer.getQuery());
if (query != null) {
entry = new ResultSelected(rl, path, 1, query);
entry.setTimestamp(time);
resultCount++;
}
}
} else if (uri.getPath().startsWith(qpp)) {
String query = getQuery(uri.getQuery());
if (query != null) {
long time = dateFormat.parse(a.dateTime + " " + a.timeZone).getTime();
entry = new Query(ql, query, 0, 0);
entry.setTimestamp(time);
queryCount++;
}
}
} catch (IllegalArgumentException e) {
// malformed
} catch (ParseException e) {
// malformed
} catch (URISyntaxException e) {
// malformed
}
notifyListener(lineCount, queryCount, resultCount, entry);
}
return new int[]{queryCount, resultCount};
} finally {
reader.close();
}
}
/**
* Notifies the progress listener if one is set.
*
* @param lineCount the number of lines parsed so far.
* @param queryCount the number of queries loaded so far.
* @param resultCount the number of results loaded so far.
* @param entry the entry read from the log or null if
* the line did not create an entry.
*/
private void notifyListener(int lineCount,
int queryCount,
int resultCount,
Entry entry) {
if (listener != null) {
listener.lineProcessed(lineCount, queryCount, resultCount, entry);
}
}
/**
* Extracts the query string from the URI query part.
*
* @param uriQueryPart the URI query part.
* @return the query string as typed in by the user.
*/
private String getQuery(String uriQueryPart) {
if (uriQueryPart == null) {
return null;
}
StringTokenizer tokenizer = new StringTokenizer(uriQueryPart, "&");
while (tokenizer.hasMoreTokens()) {
String param = tokenizer.nextToken();
if (param.startsWith(qpn + "=")) {
String query = param.substring(qpn.length() + 1);
if (query.length() > 0) {
query = query.replaceAll("\\+", " ").trim();
return query;
}
}
}
return null;
}
/**
* Sets a progress listener.
*
* @param listener the progress listener.
*/
public void setProgressListener(ProgressListener listener) {
this.listener = listener;
}
/**
* Simple interface to get notifications about the progress of reading the
* access.log file.
*/
public interface ProgressListener {
/**
* This method is called whenever a line from the access.log is
* processed.
*
* @param line the number of lines parsed so far.
* @param queryCount the number of queries loaded so far.
* @param resultCount the number of results loaded so far.
* @param entry the entry read from the log or null
* if the line did not create an entry.
*/
public void lineProcessed(int line, int queryCount,
int resultCount, Entry entry);
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy