yacy_search_server/source/de/anomic/plasma/plasmaHTCache.java
orbiter 3d4b826ca5 migration of all databases that use the deprecated BLOBTree format into the BLOBHeap format. Old databases are migrated automatically.
This removes the last very IO-intensive data structures which were still used for Wiki, Blog and Bookmarks. Old database files will still remain in the DATA subdirectory but can be deleted manually if no major bugs appear during migration. There is no need for any user action, all migration is done automatically.

git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@5986 6c8d7289-2bf4-0310-a012-ef5d649a1542
2009-05-27 15:04:04 +00:00

297 lines
10 KiB
Java

// plasmaHTCache.java
// -----------------------
// part of YaCy
// (C) by Michael Peter Christen; mc@yacy.net
// first published on http://www.anomic.de
// Frankfurt, Germany, 2004
//
// $LastChangedDate$
// $LastChangedRevision$
// $LastChangedBy$
//
// This program is free software; you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation; either version 2 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program; if not, write to the Free Software
// Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
/*
Class documentation:
This class has two purposes:
1. provide a object that carries path and header information
that shall be used as objects within a scheduler's stack
2. static methods for a cache control and cache aging
the class shall also be used to do a cache-cleaning and index creation
*/
package de.anomic.plasma;
import java.io.ByteArrayInputStream;
import java.io.File;
import java.io.IOException;
import java.io.InputStream;
import java.util.HashMap;
import java.util.Map;
import de.anomic.http.httpResponseHeader;
import de.anomic.kelondro.blob.BLOBArray;
import de.anomic.kelondro.blob.BLOBCompressor;
import de.anomic.kelondro.blob.BLOBHeap;
import de.anomic.kelondro.blob.MapView;
import de.anomic.kelondro.order.Base64Order;
import de.anomic.kelondro.util.Log;
import de.anomic.kelondro.util.FileUtils;
import de.anomic.plasma.parser.Document;
import de.anomic.yacy.yacySeedDB;
import de.anomic.yacy.yacyURL;
public final class plasmaHTCache {
public static final String RESPONSE_HEADER_DB_NAME = "responseHeader.heap";
public static final String FILE_DB_NAME = "file.array";
public static final long oneday = 1000L * 60L * 60L * 24L; // milliseconds of a day
private static MapView responseHeaderDB = null;
private static BLOBCompressor fileDB = null;
private static BLOBArray fileDBunbuffered = null;
public static long maxCacheSize = 0l;
public static File cachePath = null;
public static String prefix;
public static final Log log = new Log("HTCACHE");
// URL attributes
public static final int UA_LOCAL = 0; // URL was crawled locally
public static final int UA_TILDE = 1; // tilde appears in URL
public static final int UA_REDIRECT = 2; // The URL is a redirection
// local flag attributes
public static final char LT_LOCAL = 'L';
public static final char LT_GLOBAL = 'G';
public static void init(final File htCachePath, String peerSalt, final long CacheSizeMax) {
cachePath = htCachePath;
maxCacheSize = CacheSizeMax;
prefix = peerSalt;
// reset old HTCache ?
String[] list = cachePath.list();
if (list != null) {
File object;
for (int i = list.length - 1; i >= 0; i--) {
object = new File(cachePath, list[i]);
if (object.getName().equals("http") ||
object.getName().equals("yacy") ||
object.getName().equals("https") ||
object.getName().equals("ftp")) {
FileUtils.deletedelete(cachePath);
}
}
}
// set/make cache path
if (!htCachePath.exists()) {
htCachePath.mkdirs();
}
if (!htCachePath.isDirectory()) {
// if the cache does not exists or is a file and not a directory, panic
log.logSevere("the cache path " + htCachePath.toString() + " is not a directory or does not exists and cannot be created");
System.exit(0);
}
// open the response header database
openDB();
}
static void resetDB() {
log.logFine("reset responseHeader DB with "+ responseHeaderDB.size() +" entries");
if (responseHeaderDB != null) responseHeaderDB.close();
final File dbfile = new File(cachePath, RESPONSE_HEADER_DB_NAME);
if (dbfile.exists()) FileUtils.deletedelete(dbfile);
openDB();
}
private static void openDB() {
// open the response header database
final File dbfile = new File(cachePath, RESPONSE_HEADER_DB_NAME);
BLOBHeap blob = null;
try {
blob = new BLOBHeap(dbfile, yacySeedDB.commonHashLength, Base64Order.enhancedCoder, 1024 * 1024);
} catch (final IOException e) {
e.printStackTrace();
}
responseHeaderDB = new MapView(blob, 500, '_');
try {
fileDBunbuffered = new BLOBArray(new File(cachePath, FILE_DB_NAME), prefix, 12, Base64Order.enhancedCoder, 1024 * 1024 * 2);
fileDBunbuffered.setMaxSize(maxCacheSize);
fileDB = new BLOBCompressor(fileDBunbuffered, 2 * 1024 * 1024);
} catch (IOException e) {
e.printStackTrace();
}
}
public static int responseHeaderDBSize() {
return responseHeaderDB.size();
}
public static long fileDBSize() {
return fileDB.length();
}
/**
* This method changes the HTCache size.<br>
* @param the new cache size in bytes
*/
public static void setCacheSize(final long newCacheSize) {
maxCacheSize = newCacheSize;
fileDBunbuffered.setMaxSize(maxCacheSize);
}
/**
* This method returns the free HTCache size.<br>
* @return the cache size in bytes
*/
public static long getFreeSize() {
long l = fileDB.length();
return (l >= maxCacheSize) ? 0 : maxCacheSize - l;
}
public static void close() {
responseHeaderDB.close();
fileDB.close(true);
}
public static boolean isPicture(final String mimeType) {
if (mimeType == null) return false;
return mimeType.toUpperCase().startsWith("IMAGE");
}
public static boolean isText(final String mimeType) {
return plasmaParser.supportedMimeTypesContains(mimeType);
}
public static boolean noIndexingURL(final yacyURL url) {
if (url == null) return false;
String urlString = url.toString().toLowerCase();
//http://www.yacy.net/getimage.php?image.png
int idx = urlString.indexOf("?");
if (idx > 0) urlString = urlString.substring(0,idx);
//http://www.yacy.net/getimage.php
idx = urlString.lastIndexOf(".");
if (idx > 0) urlString = urlString.substring(idx+1);
//php
return plasmaParser.mediaExtContains(urlString);
}
// Store to Cache
public static void storeMetadata(
final httpResponseHeader responseHeader,
Document metadata
) {
if (responseHeader != null) try {
// store the response header into the header database
final HashMap<String, String> hm = new HashMap<String, String>();
hm.putAll(responseHeader);
hm.put("@@URL", metadata.url().toNormalform(false, false));
hm.put("@@DEPTH", Integer.toString(metadata.depth()));
responseHeaderDB.put(metadata.urlHash(), hm);
} catch (final Exception e) {
log.logWarning("could not write ResourceInfo: "
+ e.getClass() + ": " + e.getMessage());
resetDB();
}
}
public static void storeFile(yacyURL url, byte[] file) {
try {
fileDB.put(url.hash().getBytes("UTF-8"), file);
} catch (IOException e) {
e.printStackTrace();
}
}
/**
* Returns an object containing metadata about a cached resource
* @param url the {@link URL} of the resource
* @return an {@link IResourceInfo info object}
* @throws <b>IllegalAccessException</b> if the {@link SecurityManager} doesn't allow instantiation
* of the info object with the given protocol
* @throws <b>UnsupportedProtocolException</b> if the protocol is not supported and therefore the
* info object couldn't be created
*/
public static httpResponseHeader loadResponseHeader(final yacyURL url) throws IllegalAccessException {
// loading data from database
Map<String, String> hdb;
try {
hdb = responseHeaderDB.get(url.hash());
} catch (final IOException e) {
return null;
}
if (hdb == null) return null;
return new httpResponseHeader(null, hdb);
}
/**
* Returns the content of a cached resource as {@link InputStream}
* @param url the requested resource
* @return the resource content as {@link InputStream}. In no data
* is available or the cached file is not readable, <code>null</code>
* is returned.
*/
public static InputStream getResourceContentStream(final yacyURL url) {
// load the url as resource from the cache
byte[] b = getResourceContent(url);
if (b == null) return null;
return new ByteArrayInputStream(b);
}
public static byte[] getResourceContent(final yacyURL url) {
// load the url as resource from the cache
try {
return fileDB.get(url.hash().getBytes("UTF-8"));
} catch (IOException e) {
e.printStackTrace();
return null;
}
}
public static long getResourceContentLength(final yacyURL url) {
// load the url as resource from the cache
try {
return fileDB.length(url.hash().getBytes("UTF-8"));
} catch (IOException e) {
e.printStackTrace();
return -1;
}
}
public static void deleteFromCache(yacyURL url) throws IOException {
responseHeaderDB.remove(url.hash());
fileDB.remove(url.hash().getBytes("UTF-8"));
}
}