yacy_search_server/source/de/anomic/kelondro/blob/BLOBArray.java

758 lines
28 KiB
Java
Raw Normal View History

// kelondroBLOBArray.java
// (C) 2008 by Michael Peter Christen; mc@yacy.net, Frankfurt a. M., Germany
// first published 19.08.2008 on http://yacy.net
//
// This is a part of YaCy, a peer-to-peer based web search engine
//
// $LastChangedDate: 2006-04-02 22:40:07 +0200 (So, 02 Apr 2006) $
// $LastChangedRevision: 1986 $
// $LastChangedBy: orbiter $
//
// LICENSE
//
// 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
package de.anomic.kelondro.blob;
import java.io.File;
import java.io.IOException;
import java.text.ParseException;
import java.util.ArrayList;
import java.util.Date;
import java.util.HashSet;
import java.util.Iterator;
import java.util.List;
import java.util.TreeMap;
import java.util.concurrent.CopyOnWriteArrayList;
import de.anomic.kelondro.index.Row;
import de.anomic.kelondro.order.ByteOrder;
import de.anomic.kelondro.order.CloneableIterator;
import de.anomic.kelondro.order.NaturalOrder;
import de.anomic.kelondro.order.MergeIterator;
import de.anomic.kelondro.text.Reference;
import de.anomic.kelondro.text.ReferenceContainer;
import de.anomic.kelondro.text.ReferenceFactory;
import de.anomic.kelondro.text.ReferenceContainerCache.blobFileEntries;
import de.anomic.kelondro.util.DateFormatter;
import de.anomic.kelondro.util.FileUtils;
import de.anomic.kelondro.util.Log;
public class BLOBArray implements BLOB {
/*
* This class implements a BLOB using a set of kelondroBLOBHeap objects
* In addition to a kelondroBLOBHeap this BLOB can delete large amounts of data using a given time limit.
* This is realized by creating separate BLOB files. New Files are created when either
* - a given time limit is reached
* - a given space limit is reached
* To organize such an array of BLOB files, the following file name structure is used:
* <BLOB-Name>/<YYYYMMDDhhmm>.blob
* That means all BLOB files are inside a directory that has the name of the BLOBArray.
* To delete content that is out-dated, one special method is implemented that deletes content by a given
* time-out. Deletions are not made automatically, they must be triggered using this method.
*/
public static final long oneMonth = 1000L * 60L * 60L * 24L * 365L / 12L;
private int keylength;
private ByteOrder ordering;
private File heapLocation;
private long fileAgeLimit;
private long fileSizeLimit;
private long repositoryAgeMax;
private long repositorySizeMax;
private List<blobItem> blobs;
private String prefix;
private int buffersize;
public BLOBArray(
final File heapLocation,
final String prefix,
final int keylength,
final ByteOrder ordering,
final int buffersize) throws IOException {
this.keylength = keylength;
this.prefix = prefix;
this.ordering = ordering;
this.buffersize = buffersize;
this.heapLocation = heapLocation;
this.fileAgeLimit = oneMonth;
this.fileSizeLimit = (long) Integer.MAX_VALUE;
this.repositoryAgeMax = Long.MAX_VALUE;
this.repositorySizeMax = Long.MAX_VALUE;
// check existence of the heap directory
if (heapLocation.exists()) {
if (!heapLocation.isDirectory()) throw new IOException("the BLOBArray directory " + heapLocation.toString() + " does not exist (is blocked by a file with same name)");
} else {
if(!heapLocation.mkdirs()) throw new IOException("the BLOBArray directory " + heapLocation.toString() + " does not exist (can not be created)");
}
// register all blob files inside this directory
String[] files = heapLocation.list();
HashSet<String> fh = new HashSet<String>();
for (int i = 0; i < files.length; i++) fh.add(files[i]);
// delete unused temporary files
boolean deletions = false;
for (int i = 0; i < files.length; i++) {
if (files[i].endsWith(".tmp")) {
FileUtils.deletedelete(new File(heapLocation, files[i]));
deletions = true;
}
if (files[i].endsWith(".idx") || files[i].endsWith(".gap")) {
String s = files[i].substring(0, files[i].length() - 17);
if (!fh.contains(s)) {
FileUtils.deletedelete(new File(heapLocation, files[i]));
deletions = true;
}
}
}
if (deletions) files = heapLocation.list(); // make a fresh list
// migrate old file names
Date d;
long time;
deletions = false;
for (int i = 0; i < files.length; i++) {
if (files[i].length() >= 19 && files[i].endsWith(".blob")) {
try {
d = DateFormatter.parseShortSecond(files[i].substring(0, 14));
new File(heapLocation, files[i]).renameTo(newBLOB(d));
deletions = true;
} catch (ParseException e) {continue;}
}
}
if (deletions) files = heapLocation.list(); // make a fresh list
// find maximum time: the file with this time will be given a write buffer
TreeMap<Long, blobItem> sortedItems = new TreeMap<Long, blobItem>();
BLOB oneBlob;
File f;
long maxtime = 0;
for (int i = 0; i < files.length; i++) {
if (files[i].length() >= 22 && files[i].startsWith(prefix) && files[i].endsWith(".blob")) {
try {
d = DateFormatter.parseShortMilliSecond(files[i].substring(prefix.length() + 1, prefix.length() + 18));
time = d.getTime();
if (time > maxtime) maxtime = time;
} catch (ParseException e) {continue;}
}
}
// open all blob files
for (int i = 0; i < files.length; i++) {
if (files[i].length() >= 22 && files[i].startsWith(prefix) && files[i].endsWith(".blob")) {
try {
d = DateFormatter.parseShortMilliSecond(files[i].substring(prefix.length() + 1, prefix.length() + 18));
f = new File(heapLocation, files[i]);
time = d.getTime();
oneBlob = (time == maxtime) ? new BLOBHeap(f, keylength, ordering, buffersize) : new BLOBHeapModifier(f, keylength, ordering);
sortedItems.put(Long.valueOf(time), new blobItem(d, f, oneBlob));
} catch (ParseException e) {continue;}
}
}
// read the blob tree in a sorted way and write them into an array
blobs = new CopyOnWriteArrayList<blobItem>();
for (blobItem bi : sortedItems.values()) {
blobs.add(bi);
}
}
/**
* add a blob file to the array.
* note that this file must be generated with a file name from newBLOB()
* @param location
* @throws IOException
*/
public synchronized void mountBLOB(File location, boolean full) throws IOException {
Date d;
try {
d = DateFormatter.parseShortMilliSecond(location.getName().substring(prefix.length() + 1, prefix.length() + 18));
} catch (ParseException e) {
throw new IOException("date parse problem with file " + location.toString() + ": " + e.getMessage());
}
BLOB oneBlob = (full && buffersize > 0) ? new BLOBHeap(location, keylength, ordering, buffersize) : new BLOBHeapModifier(location, keylength, ordering);
blobs.add(new blobItem(d, location, oneBlob));
}
public synchronized void unmountBLOB(File location, boolean writeIDX) {
blobItem b;
for (int i = 0; i < this.blobs.size(); i++) {
b = this.blobs.get(i);
if (b.location.getAbsolutePath().equals(location.getAbsolutePath())) {
this.blobs.remove(i);
b.blob.close(writeIDX);
b.blob = null;
b.location = null;
return;
}
}
Log.logSevere("BLOBArray", "file " + location + " cannot be unmounted. The file " + ((location.exists()) ? "exists." : "does not exist."));
}
private File unmount(int idx) {
blobItem b = this.blobs.remove(idx);
b.blob.close(false);
b.blob = null;
File f = b.location;
b.location = null;
return f;
}
public synchronized File[] unmountBestMatch(double maxq, long maxResultSize) {
if (this.blobs.size() < 2) return null;
long l, r;
File lf, rf;
double min = Double.MAX_VALUE;
File[] bestMatch = new File[2];
maxResultSize = maxResultSize >> 1;
for (int i = 0; i < this.blobs.size() - 1; i++) {
for (int j = i + 1; j < this.blobs.size(); j++) {
lf = this.blobs.get(i).location;
rf = this.blobs.get(j).location;
l = 1 + (lf.length() >> 1);
r = 1 + (rf.length() >> 1);
if (l + r > maxResultSize) continue;
double q = Math.max((double) l, (double) r) / Math.min((double) l, (double) r);
if (q < min) {
min = q;
bestMatch[0] = lf;
bestMatch[1] = rf;
}
}
}
if (min > maxq) return null;
unmountBLOB(bestMatch[1], false);
unmountBLOB(bestMatch[0], false);
return bestMatch;
}
public synchronized File[] unmountSmallest(long maxResultSize) {
if (this.blobs.size() < 2) return null;
File f0 = smallestBLOB(null, maxResultSize);
if (f0 == null) return null;
File f1 = smallestBLOB(f0, maxResultSize - f0.length());
if (f1 == null) return null;
unmountBLOB(f0, false);
unmountBLOB(f1, false);
return new File[]{f0, f1};
}
public synchronized File unmountSmallestBLOB(long maxResultSize) {
return smallestBLOB(null, maxResultSize);
}
public synchronized File smallestBLOB(File excluding, long maxsize) {
if (this.blobs.size() == 0) return null;
File bestFile = null;
long smallest = Long.MAX_VALUE;
File f = null;
for (int i = 0; i < this.blobs.size(); i++) {
f = this.blobs.get(i).location;
if (excluding != null && f.getAbsolutePath().equals(excluding.getAbsolutePath())) continue;
if (f.length() < smallest) {
smallest = f.length();
bestFile = f;
}
}
if (smallest > maxsize) return null;
return bestFile;
}
public synchronized File unmountOldestBLOB(boolean smallestFromFirst2) {
if (this.blobs.size() == 0) return null;
int idx = 0;
if (smallestFromFirst2 && this.blobs.get(1).location.length() < this.blobs.get(0).location.length()) idx = 1;
return unmount(idx);
}
public synchronized File unmountSimilarSizeBLOB(long otherSize) {
if (this.blobs.size() == 0 || otherSize == 0) return null;
blobItem b;
double delta, bestDelta = Double.MAX_VALUE;
int bestIndex = -1;
for (int i = 0; i < this.blobs.size(); i++) {
b = this.blobs.get(i);
if (b.location.length() == 0) continue;
delta = ((double) b.location.length()) / ((double) otherSize);
if (delta < 1.0) delta = 1.0 / delta;
if (delta < bestDelta) {
bestDelta = delta;
bestIndex = i;
}
}
return unmount(bestIndex);
}
/**
* return the number of BLOB files in this array
* @return
*/
public synchronized int entries() {
return (this.blobs == null) ? 0 : this.blobs.size();
}
/**
* generate a new BLOB file name with a given date.
* This method is needed to generate a file name that matches to the name structure that is needed for parts of the array
* @param creation
* @return
*/
public synchronized File newBLOB(Date creation) {
//return new File(heapLocation, DateFormatter.formatShortSecond(creation) + "." + blobSalt + ".blob");
return new File(heapLocation, prefix + "." + DateFormatter.formatShortMilliSecond(creation) + ".blob");
}
public String name() {
return this.heapLocation.getName();
}
public void setMaxAge(long maxAge) {
this.repositoryAgeMax = maxAge;
this.fileAgeLimit = Math.min(oneMonth, maxAge / 10);
}
public void setMaxSize(long maxSize) {
this.repositorySizeMax = maxSize;
this.fileSizeLimit = Math.min((long) Integer.MAX_VALUE, maxSize / 10L);
}
private void executeLimits() {
// check if storage limits are reached and execute consequences
if (blobs.size() == 0) return;
// age limit:
while (blobs.size() > 0 && System.currentTimeMillis() - blobs.get(0).creation.getTime() - this.fileAgeLimit > this.repositoryAgeMax) {
// too old
blobItem oldestBLOB = blobs.remove(0);
oldestBLOB.blob.close(false);
oldestBLOB.blob = null;
FileUtils.deletedelete(oldestBLOB.location);
}
// size limit
while (blobs.size() > 0 && length() > this.repositorySizeMax) {
// too large
blobItem oldestBLOB = blobs.remove(0);
oldestBLOB.blob.close(false);
FileUtils.deletedelete(oldestBLOB.location);
}
}
/*
* return the size of the repository (in bytes)
*/
public synchronized long length() {
long s = 0;
for (int i = 0; i < blobs.size(); i++) s += blobs.get(i).location.length();
return s;
}
public ByteOrder ordering() {
return this.ordering;
}
private class blobItem {
Date creation;
File location;
BLOB blob;
public blobItem(Date creation, File location, BLOB blob) {
this.creation = creation;
this.location = location;
this.blob = blob;
}
public blobItem(int buffer) throws IOException {
// make a new blob file and assign it in this item
this.creation = new Date();
this.location = newBLOB(this.creation);
this.blob = (buffer == 0) ? new BLOBHeapModifier(location, keylength, ordering) : new BLOBHeap(location, keylength, ordering, buffer);
}
}
/**
* ask for the length of the primary key
* @return the length of the key
*/
public int keylength() {
return this.keylength;
}
/**
* clears the content of the database
* @throws IOException
*/
public synchronized void clear() throws IOException {
for (blobItem bi: blobs) {
bi.blob.clear();
bi.blob.close(false);
FileUtils.deletedelete(bi.location);
}
blobs.clear();
}
/**
* ask for the number of blob entries
* @return the number of entries in the table
*/
public synchronized int size() {
int s = 0;
for (blobItem bi: blobs) s += bi.blob.size();
return s;
}
/**
* iterator over all keys
* @param up
* @param rotating
* @return
* @throws IOException
*/
public synchronized CloneableIterator<byte[]> keys(boolean up, boolean rotating) throws IOException {
assert rotating = false;
final List<CloneableIterator<byte[]>> c = new ArrayList<CloneableIterator<byte[]>>(blobs.size());
final Iterator<blobItem> i = blobs.iterator();
while (i.hasNext()) {
c.add(i.next().blob.keys(up, rotating));
}
return MergeIterator.cascade(c, this.ordering, MergeIterator.simpleMerge, up);
}
/**
* iterate over all keys
* @param up
* @param firstKey
* @return
* @throws IOException
*/
public synchronized CloneableIterator<byte[]> keys(boolean up, byte[] firstKey) throws IOException {
final List<CloneableIterator<byte[]>> c = new ArrayList<CloneableIterator<byte[]>>(blobs.size());
final Iterator<blobItem> i = blobs.iterator();
while (i.hasNext()) {
c.add(i.next().blob.keys(up, firstKey));
}
return MergeIterator.cascade(c, this.ordering, MergeIterator.simpleMerge, up);
}
/**
* check if a specific key is in the database
* @param key the primary key
* @return
* @throws IOException
*/
public synchronized boolean has(byte[] key) {
for (blobItem bi: blobs) if (bi.blob.has(key)) return true;
return false;
}
/**
* retrieve the whole BLOB from the table
* @param key the primary key
* @return
* @throws IOException
*/
public synchronized byte[] get(byte[] key) throws IOException {
byte[] b;
for (blobItem bi: blobs) {
b = bi.blob.get(key);
if (b != null) return b;
}
return null;
}
/**
* get all BLOBs in the array.
* this is useful when it is not clear if an entry is unique in all BLOBs in this array.
* @param key
* @return
* @throws IOException
*/
public synchronized Iterable<byte[]> getAll(byte[] key) throws IOException {
/*
byte[] b;
ArrayList<byte[]> l = new ArrayList<byte[]>(blobs.size());
for (blobItem bi: blobs) {
b = bi.blob.get(key);
if (b != null) l.add(b);
}
return l;
*/
return new BlobValues(key);
}
public class BlobValues implements Iterator<byte[]>, Iterable<byte[]> {
private Iterator<blobItem> bii;
private byte[] next;
private byte[] key;
public BlobValues(byte[] key) {
this.bii = blobs.iterator();
this.key = key;
this.next = null;
next0();
}
private void next0() {
while (this.bii.hasNext()) {
BLOB b = this.bii.next().blob;
try {
this.next = b.get(key);
if (this.next != null) return;
} catch (IOException e) {
e.printStackTrace();
this.next = null;
return;
}
}
this.next = null;
}
public Iterator<byte[]> iterator() {
return this;
}
public boolean hasNext() {
return this.next != null;
}
public byte[] next() {
byte[] n = this.next;
next0();
return n;
}
public void remove() {
throw new UnsupportedOperationException("no remove in BlobValues");
}
}
/**
* retrieve the size of the BLOB
* @param key
* @return the size of the BLOB or -1 if the BLOB does not exist
* @throws IOException
*/
public synchronized long length(byte[] key) throws IOException {
long l;
for (blobItem bi: blobs) {
l = bi.blob.length(key);
if (l >= 0) return l;
}
return -1;
}
/**
* write a whole byte array as BLOB to the table
* @param key the primary key
* @param b
* @throws IOException
*/
public synchronized void put(byte[] key, byte[] b) throws IOException {
blobItem bi = (blobs.size() == 0) ? null : blobs.get(blobs.size() - 1);
if (bi == null)
System.out.println("bi == null");
else if (System.currentTimeMillis() - bi.creation.getTime() > this.fileAgeLimit)
System.out.println("System.currentTimeMillis() - bi.creation.getTime() > this.maxage");
else if (bi.location.length() > this.fileSizeLimit)
System.out.println("bi.location.length() > this.maxsize");
if ((bi == null) || (System.currentTimeMillis() - bi.creation.getTime() > this.fileAgeLimit) || (bi.location.length() > this.fileSizeLimit)) {
// add a new blob to the array
bi = new blobItem(buffersize);
blobs.add(bi);
}
assert bi.blob instanceof BLOBHeap;
bi.blob.put(key, b);
executeLimits();
}
/**
* replace a BLOB entry with another which must be smaller or same size
* @param key the primary key
* @throws IOException
*/
public synchronized int replace(byte[] key, Rewriter rewriter) throws IOException {
int d = 0;
for (blobItem bi: blobs) {
d += bi.blob.replace(key, rewriter);
}
return d;
}
/**
* remove a BLOB
* @param key the primary key
* @throws IOException
*/
public synchronized void remove(byte[] key) throws IOException {
for (blobItem bi: blobs) bi.blob.remove(key);
}
/**
* close the BLOB
*/
public synchronized void close(boolean writeIDX) {
for (blobItem bi: blobs) bi.blob.close(writeIDX);
blobs.clear();
blobs = null;
}
public File mergeMount(File f1, File f2, ReferenceFactory<? extends Reference> factory, Row payloadrow, File newFile, int writeBuffer) throws IOException {
Log.logInfo("BLOBArray", "merging " + f1.getName() + " with " + f2.getName());
File resultFile = mergeWorker(factory, this.keylength, this.ordering, f1, f2, payloadrow, newFile, writeBuffer);
if (resultFile == null) return null;
mountBLOB(resultFile, false);
Log.logInfo("BLOBArray", "merged " + f1.getName() + " with " + f2.getName() + " into " + resultFile);
return resultFile;
}
private static <ReferenceType extends Reference> File mergeWorker(ReferenceFactory<ReferenceType> factory, int keylength, ByteOrder order, File f1, File f2, Row payloadrow, File newFile, int writeBuffer) throws IOException {
// iterate both files and write a new one
CloneableIterator<ReferenceContainer<ReferenceType>> i1 = new blobFileEntries<ReferenceType>(f1, factory, payloadrow);
CloneableIterator<ReferenceContainer<ReferenceType>> i2 = new blobFileEntries<ReferenceType>(f2, factory, payloadrow);
if (!i1.hasNext()) {
if (i2.hasNext()) {
FileUtils.deletedelete(f1);
if (f2.renameTo(newFile)) return newFile;
return f2;
} else {
FileUtils.deletedelete(f1);
FileUtils.deletedelete(f2);
return null;
}
} else if (!i2.hasNext()) {
FileUtils.deletedelete(f2);
if (f1.renameTo(newFile)) return newFile;
return f1;
}
assert i1.hasNext();
assert i2.hasNext();
File tmpFile = new File(newFile.getParentFile(), newFile.getName() + ".tmp");
HeapWriter writer = new HeapWriter(tmpFile, newFile, keylength, order, writeBuffer);
merge(i1, i2, order, writer);
try {
writer.close(true);
// we don't need the old files any more
FileUtils.deletedelete(f1);
FileUtils.deletedelete(f2);
return newFile;
} catch (IOException e) {
FileUtils.deletedelete(tmpFile);
FileUtils.deletedelete(newFile);
e.printStackTrace();
return null;
}
}
private static <ReferenceType extends Reference> void merge(CloneableIterator<ReferenceContainer<ReferenceType>> i1, CloneableIterator<ReferenceContainer<ReferenceType>> i2, ByteOrder ordering, HeapWriter writer) throws IOException {
assert i1.hasNext();
assert i2.hasNext();
ReferenceContainer<ReferenceType> c1, c2, c1o, c2o;
c1 = i1.next();
c2 = i2.next();
int e;
while (true) {
assert c1 != null;
assert c2 != null;
e = ordering.compare(c1.getTermHash(), c2.getTermHash());
if (e < 0) {
writer.add(c1.getTermHash(), c1.exportCollection());
if (i1.hasNext()) {
c1o = c1;
c1 = i1.next();
assert ordering.compare(c1.getTermHash(), c1o.getTermHash()) > 0;
continue;
}
break;
}
if (e > 0) {
writer.add(c2.getTermHash(), c2.exportCollection());
if (i2.hasNext()) {
c2o = c2;
c2 = i2.next();
assert ordering.compare(c2.getTermHash(), c2o.getTermHash()) > 0;
continue;
}
break;
}
assert e == 0;
// merge the entries
writer.add(c1.getTermHash(), (c1.merge(c2)).exportCollection());
if (i1.hasNext() && i2.hasNext()) {
c1 = i1.next();
c2 = i2.next();
continue;
}
if (i1.hasNext()) c1 = i1.next();
if (i2.hasNext()) c2 = i2.next();
break;
}
// catch up remaining entries
assert !(i1.hasNext() && i2.hasNext());
while (i1.hasNext()) {
//System.out.println("FLUSH REMAINING 1: " + c1.getWordHash());
writer.add(c1.getTermHash(), c1.exportCollection());
if (i1.hasNext()) {
c1o = c1;
c1 = i1.next();
assert ordering.compare(c1.getTermHash(), c1o.getTermHash()) > 0;
continue;
}
break;
}
while (i2.hasNext()) {
//System.out.println("FLUSH REMAINING 2: " + c2.getWordHash());
writer.add(c2.getTermHash(), c2.exportCollection());
if (i2.hasNext()) {
c2o = c2;
c2 = i2.next();
assert ordering.compare(c2.getTermHash(), c2o.getTermHash()) > 0;
continue;
}
break;
}
// finished with writing
}
public static void main(final String[] args) {
final File f = new File("/Users/admin/blobarraytest");
try {
//f.delete();
final BLOBArray heap = new BLOBArray(f, "test", 12, NaturalOrder.naturalOrder, 512 * 1024);
heap.put("aaaaaaaaaaaa".getBytes(), "eins zwei drei".getBytes());
heap.put("aaaaaaaaaaab".getBytes(), "vier fuenf sechs".getBytes());
heap.put("aaaaaaaaaaac".getBytes(), "sieben acht neun".getBytes());
heap.put("aaaaaaaaaaad".getBytes(), "zehn elf zwoelf".getBytes());
// iterate over keys
Iterator<byte[]> i = heap.keys(true, false);
while (i.hasNext()) {
System.out.println("key_b: " + new String(i.next()));
}
heap.remove("aaaaaaaaaaab".getBytes());
heap.remove("aaaaaaaaaaac".getBytes());
heap.put("aaaaaaaaaaaX".getBytes(), "WXYZ".getBytes());
heap.close(true);
} catch (final IOException e) {
e.printStackTrace();
}
}
}