/**
Copyright (C) SYSTAP, LLC DBA Blazegraph 2006-2016. All rights reserved.
Contact:
SYSTAP, LLC DBA Blazegraph
2501 Calvert ST NW #106
Washington, DC 20008
licenses@blazegraph.com
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; version 2 of the License.
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
*/
/*
* Created on Jun 6, 2011
*/
package com.bigdata.rdf.lexicon;
import java.util.Arrays;
import java.util.UUID;
import junit.framework.TestCase2;
import org.openrdf.model.vocabulary.XMLSchema;
import com.bigdata.btree.BTree;
import com.bigdata.btree.IndexMetadata;
import com.bigdata.btree.keys.IKeyBuilder;
import com.bigdata.btree.keys.KVO;
import com.bigdata.btree.keys.KeyBuilder;
import com.bigdata.io.SerializerUtil;
import com.bigdata.rawstore.IRawStore;
import com.bigdata.rawstore.SimpleMemoryRawStore;
import com.bigdata.rdf.internal.IV;
import com.bigdata.rdf.internal.IVUtility;
import com.bigdata.rdf.internal.VTE;
import com.bigdata.rdf.internal.impl.AbstractIV;
import com.bigdata.rdf.internal.impl.BlobIV;
import com.bigdata.rdf.lexicon.BlobsWriteTask.BlobsWriteProcResultHandler;
import com.bigdata.rdf.model.BigdataBNode;
import com.bigdata.rdf.model.BigdataLiteral;
import com.bigdata.rdf.model.BigdataURI;
import com.bigdata.rdf.model.BigdataValue;
import com.bigdata.rdf.model.BigdataValueFactory;
import com.bigdata.rdf.model.BigdataValueFactoryImpl;
import com.bigdata.rdf.util.DumpLexicon;
import com.bigdata.util.BytesUtil;
/**
* Test suite for low-level operations on the BLOBS index.
*
* @author thompsonbry
*/
public class TestBlobsIndex extends TestCase2 {
public TestBlobsIndex() {
}
public TestBlobsIndex(final String name) {
super(name);
}
/**
* Unit test for generation of sort keys from {@link BigdataValue}s to be
* represented as {@link BlobIV}s.
*/
public void test_generateSortKeys() {
final BlobsIndexHelper h = new BlobsIndexHelper();
final String namespace = getName();
final BigdataValueFactory vf = BigdataValueFactoryImpl
.getInstance(namespace);
/*
* Generate Values that we will use to read and write on the TERMS
* index.
*/
final BigdataValue[] values;
{
final BigdataURI uri1 = vf
.createURI("http://www.bigdata.com/testTerm");
final BigdataLiteral lit1 = vf.createLiteral("bigdata");
final BigdataLiteral lit2 = vf.createLiteral("bigdata", "en");
final BigdataLiteral lit3 = vf.createLiteral("bigdata",
XMLSchema.STRING);
final BigdataBNode bnode1 = vf.createBNode();
final BigdataBNode bnode2 = vf.createBNode("abc");
values = new BigdataValue[] {
uri1,
lit1,
lit2,
lit3,
bnode1,
bnode2
};
}
// Generate the sort keys.
final KVO<BigdataValue>[] a = h.generateKVOs(vf
.getValueSerializer(), values, values.length);
/*
* Verify that we can decode fully formed TermIVs based on these prefix
* keys.
*/
{
final IKeyBuilder keyBuilder = h.newKeyBuilder();
for (int i = 0; i < a.length; i++) {
final KVO<BigdataValue> kvo = a[i];
final byte[] baseKey = kvo.key;
final int counter = i;
// A fully formed key.
final byte[] key = h.makeKey(keyBuilder.reset(), baseKey,
counter);
// Wrap as a TermId,
final BlobIV<?> iv = (BlobIV<?>) IVUtility.decodeFromOffset(
key, 0/* offset */);
//new TermId(key);
// Verify 1st byte decodes to the VTE of the Value.
assertEquals(VTE.valueOf(kvo.obj), AbstractIV.getVTE(KeyBuilder
.decodeByte(key[0])));
// Verify the the VTE encoding is consistent in the other
// direction as well.
assertEquals(BlobIV.toFlags(VTE.valueOf(kvo.obj)), KeyBuilder
.decodeByte(key[0]));
// Verify VTE was correctly encoded.
assertEquals(VTE.valueOf(kvo.obj), iv.getVTE());
// Verify hash code was correctly encoded.
assertEquals(kvo.obj.hashCode(), iv.hashCode());
// Verify we can decode the String value of the TermIV.
assertEquals(iv, BlobIV.fromString(iv.toString()));
}
}
}
// /**
// * Unit test for creating the TERMS index.
// */
// public void test_termsIndex_create() {
//
// final IRawStore store = new SimpleMemoryRawStore();
//
// try {
//
// final String namespace = getName();
//
// final BTree ndx = createTermsIndex(store, namespace);
//
//// final TermsIndexHelper h = new TermsIndexHelper();
////
//// final IKeyBuilder keyBuilder = h.newKeyBuilder();
////
//// for (VTE vte : VTE.values()) {
////
//// // Each VTE has an associated NullIV (mapped to a [null]).
//// assertNull(ndx.lookup(BlobIV.mockIV(vte).encode(
//// keyBuilder.reset()).getKey()));
////
//// }
////
//// // Should be one entry for each type of NullIV.
//// assertEquals(4L, ndx.rangeCount());
////
//// // Verify we visit each of those NullIVs.
//// final ITupleIterator<BigdataValue> itr = ndx.rangeIterator();
////
//// while(itr.hasNext()) {
////
//// final ITuple<BigdataValue> tuple = itr.next();
////
//// assertTrue(tuple.isNull());
////
//// // The tuple is deserialized as a [null] reference.
//// assertNull(tuple.getObject());
////
//// }
//
// } finally {
//
// store.destroy();
//
// }
//
// }
/**
* Return the {@link IndexMetadata} for the TERMS index.
*
* @param name
* The name of the index.
*
* @return The {@link IndexMetadata}.
*/
static IndexMetadata getTermsIndexMetadata(final String namespace) {
final String name = namespace + ".TERMS";
final BigdataValueFactory valueFactory = BigdataValueFactoryImpl
.getInstance(namespace);
final IndexMetadata metadata = new IndexMetadata(name, UUID
.randomUUID());
// final int m = 1024;
// final int q = 8000;
// final int ratio = 32;
// final int maxRecLen = 0;
// metadata.setNodeKeySerializer(new FrontCodedRabaCoder(ratio));
// final DefaultTupleSerializer tupleSer = new DefaultTupleSerializer(
// new DefaultKeyBuilderFactory(new Properties()),//
// new FrontCodedRabaCoder(ratio),//
// CanonicalHuffmanRabaCoder.INSTANCE
// );
//
// metadata.setTupleSerializer(tupleSer);
// enable raw record support.
metadata.setRawRecords(true);
// // set the maximum length of a byte[] value in a leaf.
// metadata.setMaxRecLen(maxRecLen);
// /*
// * increase the branching factor since leaf size is smaller w/o large
// * records.
// */
// metadata.setBranchingFactor(m);
//
// // Note: You need to give sufficient heap for this option!
// metadata.setWriteRetentionQueueCapacity(q);
metadata.setTupleSerializer(new BlobsTupleSerializer(namespace,
valueFactory));
return metadata;
}
/**
* Create a TERMS index.
*
* @param namespace
* The namespace of the TERMS index (e.g., for "kb.lex" the fully
* qualified name of the index would be "kb.lex.TERMS").
*
* @return The terms index.
*/
static BTree createTermsIndex(final IRawStore store, final String namespace) {
final IndexMetadata metadata = getTermsIndexMetadata(namespace);
final BTree ndx = BTree.create(store, metadata);
// /*
// * Insert a tuple for each kind of VTE having a ZERO hash code and a
// * ZERO counter and thus qualifying it as a NullIV. Each of these tuples
// * is mapped to a null value in the index. This reserves the possible
// * distinct NullIV keys so they can not be assigned to real Values.
// *
// * Note: The hashCode of "" is ZERO, so an empty Literal would otherwise
// * be assigned the same key as mockIV(VTE.LITERAL).
// */
//
// final IKeyBuilder keyBuilder = new TermsIndexHelper().newKeyBuilder();
//
// final byte[][] keys = new byte[][] {
// BlobIV.mockIV(VTE.URI).encode(keyBuilder.reset()).getKey(), //
// BlobIV.mockIV(VTE.BNODE).encode(keyBuilder.reset()).getKey(), //
// BlobIV.mockIV(VTE.LITERAL).encode(keyBuilder.reset()).getKey(), //
// BlobIV.mockIV(VTE.STATEMENT).encode(keyBuilder.reset()).getKey(), //
// };
// final byte[][] vals = new byte[][] { null, null, null, null };
//
// // submit the task and wait for it to complete.
// ndx.submit(0/* fromIndex */, keys.length/* toIndex */, keys, vals,
// BatchInsertConstructor.RETURN_NO_VALUES, null/* aggregator */);
return ndx;
}
/**
* Unit test for lookup and adding values to the TERMS index when blank
* nodes are NOT stored in the TERMS index.
*/
public void test_termsIndex_addLookupValues_with_standard_bnode_semantics() {
doTermsIndexAddLookupTest(false/* toldBNodes */);
}
/**
* Unit test for lookup and adding values to the TERMS index when blank
* nodes are stored in the TERMS index (told bnodes semantics).
*/
public void test_termsIndex_addLookupValue_with_toldBNodesMode() {
doTermsIndexAddLookupTest(true/* toldBNodes */);
}
/**
* Test helper exercises the basic operations on the TERMS index, including
* (a) scanning a collision buckets to resolve {@link BigdataValue}s from
* their prefix keys; (b) adding a {@link BigdataValue}s to the TERMS index;
* and (c) point lookups using an {@link IV} as a fully qualified key for
* the TERMS index.
*
* @param toldBNodes
* when <code>true</code> blank nodes will be inserted into the
* TERMS index.
*/
private void doTermsIndexAddLookupTest(final boolean toldBNodes) {
final IRawStore store = new SimpleMemoryRawStore();
try {
final String namespace = getName();
final BTree ndx = createTermsIndex(store, namespace);
final BigdataValueFactory vf = BigdataValueFactoryImpl
.getInstance(namespace);
final BlobsIndexHelper h = new BlobsIndexHelper();
/*
* Generate Values that we will use to read and write on the TERMS
* index.
*/
final BigdataValue[] values;
{
final BigdataURI uri1 = vf
.createURI("http://www.bigdata.com/testTerm");
// Note: These three literals wind up with the same hash code.
// The hash code of the literal is based only on its label.
final BigdataLiteral lit1 = vf.createLiteral("bigdata");
final BigdataLiteral lit2 = vf.createLiteral("bigdata", "en");
final BigdataLiteral lit3 = vf.createLiteral("bigdata",
XMLSchema.STRING);
final BigdataBNode bnode1 = vf.createBNode();
final BigdataBNode bnode2 = vf.createBNode("abc");
values = new BigdataValue[] {
uri1,
lit1,
lit2,
lit3,
bnode1,
bnode2
};
}
final KVO<BigdataValue>[] a = h.generateKVOs(vf
.getValueSerializer(), values, values.length);
final byte[][] keys = new byte[a.length][];
final byte[][] vals = new byte[a.length][];
for (int i = 0; i < a.length; i++) {
keys[i] = a[i].key;
vals[i] = a[i].val;
}
// First, verify that the Value(s) were not found in the index.
{
final boolean readOnly = true;
final WriteTaskStats stats = new WriteTaskStats();
final BlobsWriteProc.BlobsWriteProcConstructor ctor = new BlobsWriteProc.BlobsWriteProcConstructor(
readOnly, toldBNodes);
ndx.submit(0/* fromIndex */, values.length/* toIndex */, keys,
vals, ctor, new BlobsWriteProcResultHandler(a,
readOnly, stats));
for (int i = 0; i < a.length; i++) {
// IV was not assigned (read-only and does not pre-exist).
assertNull(a[i].obj.getIV());
}
assertEquals(a.length, stats.nunknown.get());
}
// Now, verify the IVs are assigned on insert.
{
final boolean readOnly = false;
final WriteTaskStats stats = new WriteTaskStats();
final BlobsWriteProc.BlobsWriteProcConstructor ctor = new BlobsWriteProc.BlobsWriteProcConstructor(
readOnly, toldBNodes);
ndx.submit(0/* fromIndex */, values.length/* toIndex */, keys,
vals, ctor, new BlobsWriteProcResultHandler(a,
readOnly, stats));
// Note: [nunknown] is only set on read.
assertEquals(0, stats.nunknown.get());
/*
* Verify that the IV is a fully qualified key for the TERMS
* index.
*/
final IKeyBuilder keyBuilder = h.newKeyBuilder();
for (int i = 0; i < a.length; i++) {
final BigdataValue expected = a[i].obj;
final IV<?,?> iv = expected.getIV();
// An IV was assigned to the BigdataValue.
assertNotNull(iv);
// Verify the VTE is consistent.
assertEquals(VTE.valueOf(expected), iv.getVTE());
// Verify the hash code is consistent.
assertEquals(expected.hashCode(), iv.hashCode());
// Encode the IV as a key.
final byte[] key = iv.encode(keyBuilder.reset()).getKey();
// Verify can encode/decode the IV.
{
final IV<?,?> decodedIV = IVUtility.decode(key);
assertEquals(iv, decodedIV);
}
// Point lookup using the IV as the key.
final byte[] val = ndx.lookup(key);
// Verify point lookup succeeds.
if (val == null) {
fail("Could not resolve IV against index: expectedIV="
+ iv + ", key=" + BytesUtil.toString(key));
}
// Decode the returned byte[] as a Value.
final BigdataValue actual = vf.getValueSerializer()
.deserialize(val);
// Verify BigdataValues are equal()
if (!expected.equals(actual)) {
log.error(DumpLexicon.dumpBlobs(namespace, ndx));
fail("Expected=" + expected + "(" + iv + "), actual="
+ actual + "(" + actual.getIV() + ")");
}
if (log.isInfoEnabled())
log.info("i=" + expected + ", iv=" + iv);
}
}
// Finally, verify that the assigned IVs are discovered on lookup.
{
/*
* Setup an array of the expected IVs and clear out the old IVs
* on the BigdataValue objects.
*
* Note: Since we can not clear the IV once it has been set, this
* replaces the BigdataValues in the array with new values having
* the same data.
*/
final IV[] expected = new IV[a.length];
final BigdataValueFactory vf2 = BigdataValueFactoryImpl
.getInstance(namespace + "-not-the-same");
for (int i = 0; i < a.length; i++) {
final BigdataValue tmp = a[i].obj;
assertNotNull(a[i].obj.getIV()); // IV is known (from above)
expected[i] = a[i].obj.getIV(); // make a note of it.
final BigdataValue newVal = vf.asValue(vf2.asValue(tmp));
// replace entry in a[].
a[i] = new KVO<BigdataValue>(a[i].key, a[i].val, newVal);
assertEquals(tmp, a[i].obj);// same Value.
assertNull(a[i].obj.getIV()); // but IV is not set.
}
final boolean readOnly = true;
final WriteTaskStats stats = new WriteTaskStats();
final BlobsWriteProc.BlobsWriteProcConstructor ctor = new BlobsWriteProc.BlobsWriteProcConstructor(
readOnly, toldBNodes);
ndx.submit(0/* fromIndex */, values.length/* toIndex */, keys,
vals, ctor, new BlobsWriteProcResultHandler(a,
readOnly, stats));
int nnotfound = 0;
for (int i = 0; i < a.length; i++) {
final IV<?,?> expectedIV = expected[i];
final IV<?,?> actualIV = a[i].obj.getIV();
if(expectedIV.isBNode()) {
if (toldBNodes) {
// IV is discoverable.
assertNotNull(actualIV);
assertEquals(expected[i], actualIV);
} else {
// IV is NOT discoverable (can not unify bnodes).
if (actualIV != null)
fail("Not expecting to unify blank node: expectedIV="
+ expectedIV
+ ", but actualIV="
+ actualIV + "(should be null)");
nnotfound++;
}
} else {
// IV is discoverable.
assertNotNull(actualIV);
assertEquals(expected[i], actualIV);
}
}
assertEquals(nnotfound, stats.nunknown.get());
}
} finally {
store.destroy();
}
}
/**
* Unit test with standard blank nodes semantics verifies that separate
* writes on the TERMS index using the same BNode ID result in distinct keys
* being assigned (blank nodes do not unify).
*/
public void test_blank_nodes_are_distinct() {
final boolean storeBlankNodes = false;
final IRawStore store = new SimpleMemoryRawStore();
try {
final String namespace = getName();
final IndexMetadata metadata = getTermsIndexMetadata(namespace);
final BTree ndx = BTree.create(store, metadata);
final BigdataValueFactory vf = BigdataValueFactoryImpl
.getInstance(namespace);
final BlobsIndexHelper h = new BlobsIndexHelper();
// Write on the TERMS index, obtaining IVs for those BNodes.
final IV[] ivs1;
{
/*
* Generate Values that we will use to read and write on the
* TERMS index.
*/
final BigdataValue[] values;
{
final BigdataBNode bnode1 = vf.createBNode();
final BigdataBNode bnode2 = vf.createBNode("abc");
values = new BigdataValue[] { bnode1, bnode2 };
}
final KVO<BigdataValue>[] a = h.generateKVOs(vf
.getValueSerializer(), values, values.length);
final byte[][] keys = new byte[a.length][];
final byte[][] vals = new byte[a.length][];
for (int i = 0; i < a.length; i++) {
keys[i] = a[i].key;
vals[i] = a[i].val;
}
final boolean readOnly = false;
final WriteTaskStats stats = new WriteTaskStats();
final BlobsWriteProc.BlobsWriteProcConstructor ctor = new BlobsWriteProc.BlobsWriteProcConstructor(
readOnly, storeBlankNodes);
ndx.submit(0/* fromIndex */, values.length/* toIndex */, keys,
vals, ctor, new BlobsWriteProcResultHandler(a,
readOnly, stats));
// Copy out the assigned IVs.
ivs1 = new IV[a.length];
for (int i = 0; i < a.length; i++) {
final IV<?,?> iv = a[i].obj.getIV();
assertNotNull(iv);
ivs1[i] = iv;
}
}
// Write on the TERMS index, obtaining new IVs for those BNodes.
final IV[] ivs2;
{
/*
* Generate Values that we will use to read and write on the
* TERMS index (we need distinct instances since the IV once
* set can not be cleared from the BigdataValue).
*/
final BigdataValue[] values;
{
final BigdataBNode bnode1 = vf.createBNode();
final BigdataBNode bnode2 = vf.createBNode("abc");
values = new BigdataValue[] { bnode1, bnode2 };
}
final KVO<BigdataValue>[] a = h.generateKVOs(vf
.getValueSerializer(), values, values.length);
final byte[][] keys = new byte[a.length][];
final byte[][] vals = new byte[a.length][];
for (int i = 0; i < a.length; i++) {
keys[i] = a[i].key;
vals[i] = a[i].val;
}
final boolean readOnly = false;
final WriteTaskStats stats = new WriteTaskStats();
final BlobsWriteProc.BlobsWriteProcConstructor ctor = new BlobsWriteProc.BlobsWriteProcConstructor(
readOnly, storeBlankNodes);
ndx.submit(0/* fromIndex */, values.length/* toIndex */, keys,
vals, ctor, new BlobsWriteProcResultHandler(a,
readOnly, stats));
// Copy out the assigned IVs.
ivs2 = new IV[a.length];
for (int i = 0; i < a.length; i++) {
final IV<?,?> iv = a[i].obj.getIV();
assertNotNull(iv);
ivs2[i] = iv;
}
}
/*
* Verify that all assigned IVs are distinct and that all assigned
* IVs can be used to materialize the blank nodes that we wrote onto
* the TERMS index.
*/
{
final IKeyBuilder keyBuilder = h.newKeyBuilder();
// Same #of IVs.
assertEquals(ivs1.length, ivs2.length);
for (int i = 0; i < ivs1.length; i++) {
assertNotNull(ivs1[i]);
assertNotNull(ivs2[i]);
assertNotSame(ivs1[i], ivs2[i]);
assertNotNull(h.lookup(ndx, (BlobIV<?>)ivs1[i], keyBuilder));
assertNotNull(h.lookup(ndx, (BlobIV<?>)ivs2[i], keyBuilder));
// assertNotNull(ndx.lookup(ivs1[i]));
// assertNotNull(ndx.lookup(ivs2[i]));
}
}
} finally {
store.destroy();
}
}
/*
* FIXME These tests for hash collisions no longer work because the hash
* code is being computed from the original Value which makes the override
* of generateKeys() impossible. This makes it quite difficult to unit test
* the behavior which handles hash collisions since they are otherwise quite
* rare.
*
* One way to do this would be to factor out the hash function such that we
* could override it explicitly.
*/
// /**
// * Unit test with a no collisions.
// */
// public void test_noCollisions() {
//
// doHashCollisionTest(1);
//
// }
//
// /**
// * Unit test with a small number of collisions.
// */
// public void test_someCollisions() {
//
// doHashCollisionTest(12);
//
// }
//
// /**
// * Unit test with a the maximum number of collisions.
// */
// public void test_lotsOfCollisions() {
//
// doHashCollisionTest(255);
//
// }
//
// /**
// * Unit test with a too many collisions.
// */
// public void test_tooManyCollisions() {
//
// try {
// doHashCollisionTest(257);
// fail("Expecting: " + CollisionBucketSizeException.class);
// } catch (CollisionBucketSizeException ex) {
// if (log.isInfoEnabled())
// log.info("Ignoring expected exception: " + ex);
// }
//
// }
//
// /**
// * Test helper attempts to insert the given number of {@link BigdataValue}s
// * into the terms index. If the maximum collision bucket size is reached,
// * then the exception is thrown back to the caller.
// * <p>
// * Note: This test needs to be done using mock data since it would take a
// * huge number of values to have that many collisions otherwise.
// *
// * @param ncollisions
// * The number of collisions to manufacture.
// *
// * @throws CollisionBucketSizeException
// */
// private void doHashCollisionTest(final int ncollisions) {
//
// final IRawStore store = new SimpleMemoryRawStore();
//
// try {
//
// final TermsIndexHelper h = new TermsIndexHelper();
//
// final String namespace = getName();
//
// final IndexMetadata metadata = getTermsIndexMetadata(namespace);
//
// final BTree ndx = BTree.create(store, metadata);
//
// final BigdataValueFactory vf = BigdataValueFactoryImpl
// .getInstance(namespace);
//
// /*
// * Generate Values that we will use to read and write on the TERMS
// * index.
// */
// final BigdataValue[] values;
// {
//
// final ArrayList<BigdataValue> tmp = new ArrayList<BigdataValue>(
// ncollisions);
//
// for (int i = 0; i < ncollisions; i++) {
//
// final BigdataURI uri = vf
// .createURI("http://www.bigdata.com/testTerm/" + i);
//
// tmp.add(uri);
//
// }
//
// values = tmp.toArray(new BigdataValue[ncollisions]);
//
// }
//
// final int hashCode = 12;
// final KVO<BigdataValue>[] a = mockGenerateKVOs(vf
// .getValueSerializer(), values, values.length, hashCode);
//
// final byte[][] keys = new byte[a.length][];
// final byte[][] vals = new byte[a.length][];
// for (int i = 0; i < a.length; i++) {
// keys[i] = a[i].key;
// vals[i] = a[i].val;
// // Verify 1st byte decodes to the VTE of the Value.
// assertEquals(VTE.valueOf(a[i].obj), AbstractIV
// .getVTE(KeyBuilder.decodeByte(a[i].key[0])));
// }
//
// {
//
// final boolean readOnly = false;
// final boolean storeBlankNodes = false;
// final WriteTaskStats stats = new WriteTaskStats();
//
// final TermsWriteProc.TermsWriteProcConstructor ctor = new TermsWriteProc.TermsWriteProcConstructor(
// readOnly, storeBlankNodes);
//
// ndx.submit(0/* fromIndex */, values.length/* toIndex */, keys,
// vals, ctor, new TermsWriteProcResultHandler(a,
// readOnly, stats));
//
// // Note: [nunknown] is only set on read.
// assertEquals(0, stats.nunknown.get());
//
// /*
// * Verify that the IV is a fully qualified key for the TERMS
// * index.
// */
// final IKeyBuilder keyBuilder = h.newKeyBuilder();
//
// for (int i = 0; i < a.length; i++) {
//
// final BigdataValue expectedValue = a[i].obj;
//
// final IV<?, ?> actualIV = expectedValue.getIV();
//
// // Verify an IV was assigned.
// assertNotNull(actualIV);
//
// // Verify the VTE is consistent.
// assertEquals(VTE.valueOf(expectedValue), actualIV.getVTE());
//
// // Verify that the hashCode is consistent.
// assertEquals(expectedValue.hashCode(), actualIV.hashCode());
//
// // Encode the IV as a key.
// final byte[] key = actualIV.encode(keyBuilder.reset())
// .getKey();
//
// // Verify can encode/decode the IV.
// {
//
// final IV<?, ?> decodedIV = IVUtility.decode(key);
//
// assertEquals(actualIV, decodedIV);
//
// assertEquals(key, decodedIV.encode(h.newKeyBuilder())
// .getKey());
//
// }
//
// // Point lookup using the IV as the key.
// final byte[] val = ndx.lookup(key);
//
// // Verify point lookup succeeds.
// if (val == null) {
// fail("Could not resolve IV against index: expectedValue="
// + expectedValue
// + ", actualIV="
// + actualIV
// + ", key=" + BytesUtil.toString(key));
// }
//
// // Decode the returned byte[] as a Value.
// final BigdataValue actual = vf.getValueSerializer()
// .deserialize(val);
//
// // Verify BigdataValues are equal()
// assertEquals(expectedValue, actual);
//
// if (log.isInfoEnabled())
// log.info("i=" + expectedValue + ", iv=" + actualIV);
//
// }
//
// }
//
// } finally {
//
// store.destroy();
//
// }
//
// }
/**
* Create a TERMS index, put some data into it, and verify that we can use
* the {@link BlobsTupleSerializer} to access that data, including handling
* of the NullIV.
*/
public void test_TermsTupleSerializer() {
final IRawStore store = new SimpleMemoryRawStore();
try {
final String namespace = getName();
final BTree ndx = createTermsIndex(store, namespace);
final BlobsIndexHelper h = new BlobsIndexHelper();
final BigdataValueFactory vf = BigdataValueFactoryImpl
.getInstance(namespace);
/*
* Generate Values that we will use to read and write on the TERMS
* index.
*/
final BigdataValue[] values;
{
final BigdataURI uri1 = vf
.createURI("http://www.bigdata.com/testTerm");
final BigdataLiteral lit1 = vf.createLiteral("bigdata");
final BigdataLiteral lit2 = vf.createLiteral("bigdata", "en");
final BigdataLiteral lit3 = vf.createLiteral("bigdata",
XMLSchema.STRING);
final BigdataBNode bnode1 = vf.createBNode();
final BigdataBNode bnode2 = vf.createBNode("abc");
values = new BigdataValue[] { uri1, lit1, lit2, lit3, bnode1,
bnode2 };
}
final KVO<BigdataValue>[] a = h.generateKVOs(vf
.getValueSerializer(), values, values.length);
final byte[][] keys = new byte[a.length][];
final byte[][] vals = new byte[a.length][];
for (int i = 0; i < a.length; i++) {
keys[i] = a[i].key;
vals[i] = a[i].val;
}
/*
* Write on the TERMS index, setting IVs as side-effect on
* BigdataValues.
*/
{
final boolean readOnly = false;
final boolean storeBlankNodes = true;
final WriteTaskStats stats = new WriteTaskStats();
final BlobsWriteProc.BlobsWriteProcConstructor ctor = new BlobsWriteProc.BlobsWriteProcConstructor(
readOnly, storeBlankNodes);
ndx.submit(0/* fromIndex */, values.length/* toIndex */, keys,
vals, ctor, new BlobsWriteProcResultHandler(a,
readOnly, stats));
}
/*
* Exercise the TermsTupleSerializer
*/
{
final BlobsTupleSerializer tupSer = (BlobsTupleSerializer) ndx
.getIndexMetadata().getTupleSerializer();
for(BigdataValue value : values) {
final IV<?,?> iv = value.getIV();
assertNotNull(iv);
// Test serializeKey.
final byte[] key = tupSer.serializeKey(iv);
final byte[] val = ndx.lookup(key);
final BigdataValue actualValue = vf.getValueSerializer()
.deserialize(val);
assertEquals(value, actualValue);
/*
* TODO It should be possible to test more of the tupleSer
* directly. E.g., by running an iterator over the tuples
* and visiting them.
*/
}
}
} finally {
store.destroy();
}
}
// /**
// * Mock variant of
// * {@link TermsIndexHelper#generateKVOs(BigdataValueSerializer, BigdataValue[], int)}
// * which uses a constant value for the assigned hash codes.
// *
// * @param valSer
// * The object used to generate the values to be written onto the
// * index.
// * @param terms
// * The terms whose sort keys will be generated.
// * @param numTerms
// * The #of terms in that array.
// * @param hashCode
// * The hash code to assign to each value.
// *
// * @return An array of correlated key-value-object tuples.
// */
// @SuppressWarnings("unchecked")
// static private KVO<BigdataValue>[] mockGenerateKVOs(
// final BigdataValueSerializer<BigdataValue> valSer,
// final BigdataValue[] terms, final int numTerms, final int hashCode) {
//
// if (valSer == null)
// throw new IllegalArgumentException();
// if (terms == null)
// throw new IllegalArgumentException();
// if (numTerms <= 0 || numTerms > terms.length)
// throw new IllegalArgumentException();
//
// final KVO<BigdataValue>[] a = new KVO[numTerms];
//
// final TermsIndexHelper helper = new TermsIndexHelper();
//
// final IKeyBuilder keyBuilder = helper.newKeyBuilder();
//
// final DataOutputBuffer out = new DataOutputBuffer();
//
// final ByteArrayBuffer tmp = new ByteArrayBuffer();
//
// for (int i = 0; i < numTerms; i++) {
//
// final BigdataValue term = terms[i];
//
// final VTE vte = VTE.valueOf(term);
//
// final byte[] key = helper.makePrefixKey(keyBuilder.reset(), vte,
// hashCode);
//
// final byte[] val = valSer.serialize(term, out.reset(), tmp);
//
// a[i] = new KVO<BigdataValue>(key, val, term);
//
// }
//
// return a;
//
// }
/**
* Unit test for {@link BlobsWriteProc.Result} serialization.
*/
public void test_blobsResultSerialization() {
final long totalBucketSize = 7L;
final int maxBucketSize = 3;
final int[] counters = new int[] { 1, 0, 2, BlobsIndexHelper.NOT_FOUND,
3 };
final BlobsWriteProc.Result given = new BlobsWriteProc.Result(
totalBucketSize, maxBucketSize, counters);
assertEquals("totalBucketSize", totalBucketSize, given.totalBucketSize);
assertEquals("maxBucketSize", maxBucketSize, given.maxBucketSize);
assertTrue("counters[]", Arrays.equals(counters, given.counters));
final byte[] b = SerializerUtil.serialize(given);
final BlobsWriteProc.Result actual = (BlobsWriteProc.Result) SerializerUtil
.deserialize(b);
assertEquals("totalBucketSize", totalBucketSize, actual.totalBucketSize);
assertEquals("maxBucketSize", maxBucketSize, actual.maxBucketSize);
assertTrue("counters[]", Arrays.equals(counters, actual.counters));
}
}