Menu
▾
▴
[Bigdata-commit] SF.net SVN: bigdata:[7829] branches/RDR
|
From: <mrp...@us...> - 2014-01-26 20:37:39
|
Revision: 7829
http://bigdata.svn.sourceforge.net/bigdata/?rev=7829&view=rev
Author: mrpersonick
Date: 2014-01-26 20:37:25 +0000 (Sun, 26 Jan 2014)
Log Message:
-----------
major commit of RDR. one step away from entirely removing the overloading of the context position for sids. working checkpoint.
Modified Paths:
--------------
branches/RDR/.classpath
branches/RDR/bigdata/src/resources/logging/log4j.properties
branches/RDR/bigdata-gas/src/test/com/bigdata/rdf/graph/data/smallGraph.ttl
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/IVUtility.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/LexiconConfiguration.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BNodeContextFactory.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataBNode.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataBNodeImpl.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataStatementImpl.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueFactory.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueFactoryImpl.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueImpl.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/BasicRioLoader.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/StatementBuffer.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/ntriples/BigdataNTriplesParser.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/rdfxml/BigdataRDFXMLParser.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/turtle/BigdataTurtleParser.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/sparql/ast/service/history/HistoryIndexTupleSerializer.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/spo/ISPO.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/spo/SPO.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/spo/SPOTupleSerializer.java
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/store/AbstractTripleStore.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/graph/impl/bd/TestBFS.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/internal/TestEncodeDecodeKeys.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/internal/TestEncodeDecodeMixedIVs.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/rio/TestNTriplesWithSids.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/rio/TestRDFXMLInterchangeWithStatementIdentifiers.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/rio/TestStatementBuffer.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/sparql/ast/optimizers/TestALPPinTrac773.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/sparql/ast/optimizers/TestASTNamedSubqueryOptimizer.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/sparql/ast/optimizers/TestASTUnionFiltersOptimizer.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/spo/TestSPO.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/spo/TestSPOTupleSerializer.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/spo/TestSPOValueCoders.java
branches/RDR/bigdata-rdf/src/test/com/bigdata/rdf/store/TestStatementIdentifiers.java
branches/RDR/bigdata-sails/src/test/com/bigdata/rdf/sail/TestProvenanceQuery.java
Added Paths:
-----------
branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/impl/uri/IPAddrIV.java
Modified: branches/RDR/.classpath
===================================================================
--- branches/RDR/.classpath 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/.classpath 2014-01-26 20:37:25 UTC (rev 7829)
@@ -33,7 +33,7 @@
<classpathentry kind="src" path="bigdata-gas/src/test"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/dsi-utils-1.0.6-020610.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/lgpl-utils-1.0.6-020610.jar"/>
- <classpathentry kind="lib" path="bigdata-jini/lib/apache/zookeeper-3.3.3.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-jini/lib/apache/zookeeper-3.3.3.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/jetty/jetty-continuation-7.2.2.v20101205.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/jetty/jetty-http-7.2.2.v20101205.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/jetty/jetty-io-7.2.2.v20101205.jar"/>
@@ -45,8 +45,8 @@
<classpathentry exported="true" kind="lib" path="bigdata/lib/jetty/jetty-webapp-7.2.2.v20101205.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/jetty/jetty-xml-7.2.2.v20101205.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/unimi/colt-1.2.0.jar"/>
- <classpathentry kind="lib" path="bigdata/lib/icu/icu4j-4.8.jar"/>
- <classpathentry kind="lib" path="bigdata/lib/icu/icu4j-charset-4.8.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata/lib/icu/icu4j-4.8.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata/lib/icu/icu4j-charset-4.8.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/junit-3.8.1.jar" sourcepath="/root/.m2/repository/junit/junit/3.8.1/junit-3.8.1-sources.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata-jini/lib/jini/lib/browser.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata-jini/lib/jini/lib/classserver.jar"/>
@@ -67,24 +67,24 @@
<classpathentry exported="true" kind="lib" path="bigdata/lib/unimi/fastutil-5.1.5.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/lucene/lucene-analyzers-3.0.0.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/lucene/lucene-core-3.0.0.jar"/>
- <classpathentry kind="con" path="org.eclipse.jdt.launching.JRE_CONTAINER"/>
+ <classpathentry exported="true" kind="con" path="org.eclipse.jdt.launching.JRE_CONTAINER"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/high-scale-lib-v1.1.2.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata/lib/junit-ext-1.1-b3-dev.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata-rdf/lib/slf4j-api-1.6.1.jar"/>
<classpathentry exported="true" kind="lib" path="bigdata-rdf/lib/slf4j-log4j12-1.6.1.jar"/>
- <classpathentry kind="lib" path="bigdata-sails/lib/httpcomponents/commons-codec-1.4.jar"/>
- <classpathentry kind="lib" path="bigdata-sails/lib/httpcomponents/commons-logging-1.1.1.jar"/>
- <classpathentry kind="lib" path="bigdata-sails/lib/httpcomponents/httpclient-4.1.3.jar"/>
- <classpathentry kind="lib" path="bigdata-sails/lib/httpcomponents/httpclient-cache-4.1.3.jar"/>
- <classpathentry kind="lib" path="bigdata-sails/lib/httpcomponents/httpcore-4.1.4.jar"/>
- <classpathentry kind="lib" path="bigdata-sails/lib/httpcomponents/httpmime-4.1.3.jar"/>
- <classpathentry kind="lib" path="bigdata-sails/lib/httpcomponents/commons-fileupload-1.2.2.jar"/>
- <classpathentry kind="lib" path="bigdata-sails/lib/httpcomponents/commons-io-2.1.jar"/>
- <classpathentry kind="lib" path="bigdata/lib/apache/log4j-1.2.17.jar"/>
- <classpathentry kind="lib" path="bigdata-rdf/lib/openrdf-sesame-2.6.10-onejar.jar"/>
- <classpathentry kind="lib" path="bigdata-rdf/lib/sesame-rio-testsuite-2.6.10.jar"/>
- <classpathentry kind="lib" path="bigdata-sails/lib/sesame-sparql-testsuite-2.6.10.jar"/>
- <classpathentry kind="lib" path="bigdata-sails/lib/sesame-store-testsuite-2.6.10.jar"/>
- <classpathentry kind="lib" path="bigdata-rdf/lib/nxparser-1.2.3.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-sails/lib/httpcomponents/commons-codec-1.4.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-sails/lib/httpcomponents/commons-logging-1.1.1.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-sails/lib/httpcomponents/httpclient-4.1.3.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-sails/lib/httpcomponents/httpclient-cache-4.1.3.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-sails/lib/httpcomponents/httpcore-4.1.4.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-sails/lib/httpcomponents/httpmime-4.1.3.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-sails/lib/httpcomponents/commons-fileupload-1.2.2.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-sails/lib/httpcomponents/commons-io-2.1.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata/lib/apache/log4j-1.2.17.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-rdf/lib/openrdf-sesame-2.6.10-onejar.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-rdf/lib/sesame-rio-testsuite-2.6.10.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-sails/lib/sesame-sparql-testsuite-2.6.10.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-sails/lib/sesame-store-testsuite-2.6.10.jar"/>
+ <classpathentry exported="true" kind="lib" path="bigdata-rdf/lib/nxparser-1.2.3.jar"/>
<classpathentry kind="output" path="bin"/>
</classpath>
Modified: branches/RDR/bigdata/src/resources/logging/log4j.properties
===================================================================
--- branches/RDR/bigdata/src/resources/logging/log4j.properties 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata/src/resources/logging/log4j.properties 2014-01-26 20:37:25 UTC (rev 7829)
@@ -16,9 +16,13 @@
log4j.logger.com.bigdata.rdf.store.DataLoader=INFO
log4j.logger.com.bigdata.resources.AsynchronousOverflowTask=INFO
+#log4j.logger.com.bigdata.rdf.rio.StatementBuffer=ALL
+#log4j.logger.com.bigdata.rdf.sail.TestProvenanceQuery=ALL
+
# Test suite loggers.
#log4j.logger.junit=INFO
#log4j.logger.com.bigdata.btree.AbstractBTreeTestCase=INFO
+log4j.logger.junit.framework.TestCase2=ERROR
# dest1
log4j.appender.dest1=org.apache.log4j.ConsoleAppender
Modified: branches/RDR/bigdata-gas/src/test/com/bigdata/rdf/graph/data/smallGraph.ttl
===================================================================
--- branches/RDR/bigdata-gas/src/test/com/bigdata/rdf/graph/data/smallGraph.ttl 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-gas/src/test/com/bigdata/rdf/graph/data/smallGraph.ttl 2014-01-26 20:37:25 UTC (rev 7829)
@@ -1,19 +1,19 @@
-@prefix : <http://www.bigdata.com/> .
+@prefix bd: <http://www.bigdata.com/> .
@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix foaf: <http://xmlns.com/foaf/0.1/> .
#: {
- :Mike rdf:type foaf:Person .
- :Bryan rdf:type foaf:Person .
- :Martyn rdf:type foaf:Person .
+ bd:Mike rdf:type foaf:Person .
+ bd:Bryan rdf:type foaf:Person .
+ bd:Martyn rdf:type foaf:Person .
- :Mike rdfs:label "Mike" .
- :Bryan rdfs:label "Bryan" .
- :DC rdfs:label "DC" .
+ bd:Mike rdfs:label "Mike" .
+ bd:Bryan rdfs:label "Bryan" .
+ bd:DC rdfs:label "DC" .
- :Mike foaf:knows :Bryan .
- :Bryan foaf:knows :Mike .
- :Bryan foaf:knows :Martyn .
- :Martyn foaf:knows :Bryan .
+ bd:Mike foaf:knows bd:Bryan .
+ bd:Bryan foaf:knows bd:Mike .
+ bd:Bryan foaf:knows bd:Martyn .
+ bd:Martyn foaf:knows bd:Bryan .
#}
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/IVUtility.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/IVUtility.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/IVUtility.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -35,6 +35,8 @@
import java.io.IOException;
import java.math.BigDecimal;
import java.math.BigInteger;
+import java.net.InetAddress;
+import java.net.UnknownHostException;
import java.util.LinkedList;
import java.util.List;
import java.util.UUID;
@@ -66,6 +68,7 @@
import com.bigdata.rdf.internal.impl.literal.XSDUnsignedLongIV;
import com.bigdata.rdf.internal.impl.literal.XSDUnsignedShortIV;
import com.bigdata.rdf.internal.impl.uri.FullyInlineURIIV;
+import com.bigdata.rdf.internal.impl.uri.IPAddrIV;
import com.bigdata.rdf.internal.impl.uri.PartlyInlineURIIV;
import com.bigdata.rdf.internal.impl.uri.URIExtensionIV;
import com.bigdata.rdf.internal.impl.uri.VocabURIByteIV;
@@ -437,7 +440,7 @@
final ISPO spo = SPOKeyOrder.SPO.decodeKey(key, o);
// all spos that have a sid are explicit
spo.setStatementType(StatementEnum.Explicit);
- spo.setStatementIdentifier(true);
+// spo.setStatementIdentifier(true);
// create a sid iv and return it
return new SidIV(spo);
}
@@ -535,6 +538,21 @@
// The data type
final DTE dte = AbstractIV.getDTE(flags);
switch (dte) {
+ case XSDBoolean: {
+ /*
+ * TODO Using XSDBoolean so that we can know how to decode this thing
+ * as an IPAddrIV. We need to fix the Extension mechanism for URIs.
+ * Extension is already used above.
+ */
+ try {
+ final byte[] addr = new byte[4];
+ System.arraycopy(key, o, addr, 0, 4);
+ final InetAddress ip = InetAddress.getByAddress(addr);
+ return new IPAddrIV(ip);
+ } catch (UnknownHostException ex) {
+ throw new RuntimeException(ex);
+ }
+ }
case XSDByte: {
final byte x = key[o];//KeyBuilder.decodeByte(key[o]);
return new VocabURIByteIV<BigdataURI>(x);
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/LexiconConfiguration.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/LexiconConfiguration.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/LexiconConfiguration.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -28,6 +28,7 @@
package com.bigdata.rdf.internal;
import java.math.BigInteger;
+import java.net.UnknownHostException;
import java.util.LinkedHashMap;
import java.util.Map;
import java.util.TimeZone;
@@ -58,6 +59,7 @@
import com.bigdata.rdf.internal.impl.literal.XSDUnsignedLongIV;
import com.bigdata.rdf.internal.impl.literal.XSDUnsignedShortIV;
import com.bigdata.rdf.internal.impl.uri.FullyInlineURIIV;
+import com.bigdata.rdf.internal.impl.uri.IPAddrIV;
import com.bigdata.rdf.internal.impl.uri.URIExtensionIV;
import com.bigdata.rdf.lexicon.LexiconKeyOrder;
import com.bigdata.rdf.model.BigdataBNode;
@@ -435,6 +437,22 @@
*/
private IV<BigdataURI, ?> createInlineURIIV(final URI value) {
+ try {
+
+ final String s = value.stringValue();
+
+ if (s.startsWith(IPAddrIV.NAMESPACE)) {
+
+ return new IPAddrIV(s.substring(IPAddrIV.NAMESPACE_LEN));
+
+ }
+
+ } catch (UnknownHostException ex) {
+
+ log.warn("unknown host exception, will not inline: " + value);
+
+ }
+
if (maxInlineTextLength == 0) {
return null;
Added: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/impl/uri/IPAddrIV.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/impl/uri/IPAddrIV.java (rev 0)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/impl/uri/IPAddrIV.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -0,0 +1,349 @@
+/**
+
+Copyright (C) SYSTAP, LLC 2006-2011. All rights reserved.
+
+Contact:
+ SYSTAP, LLC
+ 4501 Tower Road
+ Greensboro, NC 27410
+ lic...@bi...
+
+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
+*/
+package com.bigdata.rdf.internal.impl.uri;
+
+import java.io.Externalizable;
+import java.io.IOException;
+import java.io.ObjectInput;
+import java.io.ObjectOutput;
+import java.io.ObjectStreamException;
+import java.io.Serializable;
+import java.net.InetAddress;
+import java.net.UnknownHostException;
+
+import org.openrdf.model.URI;
+import org.openrdf.model.Value;
+
+import com.bigdata.btree.BytesUtil.UnsignedByteArrayComparator;
+import com.bigdata.btree.keys.IKeyBuilder;
+import com.bigdata.io.LongPacker;
+import com.bigdata.rdf.internal.DTE;
+import com.bigdata.rdf.internal.ILexiconConfiguration;
+import com.bigdata.rdf.internal.IV;
+import com.bigdata.rdf.internal.IVUtility;
+import com.bigdata.rdf.internal.VTE;
+import com.bigdata.rdf.internal.impl.AbstractInlineIV;
+import com.bigdata.rdf.lexicon.LexiconRelation;
+import com.bigdata.rdf.model.BigdataBNode;
+import com.bigdata.rdf.model.BigdataURI;
+import com.bigdata.rdf.model.BigdataValueFactory;
+import com.bigdata.rdf.spo.SPOKeyOrder;
+
+/**
+ * Internal value representing an inline IP address. Uses the InetAddress
+ * class to represent the IP address and perform the translation to and from
+ * byte[], which is then used directly in the IV key (after the flags).
+ * <p>
+ * This internal value has a {@link VTE} of {@link VTE#URI}.
+ * <p>
+ * {@inheritDoc}
+ */
+public class IPAddrIV<V extends BigdataURI> extends AbstractInlineIV<V, InetAddress>
+ implements Serializable, URI {
+
+ /**
+ *
+ */
+ private static final long serialVersionUID = 685148537376856907L;
+
+// private static final transient Logger log = Logger.getLogger(SidIV.class);
+
+ public static final String NAMESPACE = "ip:/";
+
+ public static final int NAMESPACE_LEN = NAMESPACE.length();
+
+ /**
+ * The inline IP address.
+ */
+ private final InetAddress value;
+
+ /**
+ * The cached string representation of this IP.
+ */
+ private transient String hostAddress;
+
+ /**
+ * The cached byte[] key for the encoding of this IV.
+ */
+ private transient byte[] key;
+
+ /**
+ * The cached materialized BigdataValue for this InetAddress.
+ */
+ private transient V uri;
+
+ public IV<V, InetAddress> clone(final boolean clearCache) {
+
+ final IPAddrIV<V> tmp = new IPAddrIV<V>(value);
+
+ // Propagate the cached byte[] key.
+ tmp.key = key;
+
+ // Propagate the cached BigdataValue.
+ tmp.uri = uri;
+
+ if (!clearCache) {
+
+ tmp.setValue(getValueCache());
+
+ }
+
+ return tmp;
+
+ }
+
+ /**
+ * Ctor with internal value specified.
+ */
+ public IPAddrIV(final InetAddress value) {
+
+ /*
+ * TODO Using XSDBoolean so that we can know how to decode this thing
+ * as an IPAddrIV. We need to fix the Extension mechanism for URIs.
+ */
+ super(VTE.URI, DTE.XSDBoolean);
+
+ this.value = value;
+
+ }
+
+ /**
+ * Ctor with host address specified.
+ */
+ public IPAddrIV(final String hostAddress) throws UnknownHostException {
+
+ /*
+ * Note: XSDBoolean happens to be assigned the code value of 0, which is
+ * the value we we want when the data type enumeration will be ignored.
+ */
+ super(VTE.URI, DTE.XSDBoolean);
+
+ this.value = InetAddress.getByName(hostAddress);
+ this.hostAddress = hostAddress;
+
+ }
+
+ /**
+ * Returns the inline value.
+ */
+ public InetAddress getInlineValue() throws UnsupportedOperationException {
+ return value;
+ }
+
+ /**
+ * Returns the URI representation of this IV.
+ */
+ public V asValue(final LexiconRelation lex) {
+ if (uri == null) {
+ uri = (V) lex.getValueFactory().createURI(getNamespace(), getLocalName());
+ uri.setIV(this);
+ }
+ return uri;
+ }
+
+ /**
+ * Return the byte length for the byte[] encoded representation of this
+ * internal value. Depends on the byte length of the encoded inline value.
+ */
+ public int byteLength() {
+ return 1 + key().length;
+ }
+
+ public String toString() {
+ return "IP("+getLocalName()+")";
+ }
+
+ public int hashCode() {
+ return value.hashCode();
+ }
+
+// /**
+// * Implements {@link BNode#getID()}.
+// * <p>
+// * This implementation uses the {@link BigInteger} class to create a unique
+// * blank node ID based on the <code>unsigned byte[]</code> key of the inline
+// * {@link SPO}.
+// */
+// @Override
+// public String getID() {
+//// // just use the hash code. can result in collisions
+//// return String.valueOf(hashCode());
+//
+// // create a big integer using the spo key. should result in unique ids
+// final byte[] key = key();
+// final int signum = key.length > 0 ? 1 : 0;
+// final BigInteger bi = new BigInteger(signum, key);
+// return 's' + bi.toString();
+// }
+
+ @Override
+ public String getNamespace() {
+ return NAMESPACE;
+ }
+
+ @Override
+ public String getLocalName() {
+ if (hostAddress == null) {
+ hostAddress = value.getHostAddress();
+ }
+ return hostAddress;
+ }
+
+ /**
+ * Two {@link IPAddrIV} are equal if their InetAddresses are equal.
+ */
+ public boolean equals(final Object o) {
+ if (this == o)
+ return true;
+ if (o instanceof IPAddrIV) {
+ final InetAddress value2 = ((IPAddrIV<?>) o).value;
+ return value.equals(value2);
+ }
+ return false;
+ }
+
+ public int _compareTo(IV o) {
+
+ /*
+ * Note: This works, but it might be more expensive.
+ */
+ return UnsignedByteArrayComparator.INSTANCE.compare(key(), ((IPAddrIV)o).key());
+
+ }
+
+ /**
+ * Encode this internal value into the supplied key builder. Emits the
+ * flags, following by the encoded byte[] representing the spo, in SPO
+ * key order.
+ * <p>
+ * {@inheritDoc}
+ */
+ @Override
+ public IKeyBuilder encode(final IKeyBuilder keyBuilder) {
+
+ // First emit the flags byte.
+ keyBuilder.appendSigned(flags());
+
+ // Then append the InetAddress byte[].
+ keyBuilder.append(key());
+
+ return keyBuilder;
+
+ }
+
+ private byte[] key() {
+
+ if (key == null) {
+
+ key = value.getAddress();
+
+ }
+
+ return key;
+
+ }
+
+ /**
+ * Object provides serialization for {@link IPAddrIV} via the write-replace
+ * and read-replace pattern.
+ */
+ private static class IPAddrIVState implements Externalizable {
+
+ private static final long serialVersionUID = -1L;
+
+// private byte flags;
+ private byte[] key;
+
+ /**
+ * De-serialization constructor.
+ */
+ public IPAddrIVState() {
+
+ }
+
+ private IPAddrIVState(final IPAddrIV iv) {
+// this.flags = flags;
+ this.key = iv.key();
+ }
+
+ public void readExternal(ObjectInput in) throws IOException,
+ ClassNotFoundException {
+// flags = in.readByte();
+ final int nbytes = LongPacker.unpackInt(in);
+ key = new byte[nbytes];
+ in.readFully(key);
+ }
+
+ public void writeExternal(ObjectOutput out) throws IOException {
+// out.writeByte(flags);
+ LongPacker.packLong(out, key.length);
+ out.write(key);
+ }
+
+ private Object readResolve() throws ObjectStreamException {
+
+ try {
+
+ final InetAddress value = InetAddress.getByAddress(key);
+
+ return new IPAddrIV(value);
+
+ } catch (UnknownHostException ex) {
+
+ throw new RuntimeException(ex);
+
+ }
+
+ }
+
+ }
+
+ private Object writeReplace() throws ObjectStreamException {
+
+ return new IPAddrIVState(this);
+
+ }
+
+ /**
+ * Implements {@link Value#stringValue()}.
+ */
+ @Override
+ public String stringValue() {
+
+ return getLocalName();
+
+ }
+
+ /**
+ * Does not need materialization to answer URI interface methods.
+ */
+ @Override
+ public boolean needsMaterialization() {
+
+ return false;
+
+ }
+
+
+}
\ No newline at end of file
Property changes on: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/internal/impl/uri/IPAddrIV.java
___________________________________________________________________
Added: svn:mime-type
+ text/plain
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BNodeContextFactory.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BNodeContextFactory.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BNodeContextFactory.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -89,6 +89,10 @@
return valueFactory.createBNode(id);
}
+ public BigdataBNode createBNode(BigdataStatement stmt) {
+ return valueFactory.createBNode(stmt);
+ }
+
public BigdataLiteral createLiteral(boolean arg0) {
return valueFactory.createLiteral(arg0);
}
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataBNode.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataBNode.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataBNode.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -67,4 +67,14 @@
*/
public boolean isStatementIdentifier();
+ /**
+ * Set the statement that this blank node models.
+ */
+ public void setStatement(BigdataStatement stmt);
+
+ /**
+ * Get the statement that this blank node models.
+ */
+ public BigdataStatement getStatement();
+
}
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataBNodeImpl.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataBNodeImpl.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataBNodeImpl.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -49,7 +49,10 @@
import org.openrdf.model.BNode;
+import com.bigdata.rdf.internal.IV;
+import com.bigdata.rdf.internal.impl.bnode.SidIV;
import com.bigdata.rdf.rio.StatementBuffer;
+import com.bigdata.rdf.rio.UnificationException;
import com.bigdata.rdf.spo.SPO;
import com.bigdata.rdf.store.AbstractTripleStore;
@@ -76,7 +79,7 @@
private static final long serialVersionUID = 2675602437833048872L;
private final String id;
-
+
/**
* Boolean flag is set during conversion from an RDF interchange syntax
* into the internal {@link SPO} model if the blank node is a statement
@@ -95,14 +98,45 @@
*/
BigdataBNodeImpl(final BigdataValueFactory valueFactory, final String id) {
+ this(valueFactory, id, null);
+
+ }
+
+ BigdataBNodeImpl(final BigdataValueFactory valueFactory, final String id,
+ final BigdataStatement stmt) {
+
super(valueFactory, null);
if (id == null)
throw new IllegalArgumentException();
this.id = id;
+
+ this.sid = stmt;
+ if (stmt != null) {
+ this.statementIdentifier = true;
+ }
}
+
+ @Override
+ public IV getIV() {
+
+ if (super.iv == null && sid != null) {
+
+ if (sid.getSubject() == this || sid.getObject() == this)
+ throw new UnificationException("illegal self-referential sid");
+
+ final IV s = sid.s();
+ final IV p = sid.p();
+ final IV o = sid.o();
+ if (s != null && p != null && o != null) {
+ setIV(new SidIV(new SPO(s, p, o)));
+ }
+ }
+
+ return super.iv;
+ }
public String toString() {
@@ -179,7 +213,7 @@
/**
* Marks this as a blank node which models the specified statement.
*
- * @param stmt
+ * @param sid
* The statement.
*/
final public void setStatement(final BigdataStatement sid) {
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataStatementImpl.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataStatementImpl.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataStatementImpl.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -63,6 +63,7 @@
protected final BigdataURI p;
protected final BigdataValue o;
protected final BigdataResource c;
+ protected IV sid = null;
private StatementEnum type;
private boolean userFlag;
private transient boolean override = false;
@@ -282,39 +283,51 @@
}
- public final void setStatementIdentifier(final boolean sidable) {
+// public final void setStatementIdentifier(final boolean sidable) {
+//
+// if (sidable && type != StatementEnum.Explicit) {
+//
+// // Only allowed for explicit statements.
+// throw new IllegalStateException();
+//
+// }
+//
+//// if (c == null) {
+////
+//// // this SHOULD not ever happen
+//// throw new IllegalStateException();
+////
+//// }
+////
+//// c.setIV(new SidIV(this));
+//
+// this.sid = new SidIV(this);
+//
+// }
- if (sidable && type != StatementEnum.Explicit) {
-
- // Only allowed for explicit statements.
- throw new IllegalStateException();
-
- }
-
- if (c == null) {
-
- // this SHOULD not ever happen
- throw new IllegalStateException();
-
- }
-
- c.setIV(new SidIV(this));
-
- }
-
public final IV getStatementIdentifier() {
- if (!hasStatementIdentifier())
- throw new IllegalStateException("No statement identifier: "
- + toString());
+// if (!hasStatementIdentifier())
+// throw new IllegalStateException("No statement identifier: "
+// + toString());
+//
+// return c.getIV();
- return c.getIV();
+ if (sid == null && type == StatementEnum.Explicit) {
+
+ sid = new SidIV(this);
+
+ }
+
+ return sid;
}
final public boolean hasStatementIdentifier() {
- return c != null && c.getIV().isStatement();
+// return c != null && c.getIV().isStatement();
+
+ return type == StatementEnum.Explicit;
}
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueFactory.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueFactory.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueFactory.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -85,6 +85,8 @@
BigdataBNode createBNode();
BigdataBNode createBNode(String id);
+
+ BigdataBNode createBNode(BigdataStatement stmt);
BigdataLiteral createLiteral(String label);
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueFactoryImpl.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueFactoryImpl.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueFactoryImpl.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -36,6 +36,7 @@
import org.openrdf.model.BNode;
import org.openrdf.model.Literal;
import org.openrdf.model.Resource;
+import org.openrdf.model.Statement;
import org.openrdf.model.URI;
import org.openrdf.model.Value;
import org.openrdf.model.datatypes.XMLDatatypeUtil;
@@ -235,6 +236,12 @@
}
+ public BigdataBNodeImpl createBNode(final BigdataStatement stmt) {
+
+ return new BigdataBNodeImpl(this, nextID(), stmt);
+
+ }
+
public BigdataLiteralImpl createLiteral(final String label) {
return new BigdataLiteralImpl(this, label, null, null);
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueImpl.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueImpl.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/model/BigdataValueImpl.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -48,7 +48,7 @@
private volatile transient BigdataValueFactory valueFactory;
- private volatile IV iv;
+ protected volatile IV iv;
public final BigdataValueFactory getValueFactory() {
@@ -115,7 +115,7 @@
}
- final public IV getIV() {
+ public IV getIV() {
return iv;
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/BasicRioLoader.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/BasicRioLoader.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/BasicRioLoader.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -153,6 +153,8 @@
final RDFParser parser = Rio.createParser(rdfFormat, valueFactory);
+ parser.setValueFactory(valueFactory);
+
return parser;
}
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/StatementBuffer.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/StatementBuffer.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/StatementBuffer.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -232,11 +232,11 @@
*/
protected final int capacity;
- /**
- * When true only distinct terms are stored in the buffer (this is always
- * true since this condition always outperforms the alternative).
- */
- protected final boolean distinct = true;
+// /**
+// * When true only distinct terms are stored in the buffer (this is always
+// * true since this condition always outperforms the alternative).
+// */
+// protected final boolean distinct = true;
public boolean isEmpty() {
@@ -347,30 +347,22 @@
this.capacity = capacity;
- values = new BigdataValue[capacity * arity];
+ values = new BigdataValue[capacity * arity + 5];
stmts = new BigdataStatement[capacity];
- if (distinct) {
-
- /*
- * initialize capacity to N times the #of statements allowed. this
- * is the maximum #of distinct terms and would only be realized if
- * each statement used distinct values. in practice the #of distinct
- * terms will be much lower. however, also note that the map will be
- * resized at .75 of the capacity so we want to over-estimate the
- * maximum likely capacity by at least 25% to avoid re-building the
- * hash map.
- */
-
- distinctTermMap = new HashMap<Value, BigdataValue>(capacity * arity);
-
- } else {
-
- distinctTermMap = null;
-
- }
+ /*
+ * initialize capacity to N times the #of statements allowed. this
+ * is the maximum #of distinct terms and would only be realized if
+ * each statement used distinct values. in practice the #of distinct
+ * terms will be much lower. however, also note that the map will be
+ * resized at .75 of the capacity so we want to over-estimate the
+ * maximum likely capacity by at least 25% to avoid re-building the
+ * hash map.
+ */
+ distinctTermMap = new HashMap<Value, BigdataValue>(capacity * arity);
+
this.statementIdentifiers = database.getStatementIdentifiers();
if(log.isInfoEnabled()) {
@@ -387,18 +379,14 @@
this.RDF_STATEMENT = valueFactory.asValue(RDF.STATEMENT);
this.RDF_TYPE = valueFactory.asValue(RDF.TYPE);
- if (distinct) {
-
- /*
- * Get the reification vocabulary into the distinct term map.
- */
- getDistinctTerm(RDF_SUBJECT);
- getDistinctTerm(RDF_PREDICATE);
- getDistinctTerm(RDF_OBJECT);
- getDistinctTerm(RDF_STATEMENT);
- getDistinctTerm(RDF_TYPE);
-
- }
+ /*
+ * Get the reification vocabulary into the distinct term map.
+ */
+ getDistinctTerm(RDF_SUBJECT, true);
+ getDistinctTerm(RDF_PREDICATE, true);
+ getDistinctTerm(RDF_OBJECT, true);
+ getDistinctTerm(RDF_STATEMENT, true);
+ getDistinctTerm(RDF_TYPE, true);
}
@@ -415,8 +403,6 @@
*/
public long flush() {
- log.info("");
-
/*
* Process deferred statements (NOP unless using statement identifiers).
*/
@@ -552,11 +538,7 @@
while(itr.hasNext()) {
final BigdataStatement stmt = itr.next();
-
- if (log.isDebugEnabled()) {
- log.debug(stmt.getSubject() + ", sid=" + ((BigdataBNode) stmt.getSubject()).isStatementIdentifier() + ", iv=" + stmt.s());
- }
-
+
if (stmt.getSubject() instanceof BNode
&& ((BigdataBNode) stmt.getSubject()).isStatementIdentifier())
continue;
@@ -569,6 +551,12 @@
log.debug("grounded: "+stmt);
}
+ if (stmt.getSubject() instanceof BNode)
+ addTerm(stmt.getSubject());
+
+ if (stmt.getObject() instanceof BNode)
+ addTerm(stmt.getObject());
+
// fully grounded so add to the buffer.
add(stmt);
@@ -703,8 +691,6 @@
*/
public void reset() {
- log.info("");
-
_clear();
/*
@@ -763,6 +749,15 @@
if (distinctTermMap != null) {
distinctTermMap.clear();
+
+ /*
+ * Get the reification vocabulary into the distinct term map.
+ */
+ getDistinctTerm(RDF_SUBJECT, true);
+ getDistinctTerm(RDF_PREDICATE, true);
+ getDistinctTerm(RDF_OBJECT, true);
+ getDistinctTerm(RDF_STATEMENT, true);
+ getDistinctTerm(RDF_TYPE, true);
}
@@ -775,6 +770,24 @@
*/
protected void incrementalWrite() {
+ if (bnodes != null) {
+
+ for (BigdataBNode bnode : bnodes.values()) {
+
+ if (bnode.isStatementIdentifier())
+ continue;
+
+ if (bnode.getIV() != null)
+ continue;
+
+ values[numValues++] = bnode;
+
+ numBNodes++;
+
+ }
+
+ }
+
final long begin = System.currentTimeMillis();
if (log.isInfoEnabled()) {
@@ -1177,9 +1190,10 @@
* @return Either the term or the pre-existing term in the buffer with the
* same data.
*/
- protected BigdataValue getDistinctTerm(final BigdataValue term) {
+ protected BigdataValue getDistinctTerm(final BigdataValue term, final boolean addIfAbsent) {
- assert distinct == true;
+ if (term == null)
+ return null;
if (term instanceof BNode) {
@@ -1193,65 +1207,125 @@
final BigdataBNode bnode = (BigdataBNode)term;
- // the BNode's ID.
- final String id = bnode.getID();
-
- if (bnodes == null) {
-
- // allocating canonicalizing map for blank nodes.
- bnodes = new HashMap<String, BigdataBNode>(capacity);
-
- // insert this blank node into the map.
- bnodes.put(id, bnode);
-
+ final BigdataStatement stmt = bnode.getStatement();
+
+ if (stmt != null) {
+
+// /*
+// * Assume for now that bnodes appearing inside the terse
+// * syntax without a statement attached are real bnodes, not
+// * sids.
+// */
+// final boolean tmp = this.statementIdentifiers;
+// this.statementIdentifiers = false;
+
+ bnode.setStatement(valueFactory.createStatement(
+ (BigdataResource) getDistinctTerm(stmt.getSubject(), true),
+ (BigdataURI) getDistinctTerm(stmt.getPredicate(), true),
+ (BigdataValue) getDistinctTerm(stmt.getObject(), true)
+ ));
+
+// this.statementIdentifiers = tmp;
+
+ /*
+ * Do not "add if absent". This is not a real term, just a
+ * composition of other terms.
+ */
+ return bnode;
+
} else {
-
- // test canonicalizing map for blank nodes.
- final BigdataBNode existingBNode = bnodes.get(id);
-
- if (existingBNode != null) {
-
- // return existing blank node with same ID.
- return existingBNode;
-
- }
-
- // insert this blank node into the map.
- bnodes.put(id, bnode);
-
+
+ // the BNode's ID.
+ final String id = bnode.getID();
+
+ if (bnodes == null) {
+
+ // allocating canonicalizing map for blank nodes.
+ bnodes = new HashMap<String, BigdataBNode>(capacity);
+
+ // insert this blank node into the map.
+ bnodes.put(id, bnode);
+
+ } else {
+
+ // test canonicalizing map for blank nodes.
+ final BigdataBNode existingBNode = bnodes.get(id);
+
+ if (existingBNode != null) {
+
+ /*
+ * Return existing blank node with same ID, do not
+ * add since not absent.
+ */
+ return existingBNode;
+
+ }
+
+ // insert this blank node into the map.
+ bnodes.put(id, bnode);
+
+ }
+
}
- return term;
+// return term;
- }
+ } else {
- /*
- * Other kinds of terms use a map whose scope is limited to the terms
- * that are currently in the buffer. This keeps down the heap demand
- * when reading very large documents.
- */
-
- final BigdataValue existingTerm = distinctTermMap.get(term);
-
- if(existingTerm != null) {
-
- // return the pre-existing term.
-
+ /*
+ * Other kinds of terms use a map whose scope is limited to the terms
+ * that are currently in the buffer. This keeps down the heap demand
+ * when reading very large documents.
+ */
+
+ final BigdataValue existingTerm = distinctTermMap.get(term);
+
+ if (existingTerm != null) {
+
+ // return the pre-existing term.
+
+ if(log.isDebugEnabled()) {
+
+ log.debug("duplicate: "+term);
+
+ }
+
+ if (equals(existingTerm, RDF_SUBJECT, RDF_PREDICATE, RDF_OBJECT, RDF_TYPE, RDF_STATEMENT)) {
+
+ if (addIfAbsent) {
+
+ addTerm(term);
+
+ }
+
+ }
+
+ /*
+ * Term already exists, do not add.
+ */
+ return existingTerm;
+
+ }
+
if(log.isDebugEnabled()) {
- log.debug("duplicate: "+term);
+ log.debug("new term: "+term);
}
- return existingTerm;
-
+ // put the new term in the map.
+ if (distinctTermMap.put(term, term) != null) {
+
+ throw new AssertionError();
+
+ }
+
}
-
- // put the new term in the map.
- if (distinctTermMap.put(term, term) != null) {
-
- throw new AssertionError();
-
+
+ if (addIfAbsent) {
+
+ addTerm(term);
+
}
// return the new term.
@@ -1259,6 +1333,37 @@
}
+ protected void addTerm(final BigdataValue term) {
+
+ if (term == null)
+ return;
+
+ if (term instanceof URI) {
+
+ numURIs++;
+
+ values[numValues++] = term;
+
+ } else if (term instanceof BNode) {
+
+// if (!statementIdentifiers) {
+//
+// numBNodes++;
+//
+// values[numValues++] = term;
+//
+// }
+
+ } else {
+
+ numLiterals++;
+
+ values[numValues++] = term;
+
+ }
+
+ }
+
/**
* Adds the values and the statement into the buffer.
*
@@ -1278,244 +1383,126 @@
*
* @see #nearCapacity()
*/
- protected void handleStatement(Resource s, URI p, Value o, Resource c,
+ protected void handleStatement(Resource _s, URI _p, Value _o, Resource _c,
StatementEnum type) {
+ if (log.isDebugEnabled()) {
+
+ log.debug("handle stmt: " + _s + ", " + _p + ", " + _o + ", " + _c);
+
+ }
+
// if (arity == 3) c = null;
- s = (Resource) valueFactory.asValue(s);
- p = (URI) valueFactory.asValue(p);
- o = valueFactory.asValue(o);
- c = (Resource) valueFactory.asValue(c);
+ final BigdataResource s = (BigdataResource)
+ getDistinctTerm(valueFactory.asValue(_s), true);
+ final BigdataURI p = (BigdataURI)
+ getDistinctTerm(valueFactory.asValue(_p), true);
+ final BigdataValue o =
+ getDistinctTerm(valueFactory.asValue(_o), true);
+ final BigdataResource c = (BigdataResource)
+ getDistinctTerm(valueFactory.asValue(_c), true);
- boolean duplicateS = false;
- boolean duplicateP = false;
- boolean duplicateO = false;
- boolean duplicateC = false;
-
- if (distinct) {
- {
- final BigdataValue tmp = getDistinctTerm((BigdataValue) s);
- if (tmp != s && !equals(tmp, RDF_SUBJECT, RDF_PREDICATE, RDF_OBJECT, RDF_TYPE, RDF_STATEMENT)) {
- duplicateS = true;
- }
- s = (Resource) tmp;
- }
- {
- final BigdataValue tmp = getDistinctTerm((BigdataValue) p);
- if (tmp != p && !equals(tmp, RDF_SUBJECT, RDF_PREDICATE, RDF_OBJECT, RDF_TYPE)) {
- duplicateP = true;
- }
- p = (URI) tmp;
- }
- {
- final BigdataValue tmp = getDistinctTerm((BigdataValue) o);
- if (tmp != o && !equals(tmp, RDF_SUBJECT, RDF_PREDICATE, RDF_OBJECT, RDF_TYPE, RDF_STATEMENT)) {
- duplicateO = true;
- }
- o = (Value) tmp;
- }
- if (c != null) {
- final BigdataValue tmp = getDistinctTerm((BigdataValue) c);
- if (tmp != c) {
- duplicateC = true;
- }
- c = (Resource) tmp;
- }
- }
-
/*
* Form the BigdataStatement object now that we have the bindings.
*/
- final BigdataStatement stmt;
- {
-
- stmt = valueFactory
- .createStatement((BigdataResource) s, (BigdataURI) p,
- (BigdataValue) o, (BigdataResource) c, type);
+ final BigdataStatement stmt = valueFactory.createStatement(s, p, o, c, type);
- if (statementIdentifiers
- && (s instanceof BNode || o instanceof BNode)) {
+ if (statementIdentifiers
+ && ((s instanceof BNode && ((BigdataBNode) s).getStatement() == null)
+// ||
+// (o instanceof BNode && ((BigdataBNode) o).getStatement() == null)
+ )) {
- /*
- * When statement identifiers are enabled a statement with a
- * blank node in the subject or object position must be deferred
- * until the end of the source so that we determine whether it
- * is being used as a statement identifier or a blank node (if
- * the blank node occurs in the context position, then we know
- * that it is being used as a statement identifier).
- */
+ /*
+ * When statement identifiers are enabled a statement with a
+ * blank node in the subject or object position must be deferred
+ * until the end of the source so that we determine whether it
+ * is being used as a statement identifier or a blank node (if
+ * the blank node occurs in the context position, then we know
+ * that it is being used as a statement identifier).
+ */
+
+ if (//s instanceof BNode &&
+ equals(p, RDF_SUBJECT, RDF_PREDICATE, RDF_OBJECT)) {
+
+ final BigdataBNodeImpl sid = (BigdataBNodeImpl) s;
+
+ if (reifiedStmts == null) {
+
+ reifiedStmts = new HashMap<BigdataBNodeImpl, ReifiedStmt>();
+
+ }
+
+ final ReifiedStmt reifiedStmt;
+ if (reifiedStmts.containsKey(sid)) {
+
+ reifiedStmt = reifiedStmts.get(sid);
+
+ } else {
+
+ reifiedStmt = new ReifiedStmt();
+
+ reifiedStmts.put(sid, reifiedStmt);
+
+ }
+
+ reifiedStmt.set(p, (BigdataValue) o);
+
+ if (log.isDebugEnabled())
+ log.debug("reified piece: "+stmt);
- log.info(stmt);
-
- if (s instanceof BNode &&
- equals((BigdataValue)p, RDF_SUBJECT, RDF_PREDICATE, RDF_OBJECT)) {
-
- final BigdataBNodeImpl sid = (BigdataBNodeImpl) s;
-
- if (reifiedStmts == null) {
-
- reifiedStmts = new HashMap<BigdataBNodeImpl, ReifiedStmt>();
-
- }
-
- final ReifiedStmt reifiedStmt;
- if (reifiedStmts.containsKey(sid)) {
-
- reifiedStmt = reifiedStmts.get(sid);
-
- } else {
-
- reifiedStmt = new ReifiedStmt();
-
- reifiedStmts.put(sid, reifiedStmt);
-
- }
-
- reifiedStmt.set(p, (BigdataValue) o);
-
- if (log.isDebugEnabled())
- log.debug("reified piece: "+stmt);
-
- } else if (s instanceof BNode &&
- equals((BigdataValue)o, RDF_STATEMENT) && equals((BigdataValue)p, RDF_TYPE)) {
-
- // ignore this statement
-
- } else {
-
- if (deferredStmts == null) {
-
- deferredStmts = new HashSet<BigdataStatement>(stmts.length);
-
- }
-
- deferredStmts.add(stmt);
-
- if (log.isDebugEnabled())
- log.debug("deferred: "+stmt);
-
- }
+ if (reifiedStmt.isFullyBound(arity)) {
+
+ sid.setStatement(reifiedStmt.toStatement(valueFactory));
+
+ reifiedStmts.remove(sid);
+
+ }
- } else {
+ return;
- // add to the buffer.
- stmts[numStmts++] = stmt;
+ }
+// else {
+//
+// if (deferredStmts == null) {
+//
+// deferredStmts = new HashSet<BigdataStatement>(stmts.length);
+//
+// }
+//
+// deferredStmts.add(stmt);
+//
+// if (log.isDebugEnabled())
+// log.debug("deferred: "+stmt);
+//
+// }
+//
+// } else {
- }
-
}
-
- /*
- * Update counters.
- */
- if (!duplicateS) {// && ((_Value) s).termId == 0L) {
+ if (statementIdentifiers && s instanceof BNode &&
+ equals(o, RDF_STATEMENT) && equals(p, RDF_TYPE)) {
+
+ // ignore this statement
+
+ return;
+
+ }
+
+ // add to the buffer.
+ stmts[numStmts++] = stmt;
- if (s instanceof URI) {
+// }
- numURIs++;
-
- values[numValues++] = (BigdataValue) s;
-
- } else {
-
- if (!statementIdentifiers) {
-
- numBNodes++;
-
- values[numValues++] = (BigdataValue) s;
-
- }
-
- }
-
+ if (c != null && statementIdentifiers && c instanceof BNode) {
+
+ ((BigdataBNodeImpl) c).setStatement(stmt);
+
}
- if (!duplicateP) {//&& ((_Value) s).termId == 0L) {
-
- values[numValues++] = (BigdataValue)p;
-
- numURIs++;
-
- }
-
- if (!duplicateO) {// && ((_Value) s).termId == 0L) {
-
- if (o instanceof URI) {
-
- numURIs++;
-
- values[numValues++] = (BigdataValue) o;
-
- } else if (o instanceof BNode) {
-
- if (!statementIdentifiers) {
-
- numBNodes++;
-
- values[numValues++] = (BigdataValue) o;
-
- }
-
- } else {
-
- numLiterals++;
-
- values[numValues++] = (BigdataValue) o;
-
- }
-
- }
-
- if (c != null && !duplicateC && ((BigdataValue) c).getIV() == null) {
-
- if (c instanceof URI) {
-
- numURIs++;
-
- values[numValues++] = (BigdataValue) c;
-
- } else {
-
- if (!database.getStatementIdentifiers()) {
-
- /*
- * We only let the context node into the buffer when
- * statement identifiers are disabled for the database.
- *
- * Note: This does NOT test [statementIdentifiers] as that
- * flag is temporarily overriden when processing deferred
- * statements.
- */
-
- values[numValues++] = (BigdataValue) c;
-
- numBNodes++;
-
- } else {
-
- /*
- * Flag the blank node as a statement identifier since it
- * appears in the context position.
- *
- * Note: the blank node is not inserted into values[] since
- * it is a statement identifier and will be assigned when we
- * insert the statement rather than based on the blank
- * node's ID.
- */
-
- // Note: done automatically by setStatement();
-// ((BigdataBNode) c).setStatementIdentifier( true);
- ((BigdataBNodeImpl) c).setStatement(stmt);
-
- }
-
- }
-
- }
-
}
private boolean equals(final BigdataValue v1, final BigdataValue... v2) {
@@ -1541,15 +1528,17 @@
private boolean _equals(final BigdataValue v1, final BigdataValue v2) {
- if (distinct) {
-
- return v1 == v2;
-
- } else {
-
- return v1.equals(v2);
-
- }
+ return v1 == v2;
+
+// if (distinct) {
+//
+// return v1 == v2;
+//
+// } else {
+//
+// return v1.equals(v2);
+//
+// }
}
@@ -1639,6 +1628,12 @@
return "<" + s + ", " + p + ", " + o + ", " + c + ">";
}
+
+ public BigdataStatement toStatement(final BigdataValueFactory vf) {
+
+ return vf.createStatement(s, p, o, c);
+
+ }
}
Modified: branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/ntriples/BigdataNTriplesParser.java
===================================================================
--- branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/ntriples/BigdataNTriplesParser.java 2014-01-24 23:39:36 UTC (rev 7828)
+++ branches/RDR/bigdata-rdf/src/java/com/bigdata/rdf/rio/ntriples/BigdataNTriplesParser.java 2014-01-26 20:37:25 UTC (rev 7829)
@@ -421,31 +421,35 @@
// Create statement.
BigdataStatement st = (BigdataStatement) createStatement(
state.subject, state.predicate, state.object);
- // Resolve against LRU map to blank node for statement.
- BigdataBNode sid = sids.get(st);
- if (sid != null) {
- state.lastSID = sid;
- } else {
- /*
- * Not found.
- *
- * TODO The use of the sid bnode in the context position should
- * go away when we migrate to sids support in both triples and
- * quads mode.
- */
- // New blank node for "sid" of this statement.
- state.lastSID = sid = (BigdataBNode) createBNode();
- // New statement using that "sid" as its context position.
- st = getValueFactory().createStatement(state.subject,
- state.predicate, state.object, sid);
- // cache it.
- sids.put(st,sid);
- // mark this blank node as a "sid".
- // st.setStatementIdentifier(true);
- ((BigdataBNodeImpl) sid).setStatement(st);
- // new statement so pass to the call back interface.
- rdfHandler.handleStatement(st);
- }
+
+ state.lastSID = ((BigdataValueFactory) valueFactory).createBNode(st);
+
+// // Resolve against LRU map to blank node for statement.
+// BigdataBNode sid = sids.get(st);
+// if (sid != null) {
+// state.lastSID = sid;
+// } else {
+// /*
+// * Not found.
+// *
+// * TODO The use of the sid bnode in the context position should
+// * go away when we migrate to sids support in both triples and
+// * quads mode.
+// */
+// // New blank node for "sid" of this statement.
+// state.lastSID = sid = (BigdataBNode) createBNode();
+// // New statement using that "sid" as its context position.
+// st = getValueFactory().createStatement(state.subject,
+// state.predicate, state.object, sid);
+// // cache it.
+// sids.put(st,sid);
+// // mark this blank node as a "sid".
+// // st.setStatementIdentifier(true);
+// ((BigdataBNodeImpl) sid).setStatement(st);
+// // new statement so pass to the call back interface.
+// rdfHandler.handleStatement(st);
+// }
+ ...
[truncated message content] |
