org.wikidata.wdtk.dumpfiles.MwRevisionDumpFileProcessor Maven / Gradle / Ivy
Show all versions of wdtk-dumpfiles Show documentation
package org.wikidata.wdtk.dumpfiles;
/*
* #%L
* Wikidata Toolkit Dump File Handling
* %%
* Copyright (C) 2014 Wikidata Toolkit Developers
* %%
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* #L%
*/
import java.io.InputStream;
import java.util.HashMap;
import java.util.Map;
import javax.xml.stream.XMLInputFactory;
import javax.xml.stream.XMLStreamConstants;
import javax.xml.stream.XMLStreamException;
import javax.xml.stream.XMLStreamReader;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
/**
* This class processes MediaWiki dumpfiles that contain lists of page revisions
* in the specific XML format used by MediaWiki for exporting pages. It extracts
* all revisions and forwards them to any registered revision processor. The
* class also keeps track of whether or not a certain article respectively
* revision has already been encountered. Therefore, no revision is processed
* twice and the registered revision processors can be informed whether the
* revision is the first of the given article or not. The first revision of an
* article that is encountered in a MediaWiki dump file is usually the most
* recent one. If multiple dump files are processed in reverse chronological
* order, the first revision that is encountered is also the most recent one
* overall.
*
* @author Markus Kroetzsch
*
*/
public class MwRevisionDumpFileProcessor implements MwDumpFileProcessor {
static final String E_MEDIAWIKI = "mediawiki";
static final String E_SITEINFO = "siteinfo";
static final String E_SITENAME = "sitename";
static final String E_BASEURL = "base";
static final String E_NAMESPACE = "namespace";
static final String A_NSKEY = "key";
static final String E_PAGE = "page";
static final String E_PAGE_TITLE = "title";
static final String E_PAGE_ID = "id";
static final String E_PAGE_NAMESPACE = "ns";
static final String E_PAGE_REVISION = "revision";
static final String E_PAGE_REDIRECT = "redirect";
static final String E_REV_ID = "id";
static final String E_REV_PARENT_ID = "parentid";
static final String E_REV_TIMESTAMP = "timestamp";
static final String E_REV_COMMENT = "comment";
static final String E_REV_MODEL = "model";
static final String E_REV_TEXT = "text";
static final String E_REV_CONTRIBUTOR = "contributor";
static final String E_REV_FORMAT = "format";
static final String E_REV_SHA1 = "sha1";
static final String E_REV_MINOR = "minor";
static final String E_CONTRIBUTOR_NAME = "username";
static final String E_CONTRIBUTOR_ID = "id";
static final String E_CONTRIBUTOR_IP = "ip";
static final Logger logger = LoggerFactory
.getLogger(MwRevisionDumpFileProcessor.class);
XMLInputFactory xmlFactory;
XMLStreamReader xmlReader;
/**
* Map from integer namespace ids to namespace prefixes. Namespace strings
* do not include the final ":" used in MediaWiki to separate namespace
* prefixes from article titles. Moreover, the prefixes use spaces, not
* underscores as in MediaWiki URLs.
*/
final Map namespaces;
/**
* Name of the site as set in the dump file.
*/
String sitename = "";
/**
* Base URL of the site as set in the dump file.
*/
String baseUrl = "";
/**
* Object used to store data about the current revision.
*/
final MwRevisionImpl mwRevision;
/**
* Object used to report all revisions to.
*/
final MwRevisionProcessor mwRevisionProcessor;
/**
* Constructor.
*
* @param mwRevisionProcessor
* the revision processor to which all revisions will be reported
*/
public MwRevisionDumpFileProcessor(MwRevisionProcessor mwRevisionProcessor) {
this.xmlFactory = XMLInputFactory.newInstance();
this.namespaces = new HashMap();
this.mwRevision = new MwRevisionImpl();
this.mwRevisionProcessor = mwRevisionProcessor;
reset();
}
/**
* Resets the internal state of the object. All information gathered from
* previously processed dumps and all related statistics will be forgotten.
* If this method is not called, then consecutive invocations of
* {@link #processDumpFileContents(InputStream, MwDumpFile)} will continue
* to add to the internal state. This is useful for processing dumps that
* are split into several parts.
*
* This will not unregister any MwRevisionProcessors.
*/
public void reset() {
this.namespaces.clear();
}
@Override
public void processDumpFileContents(InputStream inputStream,
MwDumpFile dumpFile) {
logger.info("Processing revision dump file " + dumpFile.toString());
this.namespaces.clear();
this.sitename = "";
this.baseUrl = "";
this.xmlReader = null;
try {
this.xmlReader = this.xmlFactory.createXMLStreamReader(inputStream);
processXmlMediawiki();
} catch (XMLStreamException | MwDumpFormatException e) {
MwRevisionDumpFileProcessor.logger.error(e.toString());
} finally { // unfortunately, xmlReader does not implement AutoClosable
if (this.xmlReader != null) {
try {
this.xmlReader.close();
} catch (XMLStreamException e) {
throw new RuntimeException(
"Problem closing XML Reader. This hides an earlier exception.",
e);
}
}
}
this.mwRevisionProcessor.finishRevisionProcessing();
}
/**
* Processes current XML starting from a <mediawiki> start tag up to
* the corresponding end tag. This method uses the current state of
* {@link #xmlReader} and stores its results in according member fields.
* When the method has finished, {@link #xmlReader} will be at the next
* element after the closing tag of this block.
*
* @throws XMLStreamException
* if there was a problem reading the XML or if the XML is
* malformed
* @throws MwDumpFormatException
* if the contents of the XML file did not match our
* expectations of a MediaWiki XML dump
*/
void processXmlMediawiki() throws XMLStreamException, MwDumpFormatException {
while (this.xmlReader.hasNext()) {
switch (this.xmlReader.getEventType()) {
case XMLStreamConstants.START_ELEMENT:
switch (this.xmlReader.getLocalName()) {
case MwRevisionDumpFileProcessor.E_MEDIAWIKI:
break;
case MwRevisionDumpFileProcessor.E_SITEINFO:
processXmlSiteinfo();
this.mwRevisionProcessor.startRevisionProcessing(
this.sitename, this.baseUrl, this.namespaces);
break;
case MwRevisionDumpFileProcessor.E_PAGE:
tryProcessXmlPage();
break;
}
break;
case XMLStreamConstants.END_ELEMENT:
if (!"mediawiki".equals(this.xmlReader.getLocalName())) {
throw new MwDumpFormatException("Unexpected end element .");
}
break;
}
this.xmlReader.next();
}
}
/**
* Processes current XML starting from a <siteinfo> start tag up to
* the corresponding end tag. This method uses the current state of
* {@link #xmlReader} and stores its results in according member fields.
* When the method has finished, {@link #xmlReader} will be at the next
* element after the closing tag of this block.
*
* @throws XMLStreamException
* if there was a problem reading the XML or if the XML is
* malformed
* @throws MwDumpFormatException
* if the contents of the XML file did not match our
* expectations of a MediaWiki XML dump
*/
void processXmlSiteinfo() throws XMLStreamException {
this.xmlReader.next(); // skip current start tag
while (this.xmlReader.hasNext()) {
switch (this.xmlReader.getEventType()) {
case XMLStreamConstants.START_ELEMENT:
switch (xmlReader.getLocalName()) {
case MwRevisionDumpFileProcessor.E_SITENAME:
this.sitename = this.xmlReader.getElementText();
break;
case MwRevisionDumpFileProcessor.E_NAMESPACE:
Integer namespaceKey = new Integer(
this.xmlReader.getAttributeValue(null,
MwRevisionDumpFileProcessor.A_NSKEY));
this.namespaces.put(namespaceKey,
this.xmlReader.getElementText());
break;
case MwRevisionDumpFileProcessor.E_BASEURL:
this.baseUrl = this.xmlReader.getElementText();
break;
}
break;
case XMLStreamConstants.END_ELEMENT:
if (MwRevisionDumpFileProcessor.E_SITEINFO
.equals(this.xmlReader.getLocalName())) {
return;
}
break;
}
this.xmlReader.next();
}
}
/**
* Tries to processes current XML starting from a <page> start tag up
* to the corresponding end tag using {@link #processXmlPage()}. If this
* fails for some reason, it tries to recover to read all remaining page
* blocks nonetheless.
*
* @throws XMLStreamException
* if there was a problem reading the XML
*/
void tryProcessXmlPage() throws XMLStreamException {
try {
processXmlPage();
} catch (MwDumpFormatException e) {
MwRevisionDumpFileProcessor.logger
.error("Error when trying to process revision block for page \""
+ this.mwRevision.getPrefixedTitle()
+ "\" (namespace "
+ this.mwRevision.getNamespace()
+ ", id "
+ this.mwRevision.getPageId()
+ "): "
+ e.toString());
MwRevisionDumpFileProcessor.logger.info("Trying to recover ...");
while (this.xmlReader.hasNext()) {
this.xmlReader.next();
if (this.xmlReader.getEventType() == XMLStreamConstants.END_ELEMENT
&& this.xmlReader.getLocalName().equals(MwRevisionDumpFileProcessor.E_PAGE)) {
MwRevisionDumpFileProcessor.logger
.info("... recovery successful. Continuing processing.");
return;
}
}
MwRevisionDumpFileProcessor.logger
.error("Recovery failed. Could not process remaining XML.");
}
}
/**
* Processes current XML starting from a <page> start tag up to the
* corresponding end tag. This method uses the current state of
* {@link #xmlReader} and stores its results in according member fields.
* When the method has finished, {@link #xmlReader} will be at the next
* element after the closing tag of this block.
*
* @throws XMLStreamException
* if there was a problem reading the XML or if the XML is
* malformed
* @throws MwDumpFormatException
* if the contents of the XML file did not match our
* expectations of a MediaWiki XML dump
*/
void processXmlPage() throws XMLStreamException, MwDumpFormatException {
this.mwRevision.resetCurrentPageData();
this.xmlReader.next(); // skip current start tag
while (this.xmlReader.hasNext()) {
switch (this.xmlReader.getEventType()) {
case XMLStreamConstants.START_ELEMENT:
switch (this.xmlReader.getLocalName()) {
case MwRevisionDumpFileProcessor.E_PAGE_TITLE:
this.mwRevision.prefixedTitle = this.xmlReader.getElementText();
break;
case MwRevisionDumpFileProcessor.E_PAGE_NAMESPACE:
this.mwRevision.namespace = new Integer(
this.xmlReader.getElementText());
break;
case MwRevisionDumpFileProcessor.E_PAGE_ID:
this.mwRevision.pageId = new Integer(
this.xmlReader.getElementText());
break;
case MwRevisionDumpFileProcessor.E_PAGE_REVISION:
processXmlRevision();
break;
case MwRevisionDumpFileProcessor.E_PAGE_REDIRECT:
break;
default:
throw new MwDumpFormatException("Unexpected element \""
+ this.xmlReader.getLocalName() + "\" in page.");
}
break;
case XMLStreamConstants.END_ELEMENT:
if (MwRevisionDumpFileProcessor.E_PAGE.equals(xmlReader
.getLocalName())) {
return;
}
break;
}
this.xmlReader.next();
}
}
/**
* Processes current XML starting from a <revision> start tag up to
* the corresponding end tag. This method uses the current state of
* {@link #xmlReader} and stores its results in according member fields.
* When the method has finished, {@link #xmlReader} will be at the next
* element after the closing tag of this block.
*
* @throws XMLStreamException
* if there was a problem reading the XML or if the XML is
* malformed
* @throws MwDumpFormatException
* if the contents of the XML file did not match our
* expectations of a MediaWiki XML dump
*/
void processXmlRevision() throws XMLStreamException, MwDumpFormatException {
this.mwRevision.resetCurrentRevisionData();
this.xmlReader.next(); // skip current start tag
while (this.xmlReader.hasNext()) {
switch (this.xmlReader.getEventType()) {
case XMLStreamConstants.START_ELEMENT:
switch (this.xmlReader.getLocalName()) {
case MwRevisionDumpFileProcessor.E_REV_COMMENT:
this.mwRevision.comment = this.xmlReader.getElementText();
break;
case MwRevisionDumpFileProcessor.E_REV_TEXT:
this.mwRevision.text = this.xmlReader.getElementText();
break;
case MwRevisionDumpFileProcessor.E_REV_TIMESTAMP:
this.mwRevision.timeStamp = this.xmlReader.getElementText();
break;
case MwRevisionDumpFileProcessor.E_REV_FORMAT:
this.mwRevision.format = this.xmlReader.getElementText();
break;
case MwRevisionDumpFileProcessor.E_REV_MODEL:
this.mwRevision.model = this.xmlReader.getElementText();
break;
case MwRevisionDumpFileProcessor.E_REV_CONTRIBUTOR:
processXmlContributor();
break;
case MwRevisionDumpFileProcessor.E_REV_ID:
this.mwRevision.revisionId = new Long(
this.xmlReader.getElementText());
break;
case MwRevisionDumpFileProcessor.E_REV_PARENT_ID:
case MwRevisionDumpFileProcessor.E_REV_SHA1:
case MwRevisionDumpFileProcessor.E_REV_MINOR:
break;
default:
throw new MwDumpFormatException("Unexpected element \""
+ this.xmlReader.getLocalName() + "\" in revision.");
}
break;
case XMLStreamConstants.END_ELEMENT:
if (MwRevisionDumpFileProcessor.E_PAGE_REVISION
.equals(this.xmlReader.getLocalName())) {
this.mwRevisionProcessor.processRevision(this.mwRevision);
return;
}
break;
}
this.xmlReader.next();
}
}
/**
* Processes current XML starting from a <contributor> start tag up to
* the corresponding end tag. This method uses the current state of
* {@link #xmlReader} and stores its results in according member fields.
* When the method has finished, {@link #xmlReader} will be at the next
* element after the closing tag of this block.
*
* @throws XMLStreamException
* if there was a problem reading the XML or if the XML is
* malformed
* @throws MwDumpFormatException
* if the contents of the XML file did not match our
* expectations of a MediaWiki XML dump
*/
void processXmlContributor() throws XMLStreamException,
MwDumpFormatException {
this.xmlReader.next(); // skip current start tag
while (this.xmlReader.hasNext()) {
switch (this.xmlReader.getEventType()) {
case XMLStreamConstants.START_ELEMENT:
switch (this.xmlReader.getLocalName()) {
case MwRevisionDumpFileProcessor.E_CONTRIBUTOR_NAME:
this.mwRevision.contributor = this.xmlReader
.getElementText();
break;
case MwRevisionDumpFileProcessor.E_CONTRIBUTOR_ID:
this.mwRevision.contributorId = new Integer(
this.xmlReader.getElementText());
break;
case MwRevisionDumpFileProcessor.E_CONTRIBUTOR_IP:
this.mwRevision.contributor = this.xmlReader
.getElementText();
this.mwRevision.contributorId = -1;
break;
default:
throw new MwDumpFormatException("Unexpected element \""
+ this.xmlReader.getLocalName()
+ "\" in contributor.");
}
break;
case XMLStreamConstants.END_ELEMENT:
if (MwRevisionDumpFileProcessor.E_REV_CONTRIBUTOR
.equals(this.xmlReader.getLocalName())) {
return;
}
break;
}
this.xmlReader.next();
}
}
}