1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/src/share/jaxws_classes/com/sun/tools/internal/ws/wsdl/parser/DOMForest.java Wed Apr 27 01:27:09 2016 +0800
1.3 @@ -0,0 +1,387 @@
1.4 +/*
1.5 + * Copyright (c) 1997, 2013, Oracle and/or its affiliates. All rights reserved.
1.6 + * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
1.7 + *
1.8 + * This code is free software; you can redistribute it and/or modify it
1.9 + * under the terms of the GNU General Public License version 2 only, as
1.10 + * published by the Free Software Foundation. Oracle designates this
1.11 + * particular file as subject to the "Classpath" exception as provided
1.12 + * by Oracle in the LICENSE file that accompanied this code.
1.13 + *
1.14 + * This code is distributed in the hope that it will be useful, but WITHOUT
1.15 + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
1.16 + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
1.17 + * version 2 for more details (a copy is included in the LICENSE file that
1.18 + * accompanied this code).
1.19 + *
1.20 + * You should have received a copy of the GNU General Public License version
1.21 + * 2 along with this work; if not, write to the Free Software Foundation,
1.22 + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
1.23 + *
1.24 + * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
1.25 + * or visit www.oracle.com if you need additional information or have any
1.26 + * questions.
1.27 + */
1.28 +
1.29 +package com.sun.tools.internal.ws.wsdl.parser;
1.30 +
1.31 +import com.sun.istack.internal.NotNull;
1.32 +import com.sun.tools.internal.ws.util.xml.XmlUtil;
1.33 +import com.sun.tools.internal.ws.wscompile.ErrorReceiver;
1.34 +import com.sun.tools.internal.ws.wscompile.WsimportOptions;
1.35 +import com.sun.tools.internal.ws.wsdl.document.schema.SchemaConstants;
1.36 +import com.sun.tools.internal.xjc.reader.internalizer.LocatorTable;
1.37 +import com.sun.xml.internal.bind.marshaller.DataWriter;
1.38 +import org.w3c.dom.Document;
1.39 +import org.w3c.dom.Element;
1.40 +import org.w3c.dom.NodeList;
1.41 +import org.xml.sax.ContentHandler;
1.42 +import org.xml.sax.*;
1.43 +import org.xml.sax.helpers.XMLFilterImpl;
1.44 +
1.45 +import javax.xml.parsers.DocumentBuilder;
1.46 +import javax.xml.parsers.DocumentBuilderFactory;
1.47 +import javax.xml.parsers.ParserConfigurationException;
1.48 +import javax.xml.parsers.SAXParserFactory;
1.49 +import javax.xml.transform.Transformer;
1.50 +import javax.xml.transform.TransformerException;
1.51 +import javax.xml.transform.TransformerFactory;
1.52 +import javax.xml.transform.dom.DOMSource;
1.53 +import javax.xml.transform.sax.SAXResult;
1.54 +import java.io.IOException;
1.55 +import java.io.InputStream;
1.56 +import java.io.OutputStream;
1.57 +import java.io.OutputStreamWriter;
1.58 +import java.net.*;
1.59 +import java.util.*;
1.60 +
1.61 +/**
1.62 + * @author Vivek Pandey
1.63 + */
1.64 +public class DOMForest {
1.65 + /**
1.66 + * To correctly feed documents to a schema parser, we need to remember
1.67 + * which documents (of the forest) were given as the root
1.68 + * documents, and which of them are read as included/imported
1.69 + * documents.
1.70 + * <p/>
1.71 + * <p/>
1.72 + * Set of system ids as strings.
1.73 + */
1.74 + protected final Set<String> rootDocuments = new HashSet<String>();
1.75 +
1.76 + /**
1.77 + * Contains wsdl:import(s)
1.78 + */
1.79 + protected final Set<String> externalReferences = new HashSet<String>();
1.80 +
1.81 + /**
1.82 + * actual data storage map<SystemId,Document>.
1.83 + */
1.84 + protected final Map<String, Document> core = new HashMap<String, Document>();
1.85 + protected final ErrorReceiver errorReceiver;
1.86 +
1.87 + private final DocumentBuilder documentBuilder;
1.88 + private final SAXParserFactory parserFactory;
1.89 +
1.90 + /**
1.91 + * inlined schema elements inside wsdl:type section
1.92 + */
1.93 + protected final List<Element> inlinedSchemaElements = new ArrayList<Element>();
1.94 +
1.95 +
1.96 + /**
1.97 + * Stores location information for all the trees in this forest.
1.98 + */
1.99 + public final LocatorTable locatorTable = new LocatorTable();
1.100 +
1.101 + protected final EntityResolver entityResolver;
1.102 + /**
1.103 + * Stores all the outer-most <jaxb:bindings> customizations.
1.104 + */
1.105 + public final Set<Element> outerMostBindings = new HashSet<Element>();
1.106 +
1.107 + /**
1.108 + * Schema language dependent part of the processing.
1.109 + */
1.110 + protected final InternalizationLogic logic;
1.111 + protected final WsimportOptions options;
1.112 +
1.113 + public DOMForest(InternalizationLogic logic, @NotNull EntityResolver entityResolver, WsimportOptions options, ErrorReceiver errReceiver) {
1.114 + this.options = options;
1.115 + this.entityResolver = entityResolver;
1.116 + this.errorReceiver = errReceiver;
1.117 + this.logic = logic;
1.118 + try {
1.119 + // secure xml processing can be switched off if input requires it
1.120 + boolean secureProcessingEnabled = options == null || !options.disableXmlSecurity;
1.121 + DocumentBuilderFactory dbf = XmlUtil.newDocumentBuilderFactory(secureProcessingEnabled);
1.122 + dbf.setNamespaceAware(true);
1.123 + this.documentBuilder = dbf.newDocumentBuilder();
1.124 +
1.125 + this.parserFactory = XmlUtil.newSAXParserFactory(secureProcessingEnabled);
1.126 + this.parserFactory.setNamespaceAware(true);
1.127 + } catch (ParserConfigurationException e) {
1.128 + throw new AssertionError(e);
1.129 + }
1.130 + }
1.131 +
1.132 + public List<Element> getInlinedSchemaElement() {
1.133 + return inlinedSchemaElements;
1.134 + }
1.135 +
1.136 + public @NotNull Document parse(InputSource source, boolean root) throws SAXException, IOException {
1.137 + if (source.getSystemId() == null)
1.138 + throw new IllegalArgumentException();
1.139 + return parse(source.getSystemId(), source, root);
1.140 + }
1.141 +
1.142 + /**
1.143 + * Parses an XML at the given location (
1.144 + * and XMLs referenced by it) into DOM trees
1.145 + * and stores them to this forest.
1.146 + *
1.147 + * @return the parsed DOM document object.
1.148 + */
1.149 + public Document parse(String systemId, boolean root) throws SAXException, IOException{
1.150 +
1.151 + systemId = normalizeSystemId(systemId);
1.152 +
1.153 + InputSource is = null;
1.154 +
1.155 + // allow entity resolver to find the actual byte stream.
1.156 + is = entityResolver.resolveEntity(null, systemId);
1.157 + if (is == null)
1.158 + is = new InputSource(systemId);
1.159 + else {
1.160 + resolvedCache.put(systemId, is.getSystemId());
1.161 + systemId=is.getSystemId();
1.162 + }
1.163 +
1.164 + if (core.containsKey(systemId)) {
1.165 + // this document has already been parsed. Just ignore.
1.166 + return core.get(systemId);
1.167 + }
1.168 +
1.169 + if(!root)
1.170 + addExternalReferences(systemId);
1.171 +
1.172 + // but we still use the original system Id as the key.
1.173 + return parse(systemId, is, root);
1.174 + }
1.175 + protected Map<String,String> resolvedCache = new HashMap<String,String>();
1.176 +
1.177 + public Map<String,String> getReferencedEntityMap() {
1.178 + return resolvedCache;
1.179 + }
1.180 + /**
1.181 + * Parses the given document and add it to the DOM forest.
1.182 + *
1.183 + * @return null if there was a parse error. otherwise non-null.
1.184 + */
1.185 + private @NotNull Document parse(String systemId, InputSource inputSource, boolean root) throws SAXException, IOException{
1.186 + Document dom = documentBuilder.newDocument();
1.187 +
1.188 + systemId = normalizeSystemId(systemId);
1.189 +
1.190 + // put into the map before growing a tree, to
1.191 + // prevent recursive reference from causing infinite loop.
1.192 + core.put(systemId, dom);
1.193 +
1.194 + dom.setDocumentURI(systemId);
1.195 + if (root)
1.196 + rootDocuments.add(systemId);
1.197 +
1.198 + try {
1.199 + XMLReader reader = createReader(dom);
1.200 +
1.201 + InputStream is = null;
1.202 + if(inputSource.getByteStream() == null){
1.203 + inputSource = entityResolver.resolveEntity(null, systemId);
1.204 + }
1.205 + reader.parse(inputSource);
1.206 + Element doc = dom.getDocumentElement();
1.207 + if (doc == null) {
1.208 + return null;
1.209 + }
1.210 + NodeList schemas = doc.getElementsByTagNameNS(SchemaConstants.NS_XSD, "schema");
1.211 + for (int i = 0; i < schemas.getLength(); i++) {
1.212 + inlinedSchemaElements.add((Element) schemas.item(i));
1.213 + }
1.214 + } catch (ParserConfigurationException e) {
1.215 + errorReceiver.error(e);
1.216 + throw new SAXException(e.getMessage());
1.217 + }
1.218 + resolvedCache.put(systemId, dom.getDocumentURI());
1.219 + return dom;
1.220 + }
1.221 +
1.222 + public void addExternalReferences(String ref) {
1.223 + if (!externalReferences.contains(ref))
1.224 + externalReferences.add(ref);
1.225 + }
1.226 +
1.227 +
1.228 + public Set<String> getExternalReferences() {
1.229 + return externalReferences;
1.230 + }
1.231 +
1.232 +
1.233 +
1.234 + public interface Handler extends ContentHandler {
1.235 + /**
1.236 + * Gets the DOM that was built.
1.237 + */
1.238 + public Document getDocument();
1.239 + }
1.240 +
1.241 + /**
1.242 + * Returns a {@link org.xml.sax.XMLReader} to parse a document into this DOM forest.
1.243 + * <p/>
1.244 + * This version requires that the DOM object to be created and registered
1.245 + * to the map beforehand.
1.246 + */
1.247 + private XMLReader createReader(Document dom) throws SAXException, ParserConfigurationException {
1.248 + XMLReader reader = parserFactory.newSAXParser().getXMLReader();
1.249 + DOMBuilder dombuilder = new DOMBuilder(dom, locatorTable, outerMostBindings);
1.250 + try {
1.251 + reader.setProperty("http://xml.org/sax/properties/lexical-handler", dombuilder);
1.252 + } catch(SAXException e) {
1.253 + errorReceiver.debug(e.getMessage());
1.254 + }
1.255 +
1.256 + ContentHandler handler = new WhitespaceStripper(dombuilder, errorReceiver, entityResolver);
1.257 + handler = new VersionChecker(handler, errorReceiver, entityResolver);
1.258 +
1.259 + // insert the reference finder so that
1.260 + // included/imported schemas will be also parsed
1.261 + XMLFilterImpl f = logic.createExternalReferenceFinder(this);
1.262 + f.setContentHandler(handler);
1.263 + if (errorReceiver != null)
1.264 + f.setErrorHandler(errorReceiver);
1.265 + f.setEntityResolver(entityResolver);
1.266 +
1.267 + reader.setContentHandler(f);
1.268 + if (errorReceiver != null)
1.269 + reader.setErrorHandler(errorReceiver);
1.270 + reader.setEntityResolver(entityResolver);
1.271 + return reader;
1.272 + }
1.273 +
1.274 + private String normalizeSystemId(String systemId) {
1.275 + try {
1.276 + systemId = new URI(systemId).normalize().toString();
1.277 + } catch (URISyntaxException e) {
1.278 + // leave the system ID untouched. In my experience URI is often too strict
1.279 + }
1.280 + return systemId;
1.281 + }
1.282 +
1.283 + boolean isExtensionMode() {
1.284 + return options.isExtensionMode();
1.285 + }
1.286 +
1.287 +
1.288 + /**
1.289 + * Gets the DOM tree associated with the specified system ID,
1.290 + * or null if none is found.
1.291 + */
1.292 + public Document get(String systemId) {
1.293 + Document doc = core.get(systemId);
1.294 +
1.295 + if (doc == null && systemId.startsWith("file:/") && !systemId.startsWith("file://")) {
1.296 + // As of JDK1.4, java.net.URL.toExternal method returns URLs like
1.297 + // "file:/abc/def/ghi" which is an incorrect file protocol URL according to RFC1738.
1.298 + // Some other correctly functioning parts return the correct URLs ("file:///abc/def/ghi"),
1.299 + // and this descripancy breaks DOM look up by system ID.
1.300 +
1.301 + // this extra check solves this problem.
1.302 + doc = core.get("file://" + systemId.substring(5));
1.303 + }
1.304 +
1.305 + if (doc == null && systemId.startsWith("file:")) {
1.306 + // on Windows, filenames are case insensitive.
1.307 + // perform case-insensitive search for improved user experience
1.308 + String systemPath = getPath(systemId);
1.309 + for (String key : core.keySet()) {
1.310 + if (key.startsWith("file:") && getPath(key).equalsIgnoreCase(systemPath)) {
1.311 + doc = core.get(key);
1.312 + break;
1.313 + }
1.314 + }
1.315 + }
1.316 +
1.317 + return doc;
1.318 + }
1.319 +
1.320 + /**
1.321 + * Strips off the leading 'file:///' portion from an URL.
1.322 + */
1.323 + private String getPath(String key) {
1.324 + key = key.substring(5); // skip 'file:'
1.325 + while (key.length() > 0 && key.charAt(0) == '/')
1.326 + key = key.substring(1);
1.327 + return key;
1.328 + }
1.329 +
1.330 + /**
1.331 + * Gets all the system IDs of the documents.
1.332 + */
1.333 + public String[] listSystemIDs() {
1.334 + return core.keySet().toArray(new String[core.keySet().size()]);
1.335 + }
1.336 +
1.337 + /**
1.338 + * Gets the system ID from which the given DOM is parsed.
1.339 + * <p/>
1.340 + * Poor-man's base URI.
1.341 + */
1.342 + public String getSystemId(Document dom) {
1.343 + for (Map.Entry<String, Document> e : core.entrySet()) {
1.344 + if (e.getValue() == dom)
1.345 + return e.getKey();
1.346 + }
1.347 + return null;
1.348 + }
1.349 +
1.350 + /**
1.351 + * Gets the first one (which is more or less random) in {@link #rootDocuments}.
1.352 + */
1.353 + public String getFirstRootDocument() {
1.354 + if(rootDocuments.isEmpty()) return null;
1.355 + return rootDocuments.iterator().next();
1.356 + }
1.357 +
1.358 + public Set<String> getRootDocuments() {
1.359 + return rootDocuments;
1.360 + }
1.361 +
1.362 + /**
1.363 + * Dumps the contents of the forest to the specified stream.
1.364 + * <p/>
1.365 + * This is a debug method. As such, error handling is sloppy.
1.366 + */
1.367 + public void dump(OutputStream out) throws IOException {
1.368 + try {
1.369 + // create identity transformer
1.370 + // secure xml processing can be switched off if input requires it
1.371 + boolean secureProcessingEnabled = options == null || !options.disableXmlSecurity;
1.372 + TransformerFactory tf = XmlUtil.newTransformerFactory(secureProcessingEnabled);
1.373 + Transformer it = tf.newTransformer();
1.374 +
1.375 + for (Map.Entry<String, Document> e : core.entrySet()) {
1.376 + out.write(("---<< " + e.getKey() + '\n').getBytes());
1.377 +
1.378 + DataWriter dw = new DataWriter(new OutputStreamWriter(out), null);
1.379 + dw.setIndentStep(" ");
1.380 + it.transform(new DOMSource(e.getValue()),
1.381 + new SAXResult(dw));
1.382 +
1.383 + out.write("\n\n\n".getBytes());
1.384 + }
1.385 + } catch (TransformerException e) {
1.386 + e.printStackTrace();
1.387 + }
1.388 + }
1.389 +
1.390 +}
