1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/src/share/jaxws_classes/com/sun/xml/internal/dtdparser/Resolver.java Wed Apr 27 01:27:09 2016 +0800
1.3 @@ -0,0 +1,448 @@
1.4 +/*
1.5 + * Copyright (c) 2009, 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.xml.internal.dtdparser;
1.30 +
1.31 +import org.xml.sax.EntityResolver;
1.32 +import org.xml.sax.InputSource;
1.33 +
1.34 +import java.io.File;
1.35 +import java.io.FileInputStream;
1.36 +import java.io.IOException;
1.37 +import java.io.InputStream;
1.38 +import java.net.URL;
1.39 +import java.net.URLConnection;
1.40 +import java.util.Hashtable;
1.41 +
1.42 +/**
1.43 + * This entity resolver class provides a number of utilities which can help
1.44 + * managment of external parsed entities in XML. These are commonly used
1.45 + * to hold markup declarations that are to be used as part of a Document
1.46 + * Type Declaration (DTD), or to hold text marked up with XML.
1.47 + * <p/>
1.48 + * <P> Features include: <UL>
1.49 + * <p/>
1.50 + * <LI> Static factory methods are provided for constructing SAX InputSource
1.51 + * objects from Files, URLs, or MIME objects. This eliminates a class of
1.52 + * error-prone coding in applications.
1.53 + * <p/>
1.54 + * <LI> Character encodings for XML documents are correctly supported: <UL>
1.55 + * <p/>
1.56 + * <LI> The encodings defined in the RFCs for MIME content types
1.57 + * (2046 for general MIME, and 2376 for XML in particular), are
1.58 + * supported, handling <em>charset=...</em> attributes and accepting
1.59 + * content types which are known to be safe for use with XML;
1.60 + * <p/>
1.61 + * <LI> The character encoding autodetection algorithm identified
1.62 + * in the XML specification is used, and leverages all of
1.63 + * the JDK 1.1 (and later) character encoding support.
1.64 + * <p/>
1.65 + * <LI> The use of MIME typing may optionally be disabled, forcing the
1.66 + * use of autodetection, to support web servers which don't correctly
1.67 + * report MIME types for XML. For example, they may report text that
1.68 + * is encoded in EUC-JP as being US-ASCII text, leading to fatal
1.69 + * errors during parsing.
1.70 + * <p/>
1.71 + * <LI> The InputSource objects returned by this class always
1.72 + * have a <code>java.io.Reader</code> available as the "character
1.73 + * stream" property.
1.74 + * <p/>
1.75 + * </UL>
1.76 + * <p/>
1.77 + * <LI> Catalog entries can map public identifiers to Java resources or
1.78 + * to local URLs. These are used to reduce network dependencies and loads,
1.79 + * and will often be used for external DTD components. For example, packages
1.80 + * shipping DTD files as resources in JAR files can eliminate network traffic
1.81 + * when accessing them, and sites may provide local caches of common DTDs.
1.82 + * Note that no particular catalog syntax is supported by this class, only
1.83 + * the notion of a set of entries.
1.84 + * <p/>
1.85 + * </UL>
1.86 + * <p/>
1.87 + * <P> Subclasses can perform tasks such as supporting new URI schemes for
1.88 + * URIs which are not URLs, such as URNs (see RFC 2396) or for accessing
1.89 + * MIME entities which are part of a <em>multipart/related</em> group
1.90 + * (see RFC 2387). They may also be used to support particular catalog
1.91 + * syntaxes, such as the <a href="http://www.oasis-open.org/html/a401.htm">
1.92 + * SGML/Open Catalog (SOCAT)</a> which supports the SGML notion of "Formal
1.93 + * Public Identifiers (FPIs).
1.94 + *
1.95 + * @author David Brownell
1.96 + * @author Janet Koenig
1.97 + * @version 1.3 00/02/24
1.98 + */
1.99 +public class Resolver implements EntityResolver {
1.100 + private boolean ignoringMIME;
1.101 +
1.102 + // table mapping public IDs to (local) URIs
1.103 + private Hashtable id2uri;
1.104 +
1.105 + // tables mapping public IDs to resources and classloaders
1.106 + private Hashtable id2resource;
1.107 + private Hashtable id2loader;
1.108 +
1.109 + //
1.110 + // table of MIME content types (less attributes!) known
1.111 + // to be mostly "OK" to use with XML MIME entities. the
1.112 + // idea is to rule out obvious braindamage ("image/jpg")
1.113 + // not the subtle stuff ("text/html") that might actually
1.114 + // be (or become) safe.
1.115 + //
1.116 + private static final String types [] = {
1.117 + "application/xml",
1.118 + "text/xml",
1.119 + "text/plain",
1.120 + "text/html", // commonly mis-inferred
1.121 + "application/x-netcdf", // this is often illegal XML
1.122 + "content/unknown"
1.123 + };
1.124 +
1.125 + /**
1.126 + * Constructs a resolver.
1.127 + */
1.128 + public Resolver() {
1.129 + }
1.130 +
1.131 + /**
1.132 + * Returns an input source, using the MIME type information and URL
1.133 + * scheme to statically determine the correct character encoding if
1.134 + * possible and otherwise autodetecting it. MIME carefully specifies
1.135 + * the character encoding defaults, and how attributes of the content
1.136 + * type can change it. XML further specifies two mandatory encodings
1.137 + * (UTF-8 and UTF-16), and includes an XML declaration which can be
1.138 + * used to internally label most documents encoded using US-ASCII
1.139 + * supersets (such as Shift_JIS, EUC-JP, ISO-2022-*, ISO-8859-*, and
1.140 + * more).
1.141 + * <p/>
1.142 + * <P> This method can be used to access XML documents which do not
1.143 + * have URIs (such as servlet input streams, or most JavaMail message
1.144 + * entities) and to support access methods such as HTTP POST or PUT.
1.145 + * (URLs normally return content using the GET method.)
1.146 + * <p/>
1.147 + * <P> <em> The caller should set the system ID in order for relative URIs
1.148 + * found in this document to be interpreted correctly.</em> In some cases,
1.149 + * a custom resolver will need to be used; for example, documents
1.150 + * may be grouped in a single MIME "multipart/related" bundle, and
1.151 + * relative URLs would refer to other documents in that bundle.
1.152 + *
1.153 + * @param contentType The MIME content type for the source for which
1.154 + * an InputSource is desired, such as <em>text/xml;charset=utf-8</em>.
1.155 + * @param stream The input byte stream for the input source.
1.156 + * @param checkType If true, this verifies that the content type is known
1.157 + * to support XML documents, such as <em>application/xml</em>.
1.158 + * @param scheme Unless this is "file", unspecified MIME types
1.159 + * default to US-ASCII. Files are always autodetected since most
1.160 + * file systems discard character encoding information.
1.161 + */
1.162 + public static InputSource createInputSource(String contentType,
1.163 + InputStream stream,
1.164 + boolean checkType,
1.165 + String scheme) throws IOException {
1.166 + InputSource retval;
1.167 + String charset = null;
1.168 +
1.169 + if (contentType != null) {
1.170 + int index;
1.171 +
1.172 + contentType = contentType.toLowerCase();
1.173 + index = contentType.indexOf(';');
1.174 + if (index != -1) {
1.175 + String attributes;
1.176 +
1.177 + attributes = contentType.substring(index + 1);
1.178 + contentType = contentType.substring(0, index);
1.179 +
1.180 + // use "charset=..." if it's available
1.181 + index = attributes.indexOf("charset");
1.182 + if (index != -1) {
1.183 + attributes = attributes.substring(index + 7);
1.184 + // strip out subsequent attributes
1.185 + if ((index = attributes.indexOf(';')) != -1)
1.186 + attributes = attributes.substring(0, index);
1.187 + // find start of value
1.188 + if ((index = attributes.indexOf('=')) != -1) {
1.189 + attributes = attributes.substring(index + 1);
1.190 + // strip out rfc822 comments
1.191 + if ((index = attributes.indexOf('(')) != -1)
1.192 + attributes = attributes.substring(0, index);
1.193 + // double quotes are optional
1.194 + if ((index = attributes.indexOf('"')) != -1) {
1.195 + attributes = attributes.substring(index + 1);
1.196 + attributes = attributes.substring(0,
1.197 + attributes.indexOf('"'));
1.198 + }
1.199 + charset = attributes.trim();
1.200 + // XXX "\;", "\)" etc were mishandled above
1.201 + }
1.202 + }
1.203 + }
1.204 +
1.205 + //
1.206 + // Check MIME type.
1.207 + //
1.208 + if (checkType) {
1.209 + boolean isOK = false;
1.210 + for (int i = 0; i < types.length; i++)
1.211 + if (types[i].equals(contentType)) {
1.212 + isOK = true;
1.213 + break;
1.214 + }
1.215 + if (!isOK)
1.216 + throw new IOException("Not XML: " + contentType);
1.217 + }
1.218 +
1.219 + //
1.220 + // "text/*" MIME types have hard-wired character set
1.221 + // defaults, as specified in the RFCs. For XML, we
1.222 + // ignore the system "file.encoding" property since
1.223 + // autodetection is more correct.
1.224 + //
1.225 + if (charset == null) {
1.226 + contentType = contentType.trim();
1.227 + if (contentType.startsWith("text/")) {
1.228 + if (!"file".equalsIgnoreCase(scheme))
1.229 + charset = "US-ASCII";
1.230 + }
1.231 + // "application/*" has no default
1.232 + }
1.233 + }
1.234 +
1.235 + retval = new InputSource(XmlReader.createReader(stream, charset));
1.236 + retval.setByteStream(stream);
1.237 + retval.setEncoding(charset);
1.238 + return retval;
1.239 + }
1.240 +
1.241 +
1.242 + /**
1.243 + * Creates an input source from a given URI.
1.244 + *
1.245 + * @param uri the URI (system ID) for the entity
1.246 + * @param checkType if true, the MIME content type for the entity
1.247 + * is checked for document type and character set encoding.
1.248 + */
1.249 + static public InputSource createInputSource(URL uri, boolean checkType)
1.250 + throws IOException {
1.251 +
1.252 + URLConnection conn = uri.openConnection();
1.253 + InputSource retval;
1.254 +
1.255 + if (checkType) {
1.256 + String contentType = conn.getContentType();
1.257 + retval = createInputSource(contentType, conn.getInputStream(),
1.258 + false, uri.getProtocol());
1.259 + } else {
1.260 + retval = new InputSource(XmlReader.createReader(conn.getInputStream()));
1.261 + }
1.262 + retval.setSystemId(conn.getURL().toString());
1.263 + return retval;
1.264 + }
1.265 +
1.266 +
1.267 + /**
1.268 + * Creates an input source from a given file, autodetecting
1.269 + * the character encoding.
1.270 + */
1.271 + static public InputSource createInputSource(File file)
1.272 + throws IOException {
1.273 + InputSource retval;
1.274 + String path;
1.275 +
1.276 + retval = new InputSource(XmlReader.createReader(new FileInputStream(file)));
1.277 +
1.278 + // On JDK 1.2 and later, simplify this:
1.279 + // "path = file.toURL ().toString ()".
1.280 + path = file.getAbsolutePath();
1.281 + if (File.separatorChar != '/')
1.282 + path = path.replace(File.separatorChar, '/');
1.283 + if (!path.startsWith("/"))
1.284 + path = "/" + path;
1.285 + if (!path.endsWith("/") && file.isDirectory())
1.286 + path = path + "/";
1.287 +
1.288 + retval.setSystemId("file:" + path);
1.289 + return retval;
1.290 + }
1.291 +
1.292 +
1.293 + /**
1.294 + * <b>SAX:</b>
1.295 + * Resolve the given entity into an input source. If the name can't
1.296 + * be mapped to a preferred form of the entity, the URI is used. To
1.297 + * resolve the entity, first a local catalog mapping names to URIs is
1.298 + * consulted. If no mapping is found there, a catalog mapping names
1.299 + * to java resources is consulted. Finally, if neither mapping found
1.300 + * a copy of the entity, the specified URI is used.
1.301 + * <p/>
1.302 + * <P> When a URI is used, <a href="#createInputSource">
1.303 + * createInputSource</a> is used to correctly deduce the character
1.304 + * encoding used by this entity. No MIME type checking is done.
1.305 + *
1.306 + * @param name Used to find alternate copies of the entity, when
1.307 + * this value is non-null; this is the XML "public ID".
1.308 + * @param uri Used when no alternate copy of the entity is found;
1.309 + * this is the XML "system ID", normally a URI.
1.310 + */
1.311 + public InputSource resolveEntity(String name, String uri)
1.312 + throws IOException {
1.313 + InputSource retval;
1.314 + String mappedURI = name2uri(name);
1.315 + InputStream stream;
1.316 +
1.317 + // prefer explicit URI mappings, then bundled resources...
1.318 + if (mappedURI == null && (stream = mapResource(name)) != null) {
1.319 + uri = "java:resource:" + (String) id2resource.get(name);
1.320 + retval = new InputSource(XmlReader.createReader(stream));
1.321 +
1.322 + // ...and treat all URIs the same (as URLs for now).
1.323 + } else {
1.324 + URL url;
1.325 + URLConnection conn;
1.326 +
1.327 + if (mappedURI != null)
1.328 + uri = mappedURI;
1.329 + else if (uri == null)
1.330 + return null;
1.331 +
1.332 + url = new URL(uri);
1.333 + conn = url.openConnection();
1.334 + uri = conn.getURL().toString();
1.335 + // System.out.println ("++ URI: " + url);
1.336 + if (ignoringMIME)
1.337 + retval = new InputSource(XmlReader.createReader(conn.getInputStream()));
1.338 + else {
1.339 + String contentType = conn.getContentType();
1.340 + retval = createInputSource(contentType,
1.341 + conn.getInputStream(),
1.342 + false, url.getProtocol());
1.343 + }
1.344 + }
1.345 + retval.setSystemId(uri);
1.346 + retval.setPublicId(name);
1.347 + return retval;
1.348 + }
1.349 +
1.350 +
1.351 + /**
1.352 + * Returns true if this resolver is ignoring MIME types in the documents
1.353 + * it returns, to work around bugs in how servers have reported the
1.354 + * documents' MIME types.
1.355 + */
1.356 + public boolean isIgnoringMIME() {
1.357 + return ignoringMIME;
1.358 + }
1.359 +
1.360 + /**
1.361 + * Tells the resolver whether to ignore MIME types in the documents it
1.362 + * retrieves. Many web servers incorrectly assign text documents a
1.363 + * default character encoding, even when that is incorrect. For example,
1.364 + * all HTTP text documents default to use ISO-8859-1 (used for Western
1.365 + * European languages), and other MIME sources default text documents
1.366 + * to use US-ASCII (a seven bit encoding). For XML documents which
1.367 + * include text encoding declarations (as most should do), these server
1.368 + * bugs can be worked around by ignoring the MIME type entirely.
1.369 + */
1.370 + public void setIgnoringMIME(boolean value) {
1.371 + ignoringMIME = value;
1.372 + }
1.373 +
1.374 +
1.375 + // maps the public ID to an alternate URI, if one is registered
1.376 + private String name2uri(String publicId) {
1.377 + if (publicId == null || id2uri == null)
1.378 + return null;
1.379 + return (String) id2uri.get(publicId);
1.380 + }
1.381 +
1.382 +
1.383 + /**
1.384 + * Registers the given public ID as corresponding to a particular
1.385 + * URI, typically a local copy. This URI will be used in preference
1.386 + * to ones provided as system IDs in XML entity declarations. This
1.387 + * mechanism would most typically be used for Document Type Definitions
1.388 + * (DTDs), where the public IDs are formally managed and versioned.
1.389 + *
1.390 + * @param publicId The managed public ID being mapped
1.391 + * @param uri The URI of the preferred copy of that entity
1.392 + */
1.393 + public void registerCatalogEntry(String publicId,
1.394 + String uri) {
1.395 + if (id2uri == null)
1.396 + id2uri = new Hashtable(17);
1.397 + id2uri.put(publicId, uri);
1.398 + }
1.399 +
1.400 +
1.401 + // return the resource as a stream
1.402 + private InputStream mapResource(String publicId) {
1.403 + // System.out.println ("++ PUBLIC: " + publicId);
1.404 + if (publicId == null || id2resource == null)
1.405 + return null;
1.406 +
1.407 + String resourceName = (String) id2resource.get(publicId);
1.408 + ClassLoader loader = null;
1.409 +
1.410 + if (resourceName == null)
1.411 + return null;
1.412 + // System.out.println ("++ Resource: " + resourceName);
1.413 +
1.414 + if (id2loader != null)
1.415 + loader = (ClassLoader) id2loader.get(publicId);
1.416 + // System.out.println ("++ Loader: " + loader);
1.417 + if (loader == null)
1.418 + return ClassLoader.getSystemResourceAsStream(resourceName);
1.419 + return loader.getResourceAsStream(resourceName);
1.420 + }
1.421 +
1.422 + /**
1.423 + * Registers a given public ID as corresponding to a particular Java
1.424 + * resource in a given class loader, typically distributed with a
1.425 + * software package. This resource will be preferred over system IDs
1.426 + * included in XML documents. This mechanism should most typically be
1.427 + * used for Document Type Definitions (DTDs), where the public IDs are
1.428 + * formally managed and versioned.
1.429 + * <p/>
1.430 + * <P> If a mapping to a URI has been provided, that mapping takes
1.431 + * precedence over this one.
1.432 + *
1.433 + * @param publicId The managed public ID being mapped
1.434 + * @param resourceName The name of the Java resource
1.435 + * @param loader The class loader holding the resource, or null if
1.436 + * it is a system resource.
1.437 + */
1.438 + public void registerCatalogEntry(String publicId,
1.439 + String resourceName,
1.440 + ClassLoader loader) {
1.441 + if (id2resource == null)
1.442 + id2resource = new Hashtable(17);
1.443 + id2resource.put(publicId, resourceName);
1.444 +
1.445 + if (loader != null) {
1.446 + if (id2loader == null)
1.447 + id2loader = new Hashtable(17);
1.448 + id2loader.put(publicId, loader);
1.449 + }
1.450 + }
1.451 +}
