8199329: Remove code that attempts to read bytes after connection reset reported
Reviewed-by: redestad, clanger, chegar
/*
* Copyright (c) 2013, 2017, Oracle and/or its affiliates. All rights reserved.
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* This code is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License version 2 only, as
* published by the Free Software Foundation. Oracle designates this
* particular file as subject to the "Classpath" exception as provided
* by Oracle in the LICENSE file that accompanied this code.
*
* This code 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
* version 2 for more details (a copy is included in the LICENSE file that
* accompanied this code).
*
* You should have received a copy of the GNU General Public License version
* 2 along with this work; if not, write to the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
*
* Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
* or visit www.oracle.com if you need additional information or have any
* questions.
*/
package java.net;
import java.io.ObjectInputStream;
import java.io.IOException;
import java.util.List;
import java.util.ArrayList;
import java.util.Collections;
import java.security.Permission;
/**
* Represents permission to access a resource or set of resources defined by a
* given url, and for a given set of user-settable request methods
* and request headers. The <i>name</i> of the permission is the url string.
* The <i>actions</i> string is a concatenation of the request methods and headers.
* The range of method and header names is not restricted by this class.
* <p><b>The url</b><p>
* The url string has the following expected structure.
* <pre>
* scheme : // authority [ / path ]
* </pre>
* <i>scheme</i> will typically be http or https, but is not restricted by this
* class.
* <i>authority</i> is specified as:
* <pre>
* authority = [ userinfo @ ] hostrange [ : portrange ]
* portrange = portnumber | -portnumber | portnumber-[portnumber] | *
* hostrange = ([*.] dnsname) | IPv4address | IPv6address
* </pre>
* <i>dnsname</i> is a standard DNS host or domain name, ie. one or more labels
* separated by ".". <i>IPv4address</i> is a standard literal IPv4 address and
* <i>IPv6address</i> is as defined in <a href="http://www.ietf.org/rfc/rfc2732.txt">
* RFC 2732</a>. Literal IPv6 addresses must however, be enclosed in '[]' characters.
* The <i>dnsname</i> specification can be preceded by "*." which means
* the name will match any hostname whose right-most domain labels are the same as
* this name. For example, "*.oracle.com" matches "foo.bar.oracle.com"
* <p>
* <i>portrange</i> is used to specify a port number, or a bounded or unbounded range of ports
* that this permission applies to. If portrange is absent or invalid, then a default
* port number is assumed if the scheme is {@code http} (default 80) or {@code https}
* (default 443). No default is assumed for other schemes. A wildcard may be specified
* which means all ports.
* <p>
* <i>userinfo</i> is optional. A userinfo component if present, is ignored when
* creating a URLPermission, and has no effect on any other methods defined by this class.
* <p>
* The <i>path</i> component comprises a sequence of path segments,
* separated by '/' characters. <i>path</i> may also be empty. The path is specified
* in a similar way to the path in {@link java.io.FilePermission}. There are
* three different ways as the following examples show:
* <table class="striped">
* <caption>URL Examples</caption>
* <thead>
* <tr><th scope="col">Example url</th><th scope="col">Description</th></tr>
* </thead>
* <tbody style="text-align:left">
* <tr><th scope="row" style="white-space:nowrap;">http://www.oracle.com/a/b/c.html</th>
* <td>A url which identifies a specific (single) resource</td>
* </tr>
* <tr><th scope="row">http://www.oracle.com/a/b/*</th>
* <td>The '*' character refers to all resources in the same "directory" - in
* other words all resources with the same number of path components, and
* which only differ in the final path component, represented by the '*'.
* </td>
* </tr>
* <tr><th scope="row">http://www.oracle.com/a/b/-</th>
* <td>The '-' character refers to all resources recursively below the
* preceding path (eg. http://www.oracle.com/a/b/c/d/e.html matches this
* example).
* </td>
* </tr>
* </tbody>
* </table>
* <p>
* The '*' and '-' may only be specified in the final segment of a path and must be
* the only character in that segment. Any query or fragment components of the
* url are ignored when constructing URLPermissions.
* <p>
* As a special case, urls of the form, "scheme:*" are accepted to
* mean any url of the given scheme.
* <p>
* The <i>scheme</i> and <i>authority</i> components of the url string are handled
* without regard to case. This means {@link #equals(Object)},
* {@link #hashCode()} and {@link #implies(Permission)} are case insensitive with respect
* to these components. If the <i>authority</i> contains a literal IP address,
* then the address is normalized for comparison. The path component is case sensitive.
* <p><b>The actions string</b><p>
* The actions string of a URLPermission is a concatenation of the <i>method list</i>
* and the <i>request headers list</i>. These are lists of the permitted request
* methods and permitted request headers of the permission (respectively). The two lists
* are separated by a colon ':' character and elements of each list are comma separated.
* Some examples are:
* <ul>
* <li>"POST,GET,DELETE"
* <li>"GET:X-Foo-Request,X-Bar-Request"
* <li>"POST,GET:Header1,Header2"
* </ul>
* <p>
* The first example specifies the methods: POST, GET and DELETE, but no request headers.
* The second example specifies one request method and two headers. The third
* example specifies two request methods, and two headers.
* <p>
* The colon separator need not be present if the request headers list is empty.
* No white-space is permitted in the actions string. The action strings supplied to
* the URLPermission constructors are case-insensitive and are normalized by converting
* method names to upper-case and header names to the form defines in RFC2616 (lower case
* with initial letter of each word capitalized). Either list can contain a wild-card '*'
* character which signifies all request methods or headers respectively.
* <p>
* Note. Depending on the context of use, some request methods and headers may be permitted
* at all times, and others may not be permitted at any time. For example, the
* HTTP protocol handler might disallow certain headers such as Content-Length
* from being set by application code, regardless of whether the security policy
* in force, permits it.
*
* @since 1.8
*/
public final class URLPermission extends Permission {
private static final long serialVersionUID = -2702463814894478682L;
private transient String scheme;
private transient String ssp; // scheme specific part
private transient String path;
private transient List<String> methods;
private transient List<String> requestHeaders;
private transient Authority authority;
// serialized field
private String actions;
/**
* Creates a new URLPermission from a url string and which permits the given
* request methods and user-settable request headers.
* The name of the permission is the url string it was created with. Only the scheme,
* authority and path components of the url are used internally. Any fragment or query
* components are ignored. The permissions action string is as specified above.
*
* @param url the url string
*
* @param actions the actions string
*
* @exception IllegalArgumentException if url is invalid or if actions contains white-space.
*/
public URLPermission(String url, String actions) {
super(url);
init(actions);
}
private void init(String actions) {
parseURI(getName());
int colon = actions.indexOf(':');
if (actions.lastIndexOf(':') != colon) {
throw new IllegalArgumentException(
"Invalid actions string: \"" + actions + "\"");
}
String methods, headers;
if (colon == -1) {
methods = actions;
headers = "";
} else {
methods = actions.substring(0, colon);
headers = actions.substring(colon+1);
}
List<String> l = normalizeMethods(methods);
Collections.sort(l);
this.methods = Collections.unmodifiableList(l);
l = normalizeHeaders(headers);
Collections.sort(l);
this.requestHeaders = Collections.unmodifiableList(l);
this.actions = actions();
}
/**
* Creates a URLPermission with the given url string and unrestricted
* methods and request headers by invoking the two argument
* constructor as follows: URLPermission(url, "*:*")
*
* @param url the url string
*
* @throws IllegalArgumentException if url does not result in a valid {@link URI}
*/
public URLPermission(String url) {
this(url, "*:*");
}
/**
* Returns the normalized method list and request
* header list, in the form:
* <pre>
* "method-names : header-names"
* </pre>
* <p>
* where method-names is the list of methods separated by commas
* and header-names is the list of permitted headers separated by commas.
* There is no white space in the returned String. If header-names is empty
* then the colon separator may not be present.
*/
public String getActions() {
return actions;
}
/**
* Checks if this URLPermission implies the given permission.
* Specifically, the following checks are done as if in the
* following sequence:
* <ul>
* <li>if 'p' is not an instance of URLPermission return false</li>
* <li>if any of p's methods are not in this's method list, and if
* this's method list is not equal to "*", then return false.</li>
* <li>if any of p's headers are not in this's request header list, and if
* this's request header list is not equal to "*", then return false.</li>
* <li>if this's url scheme is not equal to p's url scheme return false</li>
* <li>if the scheme specific part of this's url is '*' return true</li>
* <li>if the set of hosts defined by p's url hostrange is not a subset of
* this's url hostrange then return false. For example, "*.foo.oracle.com"
* is a subset of "*.oracle.com". "foo.bar.oracle.com" is not
* a subset of "*.foo.oracle.com"</li>
* <li>if the portrange defined by p's url is not a subset of the
* portrange defined by this's url then return false.
* <li>if the path or paths specified by p's url are contained in the
* set of paths specified by this's url, then return true
* <li>otherwise, return false</li>
* </ul>
* <p>Some examples of how paths are matched are shown below:
* <table class="plain">
* <caption>Examples of Path Matching</caption>
* <thead>
* <tr><th scope="col">this's path</th><th scope="col">p's path</th><th>match</th></tr>
* </thead>
* <tbody style="text-align:left">
* <tr><th scope="row">/a/b</th><th scope="row">/a/b</th><td>yes</td></tr>
* <tr><th scope="row" rowspan="3">/a/b/*</th><th scope="row">/a/b/c</th><td>yes</td></tr>
* <tr> <th scope="row">/a/b/c/d</th><td>no</td></tr>
* <tr> <th scope="row">/a/b/c/-</th><td>no</td></tr>
* <tr><th scope="row" rowspan="3">/a/b/-</th><th scope="row">/a/b/c/d</th><td>yes</td></tr>
* <tr> <th scope="row">/a/b/c/d/e</th><td>yes</td></tr>
* <tr> <th scope="row">/a/b/c/*</th><td>yes</td></tr>
* </tbody>
* </table>
*/
public boolean implies(Permission p) {
if (! (p instanceof URLPermission)) {
return false;
}
URLPermission that = (URLPermission)p;
if (this.methods.isEmpty() && !that.methods.isEmpty()) {
return false;
}
if (!this.methods.isEmpty() &&
!this.methods.get(0).equals("*") &&
Collections.indexOfSubList(this.methods,
that.methods) == -1) {
return false;
}
if (this.requestHeaders.isEmpty() && !that.requestHeaders.isEmpty()) {
return false;
}
if (!this.requestHeaders.isEmpty() &&
!this.requestHeaders.get(0).equals("*") &&
Collections.indexOfSubList(this.requestHeaders,
that.requestHeaders) == -1) {
return false;
}
if (!this.scheme.equals(that.scheme)) {
return false;
}
if (this.ssp.equals("*")) {
return true;
}
if (!this.authority.implies(that.authority)) {
return false;
}
if (this.path == null) {
return that.path == null;
}
if (that.path == null) {
return false;
}
if (this.path.endsWith("/-")) {
String thisprefix = this.path.substring(0, this.path.length() - 1);
return that.path.startsWith(thisprefix);
}
if (this.path.endsWith("/*")) {
String thisprefix = this.path.substring(0, this.path.length() - 1);
if (!that.path.startsWith(thisprefix)) {
return false;
}
String thatsuffix = that.path.substring(thisprefix.length());
// suffix must not contain '/' chars
if (thatsuffix.indexOf('/') != -1) {
return false;
}
if (thatsuffix.equals("-")) {
return false;
}
return true;
}
return this.path.equals(that.path);
}
/**
* Returns true if, this.getActions().equals(p.getActions())
* and p's url equals this's url. Returns false otherwise.
*/
public boolean equals(Object p) {
if (!(p instanceof URLPermission)) {
return false;
}
URLPermission that = (URLPermission)p;
if (!this.scheme.equals(that.scheme)) {
return false;
}
if (!this.getActions().equals(that.getActions())) {
return false;
}
if (!this.authority.equals(that.authority)) {
return false;
}
if (this.path != null) {
return this.path.equals(that.path);
} else {
return that.path == null;
}
}
/**
* Returns a hashcode calculated from the hashcode of the
* actions String and the url string.
*/
public int hashCode() {
return getActions().hashCode()
+ scheme.hashCode()
+ authority.hashCode()
+ (path == null ? 0 : path.hashCode());
}
private List<String> normalizeMethods(String methods) {
List<String> l = new ArrayList<>();
StringBuilder b = new StringBuilder();
for (int i=0; i<methods.length(); i++) {
char c = methods.charAt(i);
if (c == ',') {
String s = b.toString();
if (s.length() > 0)
l.add(s);
b = new StringBuilder();
} else if (c == ' ' || c == '\t') {
throw new IllegalArgumentException(
"White space not allowed in methods: \"" + methods + "\"");
} else {
if (c >= 'a' && c <= 'z') {
c += 'A' - 'a';
}
b.append(c);
}
}
String s = b.toString();
if (s.length() > 0)
l.add(s);
return l;
}
private List<String> normalizeHeaders(String headers) {
List<String> l = new ArrayList<>();
StringBuilder b = new StringBuilder();
boolean capitalizeNext = true;
for (int i=0; i<headers.length(); i++) {
char c = headers.charAt(i);
if (c >= 'a' && c <= 'z') {
if (capitalizeNext) {
c += 'A' - 'a';
capitalizeNext = false;
}
b.append(c);
} else if (c == ' ' || c == '\t') {
throw new IllegalArgumentException(
"White space not allowed in headers: \"" + headers + "\"");
} else if (c == '-') {
capitalizeNext = true;
b.append(c);
} else if (c == ',') {
String s = b.toString();
if (s.length() > 0)
l.add(s);
b = new StringBuilder();
capitalizeNext = true;
} else {
capitalizeNext = false;
b.append(c);
}
}
String s = b.toString();
if (s.length() > 0)
l.add(s);
return l;
}
private void parseURI(String url) {
int len = url.length();
int delim = url.indexOf(':');
if (delim == -1 || delim + 1 == len) {
throw new IllegalArgumentException(
"Invalid URL string: \"" + url + "\"");
}
scheme = url.substring(0, delim).toLowerCase();
this.ssp = url.substring(delim + 1);
if (!ssp.startsWith("//")) {
if (!ssp.equals("*")) {
throw new IllegalArgumentException(
"Invalid URL string: \"" + url + "\"");
}
this.authority = new Authority(scheme, "*");
return;
}
String authpath = ssp.substring(2);
delim = authpath.indexOf('/');
String auth;
if (delim == -1) {
this.path = "";
auth = authpath;
} else {
auth = authpath.substring(0, delim);
this.path = authpath.substring(delim);
}
this.authority = new Authority(scheme, auth.toLowerCase());
}
private String actions() {
// The colon separator is optional when the request headers list is
// empty.This implementation chooses to include it even when the request
// headers list is empty.
return String.join(",", methods) + ":" + String.join(",", requestHeaders);
}
/**
* restore the state of this object from stream
*/
private void readObject(ObjectInputStream s)
throws IOException, ClassNotFoundException {
ObjectInputStream.GetField fields = s.readFields();
String actions = (String)fields.get("actions", null);
init(actions);
}
static class Authority {
HostPortrange p;
Authority(String scheme, String authority) {
int at = authority.indexOf('@');
if (at == -1) {
p = new HostPortrange(scheme, authority);
} else {
p = new HostPortrange(scheme, authority.substring(at+1));
}
}
boolean implies(Authority other) {
return impliesHostrange(other) && impliesPortrange(other);
}
private boolean impliesHostrange(Authority that) {
String thishost = this.p.hostname();
String thathost = that.p.hostname();
if (p.wildcard() && thishost.equals("")) {
// this "*" implies all others
return true;
}
if (that.p.wildcard() && thathost.equals("")) {
// that "*" can only be implied by this "*"
return false;
}
if (thishost.equals(thathost)) {
// covers all cases of literal IP addresses and fixed
// domain names.
return true;
}
if (this.p.wildcard()) {
// this "*.foo.com" implies "bub.bar.foo.com"
return thathost.endsWith(thishost);
}
return false;
}
private boolean impliesPortrange(Authority that) {
int[] thisrange = this.p.portrange();
int[] thatrange = that.p.portrange();
if (thisrange[0] == -1) {
/* port not specified non http/s URL */
return true;
}
return thisrange[0] <= thatrange[0] &&
thisrange[1] >= thatrange[1];
}
boolean equals(Authority that) {
return this.p.equals(that.p);
}
public int hashCode() {
return p.hashCode();
}
}
}