Aside from some minor deviations noted below, an instance of this * class represents a URI reference as defined by * RFC 2396: Uniform * Resource Identifiers (URI): Generic Syntax, amended by RFC 2732: Format for * Literal IPv6 Addresses in URLs. The Literal IPv6 address format * also supports scope_ids. The syntax and usage of scope_ids is described * here. * This class provides constructors for creating URI instances from * their components or by parsing their string forms, methods for accessing the * various components of an instance, and methods for normalizing, resolving, * and relativizing URI instances. Instances of this class are immutable. * * *
* [scheme:]scheme-specific-part[#fragment] ** * where square brackets [...] delineate optional components and the characters * : and # stand for themselves. * *
An absolute URI specifies a scheme; a URI that is not absolute is * said to be relative. URIs are also classified according to whether * they are opaque or hierarchical. * *
An opaque URI is an absolute URI whose scheme-specific part does * not begin with a slash character ('/'). Opaque URIs are not * subject to further parsing. Some examples of opaque URIs are: * *
* **
* mailto:java-net@java.sun.com * news:comp.lang.java * urn:isbn:096139210x
A hierarchical URI is either an absolute URI whose * scheme-specific part begins with a slash character, or a relative URI, that * is, a URI that does not specify a scheme. Some examples of hierarchical * URIs are: * *
* http://java.sun.com/j2se/1.3/* *
* docs/guide/collections/designfaq.html#28
* ../../../demo/jfc/SwingSet2/src/SwingSet2.java
* file:///~/calendar *
A hierarchical URI is subject to further parsing according to the syntax * *
* [scheme:][//authority][path][?query][#fragment] ** * where the characters :, /, * ?, and # stand for themselves. The * scheme-specific part of a hierarchical URI consists of the characters * between the scheme and fragment components. * *
The authority component of a hierarchical URI is, if specified, either * server-based or registry-based. A server-based authority * parses according to the familiar syntax * *
* [user-info@]host[:port] ** * where the characters @ and : stand for * themselves. Nearly all URI schemes currently in use are server-based. An * authority component that does not parse in this way is considered to be * registry-based. * *
The path component of a hierarchical URI is itself said to be absolute * if it begins with a slash character ('/'); otherwise it is * relative. The path of a hierarchical URI that is either absolute or * specifies an authority is always absolute. * *
All told, then, a URI instance has the following nine components: * *
* * In a given instance any particular component is either undefined or * defined with a distinct value. Undefined string components are * represented by null, while undefined integer components are * represented by -1. A string component may be defined to have the * empty string as its value; this is not equivalent to that component being * undefined. * **
* Component Type * scheme String * scheme-specific-part String * authority String * user-info String * host String * port int * path String * query String * fragment String
Whether a particular component is or is not defined in an instance * depends upon the type of the URI being represented. An absolute URI has a * scheme component. An opaque URI has a scheme, a scheme-specific part, and * possibly a fragment, but has no other components. A hierarchical URI always * has a path (though it may be empty) and a scheme-specific-part (which at * least contains the path), and may have any of the other components. If the * authority component is present and is server-based then the host component * will be defined and the user-information and port components may be defined. * * *
Normalization is the process of removing unnecessary "." * and ".." segments from the path component of a hierarchical URI. * Each "." segment is simply removed. A ".." segment is * removed only if it is preceded by a non-".." segment. * Normalization has no effect upon opaque URIs. * *
Resolution is the process of resolving one URI against another, * base URI. The resulting URI is constructed from components of both * URIs in the manner specified by RFC 2396, taking components from the * base URI for those not specified in the original. For hierarchical URIs, * the path of the original is resolved against the path of the base and then * normalized. The result, for example, of resolving * *
* docs/guide/collections/designfaq.html#28 (1) ** * against the base URI http://java.sun.com/j2se/1.3/ is the result * URI * *
* http://java.sun.com/j2se/1.3/docs/guide/collections/designfaq.html#28 ** * Resolving the relative URI * *
* ../../../demo/jfc/SwingSet2/src/SwingSet2.java (2) ** * against this result yields, in turn, * *
* http://java.sun.com/j2se/1.3/demo/jfc/SwingSet2/src/SwingSet2.java ** * Resolution of both absolute and relative URIs, and of both absolute and * relative paths in the case of hierarchical URIs, is supported. Resolving * the URI file:///~calendar against any other URI simply yields the * original URI, since it is absolute. Resolving the relative URI (2) above * against the relative base URI (1) yields the normalized, but still relative, * URI * *
* demo/jfc/SwingSet2/src/SwingSet2.java ** *
Relativization, finally, is the inverse of resolution: For any * two normalized URIs u and v, * *
* u.relativize(u.resolve(v)).equals(v) and* * This operation is often useful when constructing a document containing URIs * that must be made relative to the base URI of the document wherever * possible. For example, relativizing the URI * *
* u.resolve(u.relativize(v)).equals(v) .
*
* http://java.sun.com/j2se/1.3/docs/guide/index.html ** * against the base URI * *
* http://java.sun.com/j2se/1.3 ** * yields the relative URI docs/guide/index.html. * * *
* **
* alpha *The US-ASCII alphabetic characters, * 'A' through 'Z' * and 'a' through 'z' * digit *The US-ASCII decimal digit characters, * '0' through '9' * alphanum *All alpha and digit characters * unreserved *All alphanum characters together with those in the string * "_-!.~'()*" * punct *The characters in the string ",;:$&+=" * reserved *All punct characters together with those in the string * "?/[]@" * escaped *Escaped octets, that is, triplets consisting of the percent * character ('%') followed by two hexadecimal digits * ('0'-'9', 'A'-'F', and * 'a'-'f') * other *The Unicode characters that are not in the US-ASCII character set, * are not control characters (according to the {@link * java.lang.Character#isISOControl(char) Character.isISOControl} * method), and are not space characters (according to the {@link * java.lang.Character#isSpaceChar(char) Character.isSpaceChar} * method) (Deviation from RFC 2396, which is * limited to US-ASCII)
The set of all legal URI characters consists of * the unreserved, reserved, escaped, and other * characters. * * *
To encode non-US-ASCII characters when a URI is required to * conform strictly to RFC 2396 by not containing any other * characters.
To quote characters that are otherwise illegal in a * component. The user-info, path, query, and fragment components differ * slightly in terms of which characters are considered legal and illegal. *
A character is encoded by replacing it * with the sequence of escaped octets that represent that character in the * UTF-8 character set. The Euro currency symbol ('\u20AC'), * for example, is encoded as "%E2%82%AC". (Deviation from * RFC 2396, which does not specify any particular character * set.)
An illegal character is quoted simply by * encoding it. The space character, for example, is quoted by replacing it * with "%20". UTF-8 contains US-ASCII, hence for US-ASCII * characters this transformation has exactly the effect required by * RFC 2396.
* A sequence of escaped octets is decoded by * replacing it with the sequence of characters that it represents in the * UTF-8 character set. UTF-8 contains US-ASCII, hence decoding has the * effect of de-quoting any quoted US-ASCII characters as well as that of * decoding any encoded non-US-ASCII characters. If a decoding error occurs * when decoding the escaped octets then the erroneous octets are replaced by * '\uFFFD', the Unicode replacement character.
The {@link #URI(java.lang.String) single-argument
* constructor} requires any illegal characters in its argument to be
* quoted and preserves any escaped octets and other characters that
* are present.
The {@link
* #URI(java.lang.String,java.lang.String,java.lang.String,int,java.lang.String,java.lang.String,java.lang.String)
* multi-argument constructors} quote illegal characters as
* required by the components in which they appear. The percent character
* ('%') is always quoted by these constructors. Any other
* characters are preserved.
The {@link #getRawUserInfo() getRawUserInfo}, {@link #getRawPath() * getRawPath}, {@link #getRawQuery() getRawQuery}, {@link #getRawFragment() * getRawFragment}, {@link #getRawAuthority() getRawAuthority}, and {@link * #getRawSchemeSpecificPart() getRawSchemeSpecificPart} methods return the * values of their corresponding components in raw form, without interpreting * any escaped octets. The strings returned by these methods may contain * both escaped octets and other characters, and will not contain any * illegal characters.
The {@link #getUserInfo() getUserInfo}, {@link #getPath() * getPath}, {@link #getQuery() getQuery}, {@link #getFragment() * getFragment}, {@link #getAuthority() getAuthority}, and {@link * #getSchemeSpecificPart() getSchemeSpecificPart} methods decode any escaped * octets in their corresponding components. The strings returned by these * methods may contain both other characters and illegal characters, * and will not contain any escaped octets.
The {@link #toString() toString} method returns a URI string with * all necessary quotation but which may contain other characters. *
The {@link #toASCIIString() toASCIIString} method returns a fully * quoted and encoded URI string that does not contain any other * characters.
* new URI(u.toString()).equals(u) . ** * For any URI u that does not contain redundant syntax such as two * slashes before an empty authority (as in file:///tmp/ ) or a * colon following a host name but no port (as in * http://java.sun.com: ), and that does not encode characters * except those that must be quoted, the following identities also hold: * *
* new URI(u.getScheme(),* * in all cases, * *
* u.getSchemeSpecificPart(),
* u.getFragment())
* .equals(u) *
* new URI(u.getScheme(),* * if u is hierarchical, and * *
* u.getUserInfo(), u.getAuthority(),
* u.getPath(), u.getQuery(),
* u.getFragment())
* .equals(u) *
* new URI(u.getScheme(),* * if u is hierarchical and has either no authority or a server-based * authority. * * *
* u.getUserInfo(), u.getHost(), u.getPort(),
* u.getPath(), u.getQuery(),
* u.getFragment())
* .equals(u) *
The conceptual distinction between URIs and URLs is reflected in the * differences between this class and the {@link URL} class. * *
An instance of this class represents a URI reference in the syntactic * sense defined by RFC 2396. A URI may be either absolute or relative. * A URI string is parsed according to the generic syntax without regard to the * scheme, if any, that it specifies. No lookup of the host, if any, is * performed, and no scheme-dependent stream handler is constructed. Equality, * hashing, and comparison are defined strictly in terms of the character * content of the instance. In other words, a URI instance is little more than * a structured string that supports the syntactic, scheme-independent * operations of comparison, normalization, resolution, and relativization. * *
An instance of the {@link URL} class, by contrast, represents the
* syntactic components of a URL together with some of the information required
* to access the resource that it describes. A URL must be absolute, that is,
* it must always specify a scheme. A URL string is parsed according to its
* scheme. A stream handler is always established for a URL, and in fact it is
* impossible to create a URL instance for a scheme for which no handler is
* available. Equality and hashing depend upon both the scheme and the
* Internet address of the host, if any; comparison is not defined. In other
* words, a URL is a structured string that supports the syntactic operation of
* resolution as well as the network I/O operations of looking up the host and
* opening a connection to the specified resource.
*
*
* @author Mark Reinhold
* @since 1.4
*
* @see RFC 2279: UTF-8, a transformation format of ISO 10646
* @see RFC 2373: IPv6 Addressing Architecture
* @see RFC 2396: Uniform Resource Identifiers (URI): Generic Syntax
* @see RFC 2732: Format for Literal IPv6 Addresses in URLs
*/
// Android changed: Reformat @see links.
public final class URI
implements Comparable This constructor parses the given string exactly as specified by the
* grammar in RFC 2396,
* Appendix A, except for the following deviations: An empty authority component is permitted as long as it is
* followed by a non-empty path, a query component, or a fragment
* component. This allows the parsing of URIs such as
* "file:///foo/bar", which seems to be the intent of
* RFC 2396 although the grammar does not permit it. If the
* authority component is empty then the user-information, host, and port
* components are undefined. Empty relative paths are permitted; this seems to be the
* intent of RFC 2396 although the grammar does not permit it. The
* primary consequence of this deviation is that a standalone fragment
* such as "#foo" parses as a relative URI with an empty path
* and the given fragment, and can be usefully resolved against a base URI.
*
* IPv4 addresses in host components are parsed rigorously, as
* specified by RFC 2732: Each
* element of a dotted-quad address must contain no more than three
* decimal digits. Each element is further constrained to have a value
* no greater than 255. Hostnames in host components that comprise only a single
* domain label are permitted to start with an alphanum
* character. This seems to be the intent of RFC 2396
* section 3.2.2 although the grammar does not permit it. The
* consequence of this deviation is that the authority component of a
* hierarchical URI such as s://123, will parse as a server-based
* authority. IPv6 addresses are permitted for the host component. An IPv6
* address must be enclosed in square brackets ('[' and
* ']') as specified by RFC 2732. The
* IPv6 address itself must parse according to RFC 2373. IPv6
* addresses are further constrained to describe no more than sixteen
* bytes of address information, a constraint implicit in RFC 2373
* but not expressible in the grammar. Characters in the other category are permitted wherever
* RFC 2396 permits escaped octets, that is, in the
* user-information, path, query, and fragment components, as well as in
* the authority component if the authority is registry-based. This
* allows URIs to contain Unicode characters beyond those in the US-ASCII
* character set. If a scheme is given then the path, if also given, must either be
* empty or begin with a slash character ('/'). Otherwise a
* component of the new URI may be left undefined by passing null
* for the corresponding parameter or, in the case of the port
* parameter, by passing -1.
*
* This constructor first builds a URI string from the given components
* according to the rules specified in RFC 2396,
* section 5.2, step 7: Initially, the result string is empty. If a scheme is given then it is appended to the result,
* followed by a colon character (':'). If user information, a host, or a port are given then the
* string "//" is appended. If user information is given then it is appended, followed by
* a commercial-at character ('@'). Any character not in the
* unreserved, punct, escaped, or other
* categories is quoted. If a host is given then it is appended. If the host is a
* literal IPv6 address but is not enclosed in square brackets
* ('[' and ']') then the square brackets are added.
* If a port number is given then a colon character
* (':') is appended, followed by the port number in decimal.
* If a path is given then it is appended. Any character not in
* the unreserved, punct, escaped, or other
* categories, and not equal to the slash character ('/') or the
* commercial-at character ('@'), is quoted. If a query is given then a question-mark character
* ('?') is appended, followed by the query. Any character that
* is not a legal URI character is quoted.
* Finally, if a fragment is given then a hash character
* ('#') is appended, followed by the fragment. Any character
* that is not a legal URI character is quoted. The resulting URI string is then parsed as if by invoking the {@link
* #URI(String)} constructor and then invoking the {@link
* #parseServerAuthority()} method upon the result; this may cause a {@link
* URISyntaxException} to be thrown. If a scheme is given then the path, if also given, must either be
* empty or begin with a slash character ('/'). Otherwise a
* component of the new URI may be left undefined by passing null
* for the corresponding parameter.
*
* This constructor first builds a URI string from the given components
* according to the rules specified in RFC 2396,
* section 5.2, step 7: Initially, the result string is empty. If a scheme is given then it is appended to the result,
* followed by a colon character (':'). If an authority is given then the string "//" is
* appended, followed by the authority. If the authority contains a
* literal IPv6 address then the address must be enclosed in square
* brackets ('[' and ']'). Any character not in the
* unreserved, punct, escaped, or other
* categories, and not equal to the commercial-at character
* ('@'), is quoted. If a path is given then it is appended. Any character not in
* the unreserved, punct, escaped, or other
* categories, and not equal to the slash character ('/') or the
* commercial-at character ('@'), is quoted. If a query is given then a question-mark character
* ('?') is appended, followed by the query. Any character that
* is not a legal URI character is quoted.
* Finally, if a fragment is given then a hash character
* ('#') is appended, followed by the fragment. Any character
* that is not a legal URI character is quoted. The resulting URI string is then parsed as if by invoking the {@link
* #URI(String)} constructor and then invoking the {@link
* #parseServerAuthority()} method upon the result; this may cause a {@link
* URISyntaxException} to be thrown. A component may be left undefined by passing null.
*
* This convenience constructor works as if by invoking the
* seven-argument constructor as follows:
*
* A component may be left undefined by passing null.
*
* This constructor first builds a URI in string form using the given
* components as follows: Initially, the result string is empty. If a scheme is given then it is appended to the result,
* followed by a colon character (':'). If a scheme-specific part is given then it is appended. Any
* character that is not a legal URI character
* is quoted. Finally, if a fragment is given then a hash character
* ('#') is appended to the string, followed by the fragment.
* Any character that is not a legal URI character is quoted. The resulting URI string is then parsed in order to create the new
* URI instance as if by invoking the {@link #URI(String)} constructor;
* this may cause a {@link URISyntaxException} to be thrown. This convenience factory method works as if by invoking the {@link
* #URI(String)} constructor; any {@link URISyntaxException} thrown by the
* constructor is caught and wrapped in a new {@link
* IllegalArgumentException} object, which is then thrown.
*
* This method is provided for use in situations where it is known that
* the given string is a legal URI, for example for URI constants declared
* within in a program, and so it would be considered a programming error
* for the string not to parse as such. The constructors, which throw
* {@link URISyntaxException} directly, should be used situations where a
* URI is being constructed from user input or from some other source that
* may be prone to errors. If this URI's authority component has already been recognized as
* being server-based then it will already have been parsed into
* user-information, host, and port components. In this case, or if this
* URI has no authority component, this method simply returns this URI.
*
* Otherwise this method attempts once more to parse the authority
* component into user-information, host, and port components, and throws
* an exception describing why the authority component could not be parsed
* in that way.
*
* This method is provided because the generic URI syntax specified in
* RFC 2396
* cannot always distinguish a malformed server-based authority from a
* legitimate registry-based authority. It must therefore treat some
* instances of the former as instances of the latter. The authority
* component in the URI string "//foo:bar", for example, is not a
* legal server-based authority but it is legal as a registry-based
* authority.
*
* In many common situations, for example when working URIs that are
* known to be either URNs or URLs, the hierarchical URIs being used will
* always be server-based. They therefore must either be parsed as such or
* treated as an error. In these cases a statement such as
*
* can be used to ensure that u always refers to a URI that, if
* it has an authority component, has a server-based authority with proper
* user-information, host, and port components. Invoking this method also
* ensures that if the authority could not be parsed in that way then an
* appropriate diagnostic message can be issued based upon the exception
* that is thrown. If this URI is opaque, or if its path is already in normal form,
* then this URI is returned. Otherwise a new URI is constructed that is
* identical to this URI except that its path is computed by normalizing
* this URI's path in a manner consistent with RFC 2396,
* section 5.2, step 6, sub-steps c through f; that is:
* All "." segments are removed. If a ".." segment is preceded by a non-".."
* segment then both of these segments are removed. This step is
* repeated until it is no longer applicable. If the path is relative, and if its first segment contains a
* colon character (':'), then a "." segment is
* prepended. This prevents a relative URI with a path such as
* "a:b/c/d" from later being re-parsed as an opaque URI with a
* scheme of "a" and a scheme-specific part of "b/c/d".
* (Deviation from RFC 2396) A normalized path will begin with one or more ".." segments
* if there were insufficient non-".." segments preceding them to
* allow their removal. A normalized path will begin with a "."
* segment if one was inserted by step 3 above. Otherwise, a normalized
* path will not contain any "." or ".." segments. If the given URI is already absolute, or if this URI is opaque, then
* the given URI is returned.
*
* If the given URI's fragment component is
* defined, its path component is empty, and its scheme, authority, and
* query components are undefined, then a URI with the given fragment but
* with all other components equal to those of this URI is returned. This
* allows a URI representing a standalone fragment reference, such as
* "#foo", to be usefully resolved against a base URI.
*
* Otherwise this method constructs a new hierarchical URI in a manner
* consistent with RFC 2396,
* section 5.2; that is: A new URI is constructed with this URI's scheme and the given
* URI's query and fragment components. If the given URI has an authority component then the new URI's
* authority and path are taken from the given URI. Otherwise the new URI's authority component is copied from
* this URI, and its path is computed as follows: If the given URI's path is absolute then the new URI's path
* is taken from the given URI. Otherwise the given URI's path is relative, and so the new
* URI's path is computed by resolving the path of the given URI
* against the path of this URI. This is done by concatenating all but
* the last segment of this URI's path, if any, with the given URI's
* path and then normalizing the result as if by invoking the {@link
* #normalize() normalize} method. The result of this method is absolute if, and only if, either this
* URI is absolute or the given URI is absolute. This convenience method works as if invoking it were equivalent to
* evaluating the expression {@link #resolve(java.net.URI)
* resolve}(URI.{@link #create(String) create}(str)). The relativization of the given URI against this URI is computed as
* follows: If either this URI or the given URI are opaque, or if the
* scheme and authority components of the two URIs are not identical, or
* if the path of this URI is not a prefix of the path of the given URI,
* then the given URI is returned. Otherwise a new relative hierarchical URI is constructed with
* query and fragment components taken from the given URI and with a path
* component computed by removing this URI's path from the beginning of
* the given URI's path. This convenience method works as if invoking it were equivalent to
* evaluating the expression new URL(this.toString()) after
* first checking that this URI is absolute. The scheme component of a URI, if defined, only contains characters
* in the alphanum category and in the string "-.+". A
* scheme always starts with an alpha character.
*
* The scheme component of a URI cannot contain escaped octets, hence this
* method does not perform any decoding.
*
* @return The scheme component of this URI,
* or null if the scheme is undefined
*/
public String getScheme() {
return scheme;
}
/**
* Tells whether or not this URI is absolute.
*
* A URI is absolute if, and only if, it has a scheme component. A URI is opaque if, and only if, it is absolute and its
* scheme-specific part does not begin with a slash character ('/').
* An opaque URI has a scheme, a scheme-specific part, and possibly
* a fragment; all other components are undefined. The scheme-specific part of a URI only contains legal URI
* characters. The string returned by this method is equal to that returned by the
* {@link #getRawSchemeSpecificPart() getRawSchemeSpecificPart} method
* except that all sequences of escaped octets are decoded. The authority component of a URI, if defined, only contains the
* commercial-at character ('@') and characters in the
* unreserved, punct, escaped, and other
* categories. If the authority is server-based then it is further
* constrained to have valid user-information, host, and port
* components. The string returned by this method is equal to that returned by the
* {@link #getRawAuthority() getRawAuthority} method except that all
* sequences of escaped octets are decoded. The user-information component of a URI, if defined, only contains
* characters in the unreserved, punct, escaped, and
* other categories. The string returned by this method is equal to that returned by the
* {@link #getRawUserInfo() getRawUserInfo} method except that all
* sequences of escaped octets are decoded. The host component of a URI, if defined, will have one of the
* following forms: A domain name consisting of one or more labels
* separated by period characters ('.'), optionally followed by
* a period character. Each label consists of alphanum characters
* as well as hyphen characters ('-'), though hyphens never
* occur as the first or last characters in a label. The rightmost
* label of a domain name consisting of two or more labels, begins
* with an alpha character. A dotted-quad IPv4 address of the form
* digit+.digit+.digit+.digit+,
* where no digit sequence is longer than three characters and no
* sequence has a value larger than 255. An IPv6 address enclosed in square brackets ('[' and
* ']') and consisting of hexadecimal digits, colon characters
* (':'), and possibly an embedded IPv4 address. The full
* syntax of IPv6 addresses is specified in RFC 2373: IPv6
* Addressing Architecture. The port component of a URI, if defined, is a non-negative
* integer. The path component of a URI, if defined, only contains the slash
* character ('/'), the commercial-at character ('@'),
* and characters in the unreserved, punct, escaped,
* and other categories. The string returned by this method is equal to that returned by the
* {@link #getRawPath() getRawPath} method except that all sequences of
* escaped octets are decoded. The query component of a URI, if defined, only contains legal URI
* characters. The string returned by this method is equal to that returned by the
* {@link #getRawQuery() getRawQuery} method except that all sequences of
* escaped octets are decoded. The fragment component of a URI, if defined, only contains legal URI
* characters. The string returned by this method is equal to that returned by the
* {@link #getRawFragment() getRawFragment} method except that all
* sequences of escaped octets are decoded. If the given object is not a URI then this method immediately
* returns false.
*
* For two URIs to be considered equal requires that either both are
* opaque or both are hierarchical. Their schemes must either both be
* undefined or else be equal without regard to case. Their fragments
* must either both be undefined or else be equal.
*
* For two opaque URIs to be considered equal, their scheme-specific
* parts must be equal.
*
* For two hierarchical URIs to be considered equal, their paths must
* be equal and their queries must either both be undefined or else be
* equal. Their authorities must either both be undefined, or both be
* registry-based, or both be server-based. If their authorities are
* defined and are registry-based, then they must be equal. If their
* authorities are defined and are server-based, then their hosts must be
* equal without regard to case, their port numbers must be equal, and
* their user-information components must be equal.
*
* When testing the user-information, path, query, fragment, authority,
* or scheme-specific parts of two URIs for equality, the raw forms rather
* than the encoded forms of these components are compared and the
* hexadecimal digits of escaped octets are compared without regard to
* case.
*
* This method satisfies the general contract of the {@link
* java.lang.Object#equals(Object) Object.equals} method. When comparing corresponding components of two URIs, if one
* component is undefined but the other is defined then the first is
* considered to be less than the second. Unless otherwise noted, string
* components are ordered according to their natural, case-sensitive
* ordering as defined by the {@link java.lang.String#compareTo(Object)
* String.compareTo} method. String components that are subject to
* encoding are compared by comparing their raw forms rather than their
* encoded forms.
*
* The ordering of URIs is defined as follows: Two URIs with different schemes are ordered according the
* ordering of their schemes, without regard to case. A hierarchical URI is considered to be less than an opaque URI
* with an identical scheme. Two opaque URIs with identical schemes are ordered according
* to the ordering of their scheme-specific parts. Two opaque URIs with identical schemes and scheme-specific
* parts are ordered according to the ordering of their
* fragments. Two hierarchical URIs with identical schemes are ordered
* according to the ordering of their authority components: If both authority components are server-based then the URIs
* are ordered according to their user-information components; if these
* components are identical then the URIs are ordered according to the
* ordering of their hosts, without regard to case; if the hosts are
* identical then the URIs are ordered according to the ordering of
* their ports. If one or both authority components are registry-based then
* the URIs are ordered according to the ordering of their authority
* components. Finally, two hierarchical URIs with identical schemes and
* authority components are ordered according to the ordering of their
* paths; if their paths are identical then they are ordered according to
* the ordering of their queries; if the queries are identical then they
* are ordered according to the order of their fragments. This method satisfies the general contract of the {@link
* java.lang.Comparable#compareTo(Object) Comparable.compareTo}
* method. If this URI was created by invoking one of the constructors in this
* class then a string equivalent to the original input string, or to the
* string computed from the originally-given components, as appropriate, is
* returned. Otherwise this URI was created by normalization, resolution,
* or relativization, and so a string is constructed from this URI's
* components according to the rules specified in RFC 2396,
* section 5.2, step 7. If this URI does not contain any characters in the other
* category then an invocation of this method will return the same value as
* an invocation of the {@link #toString() toString} method. Otherwise
* this method works as if by invoking that method and then encoding the result. The only serializable field of a URI instance is its string
* field. That field is given a value, if it does not have one already,
* and then the {@link java.io.ObjectOutputStream#defaultWriteObject()}
* method of the given object-output stream is invoked. The {@link java.io.ObjectInputStream#defaultReadObject()} method is
* invoked to read the value of the string field. The result is
* then parsed in the usual way.
*
* @param is The object-input stream from which this object
* is being read
*/
private void readObject(ObjectInputStream is)
throws ClassNotFoundException, IOException
{
port = -1; // Argh
is.defaultReadObject();
try {
new Parser(string).parse(false);
} catch (URISyntaxException x) {
IOException y = new InvalidObjectException("Invalid URI");
y.initCause(x);
throw y;
}
}
// -- End of public methods --
// -- Utility methods for string-field comparison and hashing --
// These methods return appropriate values for null string arguments,
// thereby simplifying the equals, hashCode, and compareTo methods.
//
// The case-ignoring methods should only be applied to strings whose
// characters are all known to be US-ASCII. Because of this restriction,
// these methods are faster than the similar methods in the String class.
// US-ASCII only
private static int toLower(char c) {
if ((c >= 'A') && (c <= 'Z'))
return c + ('a' - 'A');
return c;
}
private static boolean equal(String s, String t) {
if (s == t) return true;
if ((s != null) && (t != null)) {
if (s.length() != t.length())
return false;
if (s.indexOf('%') < 0)
return s.equals(t);
int n = s.length();
for (int i = 0; i < n;) {
char c = s.charAt(i);
char d = t.charAt(i);
if (c != '%') {
if (c != d)
return false;
i++;
continue;
}
if (d != '%')
return false;
i++;
if (toLower(s.charAt(i)) != toLower(t.charAt(i)))
return false;
i++;
if (toLower(s.charAt(i)) != toLower(t.charAt(i)))
return false;
i++;
}
return true;
}
return false;
}
// US-ASCII only
private static boolean equalIgnoringCase(String s, String t) {
if (s == t) return true;
if ((s != null) && (t != null)) {
int n = s.length();
if (t.length() != n)
return false;
for (int i = 0; i < n; i++) {
if (toLower(s.charAt(i)) != toLower(t.charAt(i)))
return false;
}
return true;
}
return false;
}
private static int hash(int hash, String s) {
if (s == null) return hash;
return hash * 127 + s.hashCode();
}
// US-ASCII only
private static int hashIgnoringCase(int hash, String s) {
if (s == null) return hash;
int h = hash;
int n = s.length();
for (int i = 0; i < n; i++)
h = 31 * h + toLower(s.charAt(i));
return h;
}
private static int compare(String s, String t) {
if (s == t) return 0;
if (s != null) {
if (t != null)
return s.compareTo(t);
else
return +1;
} else {
return -1;
}
}
// US-ASCII only
private static int compareIgnoringCase(String s, String t) {
if (s == t) return 0;
if (s != null) {
if (t != null) {
int sn = s.length();
int tn = t.length();
int n = sn < tn ? sn : tn;
for (int i = 0; i < n; i++) {
int c = toLower(s.charAt(i)) - toLower(t.charAt(i));
if (c != 0)
return c;
}
return sn - tn;
}
return +1;
} else {
return -1;
}
}
// -- String construction --
// If a scheme is given then the path, if given, must be absolute
//
private static void checkPath(String s, String scheme, String path)
throws URISyntaxException
{
if (scheme != null) {
if ((path != null)
&& ((path.length() > 0) && (path.charAt(0) != '/')))
throw new URISyntaxException(s,
"Relative path in absolute URI");
}
}
private void appendAuthority(StringBuffer sb,
String authority,
String userInfo,
String host,
int port)
{
if (host != null) {
sb.append("//");
if (userInfo != null) {
sb.append(quote(userInfo, L_USERINFO, H_USERINFO));
sb.append('@');
}
boolean needBrackets = ((host.indexOf(':') >= 0)
&& !host.startsWith("[")
&& !host.endsWith("]"));
if (needBrackets) sb.append('[');
sb.append(host);
if (needBrackets) sb.append(']');
if (port != -1) {
sb.append(':');
sb.append(port);
}
} else if (authority != null) {
sb.append("//");
if (authority.startsWith("[")) {
// authority should (but may not) contain an embedded IPv6 address
int end = authority.indexOf("]");
String doquote = authority, dontquote = "";
if (end != -1 && authority.indexOf(":") != -1) {
// the authority contains an IPv6 address
if (end == authority.length()) {
dontquote = authority;
doquote = "";
} else {
dontquote = authority.substring(0 , end + 1);
doquote = authority.substring(end + 1);
}
}
sb.append(dontquote);
sb.append(quote(doquote,
L_REG_NAME | L_SERVER,
H_REG_NAME | H_SERVER));
} else {
sb.append(quote(authority,
L_REG_NAME | L_SERVER,
H_REG_NAME | H_SERVER));
}
}
}
private void appendSchemeSpecificPart(StringBuffer sb,
String opaquePart,
String authority,
String userInfo,
String host,
int port,
String path,
String query)
{
if (opaquePart != null) {
/* check if SSP begins with an IPv6 address
* because we must not quote a literal IPv6 address
*/
if (opaquePart.startsWith("//[")) {
int end = opaquePart.indexOf("]");
if (end != -1 && opaquePart.indexOf(":")!=-1) {
String doquote, dontquote;
if (end == opaquePart.length()) {
dontquote = opaquePart;
doquote = "";
} else {
dontquote = opaquePart.substring(0,end+1);
doquote = opaquePart.substring(end+1);
}
sb.append (dontquote);
sb.append(quote(doquote, L_URIC, H_URIC));
}
} else {
sb.append(quote(opaquePart, L_URIC, H_URIC));
}
} else {
appendAuthority(sb, authority, userInfo, host, port);
if (path != null)
sb.append(quote(path, L_PATH, H_PATH));
if (query != null) {
sb.append('?');
sb.append(quote(query, L_URIC, H_URIC));
}
}
}
private void appendFragment(StringBuffer sb, String fragment) {
if (fragment != null) {
sb.append('#');
sb.append(quote(fragment, L_URIC, H_URIC));
}
}
private String toString(String scheme,
String opaquePart,
String authority,
String userInfo,
String host,
int port,
String path,
String query,
String fragment)
{
StringBuffer sb = new StringBuffer();
if (scheme != null) {
sb.append(scheme);
sb.append(':');
}
appendSchemeSpecificPart(sb, opaquePart,
authority, userInfo, host, port,
path, query);
appendFragment(sb, fragment);
return sb.toString();
}
private void defineSchemeSpecificPart() {
if (schemeSpecificPart != null) return;
StringBuffer sb = new StringBuffer();
appendSchemeSpecificPart(sb, null, getAuthority(), getUserInfo(),
host, port, getPath(), getQuery());
if (sb.length() == 0) return;
schemeSpecificPart = sb.toString();
}
private void defineString() {
if (string != null) return;
StringBuffer sb = new StringBuffer();
if (scheme != null) {
sb.append(scheme);
sb.append(':');
}
if (isOpaque()) {
sb.append(schemeSpecificPart);
} else {
if (host != null) {
sb.append("//");
if (userInfo != null) {
sb.append(userInfo);
sb.append('@');
}
boolean needBrackets = ((host.indexOf(':') >= 0)
&& !host.startsWith("[")
&& !host.endsWith("]"));
if (needBrackets) sb.append('[');
sb.append(host);
if (needBrackets) sb.append(']');
if (port != -1) {
sb.append(':');
sb.append(port);
}
} else if (authority != null) {
sb.append("//");
sb.append(authority);
}
if (path != null)
sb.append(path);
if (query != null) {
sb.append('?');
sb.append(query);
}
}
if (fragment != null) {
sb.append('#');
sb.append(fragment);
}
string = sb.toString();
}
// -- Normalization, resolution, and relativization --
// RFC2396 5.2 (6)
private static String resolvePath(String base, String child,
boolean absolute)
{
int i = base.lastIndexOf('/');
int cn = child.length();
String path = "";
if (cn == 0) {
// 5.2 (6a)
if (i >= 0)
path = base.substring(0, i + 1);
} else {
StringBuffer sb = new StringBuffer(base.length() + cn);
// 5.2 (6a)
if (i >= 0)
sb.append(base.substring(0, i + 1));
// 5.2 (6b)
sb.append(child);
path = sb.toString();
}
// 5.2 (6c-f)
String np = normalize(path, true);
// 5.2 (6g): If the result is absolute but the path begins with "../",
// then we simply leave the path as-is
return np;
}
// RFC2396 5.2
private static URI resolve(URI base, URI child) {
// check if child if opaque first so that NPE is thrown
// if child is null.
if (child.isOpaque() || base.isOpaque())
return child;
// 5.2 (2): Reference to current document (lone fragment)
if ((child.scheme == null) && (child.authority == null)
&& child.path.equals("") && (child.fragment != null)
&& (child.query == null)) {
if ((base.fragment != null)
&& child.fragment.equals(base.fragment)) {
return base;
}
URI ru = new URI();
ru.scheme = base.scheme;
ru.authority = base.authority;
ru.userInfo = base.userInfo;
ru.host = base.host;
ru.port = base.port;
ru.path = base.path;
ru.fragment = child.fragment;
ru.query = base.query;
return ru;
}
// 5.2 (3): Child is absolute
if (child.scheme != null)
return child;
URI ru = new URI(); // Resolved URI
ru.scheme = base.scheme;
ru.query = child.query;
ru.fragment = child.fragment;
// 5.2 (4): Authority
if (child.authority == null) {
ru.authority = base.authority;
ru.host = base.host;
ru.userInfo = base.userInfo;
ru.port = base.port;
if (child.path == null || child.path.isEmpty()) {
// This is an addtional path from RFC 3986 RI, which fixes following RFC 2396
// "normal" examples:
// Base: http://a/b/c/d;p?q
// "?y" = "http://a/b/c/d;p?y"
// "" = "http://a/b/c/d;p?q"
// http://b/25897693
ru.path = base.path;
ru.query = child.query != null ? child.query : base.query;
} else if ((child.path.length() > 0) && (child.path.charAt(0) == '/')) {
// 5.2 (5): Child path is absolute
//
// There is an additional step from RFC 3986 RI, requiring to remove dots for
// absolute path as well.
// http://b/25897693
ru.path = normalize(child.path, true);
} else {
// 5.2 (6): Resolve relative path
ru.path = resolvePath(base.path, child.path, base.isAbsolute());
}
} else {
ru.authority = child.authority;
ru.host = child.host;
ru.userInfo = child.userInfo;
ru.host = child.host;
ru.port = child.port;
ru.path = child.path;
}
// 5.2 (7): Recombine (nothing to do here)
return ru;
}
// If the given URI's path is normal then return the URI;
// o.w., return a new URI containing the normalized path.
//
private static URI normalize(URI u) {
if (u.isOpaque() || (u.path == null) || (u.path.length() == 0))
return u;
String np = normalize(u.path);
if (np == u.path)
return u;
URI v = new URI();
v.scheme = u.scheme;
v.fragment = u.fragment;
v.authority = u.authority;
v.userInfo = u.userInfo;
v.host = u.host;
v.port = u.port;
v.path = np;
v.query = u.query;
return v;
}
// If both URIs are hierarchical, their scheme and authority components are
// identical, and the base path is a prefix of the child's path, then
// return a relative URI that, when resolved against the base, yields the
// child; otherwise, return the child.
//
private static URI relativize(URI base, URI child) {
// check if child if opaque first so that NPE is thrown
// if child is null.
if (child.isOpaque() || base.isOpaque())
return child;
if (!equalIgnoringCase(base.scheme, child.scheme)
|| !equal(base.authority, child.authority))
return child;
String bp = normalize(base.path);
String cp = normalize(child.path);
if (!bp.equals(cp)) {
// Android-changed: The original OpenJdk implementation would append a trailing slash
// to paths like "/a/b" before relativizing them. This would relativize /a/b/c to
// "/c" against "/a/b" the android implementation did not do this. It would assume that
// "b" wasn't a directory and relativize the path to "/b/c". The spec is pretty vague
// about this but this change is being made because we have several tests that expect
// this behaviour.
if (bp.indexOf('/') != -1) {
bp = bp.substring(0, bp.lastIndexOf('/') + 1);
}
if (!cp.startsWith(bp))
return child;
}
URI v = new URI();
v.path = cp.substring(bp.length());
v.query = child.query;
v.fragment = child.fragment;
return v;
}
// -- Path normalization --
// The following algorithm for path normalization avoids the creation of a
// string object for each segment, as well as the use of a string buffer to
// compute the final result, by using a single char array and editing it in
// place. The array is first split into segments, replacing each slash
// with '\0' and creating a segment-index array, each element of which is
// the index of the first char in the corresponding segment. We then walk
// through both arrays, removing ".", "..", and other segments as necessary
// by setting their entries in the index array to -1. Finally, the two
// arrays are used to rejoin the segments and compute the final result.
//
// This code is based upon src/solaris/native/java/io/canonicalize_md.c
// Check the given path to see if it might need normalization. A path
// might need normalization if it contains duplicate slashes, a "."
// segment, or a ".." segment. Return -1 if no further normalization is
// possible, otherwise return the number of segments found.
//
// This method takes a string argument rather than a char array so that
// this test can be performed without invoking path.toCharArray().
//
static private int needsNormalization(String path) {
boolean normal = true;
int ns = 0; // Number of segments
int end = path.length() - 1; // Index of last char in path
int p = 0; // Index of next char in path
// Skip initial slashes
while (p <= end) {
if (path.charAt(p) != '/') break;
p++;
}
if (p > 1) normal = false;
// Scan segments
while (p <= end) {
// Looking at "." or ".." ?
if ((path.charAt(p) == '.')
&& ((p == end)
|| ((path.charAt(p + 1) == '/')
|| ((path.charAt(p + 1) == '.')
&& ((p + 1 == end)
|| (path.charAt(p + 2) == '/')))))) {
normal = false;
}
ns++;
// Find beginning of next segment
while (p <= end) {
if (path.charAt(p++) != '/')
continue;
// Skip redundant slashes
while (p <= end) {
if (path.charAt(p) != '/') break;
normal = false;
p++;
}
break;
}
}
return normal ? -1 : ns;
}
// Split the given path into segments, replacing slashes with nulls and
// filling in the given segment-index array.
//
// Preconditions:
// segs.length == Number of segments in path
//
// Postconditions:
// All slashes in path replaced by '\0'
// segs[i] == Index of first char in segment i (0 <= i < segs.length)
//
static private void split(char[] path, int[] segs) {
int end = path.length - 1; // Index of last char in path
int p = 0; // Index of next char in path
int i = 0; // Index of current segment
// Skip initial slashes
while (p <= end) {
if (path[p] != '/') break;
path[p] = '\0';
p++;
}
while (p <= end) {
// Note start of segment
segs[i++] = p++;
// Find beginning of next segment
while (p <= end) {
if (path[p++] != '/')
continue;
path[p - 1] = '\0';
// Skip redundant slashes
while (p <= end) {
if (path[p] != '/') break;
path[p++] = '\0';
}
break;
}
}
if (i != segs.length)
throw new InternalError(); // ASSERT
}
// Join the segments in the given path according to the given segment-index
// array, ignoring those segments whose index entries have been set to -1,
// and inserting slashes as needed. Return the length of the resulting
// path.
//
// Preconditions:
// segs[i] == -1 implies segment i is to be ignored
// path computed by split, as above, with '\0' having replaced '/'
//
// Postconditions:
// path[0] .. path[return value] == Resulting path
//
static private int join(char[] path, int[] segs) {
int ns = segs.length; // Number of segments
int end = path.length - 1; // Index of last char in path
int p = 0; // Index of next path char to write
if (path[p] == '\0') {
// Restore initial slash for absolute paths
path[p++] = '/';
}
for (int i = 0; i < ns; i++) {
int q = segs[i]; // Current segment
if (q == -1)
// Ignore this segment
continue;
if (p == q) {
// We're already at this segment, so just skip to its end
while ((p <= end) && (path[p] != '\0'))
p++;
if (p <= end) {
// Preserve trailing slash
path[p++] = '/';
}
} else if (p < q) {
// Copy q down to p
while ((q <= end) && (path[q] != '\0'))
path[p++] = path[q++];
if (q <= end) {
// Preserve trailing slash
path[p++] = '/';
}
} else
throw new InternalError(); // ASSERT false
}
return p;
}
// Remove "." segments from the given path, and remove segment pairs
// consisting of a non-".." segment followed by a ".." segment.
//
private static void removeDots(char[] path, int[] segs, boolean removeLeading) {
int ns = segs.length;
int end = path.length - 1;
for (int i = 0; i < ns; i++) {
int dots = 0; // Number of dots found (0, 1, or 2)
// Find next occurrence of "." or ".."
do {
int p = segs[i];
if (path[p] == '.') {
if (p == end) {
dots = 1;
break;
} else if (path[p + 1] == '\0') {
dots = 1;
break;
} else if ((path[p + 1] == '.')
&& ((p + 1 == end)
|| (path[p + 2] == '\0'))) {
dots = 2;
break;
}
}
i++;
} while (i < ns);
if ((i > ns) || (dots == 0))
break;
if (dots == 1) {
// Remove this occurrence of "."
segs[i] = -1;
} else {
// If there is a preceding non-".." segment, remove both that
// segment and this occurrence of ".."
int j;
for (j = i - 1; j >= 0; j--) {
if (segs[j] != -1) break;
}
if (j >= 0) {
int q = segs[j];
if (!((path[q] == '.')
&& (path[q + 1] == '.')
&& (path[q + 2] == '\0'))) {
segs[i] = -1;
segs[j] = -1;
}
} else if (removeLeading) {
// This is a leading ".." segment. Per RFC 3986 RI, this should be removed as
// well. This fixes RFC 2396 "abnormal" examples.
// http://b/25897693
segs[i] = -1;
}
}
}
}
// DEVIATION: If the normalized path is relative, and if the first
// segment could be parsed as a scheme name, then prepend a "." segment
//
private static void maybeAddLeadingDot(char[] path, int[] segs) {
if (path[0] == '\0')
// The path is absolute
return;
int ns = segs.length;
int f = 0; // Index of first segment
while (f < ns) {
if (segs[f] >= 0)
break;
f++;
}
if ((f >= ns) || (f == 0))
// The path is empty, or else the original first segment survived,
// in which case we already know that no leading "." is needed
return;
int p = segs[f];
while ((p < path.length) && (path[p] != ':') && (path[p] != '\0')) p++;
if (p >= path.length || path[p] == '\0')
// No colon in first segment, so no "." needed
return;
// At this point we know that the first segment is unused,
// hence we can insert a "." segment at that position
path[0] = '.';
path[1] = '\0';
segs[0] = 0;
}
// Normalize the given path string. A normal path string has no empty
// segments (i.e., occurrences of "//"), no segments equal to ".", and no
// segments equal to ".." that are preceded by a segment not equal to "..".
// In contrast to Unix-style pathname normalization, for URI paths we
// always retain trailing slashes.
//
private static String normalize(String ps) {
return normalize(ps, false);
}
private static String normalize(String ps, boolean removeLeading) {
// Does this path need normalization?
int ns = needsNormalization(ps); // Number of segments
if (ns < 0)
// Nope -- just return it
return ps;
char[] path = ps.toCharArray(); // Path in char-array form
// Split path into segments
int[] segs = new int[ns]; // Segment-index array
split(path, segs);
// Remove dots
removeDots(path, segs, removeLeading);
// Prevent scheme-name confusion
maybeAddLeadingDot(path, segs);
// Join the remaining segments and return the result
String s = new String(path, 0, join(path, segs));
if (s.equals(ps)) {
// string was already normalized
return ps;
}
return s;
}
// -- Character classes for parsing --
// RFC2396 precisely specifies which characters in the US-ASCII charset are
// permissible in the various components of a URI reference. We here
// define a set of mask pairs to aid in enforcing these restrictions. Each
// mask pair consists of two longs, a low mask and a high mask. Taken
// together they represent a 128-bit mask, where bit i is set iff the
// character with value i is permitted.
//
// This approach is more efficient than sequentially searching arrays of
// permitted characters. It could be made still more efficient by
// precompiling the mask information so that a character's presence in a
// given mask could be determined by a single table lookup.
// Compute the low-order mask for the characters in the given string
private static long lowMask(String chars) {
int n = chars.length();
long m = 0;
for (int i = 0; i < n; i++) {
char c = chars.charAt(i);
if (c < 64)
m |= (1L << c);
}
return m;
}
// Compute the high-order mask for the characters in the given string
private static long highMask(String chars) {
int n = chars.length();
long m = 0;
for (int i = 0; i < n; i++) {
char c = chars.charAt(i);
if ((c >= 64) && (c < 128))
m |= (1L << (c - 64));
}
return m;
}
// Compute a low-order mask for the characters
// between first and last, inclusive
private static long lowMask(char first, char last) {
long m = 0;
int f = Math.max(Math.min(first, 63), 0);
int l = Math.max(Math.min(last, 63), 0);
for (int i = f; i <= l; i++)
m |= 1L << i;
return m;
}
// Compute a high-order mask for the characters
// between first and last, inclusive
private static long highMask(char first, char last) {
long m = 0;
int f = Math.max(Math.min(first, 127), 64) - 64;
int l = Math.max(Math.min(last, 127), 64) - 64;
for (int i = f; i <= l; i++)
m |= 1L << i;
return m;
}
// Tell whether the given character is permitted by the given mask pair
private static boolean match(char c, long lowMask, long highMask) {
if (c == 0) // 0 doesn't have a slot in the mask. So, it never matches.
return false;
if (c < 64)
return ((1L << c) & lowMask) != 0;
if (c < 128)
return ((1L << (c - 64)) & highMask) != 0;
return false;
}
// Character-class masks, in reverse order from RFC2396 because
// initializers for static fields cannot make forward references.
// digit = "0" | "1" | "2" | "3" | "4" | "5" | "6" | "7" |
// "8" | "9"
private static final long L_DIGIT = lowMask('0', '9');
private static final long H_DIGIT = 0L;
// upalpha = "A" | "B" | "C" | "D" | "E" | "F" | "G" | "H" | "I" |
// "J" | "K" | "L" | "M" | "N" | "O" | "P" | "Q" | "R" |
// "S" | "T" | "U" | "V" | "W" | "X" | "Y" | "Z"
private static final long L_UPALPHA = 0L;
private static final long H_UPALPHA = highMask('A', 'Z');
// lowalpha = "a" | "b" | "c" | "d" | "e" | "f" | "g" | "h" | "i" |
// "j" | "k" | "l" | "m" | "n" | "o" | "p" | "q" | "r" |
// "s" | "t" | "u" | "v" | "w" | "x" | "y" | "z"
private static final long L_LOWALPHA = 0L;
private static final long H_LOWALPHA = highMask('a', 'z');
// alpha = lowalpha | upalpha
private static final long L_ALPHA = L_LOWALPHA | L_UPALPHA;
private static final long H_ALPHA = H_LOWALPHA | H_UPALPHA;
// alphanum = alpha | digit
private static final long L_ALPHANUM = L_DIGIT | L_ALPHA;
private static final long H_ALPHANUM = H_DIGIT | H_ALPHA;
// hex = digit | "A" | "B" | "C" | "D" | "E" | "F" |
// "a" | "b" | "c" | "d" | "e" | "f"
private static final long L_HEX = L_DIGIT;
private static final long H_HEX = highMask('A', 'F') | highMask('a', 'f');
// mark = "-" | "_" | "." | "!" | "~" | "*" | "'" |
// "(" | ")"
private static final long L_MARK = lowMask("-_.!~*'()");
private static final long H_MARK = highMask("-_.!~*'()");
// unreserved = alphanum | mark
private static final long L_UNRESERVED = L_ALPHANUM | L_MARK;
private static final long H_UNRESERVED = H_ALPHANUM | H_MARK;
// reserved = ";" | "/" | "?" | ":" | "@" | "&" | "=" | "+" |
// "$" | "," | "[" | "]"
// Added per RFC2732: "[", "]"
private static final long L_RESERVED = lowMask(";/?:@&=+$,[]");
private static final long H_RESERVED = highMask(";/?:@&=+$,[]");
// The zero'th bit is used to indicate that escape pairs and non-US-ASCII
// characters are allowed; this is handled by the scanEscape method below.
private static final long L_ESCAPED = 1L;
private static final long H_ESCAPED = 0L;
// uric = reserved | unreserved | escaped
private static final long L_URIC = L_RESERVED | L_UNRESERVED | L_ESCAPED;
private static final long H_URIC = H_RESERVED | H_UNRESERVED | H_ESCAPED;
// pchar = unreserved | escaped |
// ":" | "@" | "&" | "=" | "+" | "$" | ","
private static final long L_PCHAR
= L_UNRESERVED | L_ESCAPED | lowMask(":@&=+$,");
private static final long H_PCHAR
= H_UNRESERVED | H_ESCAPED | highMask(":@&=+$,");
// All valid path characters
private static final long L_PATH = L_PCHAR | lowMask(";/");
private static final long H_PATH = H_PCHAR | highMask(";/");
// Dash, for use in domainlabel and toplabel
private static final long L_DASH = lowMask("-");
private static final long H_DASH = highMask("-");
// UNDERSCORE, for use in domainlabel and toplabel
private static final long L_UNDERSCORE = lowMask("_");
private static final long H_UNDERSCORE = highMask("_");
// Dot, for use in hostnames
private static final long L_DOT = lowMask(".");
private static final long H_DOT = highMask(".");
// userinfo = *( unreserved | escaped |
// ";" | ":" | "&" | "=" | "+" | "$" | "," )
private static final long L_USERINFO
= L_UNRESERVED | L_ESCAPED | lowMask(";:&=+$,");
private static final long H_USERINFO
= H_UNRESERVED | H_ESCAPED | highMask(";:&=+$,");
// reg_name = 1*( unreserved | escaped | "$" | "," |
// ";" | ":" | "@" | "&" | "=" | "+" )
private static final long L_REG_NAME
= L_UNRESERVED | L_ESCAPED | lowMask("$,;:@&=+");
private static final long H_REG_NAME
= H_UNRESERVED | H_ESCAPED | highMask("$,;:@&=+");
// All valid characters for server-based authorities
private static final long L_SERVER
= L_USERINFO | L_ALPHANUM | L_DASH | lowMask(".:@[]");
private static final long H_SERVER
= H_USERINFO | H_ALPHANUM | H_DASH | highMask(".:@[]");
// Special case of server authority that represents an IPv6 address
// In this case, a % does not signify an escape sequence
private static final long L_SERVER_PERCENT
= L_SERVER | lowMask("%");
private static final long H_SERVER_PERCENT
= H_SERVER | highMask("%");
private static final long L_LEFT_BRACKET = lowMask("[");
private static final long H_LEFT_BRACKET = highMask("[");
// scheme = alpha *( alpha | digit | "+" | "-" | "." )
private static final long L_SCHEME = L_ALPHA | L_DIGIT | lowMask("+-.");
private static final long H_SCHEME = H_ALPHA | H_DIGIT | highMask("+-.");
// uric_no_slash = unreserved | escaped | ";" | "?" | ":" | "@" |
// "&" | "=" | "+" | "$" | ","
private static final long L_URIC_NO_SLASH
= L_UNRESERVED | L_ESCAPED | lowMask(";?:@&=+$,");
private static final long H_URIC_NO_SLASH
= H_UNRESERVED | H_ESCAPED | highMask(";?:@&=+$,");
// -- Escaping and encoding --
private final static char[] hexDigits = {
'0', '1', '2', '3', '4', '5', '6', '7',
'8', '9', 'A', 'B', 'C', 'D', 'E', 'F'
};
private static void appendEscape(StringBuffer sb, byte b) {
sb.append('%');
sb.append(hexDigits[(b >> 4) & 0x0f]);
sb.append(hexDigits[(b >> 0) & 0x0f]);
}
private static void appendEncoded(StringBuffer sb, char c) {
ByteBuffer bb = null;
try {
bb = ThreadLocalCoders.encoderFor("UTF-8")
.encode(CharBuffer.wrap("" + c));
} catch (CharacterCodingException x) {
assert false;
}
while (bb.hasRemaining()) {
int b = bb.get() & 0xff;
if (b >= 0x80)
appendEscape(sb, (byte)b);
else
sb.append((char)b);
}
}
// Quote any characters in s that are not permitted
// by the given mask pair
//
private static String quote(String s, long lowMask, long highMask) {
int n = s.length();
StringBuffer sb = null;
boolean allowNonASCII = ((lowMask & L_ESCAPED) != 0);
for (int i = 0; i < s.length(); i++) {
char c = s.charAt(i);
if (c < '\u0080') {
if (!match(c, lowMask, highMask)) {
if (sb == null) {
sb = new StringBuffer();
sb.append(s.substring(0, i));
}
appendEscape(sb, (byte)c);
} else {
if (sb != null)
sb.append(c);
}
} else if (allowNonASCII
&& (Character.isSpaceChar(c)
|| Character.isISOControl(c))) {
if (sb == null) {
sb = new StringBuffer();
sb.append(s.substring(0, i));
}
appendEncoded(sb, c);
} else {
if (sb != null)
sb.append(c);
}
}
return (sb == null) ? s : sb.toString();
}
// Encodes all characters >= \u0080 into escaped, normalized UTF-8 octets,
// assuming that s is otherwise legal
//
private static String encode(String s) {
int n = s.length();
if (n == 0)
return s;
// First check whether we actually need to encode
for (int i = 0;;) {
if (s.charAt(i) >= '\u0080')
break;
if (++i >= n)
return s;
}
String ns = Normalizer.normalize(s, Normalizer.Form.NFC);
ByteBuffer bb = null;
try {
bb = ThreadLocalCoders.encoderFor("UTF-8")
.encode(CharBuffer.wrap(ns));
} catch (CharacterCodingException x) {
assert false;
}
StringBuffer sb = new StringBuffer();
while (bb.hasRemaining()) {
int b = bb.get() & 0xff;
if (b >= 0x80)
appendEscape(sb, (byte)b);
else
sb.append((char)b);
}
return sb.toString();
}
private static int decode(char c) {
if ((c >= '0') && (c <= '9'))
return c - '0';
if ((c >= 'a') && (c <= 'f'))
return c - 'a' + 10;
if ((c >= 'A') && (c <= 'F'))
return c - 'A' + 10;
assert false;
return -1;
}
private static byte decode(char c1, char c2) {
return (byte)( ((decode(c1) & 0xf) << 4)
| ((decode(c2) & 0xf) << 0));
}
// Evaluates all escapes in s, applying UTF-8 decoding if needed. Assumes
// that escapes are well-formed syntactically, i.e., of the form %XX. If a
// sequence of escaped octets is not valid UTF-8 then the erroneous octets
// are replaced with '\uFFFD'.
// Exception: any "%" found between "[]" is left alone. It is an IPv6 literal
// with a scope_id
//
private static String decode(String s) {
if (s == null)
return s;
int n = s.length();
if (n == 0)
return s;
if (s.indexOf('%') < 0)
return s;
StringBuffer sb = new StringBuffer(n);
ByteBuffer bb = ByteBuffer.allocate(n);
CharBuffer cb = CharBuffer.allocate(n);
CharsetDecoder dec = ThreadLocalCoders.decoderFor("UTF-8")
.onMalformedInput(CodingErrorAction.REPLACE)
.onUnmappableCharacter(CodingErrorAction.REPLACE);
// This is not horribly efficient, but it will do for now
char c = s.charAt(0);
boolean betweenBrackets = false;
for (int i = 0; i < n;) {
assert c == s.charAt(i); // Loop invariant
if (c == '[') {
betweenBrackets = true;
} else if (betweenBrackets && c == ']') {
betweenBrackets = false;
}
if (c != '%' || betweenBrackets) {
sb.append(c);
if (++i >= n)
break;
c = s.charAt(i);
continue;
}
bb.clear();
int ui = i;
for (;;) {
assert (n - i >= 2);
bb.put(decode(s.charAt(++i), s.charAt(++i)));
if (++i >= n)
break;
c = s.charAt(i);
if (c != '%')
break;
}
bb.flip();
cb.clear();
dec.reset();
CoderResult cr = dec.decode(bb, cb, true);
assert cr.isUnderflow();
cr = dec.flush(cb);
assert cr.isUnderflow();
sb.append(cb.flip().toString());
}
return sb.toString();
}
// -- Parsing --
// For convenience we wrap the input URI string in a new instance of the
// following internal class. This saves always having to pass the input
// string as an argument to each internal scan/parse method.
private class Parser {
private String input; // URI input string
private boolean requireServerAuthority = false;
Parser(String s) {
input = s;
string = s;
}
// -- Methods for throwing URISyntaxException in various ways --
private void fail(String reason) throws URISyntaxException {
throw new URISyntaxException(input, reason);
}
private void fail(String reason, int p) throws URISyntaxException {
throw new URISyntaxException(input, reason, p);
}
private void failExpecting(String expected, int p)
throws URISyntaxException
{
fail("Expected " + expected, p);
}
private void failExpecting(String expected, String prior, int p)
throws URISyntaxException
{
fail("Expected " + expected + " following " + prior, p);
}
// -- Simple access to the input string --
// Return a substring of the input string
//
private String substring(int start, int end) {
return input.substring(start, end);
}
// Return the char at position p,
// assuming that p < input.length()
//
private char charAt(int p) {
return input.charAt(p);
}
// Tells whether start < end and, if so, whether charAt(start) == c
//
private boolean at(int start, int end, char c) {
return (start < end) && (charAt(start) == c);
}
// Tells whether start + s.length() < end and, if so,
// whether the chars at the start position match s exactly
//
private boolean at(int start, int end, String s) {
int p = start;
int sn = s.length();
if (sn > end - p)
return false;
int i = 0;
while (i < sn) {
if (charAt(p++) != s.charAt(i)) {
break;
}
i++;
}
return (i == sn);
}
// -- Scanning --
// The various scan and parse methods that follow use a uniform
// convention of taking the current start position and end index as
// their first two arguments. The start is inclusive while the end is
// exclusive, just as in the String class, i.e., a start/end pair
// denotes the left-open interval [start, end) of the input string.
//
// These methods never proceed past the end position. They may return
// -1 to indicate outright failure, but more often they simply return
// the position of the first char after the last char scanned. Thus
// a typical idiom is
//
// int p = start;
// int q = scan(p, end, ...);
// if (q > p)
// // We scanned something
// ...;
// else if (q == p)
// // We scanned nothing
// ...;
// else if (q == -1)
// // Something went wrong
// ...;
// Scan a specific char: If the char at the given start position is
// equal to c, return the index of the next char; otherwise, return the
// start position.
//
private int scan(int start, int end, char c) {
if ((start < end) && (charAt(start) == c))
return start + 1;
return start;
}
// Scan forward from the given start position. Stop at the first char
// in the err string (in which case -1 is returned), or the first char
// in the stop string (in which case the index of the preceding char is
// returned), or the end of the input string (in which case the length
// of the input string is returned). May return the start position if
// nothing matches.
//
private int scan(int start, int end, String err, String stop) {
int p = start;
while (p < end) {
char c = charAt(p);
if (err.indexOf(c) >= 0)
return -1;
if (stop.indexOf(c) >= 0)
break;
p++;
}
return p;
}
// Scan a potential escape sequence, starting at the given position,
// with the given first char (i.e., charAt(start) == c).
//
// This method assumes that if escapes are allowed then visible
// non-US-ASCII chars are also allowed.
//
private int scanEscape(int start, int n, char first)
throws URISyntaxException
{
int p = start;
char c = first;
if (c == '%') {
// Process escape pair
if ((p + 3 <= n)
&& match(charAt(p + 1), L_HEX, H_HEX)
&& match(charAt(p + 2), L_HEX, H_HEX)) {
return p + 3;
}
fail("Malformed escape pair", p);
} else if ((c > 128)
&& !Character.isSpaceChar(c)
&& !Character.isISOControl(c)) {
// Allow unescaped but visible non-US-ASCII chars
return p + 1;
}
return p;
}
// Scan chars that match the given mask pair
//
private int scan(int start, int n, long lowMask, long highMask)
throws URISyntaxException
{
int p = start;
while (p < n) {
char c = charAt(p);
if (match(c, lowMask, highMask)) {
p++;
continue;
}
if ((lowMask & L_ESCAPED) != 0) {
int q = scanEscape(p, n, c);
if (q > p) {
p = q;
continue;
}
}
break;
}
return p;
}
// Check that each of the chars in [start, end) matches the given mask
//
private void checkChars(int start, int end,
long lowMask, long highMask,
String what)
throws URISyntaxException
{
int p = scan(start, end, lowMask, highMask);
if (p < end)
fail("Illegal character in " + what, p);
}
// Check that the char at position p matches the given mask
//
private void checkChar(int p,
long lowMask, long highMask,
String what)
throws URISyntaxException
{
checkChars(p, p + 1, lowMask, highMask, what);
}
// -- Parsing --
// [
*
*
*
* @param str The string to be parsed into a URI
*
* @throws NullPointerException
* If str is null
*
* @throws URISyntaxException
* If the given string violates RFC 2396, as augmented
* by the above deviations
*/
public URI(String str) throws URISyntaxException {
new Parser(str).parse(false);
}
/**
* Constructs a hierarchical URI from the given components.
*
*
*
*
*
*
*
*
*
*
* new {@link #URI(String, String, String, int, String, String, String)
* URI}(scheme, null, host, -1, path, null, fragment);
*
*
* @param scheme Scheme name
* @param host Host name
* @param path Path
* @param fragment Fragment
*
* @throws URISyntaxException
* If the URI string constructed from the given components
* violates RFC 2396
*/
public URI(String scheme, String host, String path, String fragment)
throws URISyntaxException
{
this(scheme, null, host, -1, path, null, fragment);
}
/**
* Constructs a URI from the given components.
*
*
*
*
*
*
* URI u = new URI(str).parseServerAuthority();
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
* @param uri The URI to be relativized against this URI
* @return The resulting URI
*
* @throws NullPointerException
* If uri is null
*/
public URI relativize(URI uri) {
return relativize(this, uri);
}
/**
* Constructs a URL from this URI.
*
*
*
*
*
* The host component of a URI cannot contain escaped octets, hence this
* method does not perform any decoding.
*
* @return The host component of this URI,
* or null if the host is undefined
*/
public String getHost() {
return host;
}
/**
* Returns the port number of this URI.
*
*
*
*
*
*
*
*