info.dmtree
Class Uri

java.lang.Object
  info.dmtree.Uri

public final class Uri

extends java.lang.Object

This class contains static utility methods to manipulate DMT URIs.

Syntax of valid DMT URIs:

• A slash ('/' \u002F) is the separator of the node names. Slashes used in node name must therefore be escaped using a backslash slash ("\/"). The backslash must be escaped with a double backslash sequence. A backslash found must be ignored when it is not followed by a slash or backslash.
• The node name can be constructed using full Unicode character set (except the Supplementary code, not being supported by CLDC/CDC). However, using the full Unicode character set for node names is discouraged because the encoding in the underlying storage as well as the encoding needed in communications can create significant performance and memory usage overhead. Names that are restricted to the URI set [-a-zA-Z0-9_.!~*'()] are most efficient.
• URIs used in the DMT must be treated and interpreted as case sensitive.
• No End Slash: URI must not end with the delimiter slash ('/' \u002F). This implies that the root node must be denoted as "." and not "./".
• No parent denotation: URI must not be constructed using the character sequence "../" to traverse the tree upwards.
• Single Root: The character sequence "./" must not be used anywhere else but in the beginning of a URI.

Method Summary

static int	getMaxSegmentNameLength() Returns the maximum allowed length of a URI segment.
static int	getMaxUriLength() Returns the maximum allowed length of a URI.
static int	getMaxUriSegments() Returns the maximum allowed number of URI segments.
static boolean	isAbsoluteUri(java.lang.String uri) Checks whether the specified URI is an absolute URI.
static boolean	isValidUri(java.lang.String uri) Checks whether the specified URI is valid.
static java.lang.String	mangle(java.lang.String nodeName) Returns a node name that is valid for the tree operation methods, based on the given node name.
static java.lang.String[]	toPath(java.lang.String uri) Split the specified URI along the path separator '/' charaters and return an array of URI segments.
static java.lang.String	toUri(java.lang.String[] path) Construct a URI from the specified URI segments.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

mangle

public static java.lang.String mangle(java.lang.String nodeName)

Returns a node name that is valid for the tree operation methods, based on the given node name. This transformation is not idempotent, so it must not be called with a parameter that is the result of a previous mangle method call.

Node name mangling is needed in the following cases:

• if the name contains '/' or '\' characters
• if the length of the name exceeds the limit defined by the implementation

A node name that does not suffer from either of these problems is guaranteed to remain unchanged by this method. Therefore the client may skip the mangling if the node name is known to be valid (though it is always safe to call this method).

The method returns the normalized nodeName as described below. Invalid node names are normalized in different ways, depending on the cause. If the length of the name does not exceed the limit, but the name contains '/' or '\' characters, then these are simply escaped by inserting an additional '\' before each occurrence. If the length of the name does exceed the limit, the following mechanism is used to normalize it:

• the SHA 1 digest of the name is calculated
• the digest is encoded with the base 64 algorithm
• all '/' characters in the encoded digest are replaced with '_'
• trailing '=' signs are removed

Parameters:

nodeName - the node name to be mangled (if necessary), must not be null or empty

Returns:

the normalized node name that is valid for tree operations

Throws:

java.lang.NullPointerException - if nodeName is null

java.lang.IllegalArgumentException - if nodeName is empty

toUri

public static java.lang.String toUri(java.lang.String[] path)

Construct a URI from the specified URI segments. The segments must already be mangled.

If the specified path is an empty array then an empty URI ("") is returned.

Parameters:

path - a possibly empty array of URI segments, must not be null

Returns:

the URI created from the specified segments

Throws:

java.lang.NullPointerException - if the specified path or any of its segments are null

java.lang.IllegalArgumentException - if the specified path contains too many or malformed segments or the resulting URI is too long

toPath

public static java.lang.String[] toPath(java.lang.String uri)

Split the specified URI along the path separator '/' charaters and return an array of URI segments. Special characters in the returned segments are escaped. The returned array may be empty if the specifed URI was empty.

Parameters:

uri - the URI to be split, must not be null

Returns:

an array of URI segments created by splitting the specified URI

Throws:

java.lang.NullPointerException - if the specified URI is null

java.lang.IllegalArgumentException - if the specified URI is malformed

getMaxUriSegments

public static int getMaxUriSegments()

Returns the maximum allowed number of URI segments. The returned value is implementation specific.

The return value of Integer.MAX_VALUE indicates that there is no upper limit on the number of URI segments.

Returns:

maximum number of URI segments supported by the implementation

getMaxUriLength

public static int getMaxUriLength()

Returns the maximum allowed length of a URI. The value is implementation specific. The length of the URI is defined as the number of bytes in the unescaped, UTF-8 encoded represenation of the URI.

The return value of Integer.MAX_VALUE indicates that there is no upper limit on the length of URIs.

Returns:

maximum URI length supported by the implementation

getMaxSegmentNameLength

public static int getMaxSegmentNameLength()

Returns the maximum allowed length of a URI segment. The value is implementation specific. The length of the URI segment is defined as the number of bytes in the unescaped, UTF-8 encoded represenation of the segment.

The return value of Integer.MAX_VALUE indicates that there is no upper limit on the length of segment names.

Returns:

maximum URI segment length supported by the implementation

isAbsoluteUri

public static boolean isAbsoluteUri(java.lang.String uri)

Checks whether the specified URI is an absolute URI. An absolute URI contains the complete path to a node in the DMT starting from the DMT root (".").

Parameters:

uri - the URI to be checked, must not be null and must contain a valid URI

Returns:

whether the specified URI is absolute

Throws:

java.lang.NullPointerException - if the specified URI is null

java.lang.IllegalArgumentException - if the specified URI is malformed

isValidUri

public static boolean isValidUri(java.lang.String uri)

Checks whether the specified URI is valid. A URI is considered valid if it meets the following constraints:

• the URI is not null;
• the URI follows the syntax defined for valid DMT URIs;
• the length of the URI is not more than getMaxUriLength();
• the URI doesn't contain more than getMaxUriSegments() segments;
• the length of each segment of the URI is less than or equal to getMaxSegmentNameLength().

The exact definition of the length of a URI and its segments is given in the descriptions of the getMaxUriLength() and getMaxSegmentNameLength() methods.

Parameters:

uri - the URI to be validated

Returns:

whether the specified URI is valid