Copyright (c) 2000, 2015 IBM Corporation and others. This program and the accompanying materials are made available under the terms of the Eclipse Public License 2.0 which accompanies this distribution, and is available at https://www.eclipse.org/legal/epl-2.0/ SPDX-License-Identifier: EPL-2.0 Contributors: IBM Corporation - initial API and implementation Patrick Tasse - Add extra constructor to Path class (bug 454959)
/******************************************************************************* * Copyright (c) 2000, 2015 IBM Corporation and others. * * This program and the accompanying materials * are made available under the terms of the Eclipse Public License 2.0 * which accompanies this distribution, and is available at * https://www.eclipse.org/legal/epl-2.0/ * * SPDX-License-Identifier: EPL-2.0 * * Contributors: * IBM Corporation - initial API and implementation * Patrick Tasse - Add extra constructor to Path class (bug 454959) *******************************************************************************/
package org.eclipse.core.runtime;
A path is an ordered collection of string segments, separated by a standard separator character, "/". A path may also have a leading and/or a trailing separator. Paths may also be prefixed by an optional device id, which includes the character(s) which separate the device id from the rest of the path. For example, "C:" and "Server/Volume:" are typical device ids. A device independent path has null for a device id.

Note that paths are value objects; all operations on paths return a new path; the path that is operated on is unscathed.

UNC paths are denoted by leading double-slashes such as //Server/Volume/My/Path. When a new path is constructed all double-slashes are removed except those appearing at the beginning of the path.

This interface can be used without OSGi running.

This interface is not intended to be implemented by clients.

See Also:
  • Path
@noextendThis interface is not intended to be extended by clients.
@noimplementThis interface is not intended to be implemented by clients.
/** * A path is an ordered collection of string segments, * separated by a standard separator character, "/". * A path may also have a leading and/or a trailing separator. * Paths may also be prefixed by an optional device id, which includes * the character(s) which separate the device id from the rest * of the path. For example, "C:" and "Server/Volume:" are typical * device ids. * A device independent path has <code>null</code> for a device id. * <p> * Note that paths are value objects; all operations on paths * return a new path; the path that is operated on is unscathed. * </p> * <p> * UNC paths are denoted by leading double-slashes such * as <code>//Server/Volume/My/Path</code>. When a new path * is constructed all double-slashes are removed except those * appearing at the beginning of the path. * </p> * <p> * This interface can be used without OSGi running. * </p><p> * This interface is not intended to be implemented by clients. * </p> * @see Path * @noextend This interface is not intended to be extended by clients. * @noimplement This interface is not intended to be implemented by clients. */
public interface IPath extends Cloneable {
Path separator character constant "/" used in paths.
/** * Path separator character constant "/" used in paths. */
public static final char SEPARATOR = '/';
Device separator character constant ":" used in paths.
/** * Device separator character constant ":" used in paths. */
public static final char DEVICE_SEPARATOR = ':';
Returns a new path which is the same as this path but with the given file extension added. If this path is empty, root or has a trailing separator, this path is returned. If this path already has an extension, the existing extension is left and the given extension simply appended. Clients wishing to replace the current extension should first remove the extension and then add the desired one.

The file extension portion is defined as the string following the last period (".") character in the last segment. The given extension should not include a leading ".".

Params:
  • extension – the file extension to append
Returns:the new path
/** * Returns a new path which is the same as this path but with * the given file extension added. If this path is empty, root or has a * trailing separator, this path is returned. If this path already * has an extension, the existing extension is left and the given * extension simply appended. Clients wishing to replace * the current extension should first remove the extension and * then add the desired one. * <p> * The file extension portion is defined as the string * following the last period (".") character in the last segment. * The given extension should not include a leading ".". * </p> * * @param extension the file extension to append * @return the new path */
public IPath addFileExtension(String extension);
Returns a path with the same segments as this path but with a trailing separator added. This path must have at least one segment.

If this path already has a trailing separator, this path is returned.

See Also:
Returns:the new path
/** * Returns a path with the same segments as this path * but with a trailing separator added. * This path must have at least one segment. * <p> * If this path already has a trailing separator, * this path is returned. * </p> * * @return the new path * @see #hasTrailingSeparator() * @see #removeTrailingSeparator() */
public IPath addTrailingSeparator();
Returns the canonicalized path obtained from the concatenation of the given string path to the end of this path. The given string path must be a valid path. If it has a trailing separator, the result will have a trailing separator. The device id of this path is preserved (the one of the given string is ignored). Duplicate slashes are removed from the path except at the beginning where the path is considered to be UNC.
Params:
  • path – the string path to concatenate
See Also:
Returns:the new path
/** * Returns the canonicalized path obtained from the * concatenation of the given string path to the * end of this path. The given string path must be a valid * path. If it has a trailing separator, * the result will have a trailing separator. * The device id of this path is preserved (the one * of the given string is ignored). Duplicate slashes * are removed from the path except at the beginning * where the path is considered to be UNC. * * @param path the string path to concatenate * @return the new path * @see #isValidPath(String) */
public IPath append(String path);
Returns the canonicalized path obtained from the concatenation of the given path's segments to the end of this path. If the given path has a trailing separator, the result will have a trailing separator. The device id of this path is preserved (the one of the given path is ignored). Duplicate slashes are removed from the path except at the beginning where the path is considered to be UNC.
Params:
  • path – the path to concatenate
Returns:the new path
/** * Returns the canonicalized path obtained from the * concatenation of the given path's segments to the * end of this path. If the given path has a trailing * separator, the result will have a trailing separator. * The device id of this path is preserved (the one * of the given path is ignored). Duplicate slashes * are removed from the path except at the beginning * where the path is considered to be UNC. * * @param path the path to concatenate * @return the new path */
public IPath append(IPath path);
Returns a copy of this path.
Returns:the cloned path
/** * Returns a copy of this path. * * @return the cloned path */
public Object clone();
Returns whether this path equals the given object.

Equality for paths is defined to be: same sequence of segments, same absolute/relative status, and same device. Trailing separators are disregarded. The paths' file systems are disregarded. Paths are not generally considered equal to objects other than paths.

Params:
  • obj – the other object
Returns:true if the paths are equivalent, and false if they are not
/** * Returns whether this path equals the given object. * <p> * Equality for paths is defined to be: same sequence of segments, * same absolute/relative status, and same device. * Trailing separators are disregarded. * The paths' file systems are disregarded. * Paths are not generally considered equal to objects other than paths. * </p> * * @param obj the other object * @return <code>true</code> if the paths are equivalent, * and <code>false</code> if they are not */
@Override public boolean equals(Object obj);
Returns the device id for this path, or null if this path has no device id. Note that the result will end in ':'.
See Also:
Returns:the device id, or null
/** * Returns the device id for this path, or <code>null</code> if this * path has no device id. Note that the result will end in ':'. * * @return the device id, or <code>null</code> * @see #setDevice(String) */
public String getDevice();
Returns the file extension portion of this path, or null if there is none.

The file extension portion is defined as the string following the last period (".") character in the last segment. If there is no period in the last segment, the path has no file extension portion. If the last segment ends in a period, the file extension portion is the empty string.

Returns:the file extension or null
/** * Returns the file extension portion of this path, * or <code>null</code> if there is none. * <p> * The file extension portion is defined as the string * following the last period (".") character in the last segment. * If there is no period in the last segment, the path has no * file extension portion. If the last segment ends in a period, * the file extension portion is the empty string. * </p> * * @return the file extension or <code>null</code> */
public String getFileExtension();
Returns whether this path has a trailing separator.

Note: In the root path ("/"), the separator is considered to be leading rather than trailing.

See Also:
Returns:true if this path has a trailing separator, and false otherwise
/** * Returns whether this path has a trailing separator. * <p> * Note: In the root path ("/"), the separator is considered to * be leading rather than trailing. * </p> * * @return <code>true</code> if this path has a trailing * separator, and <code>false</code> otherwise * @see #addTrailingSeparator() * @see #removeTrailingSeparator() */
public boolean hasTrailingSeparator();
Returns whether this path is an absolute path (ignoring any device id).

Absolute paths start with a path separator. A root path, like / or C:/, is considered absolute. UNC paths are always absolute.

Returns:true if this path is an absolute path, and false otherwise
/** * Returns whether this path is an absolute path (ignoring * any device id). * <p> * Absolute paths start with a path separator. * A root path, like <code>/</code> or <code>C:/</code>, * is considered absolute. UNC paths are always absolute. * </p> * * @return <code>true</code> if this path is an absolute path, * and <code>false</code> otherwise */
public boolean isAbsolute();
Returns whether this path has no segments and is not a root path.
Returns:true if this path is empty, and false otherwise
/** * Returns whether this path has no segments and is not * a root path. * * @return <code>true</code> if this path is empty, * and <code>false</code> otherwise */
public boolean isEmpty();
Returns whether this path is a prefix of the given path. To be a prefix, this path's segments must appear in the argument path in the same order, and their device ids must match.

An empty path is a prefix of all paths with the same device; a root path is a prefix of all absolute paths with the same device.

Params:
  • anotherPath – the other path
Returns:true if this path is a prefix of the given path, and false otherwise
/** * Returns whether this path is a prefix of the given path. * To be a prefix, this path's segments must * appear in the argument path in the same order, * and their device ids must match. * <p> * An empty path is a prefix of all paths with the same device; a root path is a prefix of * all absolute paths with the same device. * </p> * @param anotherPath the other path * @return <code>true</code> if this path is a prefix of the given path, * and <code>false</code> otherwise */
public boolean isPrefixOf(IPath anotherPath);
Returns whether this path is a root path.

The root path is the absolute non-UNC path with zero segments; e.g., / or C:/. The separator is considered a leading separator, not a trailing one.

Returns:true if this path is a root path, and false otherwise
/** * Returns whether this path is a root path. * <p> * The root path is the absolute non-UNC path with zero segments; * e.g., <code>/</code> or <code>C:/</code>. * The separator is considered a leading separator, not a trailing one. * </p> * * @return <code>true</code> if this path is a root path, * and <code>false</code> otherwise */
public boolean isRoot();
Returns a boolean value indicating whether or not this path is considered to be in UNC form. Return false if this path has a device set or if the first 2 characters of the path string are not Path.SEPARATOR.
Returns:boolean indicating if this path is UNC
/** * Returns a boolean value indicating whether or not this path * is considered to be in UNC form. Return false if this path * has a device set or if the first 2 characters of the path string * are not <code>Path.SEPARATOR</code>. * * @return boolean indicating if this path is UNC */
public boolean isUNC();
Returns whether the given string is syntactically correct as a path, on this path's file system. The device id is the prefix up to and including the device separator for the file system of this path; the path proper is everything to the right of it, or the entire string if there is no device separator. When this path's file system has no meaningful device separator, the entire string is treated as the path proper. The device id is not checked for validity; the path proper is correct if each of the segments in its canonicalized form is valid.
Params:
  • path – the path to check
See Also:
Returns:true if the given string is a valid path, and false otherwise
/** * Returns whether the given string is syntactically correct as a path, on * this path's file system. The device id is the prefix up to and including * the device separator for the file system of this path; the path proper is * everything to the right of it, or the entire string if there is no device * separator. When this path's file system has no meaningful device * separator, the entire string is treated as the path proper. * The device id is not checked for validity; the path proper is correct * if each of the segments in its canonicalized form is valid. * * @param path the path to check * @return <code>true</code> if the given string is a valid path, * and <code>false</code> otherwise * @see #isValidSegment(String) */
public boolean isValidPath(String path);
Returns whether the given string is valid as a segment in this path. The rules for valid segments are as follows:
  • the empty string is not valid
  • any string containing the slash character ('/') is not valid
  • any string containing segment or device separator characters on this path's file system, such as the backslash ('\') and colon (':') on some file systems, is not valid
Params:
  • segment – the path segment to check
Returns:true if the given path segment is valid, and false otherwise
/** * Returns whether the given string is valid as a segment in this path. * The rules for valid segments are as follows: * <ul> * <li>the empty string is not valid * <li>any string containing the slash character ('/') is not valid * <li>any string containing segment or device separator characters on this * path's file system, such as the backslash ('\') and colon (':') on some * file systems, is not valid * </ul> * * @param segment the path segment to check * @return <code>true</code> if the given path segment is valid, * and <code>false</code> otherwise */
public boolean isValidSegment(String segment);
Returns the last segment of this path, or null if it does not have any segments.
Returns:the last segment of this path, or null
/** * Returns the last segment of this path, or * <code>null</code> if it does not have any segments. * * @return the last segment of this path, or <code>null</code> */
public String lastSegment();
Returns an absolute path with the segments and device id of this path. Absolute paths start with a path separator. If this path is absolute, it is simply returned.
Returns:the new path
/** * Returns an absolute path with the segments and device id of this path. * Absolute paths start with a path separator. If this path is absolute, * it is simply returned. * * @return the new path */
public IPath makeAbsolute();
Returns a relative path with the segments and device id of this path. Absolute paths start with a path separator and relative paths do not. If this path is relative, it is simply returned.
Returns:the new path
/** * Returns a relative path with the segments and device id of this path. * Absolute paths start with a path separator and relative paths do not. * If this path is relative, it is simply returned. * * @return the new path */
public IPath makeRelative();
Returns a path equivalent to this path, but relative to the given base path if possible.

The path is only made relative if the base path if both paths have the same device and have a non-zero length common prefix. If the paths have different devices, or no common prefix, then this path is simply returned. If the path is successfully made relative, then appending the returned path to the base will always produce a path equal to this path.

Params:
  • base – The base path to make this path relative to
Returns:A path relative to the base path, or this path if it could not be made relative to the given base
Since:org.eclipse.equinox.common 3.5
/** * Returns a path equivalent to this path, but relative to the given base path if possible. * <p> * The path is only made relative if the base path if both paths have the same device * and have a non-zero length common prefix. If the paths have different devices, * or no common prefix, then this path is simply returned. If the path is successfully * made relative, then appending the returned path to the base will always produce * a path equal to this path. * </p> * @param base The base path to make this path relative to * @return A path relative to the base path, or this path if it could * not be made relative to the given base * @since org.eclipse.equinox.common 3.5 */
public IPath makeRelativeTo(IPath base);
Return a new path which is the equivalent of this path converted to UNC form (if the given boolean is true) or this path not as a UNC path (if the given boolean is false). If UNC, the returned path will not have a device and the first 2 characters of the path string will be Path.SEPARATOR. If not UNC, the first 2 characters of the returned path string will not be Path.SEPARATOR.
Params:
  • toUNC – true if converting to UNC, false otherwise
Returns:the new path, either in UNC form or not depending on the boolean parameter
/** * Return a new path which is the equivalent of this path converted to UNC * form (if the given boolean is true) or this path not as a UNC path (if the given * boolean is false). If UNC, the returned path will not have a device and the * first 2 characters of the path string will be <code>Path.SEPARATOR</code>. If not UNC, the * first 2 characters of the returned path string will not be <code>Path.SEPARATOR</code>. * * @param toUNC true if converting to UNC, false otherwise * @return the new path, either in UNC form or not depending on the boolean parameter */
public IPath makeUNC(boolean toUNC);
Returns a count of the number of segments which match in this path and the given path (device ids are ignored), comparing in increasing segment number order.
Params:
  • anotherPath – the other path
Returns:the number of matching segments
/** * Returns a count of the number of segments which match in * this path and the given path (device ids are ignored), * comparing in increasing segment number order. * * @param anotherPath the other path * @return the number of matching segments */
public int matchingFirstSegments(IPath anotherPath);
Returns a new path which is the same as this path but with the file extension removed. If this path does not have an extension, this path is returned.

The file extension portion is defined as the string following the last period (".") character in the last segment. If there is no period in the last segment, the path has no file extension portion. If the last segment ends in a period, the file extension portion is the empty string.

Returns:the new path
/** * Returns a new path which is the same as this path but with * the file extension removed. If this path does not have an * extension, this path is returned. * <p> * The file extension portion is defined as the string * following the last period (".") character in the last segment. * If there is no period in the last segment, the path has no * file extension portion. If the last segment ends in a period, * the file extension portion is the empty string. * </p> * * @return the new path */
public IPath removeFileExtension();
Returns a copy of this path with the given number of segments removed from the beginning. The device id is preserved. The count must be greater or equal zero.

If the count is zero, this path is returned. This is the only case where the returned path can be absolute. Use makeRelative() if necessary.

If the count is greater than zero, the resulting path will always be a relative path.

If the count equals or exceeds the number of segments in this path, an empty relative path is returned.

Params:
  • count – the number of segments to remove
Returns:the new path
/** * Returns a copy of this path with the given number of segments * removed from the beginning. The device id is preserved. * The count must be greater or equal zero. * <p> * If the count is zero, this path is returned. This is the only case where * the returned path can be absolute. Use {@link #makeRelative()} if necessary. * </p><p> * If the count is greater than zero, the resulting path will * always be a relative path. * </p><p> * If the count equals or exceeds the number * of segments in this path, an empty relative path is returned. * </p> * * @param count the number of segments to remove * @return the new path */
public IPath removeFirstSegments(int count);
Returns a copy of this path with the given number of segments removed from the end. The device id is preserved. The number must be greater or equal zero. If the count is zero, this path is returned.

If this path has a trailing separator, it will still have a trailing separator after the last segments are removed (assuming there are some segments left). If there is no trailing separator, the result will not have a trailing separator. If the number equals or exceeds the number of segments in this path, a path with no segments is returned.

Params:
  • count – the number of segments to remove
Returns:the new path
/** * Returns a copy of this path with the given number of segments * removed from the end. The device id is preserved. * The number must be greater or equal zero. * If the count is zero, this path is returned. * <p> * If this path has a trailing separator, it will still * have a trailing separator after the last segments are removed * (assuming there are some segments left). If there is no * trailing separator, the result will not have a trailing * separator. * If the number equals or exceeds the number * of segments in this path, a path with no segments is returned. * </p> * * @param count the number of segments to remove * @return the new path */
public IPath removeLastSegments(int count);
Returns a path with the same segments as this path but with a trailing separator removed. Does nothing if this path does not have at least one segment. The device id is preserved.

If this path does not have a trailing separator, this path is returned.

See Also:
Returns:the new path
/** * Returns a path with the same segments as this path * but with a trailing separator removed. * Does nothing if this path does not have at least one segment. * The device id is preserved. * <p> * If this path does not have a trailing separator, * this path is returned. * </p> * * @return the new path * @see #addTrailingSeparator() * @see #hasTrailingSeparator() */
public IPath removeTrailingSeparator();
Returns the specified segment of this path, or null if the path does not have such a segment.
Params:
  • index – the 0-based segment index
Returns:the specified segment, or null
/** * Returns the specified segment of this path, or * <code>null</code> if the path does not have such a segment. * * @param index the 0-based segment index * @return the specified segment, or <code>null</code> */
public String segment(int index);
Returns the number of segments in this path.

Note that both root and empty paths have 0 segments.

Returns:the number of segments
/** * Returns the number of segments in this path. * <p> * Note that both root and empty paths have 0 segments. * </p> * * @return the number of segments */
public int segmentCount();
Returns the segments in this path in order.
Returns:an array of string segments
/** * Returns the segments in this path in order. * * @return an array of string segments */
public String[] segments();
Returns a new path which is the same as this path but with the given device id. The device id must end with a ":". A device independent path is obtained by passing null.

For example, "C:" and "Server/Volume:" are typical device ids.

Params:
  • device – the device id or null
See Also:
Returns:a new path
/** * Returns a new path which is the same as this path but with * the given device id. The device id must end with a ":". * A device independent path is obtained by passing <code>null</code>. * <p> * For example, "C:" and "Server/Volume:" are typical device ids. * </p> * * @param device the device id or <code>null</code> * @return a new path * @see #getDevice() */
public IPath setDevice(String device);
Returns a java.io.File corresponding to this path.
Returns:the file corresponding to this path
/** * Returns a <code>java.io.File</code> corresponding to this path. * * @return the file corresponding to this path */
public java.io.File toFile();
Returns a string representation of this path which uses the platform-dependent path separator defined by java.io.File. This method is like toString() except that the latter always uses the same separator (/) regardless of platform.

This string is suitable for passing to java.io.File(String).

Returns:a platform-dependent string representation of this path
/** * Returns a string representation of this path which uses the * platform-dependent path separator defined by <code>java.io.File</code>. * This method is like <code>toString()</code> except that the * latter always uses the same separator (<code>/</code>) regardless of platform. * <p> * This string is suitable for passing to <code>java.io.File(String)</code>. * </p> * * @return a platform-dependent string representation of this path */
public String toOSString();
Returns a platform-neutral string representation of this path. The format is not specified, except that the resulting string can be passed back to the Path#fromPortableString(String) constructor to produce the exact same path on any platform.

This string is suitable for passing to Path#fromPortableString(String).

See Also:
Returns:a platform-neutral string representation of this path
Since:3.1
/** * Returns a platform-neutral string representation of this path. The * format is not specified, except that the resulting string can be * passed back to the <code>Path#fromPortableString(String)</code> * constructor to produce the exact same path on any platform. * <p> * This string is suitable for passing to <code>Path#fromPortableString(String)</code>. * </p> * * @return a platform-neutral string representation of this path * @see Path#fromPortableString(String) * @since 3.1 */
public String toPortableString();
Returns a string representation of this path, including its device id. The same separator, "/", is used on all platforms.

Example result strings (without and with device id):

"/foo/bar.txt"
"bar.txt"
"/foo/"
"foo/"
""
"/"
"C:/foo/bar.txt"
"C:bar.txt"
"C:/foo/"
"C:foo/"
"C:"
"C:/"
This string is suitable for passing to Path(String).
See Also:
Returns:a string representation of this path
/** * Returns a string representation of this path, including its * device id. The same separator, "/", is used on all platforms. * <p> * Example result strings (without and with device id): * </p> * <pre> * "/foo/bar.txt" * "bar.txt" * "/foo/" * "foo/" * "" * "/" * "C:/foo/bar.txt" * "C:bar.txt" * "C:/foo/" * "C:foo/" * "C:" * "C:/" * </pre> * This string is suitable for passing to <code>Path(String)</code>. * * @return a string representation of this path * @see Path */
@Override public String toString();
Returns a copy of this path truncated after the given number of segments. The number must not be negative. The device id is preserved.

If this path has a trailing separator, the result will too (assuming there are some segments left). If there is no trailing separator, the result will not have a trailing separator. Copying up to segment zero simply means making an copy with no path segments.

Params:
  • count – the segment number at which to truncate the path
Returns:the new path
/** * Returns a copy of this path truncated after the * given number of segments. The number must not be negative. * The device id is preserved. * <p> * If this path has a trailing separator, the result will too * (assuming there are some segments left). If there is no * trailing separator, the result will not have a trailing * separator. * Copying up to segment zero simply means making an copy with * no path segments. * </p> * * @param count the segment number at which to truncate the path * @return the new path */
public IPath uptoSegment(int count); }