com.landawn.abacus.util.FilenameUtil Maven / Gradle / Ivy
Show all versions of abacus-common Show documentation
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.landawn.abacus.util;
import java.io.File;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Collection;
import java.util.Stack;
/**
* Note: it's copied from Apache Commons IO developed at The Apache Software Foundation (http://www.apache.org/), or under the Apache License 2.0.
*
* General filename and filepath manipulation utilities.
*
* When dealing with filenames you can hit problems when moving from a Windows
* based development machine to a Unix based production machine.
* This class aims to help avoid those problems.
*
* NOTE: You may be able to avoid using this class entirely simply by
* using JDK {@link java.io.File File} objects and the two argument constructor
* {@link java.io.File#File(java.io.File, java.lang.String) File(File,String)}.
*
* Most methods on this class are designed to work the same on both Unix and Windows.
* Those that don't include 'System', 'Unix' or 'Windows' in their name.
*
* Most methods recognise both separators (forward and back), and both
* sets of prefixes. See the javadoc of each method for details.
*
* This class defines six components within a filename
* (example C:\dev\project\file.txt):
*
* - the prefix - C:\
* - the path - dev\project\
* - the full path - C:\dev\project\
* - the name - file.txt
* - the base name - file
* - the extension - txt
*
* Note that this class works best if directory filenames end with a separator.
* If you omit the last separator, it is impossible to determine if the filename
* corresponds to a file or a directory. As a result, we have chosen to say
* it corresponds to a file.
*
* This class only supports Unix and Windows style names.
* Prefixes are matched as follows:
*
* Windows:
* a\b\c.txt --> "" --> relative
* \a\b\c.txt --> "\" --> current drive absolute
* C:a\b\c.txt --> "C:" --> drive relative
* C:\a\b\c.txt --> "C:\" --> absolute
* \\server\a\b\c.txt --> "\\server\" --> UNC
*
* Unix:
* a/b/c.txt --> "" --> relative
* /a/b/c.txt --> "/" --> absolute
* ~/a/b/c.txt --> "~/" --> current user
* ~ --> "~/" --> current user (slash added)
* ~user/a/b/c.txt --> "~user/" --> named user
* ~user --> "~user/" --> named user (slash added)
*
* Both prefix styles are matched always, irrespective of the machine that you are
* currently running on.
*
* Origin of code: Excalibur, Alexandria, Tomcat, Commons-Utils.
*
* @since 1.1
*/
public final class FilenameUtil {
private static final int NOT_FOUND = -1;
/**
* The extension separator character.
* @since 1.4
*/
public static final char EXTENSION_SEPARATOR = '.';
/**
* The extension separator String.
* @since 1.4
*/
public static final String EXTENSION_SEPARATOR_STR = Character.toString(EXTENSION_SEPARATOR);
/**
* The Unix separator character.
*/
private static final char UNIX_SEPARATOR = '/';
/**
* The Windows separator character.
*/
private static final char WINDOWS_SEPARATOR = '\\';
/**
* The system separator character.
*/
private static final char SYSTEM_SEPARATOR = File.separatorChar;
/**
* The separator character that is the opposite of the system separator.
*/
private static final char OTHER_SEPARATOR;
static {
if (IOUtil.IS_OS_WINDOWS) {
OTHER_SEPARATOR = UNIX_SEPARATOR;
} else {
OTHER_SEPARATOR = WINDOWS_SEPARATOR;
}
}
/**
* Instances should NOT be constructed in standard programming.
*/
private FilenameUtil() {
// singleton for utility class.
}
//-----------------------------------------------------------------------
/**
* Checks if the character is a separator.
*
* @param ch the character to check
* @return true if it is a separator character
*/
private static boolean isSeparator(final char ch) {
return ch == UNIX_SEPARATOR || ch == WINDOWS_SEPARATOR;
}
//-----------------------------------------------------------------------
/**
* Normalizes a path, removing double and single dot path steps.
*
* This method normalizes a path to a standard format.
* The input may contain separators in either Unix or Windows format.
* The output will contain separators in the format of the system.
*
* A trailing slash will be retained.
* A double slash will be merged to a single slash (but UNC names are handled).
* A single dot path segment will be removed.
* A double dot will cause that path segment and the one before to be removed.
* If the double dot has no parent path segment to work with, {@code null}
* is returned.
*
* The output will be the same on both Unix and Windows except
* for the separator character.
*
* /foo// --> /foo/
* /foo/./ --> /foo/
* /foo/../bar --> /bar
* /foo/../bar/ --> /bar/
* /foo/../bar/../baz --> /baz
* //foo//./bar --> /foo/bar
* /../ --> null
* ../foo --> null
* foo/bar/.. --> foo/
* foo/../../bar --> null
* foo/../bar --> bar
* //server/foo/../bar --> //server/bar
* //server/../bar --> null
* C:\foo\..\bar --> C:\bar
* C:\..\bar --> null
* ~/foo/../bar/ --> ~/bar/
* ~/../bar --> null
*
* (Note the file separator returned will be correct for Windows/Unix)
*
* @param filename the filename to normalize, null returns null
* @return the normalized filename, or null if invalid. Null bytes inside string will be removed
*/
public static String normalize(final String filename) {
return doNormalize(filename, SYSTEM_SEPARATOR, true);
}
/**
* Normalizes a path, removing double and single dot path steps.
*
* This method normalizes a path to a standard format.
* The input may contain separators in either Unix or Windows format.
* The output will contain separators in the format specified.
*
* A trailing slash will be retained.
* A double slash will be merged to a single slash (but UNC names are handled).
* A single dot path segment will be removed.
* A double dot will cause that path segment and the one before to be removed.
* If the double dot has no parent path segment to work with, {@code null}
* is returned.
*
* The output will be the same on both Unix and Windows except
* for the separator character.
*
* /foo// --> /foo/
* /foo/./ --> /foo/
* /foo/../bar --> /bar
* /foo/../bar/ --> /bar/
* /foo/../bar/../baz --> /baz
* //foo//./bar --> /foo/bar
* /../ --> null
* ../foo --> null
* foo/bar/.. --> foo/
* foo/../../bar --> null
* foo/../bar --> bar
* //server/foo/../bar --> //server/bar
* //server/../bar --> null
* C:\foo\..\bar --> C:\bar
* C:\..\bar --> null
* ~/foo/../bar/ --> ~/bar/
* ~/../bar --> null
*
* The output will be the same on both Unix and Windows including
* the separator character.
*
* @param filename the filename to normalize, null returns null
* @param unixSeparator {@code true} if a unix separator should
* be used or {@code false} if a windows separator should be used.
* @return the normalized filename, or null if invalid. Null bytes inside string will be removed
* @since 2.0
*/
public static String normalize(final String filename, final boolean unixSeparator) {
final char separator = unixSeparator ? UNIX_SEPARATOR : WINDOWS_SEPARATOR;
return doNormalize(filename, separator, true);
}
//-----------------------------------------------------------------------
/**
* Normalizes a path, removing double and single dot path steps,
* and removing any final directory separator.
*
* This method normalizes a path to a standard format.
* The input may contain separators in either Unix or Windows format.
* The output will contain separators in the format of the system.
*
* A trailing slash will be removed.
* A double slash will be merged to a single slash (but UNC names are handled).
* A single dot path segment will be removed.
* A double dot will cause that path segment and the one before to be removed.
* If the double dot has no parent path segment to work with, {@code null}
* is returned.
*
* The output will be the same on both Unix and Windows except
* for the separator character.
*
* /foo// --> /foo
* /foo/./ --> /foo
* /foo/../bar --> /bar
* /foo/../bar/ --> /bar
* /foo/../bar/../baz --> /baz
* //foo//./bar --> /foo/bar
* /../ --> null
* ../foo --> null
* foo/bar/.. --> foo
* foo/../../bar --> null
* foo/../bar --> bar
* //server/foo/../bar --> //server/bar
* //server/../bar --> null
* C:\foo\..\bar --> C:\bar
* C:\..\bar --> null
* ~/foo/../bar/ --> ~/bar
* ~/../bar --> null
*
* (Note the file separator returned will be correct for Windows/Unix)
*
* @param filename the filename to normalize, null returns null
* @return the normalized filename, or null if invalid. Null bytes inside string will be removed
*/
public static String normalizeNoEndSeparator(final String filename) {
return doNormalize(filename, SYSTEM_SEPARATOR, false);
}
/**
* Normalizes a path, removing double and single dot path steps,
* and removing any final directory separator.
*
* This method normalizes a path to a standard format.
* The input may contain separators in either Unix or Windows format.
* The output will contain separators in the format specified.
*
* A trailing slash will be removed.
* A double slash will be merged to a single slash (but UNC names are handled).
* A single dot path segment will be removed.
* A double dot will cause that path segment and the one before to be removed.
* If the double dot has no parent path segment to work with, {@code null}
* is returned.
*
* The output will be the same on both Unix and Windows including
* the separator character.
*
* /foo// --> /foo
* /foo/./ --> /foo
* /foo/../bar --> /bar
* /foo/../bar/ --> /bar
* /foo/../bar/../baz --> /baz
* //foo//./bar --> /foo/bar
* /../ --> null
* ../foo --> null
* foo/bar/.. --> foo
* foo/../../bar --> null
* foo/../bar --> bar
* //server/foo/../bar --> //server/bar
* //server/../bar --> null
* C:\foo\..\bar --> C:\bar
* C:\..\bar --> null
* ~/foo/../bar/ --> ~/bar
* ~/../bar --> null
*
*
* @param filename the filename to normalize, null returns null
* @param unixSeparator {@code true} if a unix separator should
* be used or {@code false} if a windows separator should be used.
* @return the normalized filename, or null if invalid. Null bytes inside string will be removed
* @since 2.0
*/
public static String normalizeNoEndSeparator(final String filename, final boolean unixSeparator) {
final char separator = unixSeparator ? UNIX_SEPARATOR : WINDOWS_SEPARATOR;
return doNormalize(filename, separator, false);
}
/**
* Internal method to perform the normalization.
*
* @param filename
* @param separator The separator character to use
* @param keepSeparator true to keep the final separator
* @return the normalized filename. Null bytes inside string will be removed.
*/
private static String doNormalize(final String filename, final char separator, final boolean keepSeparator) {
if (filename == null) {
return null;
}
failIfNullBytePresent(filename);
int size = filename.length();
if (size == 0) {
return filename;
}
final int prefix = getPrefixLength(filename);
if (prefix < 0) {
return null;
}
final char[] array = new char[size + 2]; // +1 for possible extra slash, +2 for arraycopy
filename.getChars(0, filename.length(), array, 0);
// fix separators throughout
final char otherSeparator = separator == SYSTEM_SEPARATOR ? OTHER_SEPARATOR : SYSTEM_SEPARATOR;
for (int i = 0; i < array.length; i++) {
if (array[i] == otherSeparator) {
array[i] = separator;
}
}
// add extra separator on the end to simplify code below
boolean lastIsDirectory = true;
if (array[size - 1] != separator) {
array[size++] = separator;
lastIsDirectory = false;
}
// adjoining slashes
for (int i = prefix + 1; i < size; i++) {
if (array[i] == separator && array[i - 1] == separator) {
System.arraycopy(array, i, array, i - 1, size - i);
size--;
i--;
}
}
// dot slash
for (int i = prefix + 1; i < size; i++) {
if (array[i] == separator && array[i - 1] == '.' && (i == prefix + 1 || array[i - 2] == separator)) {
if (i == size - 1) {
lastIsDirectory = true;
}
System.arraycopy(array, i + 1, array, i - 1, size - i);
size -= 2;
i--;
}
}
// double dot slash
outer: for (int i = prefix + 2; i < size; i++) {
if (array[i] == separator && array[i - 1] == '.' && array[i - 2] == '.' && (i == prefix + 2 || array[i - 3] == separator)) {
if (i == prefix + 2) {
return null;
}
if (i == size - 1) {
lastIsDirectory = true;
}
int j;
for (j = i - 4; j >= prefix; j--) {
if (array[j] == separator) {
// remove b/../ from a/b/../c
System.arraycopy(array, i + 1, array, j + 1, size - i);
size -= i - j;
i = j + 1;
continue outer;
}
}
// remove a/../ from a/../c
System.arraycopy(array, i + 1, array, prefix, size - i);
size -= i + 1 - prefix;
i = prefix + 1;
}
}
if (size <= 0) { // should never be less than 0
return "";
}
if ((size <= prefix) || (lastIsDirectory && keepSeparator)) {
return new String(array, 0, size); // keep trailing separator
}
return new String(array, 0, size - 1); // lose trailing separator
}
//-----------------------------------------------------------------------
/**
* Concatenates a filename to a base path using normal command line style rules.
*
* The effect is equivalent to resultant directory after changing
* directory to the first argument, followed by changing directory to
* the second argument.
*
* The first argument is the base path, the second is the path to concatenate.
* The returned path is always normalized via {@link #normalize(String)},
* thus .. is handled.
*
* If pathToAdd is absolute (has an absolute prefix), then
* it will be normalized and returned.
* Otherwise, the paths will be joined, normalized and returned.
*
* The output will be the same on both Unix and Windows except
* for the separator character.
*
* /foo/ + bar --> /foo/bar
* /foo + bar --> /foo/bar
* /foo + /bar --> /bar
* /foo + C:/bar --> C:/bar
* /foo + C:bar --> C:bar (*)
* /foo/a/ + ../bar --> foo/bar
* /foo/ + ../../bar --> null
* /foo/ + /bar --> /bar
* /foo/.. + /bar --> /bar
* /foo + bar/c.txt --> /foo/bar/c.txt
* /foo/c.txt + bar --> /foo/c.txt/bar (!)
*
* (*) Note that the Windows relative drive prefix is unreliable when
* used with this method.
* (!) Note that the first parameter must be a path. If it ends with a name, then
* the name will be built into the concatenated path. If this might be a problem,
* use {@link #getFullPath(String)} on the base path argument.
*
* @param basePath the base path to attach to, always treated as a path
* @param fullFilenameToAdd the filename (or path) to attach to the base
* @return the concatenated path, or null if invalid. Null bytes inside string will be removed
*/
public static String concat(final String basePath, final String fullFilenameToAdd) {
final int prefix = getPrefixLength(fullFilenameToAdd);
if (prefix < 0) {
return null;
}
if (prefix > 0) {
return normalize(fullFilenameToAdd);
}
if (basePath == null) {
return null;
}
final int len = basePath.length();
if (len == 0) {
return normalize(fullFilenameToAdd);
}
final char ch = basePath.charAt(len - 1);
if (isSeparator(ch)) {
return normalize(basePath + fullFilenameToAdd);
} else {
return normalize(basePath + '/' + fullFilenameToAdd);
}
}
/**
* Determines whether the {@code parent} directory contains the {@code child} element (a file or directory).
*
* The files names are expected to be normalized.
*
*
* Edge cases:
*
* - A {@code directory} must not be null: if null, throw IllegalArgumentException
* - A directory does not contain itself: return false
* - A null child file is not contained in any parent: return false
*
*
* @param canonicalParent
* the file to consider as the parent.
* @param canonicalChild
* the file to consider as the child.
* @return true is the candidate leaf is under by the specified composite. False otherwise.
* @throws IOException
* if an IO error occurs while checking the files.
* @since 2.2
* @see FileUtils#directoryContains(File, File)
*/
public static boolean directoryContains(final String canonicalParent, final String canonicalChild) throws IOException {
// Fail fast against NullPointerException
if (canonicalParent == null) {
throw new IllegalArgumentException("Directory must not be null");
}
if ((canonicalChild == null) || IOCase.SYSTEM.checkEquals(canonicalParent, canonicalChild)) {
return false;
}
return IOCase.SYSTEM.checkStartsWith(canonicalChild, canonicalParent);
}
//-----------------------------------------------------------------------
/**
* Converts all separators to the Unix separator of forward slash.
*
* @param path the path to be changed, null ignored
* @return the updated path
*/
public static String separatorsToUnix(final String path) {
if (path == null || path.indexOf(WINDOWS_SEPARATOR) == NOT_FOUND) {
return path;
}
return path.replace(WINDOWS_SEPARATOR, UNIX_SEPARATOR);
}
/**
* Converts all separators to the Windows separator of backslash.
*
* @param path the path to be changed, null ignored
* @return the updated path
*/
public static String separatorsToWindows(final String path) {
if (path == null || path.indexOf(UNIX_SEPARATOR) == NOT_FOUND) {
return path;
}
return path.replace(UNIX_SEPARATOR, WINDOWS_SEPARATOR);
}
/**
* Converts all separators to the system separator.
*
* @param path the path to be changed, null ignored
* @return the updated path
*/
public static String separatorsToSystem(final String path) {
if (path == null) {
return null;
}
if (IOUtil.IS_OS_WINDOWS) {
return separatorsToWindows(path);
} else {
return separatorsToUnix(path);
}
}
//-----------------------------------------------------------------------
/**
* Returns the length of the filename prefix, such as C:/ or ~/.
*
* This method will handle a file in either Unix or Windows format.
*
* The prefix length includes the first slash in the full filename
* if applicable. Thus, it is possible that the length returned is greater
* than the length of the input string.
*
* Windows:
* a\b\c.txt --> "" --> relative
* \a\b\c.txt --> "\" --> current drive absolute
* C:a\b\c.txt --> "C:" --> drive relative
* C:\a\b\c.txt --> "C:\" --> absolute
* \\server\a\b\c.txt --> "\\server\" --> UNC
* \\\a\b\c.txt --> error, length = -1
*
* Unix:
* a/b/c.txt --> "" --> relative
* /a/b/c.txt --> "/" --> absolute
* ~/a/b/c.txt --> "~/" --> current user
* ~ --> "~/" --> current user (slash added)
* ~user/a/b/c.txt --> "~user/" --> named user
* ~user --> "~user/" --> named user (slash added)
* //server/a/b/c.txt --> "//server/"
* ///a/b/c.txt --> error, length = -1
*
*
* The output will be the same irrespective of the machine that the code is running on.
* ie. both Unix and Windows prefixes are matched regardless.
*
* Note that a leading // (or \\) is used to indicate a UNC name on Windows.
* These must be followed by a server name, so double-slashes are not collapsed
* to a single slash at the start of the filename.
*
* @param filename the filename to find the prefix in, null returns -1
* @return the length of the prefix, -1 if invalid or null
*/
public static int getPrefixLength(final String filename) {
if (filename == null) {
return NOT_FOUND;
}
final int len = filename.length();
if (len == 0) {
return 0;
}
char ch0 = filename.charAt(0);
if (ch0 == ':') {
return NOT_FOUND;
}
if (len == 1) {
if (ch0 == '~') {
return 2; // return a length greater than the input
}
return isSeparator(ch0) ? 1 : 0;
} else {
if (ch0 == '~') {
int posUnix = filename.indexOf(UNIX_SEPARATOR, 1);
int posWin = filename.indexOf(WINDOWS_SEPARATOR, 1);
if (posUnix == NOT_FOUND && posWin == NOT_FOUND) {
return len + 1; // return a length greater than the input
}
posUnix = posUnix == NOT_FOUND ? posWin : posUnix;
posWin = posWin == NOT_FOUND ? posUnix : posWin;
return Math.min(posUnix, posWin) + 1;
}
final char ch1 = filename.charAt(1);
if (ch1 == ':') {
ch0 = Character.toUpperCase(ch0);
if (ch0 >= 'A' && ch0 <= 'Z') {
if (len == 2 || !isSeparator(filename.charAt(2))) {
return 2;
}
return 3;
} else if (ch0 == UNIX_SEPARATOR) {
return 1;
}
return NOT_FOUND;
} else if (isSeparator(ch0) && isSeparator(ch1)) {
int posUnix = filename.indexOf(UNIX_SEPARATOR, 2);
int posWin = filename.indexOf(WINDOWS_SEPARATOR, 2);
if (posUnix == NOT_FOUND && posWin == NOT_FOUND || posUnix == 2 || posWin == 2) {
return NOT_FOUND;
}
posUnix = posUnix == NOT_FOUND ? posWin : posUnix;
posWin = posWin == NOT_FOUND ? posUnix : posWin;
return Math.min(posUnix, posWin) + 1;
} else {
return isSeparator(ch0) ? 1 : 0;
}
}
}
/**
* Returns the index of the last directory separator character.
*
* This method will handle a file in either Unix or Windows format.
* The position of the last forward or backslash is returned.
*
* The output will be the same irrespective of the machine that the code is running on.
*
* @param filename the filename to find the last path separator in, null returns -1
* @return the index of the last separator character, or -1 if there
* is no such character
*/
public static int indexOfLastSeparator(final String filename) {
if (filename == null) {
return NOT_FOUND;
}
final int lastUnixPos = filename.lastIndexOf(UNIX_SEPARATOR);
final int lastWindowsPos = filename.lastIndexOf(WINDOWS_SEPARATOR);
return Math.max(lastUnixPos, lastWindowsPos);
}
/**
* Returns the index of the last extension separator character, which is a dot.
*
* This method also checks that there is no directory separator after the last dot. To do this it uses
* {@link #indexOfLastSeparator(String)} which will handle a file in either Unix or Windows format.
*
*
* The output will be the same irrespective of the machine that the code is running on.
*
*
* @param filename
* the filename to find the last extension separator in, null returns -1
* @return the index of the last extension separator character, or -1 if there is no such character
*/
public static int indexOfExtension(final String filename) {
if (filename == null) {
return NOT_FOUND;
}
final int extensionPos = filename.lastIndexOf(EXTENSION_SEPARATOR);
final int lastSeparator = indexOfLastSeparator(filename);
return lastSeparator > extensionPos ? NOT_FOUND : extensionPos;
}
//-----------------------------------------------------------------------
/**
* Gets the prefix from a full filename, such as C:/
* or ~/.
*
* This method will handle a file in either Unix or Windows format.
* The prefix includes the first slash in the full filename where applicable.
*
* Windows:
* a\b\c.txt --> "" --> relative
* \a\b\c.txt --> "\" --> current drive absolute
* C:a\b\c.txt --> "C:" --> drive relative
* C:\a\b\c.txt --> "C:\" --> absolute
* \\server\a\b\c.txt --> "\\server\" --> UNC
*
* Unix:
* a/b/c.txt --> "" --> relative
* /a/b/c.txt --> "/" --> absolute
* ~/a/b/c.txt --> "~/" --> current user
* ~ --> "~/" --> current user (slash added)
* ~user/a/b/c.txt --> "~user/" --> named user
* ~user --> "~user/" --> named user (slash added)
*
*
* The output will be the same irrespective of the machine that the code is running on.
* ie. both Unix and Windows prefixes are matched regardless.
*
* @param filename the filename to query, null returns null
* @return the prefix of the file, null if invalid. Null bytes inside string will be removed
*/
public static String getPrefix(final String filename) {
if (filename == null) {
return null;
}
final int len = getPrefixLength(filename);
if (len < 0) {
return null;
}
if (len > filename.length()) {
failIfNullBytePresent(filename + UNIX_SEPARATOR);
return filename + UNIX_SEPARATOR;
}
final String path = filename.substring(0, len);
failIfNullBytePresent(path);
return path;
}
/**
* Gets the path from a full filename, which excludes the prefix.
*
* This method will handle a file in either Unix or Windows format.
* The method is entirely text based, and returns the text before and
* including the last forward or backslash.
*
* C:\a\b\c.txt --> a\b\
* ~/a/b/c.txt --> a/b/
* a.txt --> ""
* a/b/c --> a/b/
* a/b/c/ --> a/b/c/
*
*
* The output will be the same irrespective of the machine that the code is running on.
*
* This method drops the prefix from the result.
* See {@link #getFullPath(String)} for the method that retains the prefix.
*
* @param filename the filename to query, null returns null
* @return the path of the file, an empty string if none exists, null if invalid.
* Null bytes inside string will be removed
*/
public static String getPath(final String filename) {
return doGetPath(filename, 1);
}
/**
* Gets the path from a full filename, which excludes the prefix, and
* also excluding the final directory separator.
*
* This method will handle a file in either Unix or Windows format.
* The method is entirely text based, and returns the text before the
* last forward or backslash.
*
* C:\a\b\c.txt --> a\b
* ~/a/b/c.txt --> a/b
* a.txt --> ""
* a/b/c --> a/b
* a/b/c/ --> a/b/c
*
*
* The output will be the same irrespective of the machine that the code is running on.
*
* This method drops the prefix from the result.
* See {@link #getFullPathNoEndSeparator(String)} for the method that retains the prefix.
*
* @param filename the filename to query, null returns null
* @return the path of the file, an empty string if none exists, null if invalid.
* Null bytes inside string will be removed
*/
public static String getPathNoEndSeparator(final String filename) {
return doGetPath(filename, 0);
}
/**
* Does the work of getting the path.
*
* @param filename
* @param separatorAdd 0 to omit the end separator, 1 to return it
* @return the path. Null bytes inside string will be removed
*/
private static String doGetPath(final String filename, final int separatorAdd) {
if (filename == null) {
return null;
}
final int prefix = getPrefixLength(filename);
if (prefix < 0) {
return null;
}
final int index = indexOfLastSeparator(filename);
final int endIndex = index + separatorAdd;
if (prefix >= filename.length() || index < 0 || prefix >= endIndex) {
return "";
}
final String path = filename.substring(prefix, endIndex);
failIfNullBytePresent(path);
return path;
}
/**
* Gets the full path from a full filename, which is the prefix + path.
*
* This method will handle a file in either Unix or Windows format.
* The method is entirely text based, and returns the text before and
* including the last forward or backslash.
*
* C:\a\b\c.txt --> C:\a\b\
* ~/a/b/c.txt --> ~/a/b/
* a.txt --> ""
* a/b/c --> a/b/
* a/b/c/ --> a/b/c/
* C: --> C:
* C:\ --> C:\
* ~ --> ~/
* ~/ --> ~/
* ~user --> ~user/
* ~user/ --> ~user/
*
*
* The output will be the same irrespective of the machine that the code is running on.
*
* @param filename the filename to query, null returns null
* @return the path of the file, an empty string if none exists, null if invalid
*/
public static String getFullPath(final String filename) {
return doGetFullPath(filename, true);
}
/**
* Gets the full path from a full filename, which is the prefix + path,
* and also excluding the final directory separator.
*
* This method will handle a file in either Unix or Windows format.
* The method is entirely text based, and returns the text before the
* last forward or backslash.
*
* C:\a\b\c.txt --> C:\a\b
* ~/a/b/c.txt --> ~/a/b
* a.txt --> ""
* a/b/c --> a/b
* a/b/c/ --> a/b/c
* C: --> C:
* C:\ --> C:\
* ~ --> ~
* ~/ --> ~
* ~user --> ~user
* ~user/ --> ~user
*
*
* The output will be the same irrespective of the machine that the code is running on.
*
* @param filename the filename to query, null returns null
* @return the path of the file, an empty string if none exists, null if invalid
*/
public static String getFullPathNoEndSeparator(final String filename) {
return doGetFullPath(filename, false);
}
/**
* Does the work of getting the path.
*
* @param filename
* @param includeSeparator true to include the end separator
* @return the path
*/
private static String doGetFullPath(final String filename, final boolean includeSeparator) {
if (filename == null) {
return null;
}
final int prefix = getPrefixLength(filename);
if (prefix < 0) {
return null;
}
if (prefix >= filename.length()) {
if (includeSeparator) {
return getPrefix(filename); // add end slash if necessary
} else {
return filename;
}
}
final int index = indexOfLastSeparator(filename);
if (index < 0) {
return filename.substring(0, prefix);
}
int end = index + (includeSeparator ? 1 : 0);
if (end == 0) {
end++;
}
return filename.substring(0, end);
}
/**
* Gets the name minus the path from a full filename.
*
* This method will handle a file in either Unix or Windows format.
* The text after the last forward or backslash is returned.
*
* a/b/c.txt --> c.txt
* a.txt --> a.txt
* a/b/c --> c
* a/b/c/ --> ""
*
*
* The output will be the same irrespective of the machine that the code is running on.
*
* @param filename the filename to query, null returns null
* @return the name of the file without the path, or an empty string if none exists.
* Null bytes inside string will be removed
*/
public static String getName(final String filename) {
if (filename == null) {
return null;
}
failIfNullBytePresent(filename);
final int index = indexOfLastSeparator(filename);
return filename.substring(index + 1);
}
/**
* Check the input for null bytes, a sign of unsanitized data being passed to to file level functions.
*
* This may be used for poison byte attacks.
* @param path the path to check
*/
private static void failIfNullBytePresent(final String path) {
final int len = path.length();
for (int i = 0; i < len; i++) {
if (path.charAt(i) == 0) {
throw new IllegalArgumentException("Null byte present in file/path name. There are no "
+ "known legitimate use cases for such data, but several injection attacks may use it");
}
}
}
/**
* Gets the base name, minus the full path and extension, from a full filename.
*
* This method will handle a file in either Unix or Windows format.
* The text after the last forward or backslash and before the last dot is returned.
*
* a/b/c.txt --> c
* a.txt --> a
* a/b/c --> c
* a/b/c/ --> ""
*
*
* The output will be the same irrespective of the machine that the code is running on.
*
* @param filename the filename to query, null returns null
* @return the name of the file without the path, or an empty string if none exists. Null bytes inside string
* will be removed
*/
public static String getBaseName(final String filename) {
return removeExtension(getName(filename));
}
/**
* Gets the extension of a filename.
*
* This method returns the textual part of the filename after the last dot.
* There must be no directory separator after the dot.
*
* foo.txt --> "txt"
* a/b/c.jpg --> "jpg"
* a/b.txt/c --> ""
* a/b/c --> ""
*
*
* The output will be the same irrespective of the machine that the code is running on.
*
* @param filename the filename to retrieve the extension of.
* @return the extension of the file or an empty string if none exists or {@code null}
* if the filename is {@code null}.
*/
public static String getExtension(final String filename) {
if (filename == null) {
return null;
}
final int index = indexOfExtension(filename);
if (index == NOT_FOUND) {
return "";
} else {
return filename.substring(index + 1);
}
}
//-----------------------------------------------------------------------
/**
* Removes the extension from a filename.
*
* This method returns the textual part of the filename before the last dot.
* There must be no directory separator after the dot.
*
* foo.txt --> foo
* a\b\c.jpg --> a\b\c
* a\b\c --> a\b\c
* a.b\c --> a.b\c
*
*
* The output will be the same irrespective of the machine that the code is running on.
*
* @param filename the filename to query, null returns null
* @return the filename minus the extension
*/
public static String removeExtension(final String filename) {
if (filename == null) {
return null;
}
failIfNullBytePresent(filename);
final int index = indexOfExtension(filename);
if (index == NOT_FOUND) {
return filename;
} else {
return filename.substring(0, index);
}
}
//-----------------------------------------------------------------------
/**
* Checks whether two filenames are equal exactly.
*
* No processing is performed on the filenames other than comparison,
* thus this is merely a null-safe case-sensitive equals.
*
* @param filename1 the first filename to query, may be null
* @param filename2 the second filename to query, may be null
* @return true if the filenames are equal, null equals null
* @see IOCase#SENSITIVE
*/
public static boolean equals(final String filename1, final String filename2) {
return equals(filename1, filename2, false, IOCase.SENSITIVE);
}
/**
* Checks whether two filenames are equal using the case rules of the system.
*
* No processing is performed on the filenames other than comparison.
* The check is case-sensitive on Unix and case-insensitive on Windows.
*
* @param filename1 the first filename to query, may be null
* @param filename2 the second filename to query, may be null
* @return true if the filenames are equal, null equals null
* @see IOCase#SYSTEM
*/
public static boolean equalsOnSystem(final String filename1, final String filename2) {
return equals(filename1, filename2, false, IOCase.SYSTEM);
}
//-----------------------------------------------------------------------
/**
* Checks whether two filenames are equal after both have been normalized.
*
* Both filenames are first passed to {@link #normalize(String)}.
* The check is then performed in a case-sensitive manner.
*
* @param filename1 the first filename to query, may be null
* @param filename2 the second filename to query, may be null
* @return true if the filenames are equal, null equals null
* @see IOCase#SENSITIVE
*/
public static boolean equalsNormalized(final String filename1, final String filename2) {
return equals(filename1, filename2, true, IOCase.SENSITIVE);
}
/**
* Checks whether two filenames are equal after both have been normalized
* and using the case rules of the system.
*
* Both filenames are first passed to {@link #normalize(String)}.
* The check is then performed case-sensitive on Unix and
* case-insensitive on Windows.
*
* @param filename1 the first filename to query, may be null
* @param filename2 the second filename to query, may be null
* @return true if the filenames are equal, null equals null
* @see IOCase#SYSTEM
*/
public static boolean equalsNormalizedOnSystem(final String filename1, final String filename2) {
return equals(filename1, filename2, true, IOCase.SYSTEM);
}
/**
* Checks whether two filenames are equal, optionally normalizing and providing
* control over the case-sensitivity.
*
* @param filename1 the first filename to query, may be null
* @param filename2 the second filename to query, may be null
* @param normalized whether to normalize the filenames
* @param caseSensitivity what case sensitivity rule to use, null means case-sensitive
* @return true if the filenames are equal, null equals null
* @since 1.3
*/
public static boolean equals(String filename1, String filename2, final boolean normalized, IOCase caseSensitivity) {
if (filename1 == null || filename2 == null) {
return filename1 == null && filename2 == null;
}
if (normalized) {
filename1 = normalize(filename1);
filename2 = normalize(filename2);
if (filename1 == null || filename2 == null) {
throw new NullPointerException("Error normalizing one or both of the file names");
}
}
if (caseSensitivity == null) {
caseSensitivity = IOCase.SENSITIVE;
}
return caseSensitivity.checkEquals(filename1, filename2);
}
//-----------------------------------------------------------------------
/**
* Checks whether the extension of the filename is that specified.
*
* This method obtains the extension as the textual part of the filename
* after the last dot. There must be no directory separator after the dot.
* The extension check is case-sensitive on all platforms.
*
* @param filename the filename to query, null returns false
* @param extension the extension to check for, null or empty checks for no extension
* @return true if the filename has the specified extension
* @throws java.lang.IllegalArgumentException if the supplied filename contains null bytes
*/
public static boolean isExtension(final String filename, final String extension) {
if (filename == null) {
return false;
}
failIfNullBytePresent(filename);
if (extension == null || extension.isEmpty()) {
return indexOfExtension(filename) == NOT_FOUND;
}
final String fileExt = getExtension(filename);
return fileExt.equals(extension);
}
/**
* Checks whether the extension of the filename is one of those specified.
*
* This method obtains the extension as the textual part of the filename
* after the last dot. There must be no directory separator after the dot.
* The extension check is case-sensitive on all platforms.
*
* @param filename the filename to query, null returns false
* @param extensions the extensions to check for, null checks for no extension
* @return true if the filename is one of the extensions
* @throws java.lang.IllegalArgumentException if the supplied filename contains null bytes
*/
public static boolean isExtension(final String filename, final String[] extensions) {
if (filename == null) {
return false;
}
failIfNullBytePresent(filename);
if (extensions == null || extensions.length == 0) {
return indexOfExtension(filename) == NOT_FOUND;
}
final String fileExt = getExtension(filename);
for (final String extension : extensions) {
if (fileExt.equals(extension)) {
return true;
}
}
return false;
}
/**
* Checks whether the extension of the filename is one of those specified.
*
* This method obtains the extension as the textual part of the filename
* after the last dot. There must be no directory separator after the dot.
* The extension check is case-sensitive on all platforms.
*
* @param filename the filename to query, null returns false
* @param extensions the extensions to check for, null checks for no extension
* @return true if the filename is one of the extensions
* @throws java.lang.IllegalArgumentException if the supplied filename contains null bytes
*/
public static boolean isExtension(final String filename, final Collection extensions) {
if (filename == null) {
return false;
}
failIfNullBytePresent(filename);
if (extensions == null || extensions.isEmpty()) {
return indexOfExtension(filename) == NOT_FOUND;
}
final String fileExt = getExtension(filename);
for (final String extension : extensions) {
if (fileExt.equals(extension)) {
return true;
}
}
return false;
}
//-----------------------------------------------------------------------
/**
* Checks a filename to see if it matches the specified wildcard matcher,
* always testing case-sensitive.
*
* The wildcard matcher uses the characters '?' and '*' to represent a
* single or multiple (zero or more) wildcard characters.
* This is the same as often found on Dos/Unix command lines.
* The check is case-sensitive always.
*
* wildcardMatch("c.txt", "*.txt") --> true
* wildcardMatch("c.txt", "*.jpg") --> false
* wildcardMatch("a/b/c.txt", "a/b/*") --> true
* wildcardMatch("c.txt", "*.???") --> true
* wildcardMatch("c.txt", "*.????") --> false
*
* N.B. the sequence "*?" does not work properly at present in match strings.
*
* @param filename the filename to match on
* @param wildcardMatcher the wildcard string to match against
* @return true if the filename matches the wildcard string
* @see IOCase#SENSITIVE
*/
public static boolean wildcardMatch(final String filename, final String wildcardMatcher) {
return wildcardMatch(filename, wildcardMatcher, IOCase.SENSITIVE);
}
/**
* Checks a filename to see if it matches the specified wildcard matcher
* using the case rules of the system.
*
* The wildcard matcher uses the characters '?' and '*' to represent a
* single or multiple (zero or more) wildcard characters.
* This is the same as often found on Dos/Unix command lines.
* The check is case-sensitive on Unix and case-insensitive on Windows.
*
* wildcardMatch("c.txt", "*.txt") --> true
* wildcardMatch("c.txt", "*.jpg") --> false
* wildcardMatch("a/b/c.txt", "a/b/*") --> true
* wildcardMatch("c.txt", "*.???") --> true
* wildcardMatch("c.txt", "*.????") --> false
*
* N.B. the sequence "*?" does not work properly at present in match strings.
*
* @param filename the filename to match on
* @param wildcardMatcher the wildcard string to match against
* @return true if the filename matches the wildcard string
* @see IOCase#SYSTEM
*/
public static boolean wildcardMatchOnSystem(final String filename, final String wildcardMatcher) {
return wildcardMatch(filename, wildcardMatcher, IOCase.SYSTEM);
}
/**
* Checks a filename to see if it matches the specified wildcard matcher
* allowing control over case-sensitivity.
*
* The wildcard matcher uses the characters '?' and '*' to represent a
* single or multiple (zero or more) wildcard characters.
* N.B. the sequence "*?" does not work properly at present in match strings.
*
* @param filename the filename to match on
* @param wildcardMatcher the wildcard string to match against
* @param caseSensitivity what case sensitivity rule to use, null means case-sensitive
* @return true if the filename matches the wildcard string
* @since 1.3
*/
public static boolean wildcardMatch(final String filename, final String wildcardMatcher, IOCase caseSensitivity) {
if (filename == null && wildcardMatcher == null) {
return true;
}
if (filename == null || wildcardMatcher == null) {
return false;
}
if (caseSensitivity == null) {
caseSensitivity = IOCase.SENSITIVE;
}
final String[] wcs = splitOnTokens(wildcardMatcher);
boolean anyChars = false;
int textIdx = 0;
int wcsIdx = 0;
final Stack backtrack = new Stack<>();
// loop around a backtrack stack, to handle complex * matching
do {
if (backtrack.size() > 0) {
final int[] array = backtrack.pop();
wcsIdx = array[0];
textIdx = array[1];
anyChars = true;
}
// loop whilst tokens and text left to process
while (wcsIdx < wcs.length) {
if (wcs[wcsIdx].equals("?")) {
// ? so move to next text char
textIdx++;
if (textIdx > filename.length()) {
break;
}
anyChars = false;
} else if (wcs[wcsIdx].equals("*")) {
// set any chars status
anyChars = true;
if (wcsIdx == wcs.length - 1) {
textIdx = filename.length();
}
} else {
// matching text token
if (anyChars) {
// any chars then try to locate text token
textIdx = caseSensitivity.checkIndexOf(filename, textIdx, wcs[wcsIdx]);
if (textIdx == NOT_FOUND) {
// token not found
break;
}
final int repeat = caseSensitivity.checkIndexOf(filename, textIdx + 1, wcs[wcsIdx]);
if (repeat >= 0) {
backtrack.push(new int[] { wcsIdx, repeat });
}
} else {
// matching from current position
if (!caseSensitivity.checkRegionMatches(filename, textIdx, wcs[wcsIdx])) {
// couldnt match token
break;
}
}
// matched text token, move text index to end of matched token
textIdx += wcs[wcsIdx].length();
anyChars = false;
}
wcsIdx++;
}
// full match
if (wcsIdx == wcs.length && textIdx == filename.length()) {
return true;
}
} while (backtrack.size() > 0);
return false;
}
/**
* Splits a string into a number of tokens.
* The text is split by '?' and '*'.
* Where multiple '*' occur consecutively they are collapsed into a single '*'.
*
* @param text the text to split
* @return the array of tokens, never null
*/
static String[] splitOnTokens(final String text) {
// used by wildcardMatch
// package level so a unit test may run on this
if (text.indexOf('?') == NOT_FOUND && text.indexOf('*') == NOT_FOUND) {
return new String[] { text };
}
final char[] array = text.toCharArray();
final ArrayList list = new ArrayList<>();
final StringBuilder buffer = new StringBuilder();
char prevChar = 0;
for (final char ch : array) {
if (ch == '?' || ch == '*') {
if (buffer.length() != 0) {
list.add(buffer.toString());
buffer.setLength(0);
}
if (ch == '?') {
list.add("?");
} else if (prevChar != '*') {// ch == '*' here; check if previous char was '*'
list.add("*");
}
} else {
buffer.append(ch);
}
prevChar = ch;
}
if (buffer.length() != 0) {
list.add(buffer.toString());
}
return list.toArray(new String[list.size()]);
}
}