001 /*
002 * Licensed to the Apache Software Foundation (ASF) under one or more
003 * contributor license agreements. See the NOTICE file distributed with
004 * this work for additional information regarding copyright ownership.
005 * The ASF licenses this file to You under the Apache License, Version 2.0
006 * (the "License"); you may not use this file except in compliance with
007 * the License. You may obtain a copy of the License at
008 *
009 * http://www.apache.org/licenses/LICENSE-2.0
010 *
011 * Unless required by applicable law or agreed to in writing, software
012 * distributed under the License is distributed on an "AS IS" BASIS,
013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014 * See the License for the specific language governing permissions and
015 * limitations under the License.
016 */
017
018 package org.apache.commons.net.ftp.parser;
019 import java.text.ParseException;
020 import java.util.List;
021 import java.util.ListIterator;
022
023 import org.apache.commons.net.ftp.FTPClientConfig;
024 import org.apache.commons.net.ftp.FTPFile;
025
026 /**
027 * Implementation FTPFileEntryParser and FTPFileListParser for standard
028 * Unix Systems.
029 *
030 * This class is based on the logic of Daniel Savarese's
031 * DefaultFTPListParser, but adapted to use regular expressions and to fit the
032 * new FTPFileEntryParser interface.
033 * @version $Id: UnixFTPEntryParser.java 1299238 2012-03-10 17:12:28Z sebb $
034 * @see org.apache.commons.net.ftp.FTPFileEntryParser FTPFileEntryParser (for usage instructions)
035 */
036 public class UnixFTPEntryParser extends ConfigurableFTPFileEntryParserImpl
037 {
038
039 static final String DEFAULT_DATE_FORMAT
040 = "MMM d yyyy"; //Nov 9 2001
041
042 static final String DEFAULT_RECENT_DATE_FORMAT
043 = "MMM d HH:mm"; //Nov 9 20:06
044
045 static final String NUMERIC_DATE_FORMAT
046 = "yyyy-MM-dd HH:mm"; //2001-11-09 20:06
047
048 /**
049 * Some Linux distributions are now shipping an FTP server which formats
050 * file listing dates in an all-numeric format:
051 * <code>"yyyy-MM-dd HH:mm</code>.
052 * This is a very welcome development, and hopefully it will soon become
053 * the standard. However, since it is so new, for now, and possibly
054 * forever, we merely accomodate it, but do not make it the default.
055 * <p>
056 * For now end users may specify this format only via
057 * <code>UnixFTPEntryParser(FTPClientConfig)</code>.
058 * Steve Cohen - 2005-04-17
059 */
060 public static final FTPClientConfig NUMERIC_DATE_CONFIG =
061 new FTPClientConfig(
062 FTPClientConfig.SYST_UNIX,
063 NUMERIC_DATE_FORMAT,
064 null, null, null, null);
065
066 /**
067 * this is the regular expression used by this parser.
068 *
069 * Permissions:
070 * r the file is readable
071 * w the file is writable
072 * x the file is executable
073 * - the indicated permission is not granted
074 * L mandatory locking occurs during access (the set-group-ID bit is
075 * on and the group execution bit is off)
076 * s the set-user-ID or set-group-ID bit is on, and the corresponding
077 * user or group execution bit is also on
078 * S undefined bit-state (the set-user-ID bit is on and the user
079 * execution bit is off)
080 * t the 1000 (octal) bit, or sticky bit, is on [see chmod(1)], and
081 * execution is on
082 * T the 1000 bit is turned on, and execution is off (undefined bit-
083 * state)
084 * e z/OS external link bit
085 */
086 private static final String REGEX =
087 "([bcdelfmpSs-])"
088 +"(((r|-)(w|-)([xsStTL-]))((r|-)(w|-)([xsStTL-]))((r|-)(w|-)([xsStTL-])))\\+?\\s*"
089 + "(\\d+)\\s+" // link count
090 + "(?:(\\S+(?:\\s\\S+)*?)\\s+)?" // owner name (optional spaces)
091 + "(?:(\\S+(?:\\s\\S+)*)\\s+)?" // group name (optional spaces)
092 + "(\\d+(?:,\\s*\\d+)?)\\s+" // size or n,m
093 /*
094 * numeric or standard format date:
095 * yyyy-mm-dd (expecting hh:mm to follow)
096 * MMM [d]d
097 * [d]d MMM
098 * N.B. use non-space for MMM to allow for languages such as German which use
099 * diacritics (e.g. umlaut) in some abbreviations.
100 */
101 + "((?:\\d+[-/]\\d+[-/]\\d+)|(?:\\S{3}\\s+\\d{1,2})|(?:\\d{1,2}\\s+\\S{3}))\\s+"
102 /*
103 year (for non-recent standard format) - yyyy
104 or time (for numeric or recent standard format) [h]h:mm
105 */
106 + "(\\d+(?::\\d+)?)\\s+"
107
108 + "(\\S*)(\\s*.*)"; // the rest
109
110
111 /**
112 * The default constructor for a UnixFTPEntryParser object.
113 *
114 * @exception IllegalArgumentException
115 * Thrown if the regular expression is unparseable. Should not be seen
116 * under normal conditions. It it is seen, this is a sign that
117 * <code>REGEX</code> is not a valid regular expression.
118 */
119 public UnixFTPEntryParser()
120 {
121 this(null);
122 }
123
124 /**
125 * This constructor allows the creation of a UnixFTPEntryParser object with
126 * something other than the default configuration.
127 *
128 * @param config The {@link FTPClientConfig configuration} object used to
129 * configure this parser.
130 * @exception IllegalArgumentException
131 * Thrown if the regular expression is unparseable. Should not be seen
132 * under normal conditions. It it is seen, this is a sign that
133 * <code>REGEX</code> is not a valid regular expression.
134 * @since 1.4
135 */
136 public UnixFTPEntryParser(FTPClientConfig config)
137 {
138 super(REGEX);
139 configure(config);
140 }
141
142 /**
143 * Preparse the list to discard "total nnn" lines
144 */
145 @Override
146 public List<String> preParse(List<String> original) {
147 ListIterator<String> iter = original.listIterator();
148 while (iter.hasNext()) {
149 String entry = iter.next();
150 if (entry.matches("^total \\d+$")) { // NET-389
151 iter.remove();
152 }
153 }
154 return original;
155 }
156
157 /**
158 * Parses a line of a unix (standard) FTP server file listing and converts
159 * it into a usable format in the form of an <code> FTPFile </code>
160 * instance. If the file listing line doesn't describe a file,
161 * <code> null </code> is returned, otherwise a <code> FTPFile </code>
162 * instance representing the files in the directory is returned.
163 * <p>
164 * @param entry A line of text from the file listing
165 * @return An FTPFile instance corresponding to the supplied entry
166 */
167 public FTPFile parseFTPEntry(String entry) {
168 FTPFile file = new FTPFile();
169 file.setRawListing(entry);
170 int type;
171 boolean isDevice = false;
172
173 if (matches(entry))
174 {
175 String typeStr = group(1);
176 String hardLinkCount = group(15);
177 String usr = group(16);
178 String grp = group(17);
179 String filesize = group(18);
180 String datestr = group(19) + " " + group(20);
181 String name = group(21);
182 String endtoken = group(22);
183
184 try
185 {
186 file.setTimestamp(super.parseTimestamp(datestr));
187 }
188 catch (ParseException e)
189 {
190 // intentionally do nothing
191 }
192
193 // A 'whiteout' file is an ARTIFICIAL entry in any of several types of
194 // 'translucent' filesystems, of which a 'union' filesystem is one.
195
196 // bcdelfmpSs-
197 switch (typeStr.charAt(0))
198 {
199 case 'd':
200 type = FTPFile.DIRECTORY_TYPE;
201 break;
202 case 'e': // NET-39 => z/OS external link
203 type = FTPFile.SYMBOLIC_LINK_TYPE;
204 break;
205 case 'l':
206 type = FTPFile.SYMBOLIC_LINK_TYPE;
207 break;
208 case 'b':
209 case 'c':
210 isDevice = true;
211 type = FTPFile.FILE_TYPE; // TODO change this if DEVICE_TYPE implemented
212 break;
213 case 'f':
214 case '-':
215 type = FTPFile.FILE_TYPE;
216 break;
217 default: // e.g. ? and w = whiteout
218 type = FTPFile.UNKNOWN_TYPE;
219 }
220
221 file.setType(type);
222
223 int g = 4;
224 for (int access = 0; access < 3; access++, g += 4)
225 {
226 // Use != '-' to avoid having to check for suid and sticky bits
227 file.setPermission(access, FTPFile.READ_PERMISSION,
228 (!group(g).equals("-")));
229 file.setPermission(access, FTPFile.WRITE_PERMISSION,
230 (!group(g + 1).equals("-")));
231
232 String execPerm = group(g + 2);
233 if (!execPerm.equals("-") && !Character.isUpperCase(execPerm.charAt(0)))
234 {
235 file.setPermission(access, FTPFile.EXECUTE_PERMISSION, true);
236 }
237 else
238 {
239 file.setPermission(access, FTPFile.EXECUTE_PERMISSION, false);
240 }
241 }
242
243 if (!isDevice)
244 {
245 try
246 {
247 file.setHardLinkCount(Integer.parseInt(hardLinkCount));
248 }
249 catch (NumberFormatException e)
250 {
251 // intentionally do nothing
252 }
253 }
254
255 file.setUser(usr);
256 file.setGroup(grp);
257
258 try
259 {
260 file.setSize(Long.parseLong(filesize));
261 }
262 catch (NumberFormatException e)
263 {
264 // intentionally do nothing
265 }
266
267 if (null == endtoken)
268 {
269 file.setName(name);
270 }
271 else
272 {
273 // oddball cases like symbolic links, file names
274 // with spaces in them.
275 name += endtoken;
276 if (type == FTPFile.SYMBOLIC_LINK_TYPE)
277 {
278
279 int end = name.indexOf(" -> ");
280 // Give up if no link indicator is present
281 if (end == -1)
282 {
283 file.setName(name);
284 }
285 else
286 {
287 file.setName(name.substring(0, end));
288 file.setLink(name.substring(end + 4));
289 }
290
291 }
292 else
293 {
294 file.setName(name);
295 }
296 }
297 return file;
298 }
299 return null;
300 }
301
302 /**
303 * Defines a default configuration to be used when this class is
304 * instantiated without a {@link FTPClientConfig FTPClientConfig}
305 * parameter being specified.
306 * @return the default configuration for this parser.
307 */
308 @Override
309 protected FTPClientConfig getDefaultConfiguration() {
310 return new FTPClientConfig(
311 FTPClientConfig.SYST_UNIX,
312 DEFAULT_DATE_FORMAT,
313 DEFAULT_RECENT_DATE_FORMAT,
314 null, null, null);
315 }
316
317 }