View Javadoc

1   /*
2    * Licensed to the Apache Software Foundation (ASF) under one or more
3    * contributor license agreements.  See the NOTICE file distributed with
4    * this work for additional information regarding copyright ownership.
5    * The ASF licenses this file to You under the Apache License, Version 2.0
6    * (the "License"); you may not use this file except in compliance with
7    * the License.  You may obtain a copy of the License at
8    *
9    *      http://www.apache.org/licenses/LICENSE-2.0
10   *
11   * Unless required by applicable law or agreed to in writing, software
12   * distributed under the License is distributed on an "AS IS" BASIS,
13   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14   * See the License for the specific language governing permissions and
15   * limitations under the License.
16   */
17  
18  package org.apache.commons.net.ftp.parser;
19  
20  import java.text.ParseException;
21  import java.util.List;
22  
23  import org.apache.commons.net.ftp.FTPClientConfig;
24  import org.apache.commons.net.ftp.FTPFile;
25  
26  /**
27   * Implementation of FTPFileEntryParser and FTPFileListParser for IBM zOS/MVS
28   * Systems.
29   *
30   * @author <a href="henrik.sorensen@balcab.ch">Henrik Sorensen</a>
31   * @author <a href="jnadler@srcginc.com">Jeff Nadler</a>
32   * @author <a href="wnoto@openfinance.com">William Noto</a>
33   *
34   * @version $Id: MVSFTPEntryParser.java 1490237 2013-06-06 11:17:23Z sebb $
35   * @see org.apache.commons.net.ftp.FTPFileEntryParser FTPFileEntryParser (for
36   *      usage instructions)
37   */
38  public class MVSFTPEntryParser extends ConfigurableFTPFileEntryParserImpl {
39  
40      static final int UNKNOWN_LIST_TYPE = -1;
41      static final int FILE_LIST_TYPE = 0;
42      static final int MEMBER_LIST_TYPE = 1;
43      static final int UNIX_LIST_TYPE = 2;
44      static final int JES_LEVEL_1_LIST_TYPE = 3;
45      static final int JES_LEVEL_2_LIST_TYPE = 4;
46  
47      private int isType = UNKNOWN_LIST_TYPE;
48  
49      /**
50       * Fallback parser for Unix-style listings
51       */
52      private UnixFTPEntryParser unixFTPEntryParser;
53  
54      /**
55       * Dates are ignored for file lists, but are used for member lists where
56       * possible
57       */
58      static final String DEFAULT_DATE_FORMAT = "yyyy/MM/dd HH:mm"; // 2001/09/18
59                                                                      // 13:52
60  
61      /**
62       * Matches these entries: Volume Unit Referred Ext Used Recfm Lrecl BlkSz
63       * Dsorg Dsname B10142 3390 2006/03/20 2 31 F 80 80 PS MDI.OKL.WORK
64       *
65       */
66      static final String FILE_LIST_REGEX = "\\S+\\s+" + // volume
67                                                                  // ignored
68              "\\S+\\s+" + // unit - ignored
69              "\\S+\\s+" + // access date - ignored
70              "\\S+\\s+" + // extents -ignored
71              "\\S+\\s+" + // used - ignored
72              "[FV]\\S*\\s+" + // recfm - must start with F or V
73              "\\S+\\s+" + // logical record length -ignored
74              "\\S+\\s+" + // block size - ignored
75              "(PS|PO|PO-E)\\s+" + // Dataset organisation. Many exist
76              // but only support: PS, PO, PO-E
77              "(\\S+)\\s*"; // Dataset Name (file name)
78  
79      /**
80       * Matches these entries: Name VV.MM Created Changed Size Init Mod Id
81       * TBSHELF 01.03 2002/09/12 2002/10/11 09:37 11 11 0 KIL001
82       */
83      static final String MEMBER_LIST_REGEX = "(\\S+)\\s+" + // name
84              "\\S+\\s+" + // version, modification (ignored)
85              "\\S+\\s+" + // create date (ignored)
86              "(\\S+)\\s+" + // modification date
87              "(\\S+)\\s+" + // modification time
88              "\\S+\\s+" + // size in lines (ignored)
89              "\\S+\\s+" + // size in lines at creation(ignored)
90              "\\S+\\s+" + // lines modified (ignored)
91              "\\S+\\s*"; // id of user who modified (ignored)
92  
93      /**
94       * Matches these entries, note: no header: IBMUSER1 JOB01906 OUTPUT 3 Spool
95       * Files 012345678901234567890123456789012345678901234 1 2 3 4
96       */
97      static final String JES_LEVEL_1_LIST_REGEX = "(\\S+)\\s+" + // job
98                                                                          // name
99                                                                          // ignored
100             "(\\S+)\\s+" + // job number
101             "(\\S+)\\s+" + // job status (OUTPUT,INPUT,ACTIVE)
102             "(\\S+)\\s+" + // number of spool files
103             "(\\S+)\\s+" + // Text "Spool" ignored
104             "(\\S+)\\s*" // Text "Files" ignored
105     ;
106 
107     /**
108      * JES INTERFACE LEVEL 2 parser Matches these entries: JOBNAME JOBID OWNER
109      * STATUS CLASS IBMUSER1 JOB01906 IBMUSER OUTPUT A RC=0000 3 spool files
110      * IBMUSER TSU01830 IBMUSER OUTPUT TSU ABEND=522 3 spool files
111      * 012345678901234567890123456789012345678901234 1 2 3 4
112      * 012345678901234567890123456789012345678901234567890
113      */
114 
115     static final String JES_LEVEL_2_LIST_REGEX = "(\\S+)\\s+" + // job
116                                                                         // name
117                                                                         // ignored
118             "(\\S+)\\s+" + // job number
119             "(\\S+)\\s+" + // owner ignored
120             "(\\S+)\\s+" + // job status (OUTPUT,INPUT,ACTIVE) ignored
121             "(\\S+)\\s+" + // job class ignored
122             "(\\S+).*" // rest ignored
123     ;
124 
125     /*
126      * ---------------------------------------------------------------------
127      * Very brief and incomplete description of the zOS/MVS-filesystem. (Note:
128      * "zOS" is the operating system on the mainframe, and is the new name for
129      * MVS)
130      *
131      * The filesystem on the mainframe does not have hierarchal structure as for
132      * example the unix filesystem. For a more comprehensive description, please
133      * refer to the IBM manuals
134      *
135      * @LINK:
136      * http://publibfp.boulder.ibm.com/cgi-bin/bookmgr/BOOKS/dgt2d440/CONTENTS
137      *
138      *
139      * Dataset names =============
140      *
141      * A dataset name consist of a number of qualifiers separated by '.', each
142      * qualifier can be at most 8 characters, and the total length of a dataset
143      * can be max 44 characters including the dots.
144      *
145      *
146      * Dataset organisation ====================
147      *
148      * A dataset represents a piece of storage allocated on one or more disks.
149      * The structure of the storage is described with the field dataset
150      * organinsation (DSORG). There are a number of dataset organisations, but
151      * only two are usable for FTP transfer.
152      *
153      * DSORG: PS: sequential, or flat file PO: partitioned dataset PO-E:
154      * extended partitioned dataset
155      *
156      * The PS file is just a flat file, as you would find it on the unix file
157      * system.
158      *
159      * The PO and PO-E files, can be compared to a single level directory
160      * structure. A PO file consist of a number of dataset members, or files if
161      * you will. It is possible to CD into the file, and to retrieve the
162      * individual members.
163      *
164      *
165      * Dataset record format =====================
166      *
167      * The physical layout of the dataset is described on the dataset itself.
168      * There are a number of record formats (RECFM), but just a few is relavant
169      * for the FTP transfer.
170      *
171      * Any one beginning with either F or V can safely used by FTP transfer. All
172      * others should only be used with great care, so this version will just
173      * ignore the other record formats. F means a fixed number of records per
174      * allocated storage, and V means a variable number of records.
175      *
176      *
177      * Other notes ===========
178      *
179      * The file system supports automatically backup and retrieval of datasets.
180      * If a file is backed up, the ftp LIST command will return: ARCIVE Not
181      * Direct Access Device KJ.IOP998.ERROR.PL.UNITTEST
182      *
183      *
184      * Implementation notes ====================
185      *
186      * Only datasets that have dsorg PS, PO or PO-E and have recfm beginning
187      * with F or V, is fully parsed.
188      *
189      * The following fields in FTPFile is used: FTPFile.Rawlisting: Always set.
190      * FTPFile.Type: DIRECTORY_TYPE or FILE_TYPE or UNKNOWN FTPFile.Name: name
191      * FTPFile.Timestamp: change time or null
192      *
193      *
194      *
195      * Additional information ======================
196      *
197      * The MVS ftp server supports a number of features via the FTP interface.
198      * The features are controlled with the FTP command quote site filetype=<SEQ|JES|DB2>
199      * SEQ is the default and used for normal file transfer JES is used to
200      * interact with the Job Entry Subsystem (JES) similar to a job scheduler
201      * DB2 is used to interact with a DB2 subsystem
202      *
203      * This parser supports SEQ and JES.
204      *
205      *
206      *
207      *
208      *
209      *
210      */
211 
212     /**
213      * The sole constructor for a MVSFTPEntryParser object.
214      *
215      */
216     public MVSFTPEntryParser() {
217         super(""); // note the regex is set in preParse.
218         super.configure(null); // configure parser with default configurations
219     }
220 
221     /**
222      * Parses a line of an z/OS - MVS FTP server file listing and converts it
223      * into a usable format in the form of an <code> FTPFile </code> instance.
224      * If the file listing line doesn't describe a file, then
225      * <code> null </code> is returned. Otherwise a <code> FTPFile </code>
226      * instance representing the file is returned.
227      *
228      * @param entry
229      *            A line of text from the file listing
230      * @return An FTPFile instance corresponding to the supplied entry
231      */
232 //    @Override
233     public FTPFile parseFTPEntry(String entry) {
234         boolean isParsed = false;
235         FTPFile f = new FTPFile();
236 
237         if (isType == FILE_LIST_TYPE) {
238             isParsed = parseFileList(f, entry);
239         } else if (isType == MEMBER_LIST_TYPE) {
240             isParsed = parseMemberList(f, entry);
241             if (!isParsed) {
242                 isParsed = parseSimpleEntry(f, entry);
243             }
244         } else if (isType == UNIX_LIST_TYPE) {
245             isParsed = parseUnixList(f, entry);
246         } else if (isType == JES_LEVEL_1_LIST_TYPE) {
247             isParsed = parseJeslevel1List(f, entry);
248         } else if (isType == JES_LEVEL_2_LIST_TYPE) {
249             isParsed = parseJeslevel2List(f, entry);
250         }
251 
252         if (!isParsed) {
253             f = null;
254         }
255 
256         return f;
257     }
258 
259     /**
260      * Parse entries representing a dataset list. Only datasets with DSORG PS or
261      * PO or PO-E and with RECFM F* or V* will be parsed.
262      *
263      * Format of ZOS/MVS file list: 1 2 3 4 5 6 7 8 9 10 Volume Unit Referred
264      * Ext Used Recfm Lrecl BlkSz Dsorg Dsname B10142 3390 2006/03/20 2 31 F 80
265      * 80 PS MDI.OKL.WORK ARCIVE Not Direct Access Device
266      * KJ.IOP998.ERROR.PL.UNITTEST B1N231 3390 2006/03/20 1 15 VB 256 27998 PO
267      * PLU B1N231 3390 2006/03/20 1 15 VB 256 27998 PO-E PLB
268      *
269      * ----------------------------------- Group within Regex [1] Volume [2]
270      * Unit [3] Referred [4] Ext: number of extents [5] Used [6] Recfm: Record
271      * format [7] Lrecl: Logical record length [8] BlkSz: Block size [9] Dsorg:
272      * Dataset organisation. Many exists but only support: PS, PO, PO-E [10]
273      * Dsname: Dataset name
274      *
275      * Note: When volume is ARCIVE, it means the dataset is stored somewhere in
276      * a tape archive. These entries is currently not supported by this parser.
277      * A null value is returned.
278      *
279      * @param file
280      *            will be updated with Name, Type, Timestamp if parsed.
281      * @param entry zosDirectoryEntry
282      * @return true: entry was parsed, false: entry was not parsed.
283      */
284     private boolean parseFileList(FTPFile file, String entry) {
285         if (matches(entry)) {
286             file.setRawListing(entry);
287             String name = group(2);
288             String dsorg = group(1);
289             file.setName(name);
290 
291             // DSORG
292             if ("PS".equals(dsorg)) {
293                 file.setType(FTPFile.FILE_TYPE);
294             }
295             else if ("PO".equals(dsorg) || "PO-E".equals(dsorg)) {
296                 // regex already ruled out anything other than PO or PO-E
297                 file.setType(FTPFile.DIRECTORY_TYPE);
298             }
299             else {
300                 return false;
301             }
302 
303             return true;
304         }
305 
306         return false;
307     }
308 
309     /**
310      * Parse entries within a partitioned dataset.
311      *
312      * Format of a memberlist within a PDS: 1 2 3 4 5 6 7 8 9 Name VV.MM Created
313      * Changed Size Init Mod Id TBSHELF 01.03 2002/09/12 2002/10/11 09:37 11 11
314      * 0 KIL001 TBTOOL 01.12 2002/09/12 2004/11/26 19:54 51 28 0 KIL001
315      *
316      * ------------------------------------------- [1] Name [2] VV.MM: Version .
317      * modification [3] Created: yyyy / MM / dd [4,5] Changed: yyyy / MM / dd
318      * HH:mm [6] Size: number of lines [7] Init: number of lines when first
319      * created [8] Mod: number of modified lines a last save [9] Id: User id for
320      * last update
321      *
322      *
323      * @param file
324      *            will be updated with Name, Type and Timestamp if parsed.
325      * @param entry zosDirectoryEntry
326      * @return true: entry was parsed, false: entry was not parsed.
327      */
328     private boolean parseMemberList(FTPFile file, String entry) {
329         if (matches(entry)) {
330             file.setRawListing(entry);
331             String name = group(1);
332             String datestr = group(2) + " " + group(3);
333             file.setName(name);
334             file.setType(FTPFile.FILE_TYPE);
335             try {
336                 file.setTimestamp(super.parseTimestamp(datestr));
337             } catch (ParseException e) {
338                 e.printStackTrace();
339                 // just ignore parsing errors.
340                 // TODO check this is ok
341                 return false; // this is a parsing failure too.
342             }
343             return true;
344         }
345 
346         return false;
347     }
348 
349     /**
350      * Assigns the name to the first word of the entry. Only to be used from a
351      * safe context, for example from a memberlist, where the regex for some
352      * reason fails. Then just assign the name field of FTPFile.
353      *
354      * @param file
355      * @param entry
356      * @return true if the entry string is non-null and non-empty
357      */
358     private boolean parseSimpleEntry(FTPFile file, String entry) {
359         if (entry != null && entry.trim().length() > 0) {
360             file.setRawListing(entry);
361             String name = entry.split(" ")[0];
362             file.setName(name);
363             file.setType(FTPFile.FILE_TYPE);
364             return true;
365         }
366         return false;
367     }
368 
369     /**
370      * Parse the entry as a standard unix file. Using the UnixFTPEntryParser.
371      *
372      * @param file
373      * @param entry
374      * @return true: entry is parsed, false: entry could not be parsed.
375      */
376     private boolean parseUnixList(FTPFile file, String entry) {
377         file = unixFTPEntryParser.parseFTPEntry(entry);
378         if (file == null) {
379             return false;
380         }
381         return true;
382     }
383 
384     /**
385      * Matches these entries, note: no header: [1] [2] [3] [4] [5] IBMUSER1
386      * JOB01906 OUTPUT 3 Spool Files
387      * 012345678901234567890123456789012345678901234 1 2 3 4
388      * ------------------------------------------- Group in regex [1] Job name
389      * [2] Job number [3] Job status (INPUT,ACTIVE,OUTPUT) [4] Number of sysout
390      * files [5] The string "Spool Files"
391      *
392      *
393      * @param file
394      *            will be updated with Name, Type and Timestamp if parsed.
395      * @param entry zosDirectoryEntry
396      * @return true: entry was parsed, false: entry was not parsed.
397      */
398     private boolean parseJeslevel1List(FTPFile file, String entry) {
399         if (matches(entry)) {
400             if (group(3).equalsIgnoreCase("OUTPUT")) {
401                 file.setRawListing(entry);
402                 String name = group(2); /* Job Number, used by GET */
403                 file.setName(name);
404                 file.setType(FTPFile.FILE_TYPE);
405                 return true;
406             }
407         }
408 
409         return false;
410     }
411 
412     /**
413      * Matches these entries, note: no header: [1] [2] [3] [4] [5] JOBNAME JOBID
414      * OWNER STATUS CLASS IBMUSER1 JOB01906 IBMUSER OUTPUT A RC=0000 3 spool
415      * files IBMUSER TSU01830 IBMUSER OUTPUT TSU ABEND=522 3 spool files
416      * 012345678901234567890123456789012345678901234 1 2 3 4
417      * ------------------------------------------- Group in regex [1] Job name
418      * [2] Job number [3] Owner [4] Job status (INPUT,ACTIVE,OUTPUT) [5] Job
419      * Class [6] The rest
420      *
421      *
422      * @param file
423      *            will be updated with Name, Type and Timestamp if parsed.
424      * @param entry zosDirectoryEntry
425      * @return true: entry was parsed, false: entry was not parsed.
426      */
427     private boolean parseJeslevel2List(FTPFile file, String entry) {
428         if (matches(entry)) {
429             if (group(4).equalsIgnoreCase("OUTPUT")) {
430                 file.setRawListing(entry);
431                 String name = group(2); /* Job Number, used by GET */
432                 file.setName(name);
433                 file.setType(FTPFile.FILE_TYPE);
434                 return true;
435             }
436         }
437 
438         return false;
439     }
440 
441     /**
442      * preParse is called as part of the interface. Per definition is is called
443      * before the parsing takes place. Three kind of lists is recognize:
444      * z/OS-MVS File lists z/OS-MVS Member lists unix file lists
445      * @since 2.0
446      */
447     @Override
448     public List<String> preParse(List<String> orig) {
449         // simply remove the header line. Composite logic will take care of the
450         // two different types of
451         // list in short order.
452         if (orig != null && orig.size() > 0) {
453             String header = orig.get(0);
454             if (header.indexOf("Volume") >= 0 && header.indexOf("Dsname") >= 0) {
455                 setType(FILE_LIST_TYPE);
456                 super.setRegex(FILE_LIST_REGEX);
457             } else if (header.indexOf("Name") >= 0 && header.indexOf("Id") >= 0) {
458                 setType(MEMBER_LIST_TYPE);
459                 super.setRegex(MEMBER_LIST_REGEX);
460             } else if (header.indexOf("total") == 0) {
461                 setType(UNIX_LIST_TYPE);
462                 unixFTPEntryParser = new UnixFTPEntryParser();
463             } else if (header.indexOf("Spool Files") >= 30) {
464                 setType(JES_LEVEL_1_LIST_TYPE);
465                 super.setRegex(JES_LEVEL_1_LIST_REGEX);
466             } else if (header.indexOf("JOBNAME") == 0
467                     && header.indexOf("JOBID") > 8) {// header contains JOBNAME JOBID OWNER // STATUS CLASS
468                 setType(JES_LEVEL_2_LIST_TYPE);
469                 super.setRegex(JES_LEVEL_2_LIST_REGEX);
470             } else {
471                 setType(UNKNOWN_LIST_TYPE);
472             }
473 
474             if (isType != JES_LEVEL_1_LIST_TYPE) { // remove header is necessary
475                 orig.remove(0);
476             }
477         }
478 
479         return orig;
480     }
481 
482     /**
483      * Explicitly set the type of listing being processed.
484      * @param type The listing type.
485      */
486     void setType(int type) {
487         isType = type;
488     }
489 
490     /*
491      * @return
492      */
493     @Override
494     protected FTPClientConfig getDefaultConfiguration() {
495         return new FTPClientConfig(FTPClientConfig.SYST_MVS,
496                 DEFAULT_DATE_FORMAT, null, null, null, null);
497     }
498 
499 }