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
018package org.apache.commons.net.ftp.parser;
019
020import java.text.ParseException;
021import java.util.List;
022
023import org.apache.commons.net.ftp.FTPClientConfig;
024import org.apache.commons.net.ftp.FTPFile;
025
026/**
027 * Implementation of FTPFileEntryParser and FTPFileListParser for IBM zOS/MVS Systems.
028 *
029 * @see org.apache.commons.net.ftp.FTPFileEntryParser FTPFileEntryParser (for usage instructions)
030 */
031public class MVSFTPEntryParser extends ConfigurableFTPFileEntryParserImpl {
032
033    static final int UNKNOWN_LIST_TYPE = -1;
034    static final int FILE_LIST_TYPE = 0;
035    static final int MEMBER_LIST_TYPE = 1;
036    static final int UNIX_LIST_TYPE = 2;
037    static final int JES_LEVEL_1_LIST_TYPE = 3;
038    static final int JES_LEVEL_2_LIST_TYPE = 4;
039
040    /**
041     * Dates are ignored for file lists, but are used for member lists where possible
042     */
043    static final String DEFAULT_DATE_FORMAT = "yyyy/MM/dd HH:mm"; // 2001/09/18
044                                                                  // 13:52
045
046    /**
047     * Matches these entries:
048     *
049     * <pre>
050     *  Volume Unit    Referred Ext Used Recfm Lrecl BlkSz Dsorg Dsname
051     *  B10142 3390   2006/03/20  2   31  F       80    80  PS   MDI.OKL.WORK
052     * </pre>
053     *
054     * @see <a href= "https://www.ibm.com/support/knowledgecenter/zosbasics/com.ibm.zos.zconcepts/zconcepts_159.htm">Data set record formats</a>
055     */
056    static final String FILE_LIST_REGEX = "\\S+\\s+" + // volume
057                                                       // ignored
058            "\\S+\\s+" + // unit - ignored
059            "\\S+\\s+" + // access date - ignored
060            "\\S+\\s+" + // extents -ignored
061            // If the values are too large, the fields may be merged (NET-639)
062            "(?:\\S+\\s+)?" + // used - ignored
063            "\\S+\\s+" + // recfm - ignored
064            "\\S+\\s+" + // logical record length -ignored
065            "\\S+\\s+" + // block size - ignored
066            "(PS|PO|PO-E)\\s+" + // Dataset organisation. Many exist
067            // but only support: PS, PO, PO-E
068            "(\\S+)\\s*"; // Dataset Name (file name)
069
070    /**
071     * Matches these entries:
072     *
073     * <pre>
074     *   Name      VV.MM   Created       Changed      Size  Init   Mod   Id
075     *   TBSHELF   01.03 2002/09/12 2002/10/11 09:37    11    11     0 KIL001
076     * </pre>
077     */
078    static final String MEMBER_LIST_REGEX = "(\\S+)\\s+" + // name
079            "\\S+\\s+" + // version, modification (ignored)
080            "\\S+\\s+" + // create date (ignored)
081            "(\\S+)\\s+" + // modification date
082            "(\\S+)\\s+" + // modification time
083            "\\S+\\s+" + // size in lines (ignored)
084            "\\S+\\s+" + // size in lines at creation(ignored)
085            "\\S+\\s+" + // lines modified (ignored)
086            "\\S+\\s*"; // id of user who modified (ignored)
087
088    /**
089     * Matches these entries, note: no header:
090     *
091     * <pre>
092     *   IBMUSER1  JOB01906  OUTPUT    3 Spool Files
093     *   012345678901234567890123456789012345678901234
094     *             1         2         3         4
095     * </pre>
096     */
097    static final String JES_LEVEL_1_LIST_REGEX = "(\\S+)\\s+" + // job name ignored
098            "(\\S+)\\s+" + // job number
099            "(\\S+)\\s+" + // job status (OUTPUT,INPUT,ACTIVE)
100            "(\\S+)\\s+" + // number of spool files
101            "(\\S+)\\s+" + // Text "Spool" ignored
102            "(\\S+)\\s*" // Text "Files" ignored
103    ;
104
105    /**
106     * JES INTERFACE LEVEL 2 parser Matches these entries:
107     *
108     * <pre>
109     * JOBNAME  JOBID    OWNER    STATUS CLASS
110     * IBMUSER1 JOB01906 IBMUSER  OUTPUT A        RC=0000 3 spool files
111     * IBMUSER  TSU01830 IBMUSER  OUTPUT TSU      ABEND=522 3 spool files
112     * </pre>
113     *
114     * Sample output from FTP session:
115     *
116     * <pre>
117     * ftp> quote site filetype=jes
118     * 200 SITE command was accepted
119     * ftp> ls
120     * 200 Port request OK.
121     * 125 List started OK for JESJOBNAME=IBMUSER*, JESSTATUS=ALL and JESOWNER=IBMUSER
122     * JOBNAME  JOBID    OWNER    STATUS CLASS
123     * IBMUSER1 JOB01906 IBMUSER  OUTPUT A        RC=0000 3 spool files
124     * IBMUSER  TSU01830 IBMUSER  OUTPUT TSU      ABEND=522 3 spool files
125     * 250 List completed successfully.
126     * ftp> ls job01906
127     * 200 Port request OK.
128     * 125 List started OK for JESJOBNAME=IBMUSER*, JESSTATUS=ALL and JESOWNER=IBMUSER
129     * JOBNAME  JOBID    OWNER    STATUS CLASS
130     * IBMUSER1 JOB01906 IBMUSER  OUTPUT A        RC=0000
131     * --------
132     * ID  STEPNAME PROCSTEP C DDNAME   BYTE-COUNT
133     * 001 JES2              A JESMSGLG       858
134     * 002 JES2              A JESJCL         128
135     * 003 JES2              A JESYSMSG       443
136     * 3 spool files
137     * 250 List completed successfully.
138     * </pre>
139     */
140
141    static final String JES_LEVEL_2_LIST_REGEX = "(\\S+)\\s+" + // job name ignored
142            "(\\S+)\\s+" + // job number
143            "(\\S+)\\s+" + // owner ignored
144            "(\\S+)\\s+" + // job status (OUTPUT,INPUT,ACTIVE) ignored
145            "(\\S+)\\s+" + // job class ignored
146            "(\\S+).*" // rest ignored
147    ;
148
149    private int isType = UNKNOWN_LIST_TYPE;
150
151    /**
152     * Fallback parser for Unix-style listings
153     */
154    private UnixFTPEntryParser unixFTPEntryParser;
155
156    /*
157     * --------------------------------------------------------------------- Very brief and incomplete description of the zOS/MVS-file system. (Note: "zOS" is
158     * the operating system on the mainframe, and is the new name for MVS)
159     *
160     * The file system on the mainframe does not have hierarchical structure as for example the unix file system. For a more comprehensive description,
161     * please refer to the IBM manuals
162     *
163     * @LINK: http://publibfp.boulder.ibm.com/cgi-bin/bookmgr/BOOKS/dgt2d440/CONTENTS
164     *
165     *
166     * Dataset names =============
167     *
168     * A dataset name consist of a number of qualifiers separated by '.', each qualifier can be at most 8 characters, and the total length of a dataset can be
169     * max 44 characters including the dots.
170     *
171     *
172     * Dataset organisation ====================
173     *
174     * A dataset represents a piece of storage allocated on one or more disks. The structure of the storage is described with the field dataset organisation
175     * (DSORG). There are a number of dataset organisations, but only two are usable for FTP transfer.
176     *
177     * DSORG: PS: sequential, or flat file PO: partitioned dataset PO-E: extended partitioned dataset
178     *
179     * The PS file is just a flat file, as you would find it on the unix file system.
180     *
181     * The PO and PO-E files, can be compared to a single level directory structure. A PO file consist of a number of dataset members, or files if you will. It
182     * is possible to CD into the file, and to retrieve the individual members.
183     *
184     *
185     * Dataset record format =====================
186     *
187     * The physical layout of the dataset is described on the dataset itself. There are a number of record formats (RECFM), but just a few is relevant for the
188     * FTP transfer.
189     *
190     * Any one beginning with either F or V can safely be used by FTP transfer. All others should only be used with great care. F means a fixed number of
191     * records per allocated storage, and V means a variable number of records.
192     *
193     *
194     * Other notes ===========
195     *
196     * The file system supports automatically backup and retrieval of datasets. If a file is backed up, the ftp LIST command will return: ARCIVE Not Direct
197     * Access Device KJ.IOP998.ERROR.PL.UNITTEST
198     *
199     *
200     * Implementation notes ====================
201     *
202     * Only datasets that have dsorg PS, PO or PO-E and have recfm beginning with F or V or U, is fully parsed.
203     *
204     * The following fields in FTPFile is used: FTPFile.Rawlisting: Always set. FTPFile.Type: DIRECTORY_TYPE or FILE_TYPE or UNKNOWN FTPFile.Name: name
205     * FTPFile.Timestamp: change time or null
206     *
207     *
208     *
209     * Additional information ======================
210     *
211     * The MVS ftp server supports a number of features via the FTP interface. The features are controlled with the FTP command quote site
212     * filetype=<SEQ|JES|DB2> SEQ is the default and used for normal file transfer JES is used to interact with the Job Entry Subsystem (JES) similar to a job
213     * scheduler DB2 is used to interact with a DB2 subsystem
214     *
215     * This parser supports SEQ and JES.
216     */
217
218    /**
219     * The sole constructor for a MVSFTPEntryParser object.
220     */
221    public MVSFTPEntryParser() {
222        super(""); // note the regex is set in preParse.
223        super.configure(null); // configure parser with default configurations
224    }
225
226    /*
227     * @return
228     */
229    @Override
230    protected FTPClientConfig getDefaultConfiguration() {
231        return new FTPClientConfig(FTPClientConfig.SYST_MVS, DEFAULT_DATE_FORMAT, null);
232    }
233
234    /**
235     * Parses entries representing a dataset list.
236     * <pre>
237     * Format of ZOS/MVS file list: 1 2 3 4 5 6 7 8 9 10
238     * Volume Unit Referred Ext Used Recfm Lrecl BlkSz Dsorg Dsname
239     * B10142 3390 2006/03/20 2 31 F 80 80 PS MDI.OKL.WORK
240     * ARCIVE Not Direct Access Device KJ.IOP998.ERROR.PL.UNITTEST
241     * B1N231 3390 2006/03/20 1 15 VB 256 27998 PO PLU
242     * B1N231 3390 2006/03/20 1 15 VB 256 27998 PO-E PLB
243     * Migrated                                                HLQ.DATASET.NAME
244     * </pre>
245     * <pre>
246     * ----------------------------------- Group within Regex [1] Volume [2] Unit [3] Referred [4] Ext: number of extents [5] Used [6] Recfm: Record format [7]
247     * Lrecl: Logical record length [8] BlkSz: Block size [9] Dsorg: Dataset organisation. Many exists but only support: PS, PO, PO-E [10] Dsname: Dataset name
248     * </pre>
249     *
250     * @param entry zosDirectoryEntry
251     * @return null: entry was not parsed.
252     */
253    private FTPFile parseFileList(final String entry) {
254        if (matches(entry)) {
255            final FTPFile file = new FTPFile();
256            file.setRawListing(entry);
257            final String name = group(2);
258            final String dsorg = group(1);
259            file.setName(name);
260
261            // DSORG
262            if ("PS".equals(dsorg)) {
263                file.setType(FTPFile.FILE_TYPE);
264            } else if ("PO".equals(dsorg) || "PO-E".equals(dsorg)) {
265                // regex already ruled out anything other than PO or PO-E
266                file.setType(FTPFile.DIRECTORY_TYPE);
267            } else {
268                return null;
269            }
270
271            return file;
272        }
273
274        final boolean migrated = entry.startsWith("Migrated");
275        if (migrated || entry.startsWith("ARCIVE")) {
276            // Type of file is unknown for migrated datasets
277            final FTPFile file = new FTPFile();
278            file.setRawListing(entry);
279            file.setType(FTPFile.UNKNOWN_TYPE);
280            file.setName(entry.split("\\s+")[migrated ? 1 : 5]);
281            return file;
282        }
283
284        return null;
285    }
286
287    /**
288     * Parses a line of a z/OS - MVS FTP server file listing and converts it into a usable format in the form of an <code> FTPFile </code> instance. If the
289     * file listing line doesn't describe a file, then <code> null </code> is returned. Otherwise, a <code> FTPFile </code> instance representing the file is
290     * returned.
291     *
292     * @param entry A line of text from the file listing
293     * @return An FTPFile instance corresponding to the supplied entry
294     */
295    @Override
296    public FTPFile parseFTPEntry(final String entry) {
297        switch (isType) {
298        case FILE_LIST_TYPE:
299            return parseFileList(entry);
300        case MEMBER_LIST_TYPE:
301            return parseMemberList(entry);
302        case UNIX_LIST_TYPE:
303            return unixFTPEntryParser.parseFTPEntry(entry);
304        case JES_LEVEL_1_LIST_TYPE:
305            return parseJeslevel1List(entry);
306        case JES_LEVEL_2_LIST_TYPE:
307            return parseJeslevel2List(entry);
308        default:
309            break;
310        }
311
312        return null;
313    }
314
315    /**
316     * Matches these entries, note: no header:
317     *
318     * <pre>
319     * [1]      [2]      [3]   [4] [5]
320     * IBMUSER1 JOB01906 OUTPUT 3 Spool Files
321     * 012345678901234567890123456789012345678901234
322     *           1         2         3         4
323     * -------------------------------------------
324     * Group in regex
325     * [1] Job name
326     * [2] Job number
327     * [3] Job status (INPUT,ACTIVE,OUTPUT)
328     * [4] Number of sysout files
329     * [5] The string "Spool Files"
330     * </pre>
331     *
332     * @param entry zosDirectoryEntry
333     * @return null: entry was not parsed.
334     */
335    private FTPFile parseJeslevel1List(final String entry) {
336        return parseJeslevelList(entry, 3);
337    }
338
339    /**
340     * Matches these entries:
341     *
342     * <pre>
343     * [1]      [2]      [3]     [4]    [5]
344     * JOBNAME  JOBID    OWNER   STATUS CLASS
345     * IBMUSER1 JOB01906 IBMUSER OUTPUT A       RC=0000 3 spool files
346     * IBMUSER  TSU01830 IBMUSER OUTPUT TSU     ABEND=522 3 spool files
347     * 012345678901234567890123456789012345678901234
348     *           1         2         3         4
349     * -------------------------------------------
350     * Group in regex
351     * [1] Job name
352     * [2] Job number
353     * [3] Owner
354     * [4] Job status (INPUT,ACTIVE,OUTPUT)
355     * [5] Job Class
356     * [6] The rest
357     * </pre>
358     *
359     * @param entry zosDirectoryEntry
360     * @return null: entry was not parsed.
361     */
362    private FTPFile parseJeslevel2List(final String entry) {
363        return parseJeslevelList(entry, 4);
364    }
365
366    private FTPFile parseJeslevelList(final String entry, final int matchNum) {
367        if (matches(entry)) {
368            final FTPFile file = new FTPFile();
369            if (group(matchNum).equalsIgnoreCase("OUTPUT")) {
370                file.setRawListing(entry);
371                final String name = group(2); /* Job Number, used by GET */
372                file.setName(name);
373                file.setType(FTPFile.FILE_TYPE);
374                return file;
375            }
376        }
377        return null;
378    }
379
380    /**
381     * Parses entries within a partitioned dataset.
382     *
383     * Format of a memberlist within a PDS:
384     *
385     * <pre>
386     *    0         1        2          3        4     5     6      7    8
387     *   Name      VV.MM   Created       Changed      Size  Init   Mod   Id
388     *   TBSHELF   01.03 2002/09/12 2002/10/11 09:37    11    11     0 KIL001
389     *   TBTOOL    01.12 2002/09/12 2004/11/26 19:54    51    28     0 KIL001
390     *
391     * -------------------------------------------
392     * [1] Name
393     * [2] VV.MM: Version . modification
394     * [3] Created: yyyy / MM / dd
395     * [4,5] Changed: yyyy / MM / dd HH:mm
396     * [6] Size: number of lines
397     * [7] Init: number of lines when first created
398     * [8] Mod: number of modified lines a last save
399     * [9] Id: User id for last update
400     * </pre>
401     *
402     * @param entry zosDirectoryEntry
403     * @return null: entry was not parsed.
404     */
405    private FTPFile parseMemberList(final String entry) {
406        final FTPFile file = new FTPFile();
407        if (matches(entry)) {
408            file.setRawListing(entry);
409            final String name = group(1);
410            final String datestr = group(2) + " " + group(3);
411            file.setName(name);
412            file.setType(FTPFile.FILE_TYPE);
413            try {
414                file.setTimestamp(super.parseTimestamp(datestr));
415            } catch (final ParseException e) {
416                // just ignore parsing errors.
417                // TODO check this is ok
418                // Drop thru to try simple parser
419            }
420            return file;
421        }
422
423        /*
424         * Assigns the name to the first word of the entry. Only to be used from a safe context, for example from a memberlist, where the regex for some reason
425         * fails. Then just assign the name field of FTPFile.
426         */
427        if (entry != null && !entry.trim().isEmpty()) {
428            file.setRawListing(entry);
429            final String name = entry.split(" ")[0];
430            file.setName(name);
431            file.setType(FTPFile.FILE_TYPE);
432            return file;
433        }
434        return null;
435    }
436
437    /**
438     * Pre-parses is called as part of the interface. Per definition, it is called before the parsing takes place. Three kinds of lists are recognized:
439     * <ul>
440     *     <li>z/OS-MVS File lists,</li>
441     *     <li>z/OS-MVS Member lists,</li>
442     *     <li>unix file lists.</li>
443     * </ul>
444     * @since 2.0
445     */
446    @Override
447    public List<String> preParse(final List<String> orig) {
448        // simply remove the header line. Composite logic will take care of the
449        // two different types of
450        // list in short order.
451        if (orig != null && !orig.isEmpty()) {
452            final String header = orig.get(0);
453            if (header.contains("Volume") && header.contains("Dsname")) {
454                setType(FILE_LIST_TYPE);
455                super.setRegex(FILE_LIST_REGEX);
456            } else if (header.contains("Name") && header.contains("Id")) {
457                setType(MEMBER_LIST_TYPE);
458                super.setRegex(MEMBER_LIST_REGEX);
459            } else if (header.indexOf("total") == 0) {
460                setType(UNIX_LIST_TYPE);
461                unixFTPEntryParser = new UnixFTPEntryParser();
462            } else if (header.indexOf("Spool Files") >= 30) {
463                setType(JES_LEVEL_1_LIST_TYPE);
464                super.setRegex(JES_LEVEL_1_LIST_REGEX);
465            } else if (header.indexOf("JOBNAME") == 0 && header.indexOf("JOBID") > 8) {// header contains JOBNAME JOBID OWNER // STATUS CLASS
466                setType(JES_LEVEL_2_LIST_TYPE);
467                super.setRegex(JES_LEVEL_2_LIST_REGEX);
468            } else {
469                setType(UNKNOWN_LIST_TYPE);
470            }
471
472            if (isType != JES_LEVEL_1_LIST_TYPE) { // remove header is necessary
473                orig.remove(0);
474            }
475        }
476
477        return orig;
478    }
479
480    /**
481     * Sets the type of listing being processed.
482     *
483     * @param type The listing type.
484     */
485    void setType(final int type) {
486        isType = type;
487    }
488
489}