root/main/trunk/greenstone2/perllib/plugouts/MySQLPlugout.pm @ 32522

Revision 32522, 21.9 KB (checked in by ak19, 22 months ago)

Fixing some syntax errors

Line 
1###########################################################################
2#
3# GreenstoneXMLPlugout.pm -- the plugout module for Greenstone Archives
4# A component of the Greenstone digital library software
5# from the New Zealand Digital Library Project at the
6# University of Waikato, New Zealand.
7#
8# Copyright (C) 2006 New Zealand Digital Library Project
9#
10# This program is free software; you can redistribute it and/or modify
11# it under the terms of the GNU General Public License as published by
12# the Free Software Foundation; either version 2 of the License, or
13# (at your option) any later version.
14#
15# This program is distributed in the hope that it will be useful,
16# but WITHOUT ANY WARRANTY; without even the implied warranty of
17# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
18# GNU General Public License for more details.
19#
20# You should have received a copy of the GNU General Public License
21# along with this program; if not, write to the Free Software
22# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
23#
24###########################################################################
25
26package MySQLPlugout;
27
28use strict;
29no strict 'refs';
30no strict 'subs';
31
32use util;
33use FileUtils;
34use GreenstoneXMLPlugout;
35use docprint;
36
37use IPC::Open2;
38use POSIX ":sys_wait_h"; # for waitpid, http://perldoc.perl.org/functions/waitpid.html
39
40# TODO: SIGTERM rollback and disconnect?
41
42
43# this plugout does not output xml to a file, but outputs rows into a mysql table
44sub BEGIN {
45    @MySQLPlugout::ISA = ('GreenstoneXMLPlugout');
46}
47
48
49# TODO: deal with -removeold and everything? Or type out instructions for user
50
51# TODO Q: what is "group" in GreenstoneXMLPlugout?
52# TODO Q: site_name only exists for GS3. What about GS2?
53
54my $process_mode_list =
55    [ { 'name' => "meta_only",
56        'desc' => "{MySQLPlugout.process_mode.meta_only}" },     
57      { 'name' => "text_only",
58        'desc' => "{MySQLPlugout.process_mode.text_only}" },
59      { 'name' => "all",
60        'desc' => "{MySQLPlugout.process_mode.all}" } ];
61
62my $arguments = [
63       { 'name' => "process_mode",
64     'desc' => "{MySQLPlugout.process_mode}",
65     'type' => "enum",
66     'list' => $process_mode_list,
67     'deft' => "all",
68     'reqd' => "no",
69     'hiddengli' => "no"} ];
70
71my $options = { 'name'     => "MySQLPlugout",
72        'desc'     => "{MySQLPlugout.desc}",
73        'abstract' => "no",
74        'inherits' => "yes",
75        'args'     => $arguments };
76
77sub new {
78    my ($class) = shift (@_);
79    my ($plugoutlist, $inputargs,$hashArgOptLists) = @_;
80    push(@$plugoutlist, $class);
81
82    push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
83    push(@{$hashArgOptLists->{"OptList"}},$options);
84
85    my $self = new GreenstoneXMLPlugout($plugoutlist,$inputargs,$hashArgOptLists);
86   
87    if ($self->{'info_only'}) {
88        # don't worry about any options etc
89        return bless $self, $class;
90    }
91    print STDERR "***** MySQLPlugout process mode = \"", $self->{'process_mode'}, "\"\n";
92   
93    return bless $self, $class;
94}
95
96# connect here and ensure all tables and databases exist
97sub begin {
98
99    my $self= shift (@_);
100
101    ########### TODO: these should be set from cmdline/GLI options to plugout #########
102    $self->{'db_driver'} = "mysql";
103    $self->{'site_name'} = "localsite";   
104    $self->{'db_client_user'} = "root";
105    $self->{'db_client_pwd'} = "6reenstone3";
106    #$self->{'db_host'} = "127.0.0.1";
107    #$self->{'db_encoding'} = "utf8";
108    #TODO: proc_mode is also a saveas option
109   
110    ############ LOAD NECESSARY OPTIONS ###########
111    print STDERR "########## COLLECTION: ". $ENV{'GSDLCOLLECTION'}."\n";
112    $self->{'collection_name'} = $ENV{'GSDLCOLLECTION'};
113   
114    if(!$self->connect_to_db()) {
115    # This is fatal for the plugout, let's terminate here
116    # PrintError would already have displayed the warning message on connection fail   
117    die("Could not connect to db. Can't proceed.\n");
118    }
119    if(!$self->load_db_and_tables()) {
120    # This is fatal for the plugout, let's terminate here
121    # PrintError would already have displayed the warning message on connection fail   
122    die("Could not use db or prepare its tables. Can't proceed.\n");
123    }
124
125    # prepare the shared/common HANDLES to SQL insert statements that contain placeholders
126    # and which we will reuse repeatedly when actually executing the insert statements
127    my $proc_mode = $self->{'process_mode'};
128    if($proc_mode eq "all" || $proc_mode eq "meta_only" ) {
129    $self->{'metadata_prepared_insert_statement_handle'} = $self->prepare_insert_metadata_row_stmthandle();
130    }
131    if($proc_mode eq "all" || $proc_mode eq "text_only" ) {
132    $self->{'fulltxt_prepared_insert_statement_handle'} = $self->prepare_insert_fulltxt_row_stmthandle();
133    }
134
135    # if setting up to work with sql db failed, we'd have terminated and wouldn't come up to here:
136    # won't bother preparing GreenstoneXMLPlugout by calling superclass' begin()
137    # finally, can call begin on super - important as doc.xml is opened as a group etc
138
139    if($proc_mode ne "all") { # TODO Q
140    $self->GreenstoneXMLPlugout::begin(@_);
141    }
142}
143
144# disconnect from database here, see inexport.pm
145sub end
146{
147    my $self = shift(@_);
148
149    # do the superclass stuff first, as any sql db failures should not prevent superclass cleanup
150   
151    if($self->{'process_mode'} ne "all") {# TODO Q
152    $self->GreenstoneXMLPlugout::end(@_);
153    }
154   
155    $self->disconnect_from_db() || warn("Unable to disconnect from database " . $self->{'site_name'} . "\n"); # disconnect_from_db() will also issue a warning, but this may be clearer
156}
157 
158sub close_group_output {
159    my $self = shift (@_);
160   
161    if($self->{'process_mode'} ne "all") { # TODO Q
162    $self->GreenstoneXMLPlugout::close_group_output();
163    }
164}
165   
166# TODO: check arc-inf.db for whether each entry is to be deleted/indexed/reindexed/been indexed
167sub saveas {
168    my $self = shift (@_);
169    my ($doc_obj, $doc_dir) = @_;
170
171    my $proc_mode = $self->{'process_mode'};
172   
173    # 1. pre save out and saving debug handle
174    my ($docxml_outhandler, $output_file);
175    if($proc_mode ne "all") { # TODO Q
176    ($docxml_outhandler, $output_file) = $self->GreenstoneXMLPlugout::pre_saveas(@_);
177
178    $self->{'debug_outhandle'} = $docxml_outhandler if ($self->{'debug'}); # STDOUT if debug
179    }
180    elsif ($self->{'debug'}) { # if we didn't call superclass::pre_saveas, then set debug handle
181    $self->{'debug_outhandle'} = STDOUT; # superclass also debugs to STDOUT
182    # TODO: also set debugging in begin()? Then stmts creating db and tables also sent to debug out and not executed
183    }
184   
185   
186    # 2. overriding saving behaviour to do what the superclass does PLUS saving to sql db   
187
188    if($proc_mode ne "all" ) { # if all, then nothing to go into doc.xml, otherwise:
189   
190    # write the INVERSE into doc.xml as to what is written to the db
191   
192    my $docxml_output_options = { 'output' => docprint::OUTPUT_NONE }; # TODO Q
193   
194    if($proc_mode eq "meta_only" ) { # since only meta to go into MySQL db, text will go into docxml
195        $docxml_output_options->{'output'} = docprint::OUTPUT_TEXT_ONLY;
196    } elsif($proc_mode eq "text_only" ) { # since only full text to go into MySQL db, meta will go into docxml
197        $docxml_output_options->{'output'} = docprint::OUTPUT_META_ONLY;
198    }
199
200    # now we've prepared to write out whatever is meant to go into docxml
201    # and can do actual the steps superclass GreenstoneXMLPlugout carries out to write out docxml
202    # So: write out the doc xml file for the current document
203    my $section_text = &docprint::get_section_xml($doc_obj, $docxml_output_options);
204    print $docxml_outhandler $section_text;   
205    }
206   
207    # We also write out whatever needs to go into the MySQL database
208    $self->write_meta_and_text($doc_obj);
209
210   
211    # 3. post save out
212    if($proc_mode ne "all" ) { # TODO Q
213    #$self->GreenstoneXMLPlugout::post_saveas(@_);
214    $self->GreenstoneXMLPlugout::post_saveas($doc_obj, $doc_dir, $docxml_outhandler, $output_file);
215    }
216   
217    # database connection is closed in end() method
218    # so we don't open and close over and over for each doc during a single build
219}
220
221
222# write meta and/or text PER DOC out to DB
223sub write_meta_and_text {
224    my $self = shift (@_);
225    my ($doc_obj) = @_;
226    my $root_section = $doc_obj->get_top_section();
227    my $doc_oid = $doc_obj->get_OID(); # we're processing a single doc at a time, so single OID
228
229    # load the prepared INSERT statement handles for both tables (can be undef for any table depending on whether meta_only or txt_only are set)
230    my $metadata_table_sth = $self->{'metadata_prepared_insert_statement_handle'};
231    my $fulltxt_table_sth = $self->{'fulltxt_prepared_insert_statement_handle'};
232   
233    $self->recursive_write_meta_and_text($doc_obj, $root_section, $metadata_table_sth, $fulltxt_table_sth);
234}
235
236# Perl: Reading or Writing to Another Program
237# https://nnc3.com/mags/Perl3/cookbook/ch16_05.htm
238sub recursive_write_meta_and_text {
239    my $self = shift (@_);
240    my ($doc_obj, $doc_oid, $section, $metadata_table_sth, $fulltxt_table_sth) = @_;   
241
242    # If section=ROOT, write "root" as section name into table
243    # doc->get_top_section() is the name of the doc root section, which is ""
244    my $section_name = ($section eq "") ? "root" : $section;
245   
246    my $section_ptr = $doc_obj->_lookup_section ($section);
247    return "" unless defined $section_ptr;
248
249    my $debug_out = $self->{'debug_outhandle'};
250   
251    #my $proc_mode = $self->{'process_mode'};
252    #if($proc_mode eq "all" || $proc_mode eq "meta_only" ) {
253    if($metadata_table_sth) { # meta insert statement handle will be undef if not writing meta
254   
255    foreach my $data (@{$section_ptr->{'metadata'}}) {
256        my $meta_name = $data->[0];
257        my $escaped_meta_value = &escape_text($data->[1]);
258
259        # Write out the current section's meta to collection db's METADATA table       
260       
261        # for each set of values to write to meta table, execute the prepared statement, filling in the values
262
263        if($self->{'debug'}) {
264        # just print the statement we were going to execute
265
266        print $debug_out $metadata_table_sth->Statement . "($doc_oid, $section_name, $meta_name, $escaped_meta_value)\n";
267        }
268        else {
269       
270        $metadata_table_sth->execute($doc_oid, $section_name, $meta_name, $escaped_meta_value);
271        #|| warn ("Unable to write metadata row to db:\n\tOID $doc_oid, section $section_name,\n\tmeta name: $meta_name, val: $escaped_meta_value");
272        # Execution failure will print out info anyway: since db connection sets PrintError
273        }
274    }
275    }
276   
277    #if($proc_mode eq "all" || $proc_mode eq "text_only" ) {
278    if($fulltxt_table_sth) { # fulltxt insert statement handle will be undef if not writing fulltxt
279
280    if($self->{'debug'}) {
281        # just print the statement we were going to execute, minus the fulltxt value
282        my $txt_repr = $section_ptr->{'text'} ? "<TXT>" : "NULL";
283        print $debug_out $fulltxt_table_sth->Statement . "($doc_oid, $section_name, $txt_repr)\n";
284    } else {
285        my $section_text = &escape_text($section_ptr->{'text'});
286       
287        # fulltxt column can be SQL NULL. undef value gets written out as NULL:
288        # https://stackoverflow.com/questions/12708633/which-one-represents-null-undef-or-empty-string
289       
290        # Write out the current section's text to collection db's FULLTeXT table
291        $fulltxt_table_sth->execute($doc_oid, $section_name, $section_text);
292        #|| warn ("Unable to write fulltxt row to db for row:\n\tOID $doc_oid, section $section_name");
293        # Execution failure will print out info anyway: since db connection sets PrintError
294    }
295    }
296   
297    # output all subsections: RECURSIVE CALL
298    foreach my $subsection (@{$section_ptr->{'subsection_order'}}) {
299    &recursive_write_meta_and_text($doc_obj, $doc_oid, "$section.$subsection", $metadata_table_sth, $fulltxt_table_sth);
300    }
301}
302
303#################################
304
305# Database access related functions
306# http://g2pc1.bu.edu/~qzpeng/manual/MySQL%20Commands.htm
307# https://www.guru99.com/insert-into.html
308
309# TODO Q: What on cancelling a build: delete table? But what if it was a rebuild and the rebuild is cancelled (not the original build)?
310# Do we create a copy of the orig database as backup, then start populating current db, and if cancelled, delete current db and RENAME backup table to current?
311# https://stackoverflow.com/questions/3280006/duplicating-a-mysql-table-indexes-and-data
312# BUT what if the table is HUGE? (Think of a collection with millions of docs.) Huge overhead in copying?
313# The alternative is we just quit on cancel, but then: cancel could leave the table in a partial committed state, with no way of rolling back.
314# Unless they do a full rebuild, which will recreate the table from scratch?
315# SOLUTION-> rollback transaction on error, see https://www.effectiveperlprogramming.com/2010/07/set-custom-dbi-error-handlers/
316# But then should set AutoCommit to off on connection, and remember to commit every time
317
318#################
319# Database functions that use the perl DBI module (with the DBD driver module for mysql)
320#################
321
322# THE NEW DB FUNCTIONS
323# NOTE: FULLTEXT is a reserved keyword in (My)SQL. So we can't name a table or any of its columns "fulltext".
324# https://dev.mysql.com/doc/refman/5.5/en/keywords.html
325
326# TODO: Consider AutoCommit status (and Autocommit off allowing commit or rollback for GS coll build cancel) later
327
328sub connect_to_db {
329    my $self= shift (@_);
330   
331    my $db_driver = $self->{'db_driver'};
332    my $db_user = $self->{'db_client_user'} || "root";
333    my $db_pwd = $self->{'db_client_pwd'};
334    my $db_host = $self->{'db_host'} || "127.0.0.1";
335    my $db_enc = $self->{'db_encoding'} || "utf8";
336   
337    #my $db_name = $self->{'site_name'};
338   
339    # try connecting to the mysql db, if that fails it will die
340    # so don't bother preparing GreenstoneXMLPlugout by calling superclass' begin()
341
342    # localhost doesn't work for us, but 127.0.0.1 works
343    # https://metacpan.org/pod/DBD::mysql
344    # "The hostname, if not specified or specified as '' or 'localhost', will default to a MySQL server
345    # running on the local machine using the default for the UNIX socket. To connect to a MySQL server
346    # on the local machine via TCP, you must specify the loopback IP address (127.0.0.1) as the host."
347    #my $connect_str = "dbi:$db_driver:database=$db_name;host=$db_host";
348    my $connect_str = "dbi:$db_driver:host=$db_host"; # don't provide db, so we can check the db is there
349    my $dbh = DBI->connect("$connect_str", $db_user, $db_pwd,
350               {
351                   ShowErrorStatement => 1, # more informative as DBI will append failed SQL stmt to error message
352                   PrintError => 1, # on by default, but being explicit
353                   RaiseError => 0, # off by default, but being explicit
354                   AutoCommit => 1, # on by default, but being explicit
355               });
356
357    if(!$dbh) {
358    # NOTE, despite handle dbh being undefined, error code will be in DBI->err
359    return 0;   
360    }
361
362    # set encoding https://metacpan.org/pod/DBD::mysql
363    # https://dev.mysql.com/doc/refman/5.7/en/charset.html
364    # https://dev.mysql.com/doc/refman/5.7/en/charset-conversion.html
365    # Setting the encoding at db server level.
366    # Not sure if this command is mysql specific:
367    my $stmt = "set NAMES '" . $db_enc . "'";
368    $dbh->do($stmt) || warn("Unable to set charset encoding at db server level to: " . $db_enc . "\n");
369   
370    # if we're here, then connection succeeded, store handle
371    $self->{'db_handle'} = $dbh;
372    return 1;
373}
374
375sub load_db_and_tables {
376    my $self= shift (@_);
377    my $db_name = $self->{'site_name'}; # one database per GS3 site
378    my $dbh = $self->{'db_handle'};
379   
380    # perl DBI switch database: https://www.perlmonks.org/?node_id=995434
381    # do() returns undef on error.
382    # connection succeeded, try to load our database. If that didn't work, attempt to create db
383    my $success = $dbh->do("use $db_name");
384   
385    if(!$success && $dbh->err == 1049) { # "Unknown database" error has code 1049 (mysql only?) meaning db doesn't exist yet
386    # attempt to create the db and its tables
387    $self->create_db($db_name) || return 0;
388
389    # once more attempt to use db, now that it exists
390    $dbh->do("use $db_name") || return 0;
391    #$dbh->do("use localsite") or die "Error (code" . $dbh->err ."): " . $dbh->errstr . "\n";
392
393    # attempt to create tables in current db
394    $self->create_metadata_table() || return 0;
395    $self->create_fulltext_table() || return 0;
396
397    $success = 1;
398    }
399    elsif($success) { # database existed and loaded successfully, but
400    # before proceeding check that the current collection's tables exist
401   
402    # attempt to create tables in current db
403    if(!$self->table_exists($self->{'collection_name'} . "metadata")) {
404        $self->create_metadata_table() || return 0;
405    }
406    if(!$self->table_exists($self->{'collection_name'} .  "fulltxt")) {
407        $self->create_fulltext_table() || return 0;
408    }
409    }
410   
411    return $success; # could still return 0, if database failed to load with an error code != 1049
412}
413
414# disconnect from db - https://metacpan.org/pod/DBI#disconnect
415# TODO: make sure to have committed or rolled back before disconnect
416# and that you've call finish() on statement handles if any fetch remnants remain
417sub disconnect_from_db {
418    my $self= shift (@_);   
419    my $dbh = $self->{'db_handle'};
420
421    # make sure any active stmt handles are finished
422    # NO: "When all the data has been fetched from a SELECT statement, the driver will automatically call finish for you. So you should not call it explicitly except when you know that you've not fetched all the data from a statement handle and the handle won't be destroyed soon."
423   
424    #$meta_sth = $self->{'metadata_prepared_insert_statement_handle'};
425    #$txt_sth = $self->{'fulltxt_prepared_insert_statement_handle'};
426    #$meta_sth->finish() if($meta_sth);
427    #$txt_sth->finish() if($txt_sth);
428   
429    my $rc = $dbh->disconnect or warn $dbh->errstr; # The handle is of little use after disconnecting. Possibly PrintError already prints a warning and this duplicates it?
430    return $rc;
431}
432
433sub create_db {
434    my $self= shift (@_);
435    my $db_name = $self->{'site_name'};
436    my $dbh = $self->{'db_handle'};
437   
438    # https://stackoverflow.com/questions/5025768/how-can-i-create-a-mysql-database-from-a-perl-script
439    return $dbh->do("create database $db_name"); # do() will return undef on fail, https://metacpan.org/pod/DBI#do
440}
441
442sub create_metadata_table {
443    my $self= shift (@_);
444    my $dbh = $self->{'db_handle'};
445   
446    my $table_name = $self->{'collection_name'} . "metadata";
447
448    # If using an auto incremented primary key:
449    my $stmt = "CREATE TABLE $table_name (id INT NOT NULL AUTO_INCREMENT, did VARCHAR(63) NOT NULL, sid VARCHAR(63) NOT NULL, metaname VARCHAR(127) NOT NULL, metavalue VARCHAR(1023) NOT NULL, PRIMARY KEY(id));";
450    return $dbh->do($stmt);
451}
452
453# TODO: Investigate: https://dev.mysql.com/doc/search/?d=10&p=1&q=FULLTEXT
454# 12.9.1 Natural Language Full-Text Searches
455# to see whether we have to index the 'fulltxt' column of the 'fulltext' tables
456# or let user edit this file, or add it as another option
457sub create_fulltext_table {
458    my $self= shift (@_);
459    my $dbh = $self->{'db_handle'};
460   
461    my $table_name = $self->{'collection_name'} . "fulltxt";
462
463    # If using an auto incremented primary key:
464    my $stmt = "CREATE TABLE $table_name (id INT NOT NULL AUTO_INCREMENT, did VARCHAR(63) NOT NULL, sid VARCHAR(63) NOT NULL, fulltxt LONGTEXT, PRIMARY KEY(id));";
465    return $dbh->do($stmt);
466
467}
468
469
470# USEFUL: https://metacpan.org/pod/DBI
471# "Many methods have an optional \%attr parameter which can be used to pass information to the driver implementing the method. Except where specifically documented, the \%attr parameter can only be used to pass driver specific hints. In general, you can ignore \%attr parameters or pass it as undef."
472
473
474# https://www.guru99.com/insert-into.html
475# and https://dev.mysql.com/doc/refman/8.0/en/example-auto-increment.html
476#     for inserting multiple rows at once
477# https://www.perlmonks.org/bare/?node_id=316183
478# https://metacpan.org/pod/DBI#do
479# https://www.quora.com/What-is-the-difference-between-prepare-and-do-statements-in-Perl-while-we-make-a-connection-to-the-database-for-executing-the-query
480# https://docstore.mik.ua/orelly/linux/dbi/ch05_05.htm
481
482# https://metacpan.org/pod/DBI#performance
483# 'The q{...} style quoting used in this example avoids clashing with quotes that may be used in the SQL statement. Use the double-quote like qq{...} operator if you want to interpolate variables into the string. See "Quote and Quote-like Operators" in perlop for more details.'
484sub prepare_insert_metadata_row_stmthandle {
485    my $self = shift (@_);   
486    my ($did, $sid, $metaname, $metavalue) = @_;
487    my $dbh = $self->{'db_handle'};
488   
489    my $tablename = $self->{'colname'}."_metadata";
490
491    #my $stmt = "INSERT INTO $tablename (did, sid, metaname, metavalue) VALUES ('$did', '$sid', '$metaname', '$metavalue');"; # ?, ?, ?, ?
492
493    # using qq{} since we want $tablename placeholder to be filled in
494    # returns Statement Handle object!
495    my $sth = $dbh->prepare(qq{INSERT INTO $tablename (did, sid, metaname, metavalue) VALUES (?, ?, ?, ?)"}) || warn("Could not prepare insert statement for metadata table\n");
496   
497    return $sth;
498}
499
500sub prepare_insert_fulltxt_row_stmthandle {
501    my $self = shift (@_);
502    my ($did, $sid, $fulltext) = @_;
503    my $dbh = $self->{'db_handle'};
504   
505    my $tablename = $self->{'colname'}."_fulltxt";
506
507    #my $stmt = "INSERT INTO $tablename (did, sid, fulltxt) VALUES ('$did', '$sid', '$fulltext');"; ?, ?, ?
508
509    # using qq{} since we want $tablename placeholder to be filled in
510    # returns Statement Handle object!
511    my $sth = $dbh->prepare(qq{INSERT INTO $tablename (did, sid, fulltxt) VALUES (?, ?, ?)"}) || warn("Could not prepare insert statement for fulltxt table\n");
512   
513    return $sth;
514}
515
516# I can get my version of table_exists to work, but it's not so ideal
517# Interesting that MySQL has non-standard command to CREATE TABLE IF NOT EXISTS and DROP TABLE IF EXISTS,
518# see https://www.perlmonks.org/bare/?node=DBI%20Recipes
519#    The page further has a table_exists function that could work with proper comparison
520# Couldn't get the first solution at https://www.perlmonks.org/bare/?node_id=500050 to work though
521sub table_exists {
522    my ($dbh,$table_name) = @_;
523
524    my @table_list = $dbh->tables;
525    #my $tables_str = @table_list[0];
526    foreach my $table (@table_list) {
527    return 1 if ($table =~ m/$table_name/);
528    }
529    return 0;
530}
531
5321;
Note: See TracBrowser for help on using the browser.