1 |
#!/usr/bin/perl -w |
2 |
|
3 |
use strict; |
4 |
|
5 |
use Cwd qw/abs_path/; |
6 |
use File::Temp qw/tempdir/; |
7 |
use lib './lib'; |
8 |
|
9 |
use WebPAC::Common 0.02; |
10 |
use WebPAC::Parser 0.08; |
11 |
use WebPAC::Input 0.16; |
12 |
use WebPAC::Store 0.15; |
13 |
use WebPAC::Normalize 0.22; |
14 |
use WebPAC::Output::TT; |
15 |
use WebPAC::Validate 0.11; |
16 |
use WebPAC::Output::MARC; |
17 |
use WebPAC::Config; |
18 |
use Getopt::Long; |
19 |
use File::Path; |
20 |
use Time::HiRes qw/time/; |
21 |
use File::Slurp; |
22 |
use Data::Dump qw/dump/; |
23 |
use Storable qw/dclone/; |
24 |
use Pod::Usage qw/pod2usage/; |
25 |
|
26 |
use Proc::Queue size => 1; |
27 |
use POSIX ":sys_wait_h"; # imports WNOHANG |
28 |
|
29 |
=head1 NAME |
30 |
|
31 |
run.pl - start WebPAC indexing |
32 |
|
33 |
B<this command will probably go away. Don't get used to it!> |
34 |
|
35 |
=head1 OPTIONS |
36 |
|
37 |
=over 4 |
38 |
|
39 |
=item --offset 42 |
40 |
|
41 |
start loading (all) databases at offset 42 |
42 |
|
43 |
=item --limit 100 |
44 |
|
45 |
limit loading to 100 records |
46 |
|
47 |
=item --clean |
48 |
|
49 |
remove database and Hyper Estraier index before indexing |
50 |
|
51 |
=item --only=database_name/input_filter |
52 |
|
53 |
reindex just single database (legacy name is --one) |
54 |
|
55 |
C</input_filter> is optional part which can be C<name> |
56 |
or C<type> from input |
57 |
|
58 |
=item --config conf/config.yml |
59 |
|
60 |
path to YAML configuration file |
61 |
|
62 |
=item --stats |
63 |
|
64 |
disable indexing, modify_* in configuration and dump statistics about field |
65 |
and subfield usage for each input |
66 |
|
67 |
=item --validate path/to/validation_file |
68 |
|
69 |
turn on extra validation of imput records, see L<WebPAC::Validation> |
70 |
|
71 |
You can use special variables C<$database> and $C<$input> in this parametar |
72 |
like C<--validate 'conf/validate/$database-$input'> to construct filename |
73 |
|
74 |
=item --validate-delimiters path/to/validate_delimiters_file |
75 |
|
76 |
this option is used with C<--validate> to turn on extra validation of |
77 |
delimiters. If file is non existant, it will be created on first run. |
78 |
|
79 |
=item --marc-generate |
80 |
|
81 |
Generate MARC file. This will automatically be on if file contains C<marc*> directives. |
82 |
You can use this option as C<--no-marc-generate> to disable MARC generation. |
83 |
|
84 |
=item --marc-lint |
85 |
|
86 |
By default turned on if normalisation file has C<marc*> directives. You can disable lint |
87 |
messages with C<--no-marc-lint>. |
88 |
|
89 |
=item --marc-dump |
90 |
|
91 |
Force dump or input and marc record for debugging. |
92 |
|
93 |
=item --parallel 4 |
94 |
|
95 |
Run databases in parallel (aproximatly same as number of processors in |
96 |
machine if you want to use full load) |
97 |
|
98 |
=item --only-links |
99 |
|
100 |
Create just links |
101 |
|
102 |
=item --merge |
103 |
|
104 |
Create merged index of databases which have links |
105 |
|
106 |
=back |
107 |
|
108 |
=cut |
109 |
|
110 |
my $offset; |
111 |
my $limit; |
112 |
|
113 |
my $clean = 0; |
114 |
my $config_path; |
115 |
my $debug = 0; |
116 |
my $only_filter; |
117 |
my $stats = 0; |
118 |
my $validate_path; |
119 |
my $validate_delimiters_path; |
120 |
my $marc_generate = 1; |
121 |
my $marc_lint = 1; |
122 |
my $marc_dump = 0; |
123 |
my $parallel = 0; |
124 |
my $only_links = 0; |
125 |
my $merge = 0; |
126 |
my $help; |
127 |
|
128 |
my $log = _new WebPAC::Common()->_get_logger(); |
129 |
|
130 |
GetOptions( |
131 |
"limit=i" => \$limit, |
132 |
"offset=i" => \$offset, |
133 |
"clean" => \$clean, |
134 |
"one=s" => \$only_filter, |
135 |
"only=s" => \$only_filter, |
136 |
"config=s" => \$config_path, |
137 |
"debug+" => \$debug, |
138 |
"stats" => \$stats, |
139 |
"validate=s" => \$validate_path, |
140 |
"validate-delimiters=s" => \$validate_delimiters_path, |
141 |
"marc-generate!" => \$marc_generate, |
142 |
"marc-lint!" => \$marc_lint, |
143 |
"marc-dump!" => \$marc_dump, |
144 |
"parallel=i" => \$parallel, |
145 |
"only-links!" => \$only_links, |
146 |
"merge" => \$merge, |
147 |
"help" => \$help, |
148 |
); |
149 |
|
150 |
$marc_generate = 0 if ( $validate_delimiters_path ); |
151 |
|
152 |
pod2usage(-verbose => 2) if ($help); |
153 |
|
154 |
my $config = new WebPAC::Config( path => $config_path ); |
155 |
|
156 |
WebPAC::Normalize::_debug( $debug - 1 ) if $debug > 1; |
157 |
|
158 |
#print "config = ",dump($config) if ($debug); |
159 |
|
160 |
die "no databases in config file!\n" unless ($config->databases); |
161 |
|
162 |
$log->info( "-" x 79 ); |
163 |
|
164 |
my $log_file = 'log'; |
165 |
|
166 |
if (-e $log_file ) { # && -s $log_file > 5 * 1024 * 1024) { |
167 |
$log->info("moved old log with ", -s $log_file, " bytes to '${log_file}.old'"); |
168 |
rename $log_file, "${log_file}.old" || $log->logwarn("can't rename $log_file to ${log_file}.old: $!"); |
169 |
} |
170 |
|
171 |
my $estcmd_fh; |
172 |
my $estcmd_path = './estcmd-merge.sh'; |
173 |
if ($merge) { |
174 |
open($estcmd_fh, '>', $estcmd_path) || $log->logdie("can't open $estcmd_path: $!"); |
175 |
print $estcmd_fh 'cd /data/estraier/_node/ || exit 1',$/; |
176 |
print $estcmd_fh 'sudo /etc/init.d/hyperestraier stop',$/; |
177 |
$log->info("created merge batch file $estcmd_path"); |
178 |
} |
179 |
|
180 |
my $validate; |
181 |
$validate = new WebPAC::Validate( |
182 |
delimiters => $config->webpac('delimiters'), |
183 |
) if ($validate_path || $validate_delimiters_path); |
184 |
|
185 |
my $use_indexer = $config->use_indexer; |
186 |
$stats ||= $validate; |
187 |
if ($stats) { |
188 |
$log->debug("disabled indexing for stats collection"); |
189 |
$use_indexer = undef; |
190 |
} elsif ( $use_indexer ) { |
191 |
$log->info("using $use_indexer indexing engine..."); |
192 |
} |
193 |
|
194 |
# parse normalize files and create source files for lookup and normalization |
195 |
|
196 |
my ($only_database,$only_input) = split(m#/#, $only_filter) if $only_filter; |
197 |
|
198 |
my $parser = new WebPAC::Parser( |
199 |
config => $config, |
200 |
only_database => $only_database, |
201 |
only_input => $only_input, |
202 |
); |
203 |
|
204 |
my $total_rows = 0; |
205 |
my $start_t = time(); |
206 |
|
207 |
my @links; |
208 |
|
209 |
if ($parallel) { |
210 |
$log->info("Using $parallel processes for speedup"); |
211 |
Proc::Queue::size($parallel); |
212 |
} |
213 |
|
214 |
sub create_ds_config { |
215 |
my ($db_config, $database, $input, $mfn) = @_; |
216 |
my $c = dclone( $db_config ); |
217 |
$c->{_} = $database || $log->logconfess("need database"); |
218 |
$c->{_mfn} = $mfn || $log->logconfess("need mfn"); |
219 |
$c->{input} = $input || $log->logconfess("need input"); |
220 |
return $c; |
221 |
} |
222 |
|
223 |
foreach my $database ( sort keys %{ $config->databases } ) { |
224 |
my $db_config = $config->databases->{$database}; |
225 |
|
226 |
next if ($only_database && $database !~ m/$only_database/i); |
227 |
|
228 |
if ($parallel) { |
229 |
my $f=fork; |
230 |
if(defined ($f) and $f==0) { |
231 |
$log->info("Created processes $$ for speedup"); |
232 |
} else { |
233 |
next; |
234 |
} |
235 |
} |
236 |
|
237 |
my $indexer; |
238 |
if ($use_indexer && $parser->have_rules( 'search', $database )) { |
239 |
|
240 |
my $cfg_name = $use_indexer; |
241 |
$cfg_name =~ s/\-.*$//; |
242 |
|
243 |
my $indexer_config = $config->get( $cfg_name ) || $log->logdie("can't find '$cfg_name' part in confguration"); |
244 |
$indexer_config->{database} = $database; |
245 |
$indexer_config->{clean} = $clean; |
246 |
$indexer_config->{label} = $db_config->{name}; |
247 |
|
248 |
# force clean if database has links |
249 |
$indexer_config->{clean} = 1 if ($db_config->{links}); |
250 |
|
251 |
if ($use_indexer eq 'hyperestraier') { |
252 |
|
253 |
# open Hyper Estraier database |
254 |
require WebPAC::Output::Estraier; |
255 |
$indexer = new WebPAC::Output::Estraier( %{ $indexer_config } ); |
256 |
|
257 |
} elsif ($use_indexer eq 'hyperestraier-native') { |
258 |
|
259 |
# open Hyper Estraier database |
260 |
require WebPAC::Output::EstraierNative; |
261 |
$indexer = new WebPAC::Output::EstraierNative( %{ $indexer_config } ); |
262 |
|
263 |
} elsif ($use_indexer eq 'kinosearch') { |
264 |
|
265 |
die "no longer supported"; |
266 |
|
267 |
} else { |
268 |
$log->logdie("unknown use_indexer: $use_indexer"); |
269 |
} |
270 |
|
271 |
$log->logdie("can't continue without valid indexer") unless ($indexer); |
272 |
} |
273 |
|
274 |
|
275 |
# |
276 |
# store Hyper Estraier links to other databases |
277 |
# |
278 |
if (ref($db_config->{links}) eq 'ARRAY' && $use_indexer) { |
279 |
foreach my $link (@{ $db_config->{links} }) { |
280 |
if ($use_indexer eq 'hyperestraier') { |
281 |
if ($merge) { |
282 |
print $estcmd_fh 'sudo -u www-data estcmd merge ' . $database . ' ' . $link->{to},$/; |
283 |
} else { |
284 |
$log->info("saving link $database -> $link->{to} [$link->{credit}]"); |
285 |
push @links, sub { |
286 |
$log->info("adding link $database -> $link->{to} [$link->{credit}]"); |
287 |
$indexer->add_link( |
288 |
from => $database, |
289 |
to => $link->{to}, |
290 |
credit => $link->{credit}, |
291 |
); |
292 |
}; |
293 |
} |
294 |
} else { |
295 |
$log->warn("NOT IMPLEMENTED WITH $use_indexer: adding link $database -> $link->{to} [$link->{credit}]"); |
296 |
} |
297 |
} |
298 |
} |
299 |
next if ($only_links); |
300 |
|
301 |
|
302 |
# |
303 |
# now WebPAC::Store |
304 |
# |
305 |
my $store = new WebPAC::Store({ |
306 |
debug => $debug, |
307 |
}); |
308 |
|
309 |
|
310 |
# |
311 |
# prepare output |
312 |
# |
313 |
my @outputs = force_array( $db_config->{output}, sub { |
314 |
$log->error("Database $database doesn't have any outputs defined. Do you want to remove it from configuration?" ); |
315 |
} ); |
316 |
|
317 |
my @output_modules; |
318 |
|
319 |
foreach my $output ( @outputs ) { |
320 |
|
321 |
#warn '## output = ',dump( $output ); |
322 |
|
323 |
my $module = $output->{module} || $log->logdie("need module in output section of $database"); |
324 |
$module = 'WebPAC::Output::' . $module unless $module =~ m/::/; |
325 |
|
326 |
$log->debug("loading output module $module"); |
327 |
eval "require $module"; |
328 |
|
329 |
# add database to arugemnts for output filter |
330 |
$output->{database} = $database; |
331 |
|
332 |
$log->debug("calling $module->new(",dump( $output ),")"); |
333 |
my $out = new $module->new( $output ); |
334 |
if ( $out->init ) { |
335 |
push @output_modules, $out; |
336 |
} else { |
337 |
$log->warn("SKIPPED $module"); |
338 |
} |
339 |
} |
340 |
|
341 |
|
342 |
# |
343 |
# now, iterate through input formats |
344 |
# |
345 |
|
346 |
|
347 |
my @inputs = force_array( $db_config->{input}, sub { |
348 |
$log->info("database $database doesn't have inputs defined"); |
349 |
} ); |
350 |
|
351 |
foreach my $input (@inputs) { |
352 |
|
353 |
my $input_name = $input->{name} || $log->logdie("input without a name isn't valid: ",dump($input)); |
354 |
|
355 |
next if ($only_input && ($input_name !~ m#$only_input#i && $input->{type} !~ m#$only_input#i)); |
356 |
|
357 |
my $type = lc($input->{type}); |
358 |
|
359 |
# FIXME check if input module exists |
360 |
my $input_module = $input->{module}; |
361 |
|
362 |
if ( ! $input_module ) { |
363 |
if ( grep(/$type/, $config->webpac('inputs')) ) { |
364 |
$input_module = $config->webpac('inputs')->{$type}; |
365 |
} else { |
366 |
$log->logdie("I know only how to handle input types ", join(",", $config->webpac('inputs') ), " not '$type'!" ); |
367 |
} |
368 |
} |
369 |
|
370 |
my @lookups = $parser->have_lookup_create($database, $input); |
371 |
|
372 |
$log->info("working on input '$input_name' in $input->{path} [type: $input->{type}] using $input_module", |
373 |
@lookups ? " creating lookups: ".join(", ", @lookups) : "" |
374 |
); |
375 |
|
376 |
if ($stats) { |
377 |
# disable modification of records if --stats is in use |
378 |
delete($input->{modify_records}); |
379 |
delete($input->{modify_file}); |
380 |
} |
381 |
|
382 |
my $input_db = new WebPAC::Input( |
383 |
module => $input_module, |
384 |
encoding => $config->webpac('webpac_encoding'), |
385 |
limit => $limit || $input->{limit}, |
386 |
offset => $offset, |
387 |
recode => $input->{recode}, |
388 |
stats => $stats, |
389 |
modify_records => $input->{modify_records}, |
390 |
modify_file => $input->{modify_file}, |
391 |
input_config => $input, |
392 |
); |
393 |
$log->logdie("can't create input using $input_module") unless ($input); |
394 |
|
395 |
if (defined( $input->{lookup} )) { |
396 |
$log->warn("$database/$input_name has depriciated lookup definition, removing it..."); |
397 |
delete( $input->{lookup} ); |
398 |
} |
399 |
|
400 |
my $lookup_coderef; |
401 |
|
402 |
if (@lookups) { |
403 |
|
404 |
my $rules = $parser->lookup_create_rules($database, $input) || $log->logdie("no rules found for $database/$input"); |
405 |
|
406 |
$lookup_coderef = sub { |
407 |
my $rec = shift || die "need rec!"; |
408 |
my $mfn = $rec->{'000'}->[0] || die "need mfn in 000"; |
409 |
|
410 |
WebPAC::Normalize::data_structure( |
411 |
row => $rec, |
412 |
rules => $rules, |
413 |
config => create_ds_config( $db_config, $database, $input, $mfn ), |
414 |
); |
415 |
|
416 |
#warn "current lookup: ", dump(WebPAC::Normalize::_get_lookup()); |
417 |
}; |
418 |
|
419 |
WebPAC::Normalize::_set_lookup( undef ); |
420 |
|
421 |
$log->debug("created lookup_coderef using:\n$rules"); |
422 |
|
423 |
}; |
424 |
|
425 |
my $lookup_jar; |
426 |
|
427 |
my $maxmfn = $input_db->open( |
428 |
path => $input->{path}, |
429 |
code_page => $input->{encoding}, # database encoding |
430 |
lookup_coderef => $lookup_coderef, |
431 |
lookup => $lookup_jar, |
432 |
%{ $input }, |
433 |
load_row => sub { |
434 |
my $a = shift; |
435 |
return $store->load_row( |
436 |
database => $database, |
437 |
input => $input_name, |
438 |
id => $a->{id}, |
439 |
); |
440 |
}, |
441 |
save_row => sub { |
442 |
my $a = shift; |
443 |
return $store->save_row( |
444 |
database => $database, |
445 |
input => $input_name, |
446 |
id => $a->{id}, |
447 |
row => $a->{row}, |
448 |
); |
449 |
}, |
450 |
|
451 |
); |
452 |
|
453 |
my $lookup_data = WebPAC::Normalize::_get_lookup(); |
454 |
|
455 |
if (defined( $lookup_data->{$database}->{$input_name} )) { |
456 |
$log->debug("created following lookups: ", sub { dump( $lookup_data ) } ); |
457 |
|
458 |
foreach my $key (keys %{ $lookup_data->{$database}->{$input_name} }) { |
459 |
$store->save_lookup( |
460 |
database => $database, |
461 |
input => $input_name, |
462 |
key => $key, |
463 |
data => $lookup_data->{$database}->{$input_name}->{$key}, |
464 |
); |
465 |
} |
466 |
} |
467 |
|
468 |
my $report_fh; |
469 |
if ($stats || $validate) { |
470 |
my $path = "out/report/${database}-${input_name}.txt"; |
471 |
open($report_fh, '>', $path) || $log->logdie("can't open $path: $!"); |
472 |
|
473 |
print $report_fh "Report for database '$database' input '$input_name' records ", |
474 |
$offset || 1, "-", $limit || $input->{limit} || $maxmfn, "\n\n"; |
475 |
$log->info("Generating report file $path"); |
476 |
|
477 |
if ( $validate ) { |
478 |
$validate->read_validate_file( $validate->fill_in( $validate_path, database => $database, input => $input_name ) ) if ( $validate_path ); |
479 |
$validate->read_validate_delimiters_file( $validate->fill_in( $validate_delimiters_path, database => $database, input => $input_name ) ) if ( $validate_delimiters_path ); |
480 |
} |
481 |
} |
482 |
|
483 |
my $marc; |
484 |
if ($marc_generate && $parser->have_rules( 'marc', $database, $input_name )) { |
485 |
$marc = new WebPAC::Output::MARC( |
486 |
path => "out/marc/${database}-${input_name}.marc", |
487 |
lint => $marc_lint, |
488 |
dump => $marc_dump, |
489 |
); |
490 |
} |
491 |
|
492 |
my $rules = $parser->normalize_rules($database,$input_name) || $log->logdie("no normalize rules found for $database/$input_name"); |
493 |
$log->debug("parsed normalize rules:\n$rules"); |
494 |
|
495 |
# reset position in database |
496 |
$input_db->seek(1); |
497 |
|
498 |
# generate name of config key for indexer (strip everything after -) |
499 |
my $indexer_config = $use_indexer; |
500 |
$indexer_config =~ s/^(\w+)-?.*$/$1/g if ($indexer_config); |
501 |
|
502 |
my $lookup_hash; |
503 |
my $depends = $parser->depends($database,$input_name); |
504 |
|
505 |
if ($depends) { |
506 |
$log->debug("$database/$input_name depends on: ", dump($depends)) if ($depends); |
507 |
$log->logdie("parser->depends didn't return HASH") unless (ref($depends) eq 'HASH'); |
508 |
|
509 |
foreach my $db (keys %$depends) { |
510 |
foreach my $i (keys %{$depends->{$db}}) { |
511 |
foreach my $k (keys %{$depends->{$db}->{$i}}) { |
512 |
my $t = time(); |
513 |
$log->debug("loading lookup $db/$i"); |
514 |
$lookup_hash->{$db}->{$i}->{$k} = $store->load_lookup( |
515 |
database => $db, |
516 |
input => $i, |
517 |
key => $k, |
518 |
); |
519 |
$log->debug(sprintf("lookup $db/$i took %.2fs", time() - $t)); |
520 |
} |
521 |
} |
522 |
} |
523 |
|
524 |
$log->debug("lookup_hash = ", sub { dump( $lookup_hash ) }); |
525 |
} |
526 |
|
527 |
|
528 |
# setup input name for all output filters |
529 |
foreach my $out ( @output_modules ) { |
530 |
if ( $out->can('input') ) { |
531 |
$out->input( $input_name ); |
532 |
} else { |
533 |
$log->warn("output filter ",ref($out)," doesn't support input name"); |
534 |
} |
535 |
} |
536 |
|
537 |
|
538 |
foreach my $pos ( 0 ... $input_db->size ) { |
539 |
|
540 |
my $row = $input_db->fetch || next; |
541 |
|
542 |
$total_rows++; |
543 |
|
544 |
my $mfn = $row->{'000'}->[0]; |
545 |
|
546 |
if (! $mfn || $mfn !~ m{^\d+$}) { |
547 |
$log->warn("record $pos doesn't have valid MFN but '$mfn', using $pos"); |
548 |
$mfn = $pos; |
549 |
push @{ $row->{'000'} }, $pos; |
550 |
} |
551 |
|
552 |
|
553 |
if ($validate) { |
554 |
if ( my $errors = $validate->validate_rec( $row, $input_db->dump_ascii ) ) { |
555 |
$log->error( "MFN $mfn validation error:\n", |
556 |
$validate->report_error( $errors ) |
557 |
); |
558 |
} |
559 |
next; # validation doesn't create any output |
560 |
} |
561 |
|
562 |
my $ds = WebPAC::Normalize::data_structure( |
563 |
row => $row, |
564 |
rules => $rules, |
565 |
lookup => $lookup_hash, |
566 |
config => create_ds_config( $db_config, $database, $input, $mfn ), |
567 |
marc_encoding => 'utf-8', |
568 |
load_row_coderef => sub { |
569 |
my ($database,$input,$mfn) = @_; |
570 |
#warn "### load_row($database,$input,$mfn) from data_structure\n"; |
571 |
return $store->load_row( |
572 |
database => $database, |
573 |
input => $input, |
574 |
id => $mfn, |
575 |
); |
576 |
}, |
577 |
); |
578 |
|
579 |
$log->debug("ds = ", sub { dump($ds) }); |
580 |
|
581 |
if ( $ds ) { |
582 |
|
583 |
$store->save_ds( |
584 |
database => $database, |
585 |
input => $input_name, |
586 |
id => $mfn, |
587 |
ds => $ds, |
588 |
) if !$stats; |
589 |
|
590 |
$indexer->add( |
591 |
id => "${input_name}/${mfn}", |
592 |
ds => $ds, |
593 |
type => $config->get($indexer_config)->{type}, |
594 |
) if $indexer; |
595 |
|
596 |
foreach my $out ( @output_modules ) { |
597 |
$out->add( $mfn, $ds ) if $out->can('add'); |
598 |
} |
599 |
|
600 |
} else { |
601 |
$log->warn("record $pos didn't produce any output after normalization rules!") unless $marc; |
602 |
} |
603 |
|
604 |
if ($marc) { |
605 |
my $i = 0; |
606 |
|
607 |
while (my $fields = WebPAC::Normalize::MARC::_get_marc_fields( fetch_next => 1 ) ) { |
608 |
$marc->add( |
609 |
id => $mfn . ( $i ? "/$i" : '' ), |
610 |
fields => $fields, |
611 |
leader => WebPAC::Normalize::MARC::_get_marc_leader(), |
612 |
row => $row, |
613 |
); |
614 |
$i++; |
615 |
} |
616 |
|
617 |
$log->info("Created $i instances of MFN $mfn\n") if ($i > 1); |
618 |
} |
619 |
|
620 |
} |
621 |
|
622 |
if ($validate) { |
623 |
my $errors = $validate->report; |
624 |
if ($errors) { |
625 |
$log->info("validation errors:\n$errors\n" ); |
626 |
print $report_fh "$errors\n" if ($report_fh); |
627 |
} |
628 |
|
629 |
print $report_fh "\nAll possible subfields/delimiter templates:\n", $validate->delimiters_templates( report => 1, current_input => 1 ), "\n\n"; |
630 |
|
631 |
# must be last thing that touches $validate for this input |
632 |
$validate->reset; |
633 |
} |
634 |
|
635 |
if ($stats) { |
636 |
my $s = $input_db->stats; |
637 |
$log->info("statistics of fields usage:\n$s"); |
638 |
print $report_fh "Statistics of fields usage:\n$s" if ($report_fh); |
639 |
} |
640 |
|
641 |
# close MARC file |
642 |
$marc->finish if ($marc); |
643 |
|
644 |
# close report |
645 |
close($report_fh) if ($report_fh); |
646 |
} |
647 |
|
648 |
eval { $indexer->finish } if ($indexer && $indexer->can('finish')); |
649 |
|
650 |
foreach my $out ( @output_modules ) { |
651 |
$out->finish if $out->can('finish'); |
652 |
} |
653 |
|
654 |
my $dt = time() - $start_t; |
655 |
$log->info("$total_rows records ", $indexer ? "indexed " : "", |
656 |
sprintf("in %.2f sec [%.2f rec/sec]", |
657 |
$dt, ($total_rows / $dt) |
658 |
) |
659 |
); |
660 |
|
661 |
|
662 |
# end forked process |
663 |
if ($parallel) { |
664 |
$log->info("parallel process $$ finished"); |
665 |
exit(0); |
666 |
} |
667 |
|
668 |
} |
669 |
|
670 |
if ($parallel) { |
671 |
# wait all children to finish |
672 |
sleep(1) while wait != -1; |
673 |
$log->info("all parallel processes finished"); |
674 |
} |
675 |
|
676 |
# save new delimiters if needed |
677 |
$validate->save_delimiters_templates if ( $validate_delimiters_path ); |
678 |
|
679 |
# |
680 |
# handle links or merge after indexing |
681 |
# |
682 |
|
683 |
if ($merge) { |
684 |
print $estcmd_fh 'sudo /etc/init.d/hyperestraier start',$/; |
685 |
close($estcmd_fh); |
686 |
chmod 0700, $estcmd_path || $log->warn("can't chmod 0700 $estcmd_path: $!"); |
687 |
system $estcmd_path; |
688 |
} else { |
689 |
foreach my $link (@links) { |
690 |
$log->logdie("coderef in link ", Dumper($link), " is ", ref($link), " and not CODE") unless (ref($link) eq 'CODE'); |
691 |
$link->(); |
692 |
} |
693 |
} |