4 |
|
|
5 |
use Cwd qw/abs_path/; |
use Cwd qw/abs_path/; |
6 |
use File::Temp qw/tempdir/; |
use File::Temp qw/tempdir/; |
|
use Data::Dumper; |
|
7 |
use lib './lib'; |
use lib './lib'; |
8 |
|
|
9 |
use WebPAC::Common 0.02; |
use WebPAC::Common 0.02; |
10 |
use WebPAC::Lookup; |
use WebPAC::Parser 0.04; |
11 |
use WebPAC::Input 0.03; |
use WebPAC::Lookup 0.03; |
12 |
|
use WebPAC::Input 0.11; |
13 |
use WebPAC::Store 0.03; |
use WebPAC::Store 0.03; |
14 |
use WebPAC::Normalize::XML; |
use WebPAC::Normalize 0.11; |
|
use WebPAC::Normalize::Set; |
|
15 |
use WebPAC::Output::TT; |
use WebPAC::Output::TT; |
16 |
use WebPAC::Validate; |
use WebPAC::Validate 0.06; |
17 |
use YAML qw/LoadFile/; |
use WebPAC::Output::MARC; |
18 |
|
use WebPAC::Config; |
19 |
use Getopt::Long; |
use Getopt::Long; |
20 |
use File::Path; |
use File::Path; |
21 |
use Time::HiRes qw/time/; |
use Time::HiRes qw/time/; |
22 |
use File::Slurp; |
use File::Slurp; |
23 |
|
use Data::Dump qw/dump/; |
24 |
|
use Storable qw/dclone/; |
25 |
|
|
26 |
|
use Proc::Queue size => 1; |
27 |
|
use POSIX ":sys_wait_h"; # imports WNOHANG |
28 |
|
|
29 |
=head1 NAME |
=head1 NAME |
30 |
|
|
59 |
|
|
60 |
path to YAML configuration file |
path to YAML configuration file |
61 |
|
|
|
=item --force-set |
|
|
|
|
|
force conversion C<< normalize->path >> in C<config.yml> from |
|
|
C<.xml> to C<.pl> |
|
|
|
|
62 |
=item --stats |
=item --stats |
63 |
|
|
64 |
disable indexing and dump statistics about field and subfield |
disable indexing, modify_* in configuration and dump statistics about field |
65 |
usage for each input |
and subfield usage for each input |
66 |
|
|
67 |
=item --validate path/to/validation_file |
=item --validate path/to/validation_file |
68 |
|
|
69 |
turn on extra validation of imput records, see L<WebPAC::Validation> |
turn on extra validation of imput records, see L<WebPAC::Validation> |
70 |
|
|
71 |
|
=item --marc-normalize conf/normalize/mapping.pl |
72 |
|
|
73 |
|
This option specifies normalisation file for MARC creation |
74 |
|
|
75 |
|
=item --marc-output out/marc/test.marc |
76 |
|
|
77 |
|
Optional path to output file |
78 |
|
|
79 |
|
=item --marc-lint |
80 |
|
|
81 |
|
By default turned on if C<--marc-normalize> is used. You can disable lint |
82 |
|
messages with C<--no-marc-lint>. |
83 |
|
|
84 |
|
=item --marc-dump |
85 |
|
|
86 |
|
Force dump or input and marc record for debugging. |
87 |
|
|
88 |
|
=item --parallel 4 |
89 |
|
|
90 |
|
Run databases in parallel (aproximatly same as number of processors in |
91 |
|
machine if you want to use full load) |
92 |
|
|
93 |
|
=item --only-links |
94 |
|
|
95 |
|
Create just links |
96 |
|
|
97 |
|
=item --merge |
98 |
|
|
99 |
|
Create merged index of databases which have links |
100 |
|
|
101 |
=back |
=back |
102 |
|
|
103 |
=cut |
=cut |
106 |
my $limit; |
my $limit; |
107 |
|
|
108 |
my $clean = 0; |
my $clean = 0; |
109 |
my $config = 'conf/config.yml'; |
my $config_path; |
110 |
my $debug = 0; |
my $debug = 0; |
111 |
my $only_filter; |
my $only_filter; |
|
my $force_set = 0; |
|
112 |
my $stats = 0; |
my $stats = 0; |
113 |
my $validate_path; |
my $validate_path; |
114 |
|
my ($marc_normalize, $marc_output); |
115 |
|
my $marc_lint = 1; |
116 |
|
my $marc_dump = 0; |
117 |
|
my $parallel = 0; |
118 |
|
my $only_links = 0; |
119 |
|
my $merge = 0; |
120 |
|
|
121 |
|
my $log = _new WebPAC::Common()->_get_logger(); |
122 |
|
|
123 |
GetOptions( |
GetOptions( |
124 |
"limit=i" => \$limit, |
"limit=i" => \$limit, |
126 |
"clean" => \$clean, |
"clean" => \$clean, |
127 |
"one=s" => \$only_filter, |
"one=s" => \$only_filter, |
128 |
"only=s" => \$only_filter, |
"only=s" => \$only_filter, |
129 |
"config" => \$config, |
"config" => \$config_path, |
130 |
"debug" => \$debug, |
"debug+" => \$debug, |
|
"force-set" => \$force_set, |
|
131 |
"stats" => \$stats, |
"stats" => \$stats, |
132 |
"validate=s" => \$validate_path, |
"validate=s" => \$validate_path, |
133 |
|
"marc-normalize=s" => \$marc_normalize, |
134 |
|
"marc-output=s" => \$marc_output, |
135 |
|
"marc-lint!" => \$marc_lint, |
136 |
|
"marc-dump!" => \$marc_dump, |
137 |
|
"parallel=i" => \$parallel, |
138 |
|
"only-links!" => \$only_links, |
139 |
|
"merge" => \$merge, |
140 |
); |
); |
141 |
|
|
142 |
$config = LoadFile($config); |
my $config = new WebPAC::Config( path => $config_path ); |
143 |
|
|
144 |
print "config = ",Dumper($config) if ($debug); |
#print "config = ",dump($config) if ($debug); |
145 |
|
|
146 |
die "no databases in config file!\n" unless ($config->{databases}); |
die "no databases in config file!\n" unless ($config->databases); |
147 |
|
|
|
my $log = _new WebPAC::Common()->_get_logger(); |
|
148 |
$log->info( "-" x 79 ); |
$log->info( "-" x 79 ); |
149 |
|
|
150 |
|
|
151 |
|
my $estcmd_fh; |
152 |
|
my $estcmd_path = './estcmd-merge.sh'; |
153 |
|
if ($merge) { |
154 |
|
open($estcmd_fh, '>', $estcmd_path) || $log->logdie("can't open $estcmd_path: $!"); |
155 |
|
print $estcmd_fh 'cd /data/estraier/_node/ || exit 1',$/; |
156 |
|
print $estcmd_fh 'sudo /etc/init.d/hyperestraier stop',$/; |
157 |
|
$log->info("created merge batch file $estcmd_path"); |
158 |
|
} |
159 |
|
|
160 |
|
|
161 |
my $validate; |
my $validate; |
162 |
$validate = new WebPAC::Validate( |
$validate = new WebPAC::Validate( |
163 |
path => $validate_path, |
path => $validate_path, |
164 |
) if ($validate_path); |
) if ($validate_path); |
165 |
|
|
166 |
my $use_indexer = $config->{use_indexer} || 'hyperestraier'; |
|
167 |
|
my $use_indexer = $config->use_indexer; |
168 |
if ($stats) { |
if ($stats) { |
169 |
$log->debug("option --stats disables update of indexing engine..."); |
$log->debug("option --stats disables update of indexing engine..."); |
170 |
$use_indexer = undef; |
$use_indexer = undef; |
172 |
$log->info("using $use_indexer indexing engine..."); |
$log->info("using $use_indexer indexing engine..."); |
173 |
} |
} |
174 |
|
|
175 |
|
# disable indexing when creating marc |
176 |
|
$use_indexer = undef if ($marc_normalize); |
177 |
|
|
178 |
|
# parse normalize files and create source files for lookup and normalization |
179 |
|
|
180 |
|
my $parser = new WebPAC::Parser( config => $config ); |
181 |
|
|
182 |
my $total_rows = 0; |
my $total_rows = 0; |
183 |
my $start_t = time(); |
my $start_t = time(); |
184 |
|
|
185 |
while (my ($database, $db_config) = each %{ $config->{databases} }) { |
my @links; |
186 |
|
|
187 |
|
if ($parallel) { |
188 |
|
$log->info("Using $parallel processes for speedup"); |
189 |
|
Proc::Queue::size($parallel); |
190 |
|
} |
191 |
|
|
192 |
|
while (my ($database, $db_config) = each %{ $config->databases }) { |
193 |
|
|
194 |
my ($only_database,$only_input) = split(m#/#, $only_filter) if ($only_filter); |
my ($only_database,$only_input) = split(m#/#, $only_filter) if ($only_filter); |
195 |
next if ($only_database && $database !~ m/$only_database/i); |
next if ($only_database && $database !~ m/$only_database/i); |
196 |
|
|
197 |
my $indexer; |
if ($parallel) { |
198 |
|
my $f=fork; |
199 |
|
if(defined ($f) and $f==0) { |
200 |
|
$log->info("Created processes $$ for speedup"); |
201 |
|
} else { |
202 |
|
next; |
203 |
|
} |
204 |
|
} |
205 |
|
|
206 |
|
my $indexer; |
207 |
if ($use_indexer) { |
if ($use_indexer) { |
208 |
my $indexer_config = $config->{$use_indexer} || $log->logdie("can't find '$use_indexer' part in confguration"); |
|
209 |
|
my $cfg_name = $use_indexer; |
210 |
|
$cfg_name =~ s/\-.*$//; |
211 |
|
|
212 |
|
my $indexer_config = $config->get( $cfg_name ) || $log->logdie("can't find '$cfg_name' part in confguration"); |
213 |
$indexer_config->{database} = $database; |
$indexer_config->{database} = $database; |
214 |
$indexer_config->{clean} = $clean; |
$indexer_config->{clean} = $clean; |
215 |
$indexer_config->{label} = $db_config->{name}; |
$indexer_config->{label} = $db_config->{name}; |
216 |
|
|
217 |
|
# force clean if database has links |
218 |
|
$indexer_config->{clean} = 1 if ($db_config->{links}); |
219 |
|
|
220 |
if ($use_indexer eq 'hyperestraier') { |
if ($use_indexer eq 'hyperestraier') { |
221 |
|
|
222 |
# open Hyper Estraier database |
# open Hyper Estraier database |
223 |
use WebPAC::Output::Estraier '0.10'; |
use WebPAC::Output::Estraier '0.10'; |
224 |
$indexer = new WebPAC::Output::Estraier( %{ $indexer_config } ); |
$indexer = new WebPAC::Output::Estraier( %{ $indexer_config } ); |
225 |
|
|
226 |
|
} elsif ($use_indexer eq 'hyperestraier-native') { |
227 |
|
|
228 |
|
# open Hyper Estraier database |
229 |
|
use WebPAC::Output::EstraierNative; |
230 |
|
$indexer = new WebPAC::Output::EstraierNative( %{ $indexer_config } ); |
231 |
|
|
232 |
} elsif ($use_indexer eq 'kinosearch') { |
} elsif ($use_indexer eq 'kinosearch') { |
233 |
|
|
234 |
# open KinoSearch |
# open KinoSearch |
245 |
|
|
246 |
|
|
247 |
# |
# |
248 |
|
# store Hyper Estraier links to other databases |
249 |
|
# |
250 |
|
if (ref($db_config->{links}) eq 'ARRAY' && $use_indexer) { |
251 |
|
foreach my $link (@{ $db_config->{links} }) { |
252 |
|
if ($use_indexer eq 'hyperestraier') { |
253 |
|
if ($merge) { |
254 |
|
print $estcmd_fh 'sudo -u www-data estcmd merge ' . $database . ' ' . $link->{to},$/; |
255 |
|
} else { |
256 |
|
$log->info("saving link $database -> $link->{to} [$link->{credit}]"); |
257 |
|
push @links, sub { |
258 |
|
$log->info("adding link $database -> $link->{to} [$link->{credit}]"); |
259 |
|
$indexer->add_link( |
260 |
|
from => $database, |
261 |
|
to => $link->{to}, |
262 |
|
credit => $link->{credit}, |
263 |
|
); |
264 |
|
}; |
265 |
|
} |
266 |
|
} else { |
267 |
|
$log->warn("NOT IMPLEMENTED WITH $use_indexer: adding link $database -> $link->{to} [$link->{credit}]"); |
268 |
|
} |
269 |
|
} |
270 |
|
} |
271 |
|
next if ($only_links); |
272 |
|
|
273 |
|
|
274 |
|
# |
275 |
# now WebPAC::Store |
# now WebPAC::Store |
276 |
# |
# |
277 |
my $abs_path = abs_path($0); |
my $abs_path = abs_path($0); |
278 |
$abs_path =~ s#/[^/]*$#/#; |
$abs_path =~ s#/[^/]*$#/#; |
279 |
|
|
280 |
my $db_path = $config->{webpac}->{db_path} . '/' . $database; |
my $db_path = $config->get('webpac')->{db_path} . '/' . $database; |
281 |
|
|
282 |
if ($clean) { |
if ($clean) { |
283 |
$log->info("creating new database '$database' in $db_path"); |
$log->info("creating new database '$database' in $db_path"); |
306 |
$log->info("database $database doesn't have inputs defined"); |
$log->info("database $database doesn't have inputs defined"); |
307 |
} |
} |
308 |
|
|
|
my @supported_inputs = keys %{ $config->{webpac}->{inputs} }; |
|
|
|
|
309 |
foreach my $input (@inputs) { |
foreach my $input (@inputs) { |
310 |
|
|
311 |
next if ($only_input && ($input->{name} !~ m#$only_input#i && $input->{type} !~ m#$only_input#i)); |
next if ($only_input && ($input->{name} !~ m#$only_input#i && $input->{type} !~ m#$only_input#i)); |
312 |
|
|
313 |
my $type = lc($input->{type}); |
my $type = lc($input->{type}); |
314 |
|
|
315 |
die "I know only how to handle input types ", join(",", @supported_inputs), " not '$type'!\n" unless (grep(/$type/, @supported_inputs)); |
die "I know only how to handle input types ", join(",", $config->webpac('inputs') ), " not '$type'!\n" unless (grep(/$type/, $config->webpac('inputs'))); |
316 |
|
|
317 |
my $lookup = new WebPAC::Lookup( |
my $input_module = $config->webpac('inputs')->{$type}; |
318 |
lookup_file => $input->{lookup}, |
|
319 |
|
$log->info("working on input '$input->{name}' in $input->{path} [type: $input->{type}] using $input_module", |
320 |
|
$input->{lookup} ? "lookup '$input->{lookup}'" : "" |
321 |
); |
); |
322 |
|
|
323 |
my $input_module = $config->{webpac}->{inputs}->{$type}; |
if ($stats) { |
324 |
|
# disable modification of records if --stats is in use |
325 |
|
delete($input->{modify_records}); |
326 |
|
delete($input->{modify_file}); |
327 |
|
} |
328 |
|
|
329 |
|
warn "parser->depends = ", dump( $parser->{depends} ); |
330 |
|
warn "depends on: ", dump( $parser->depends($database, $input->{name}) ); |
331 |
|
warn "lookup_create_rules = ", dump( $parser->lookup_create_rules($database, $input->{name}) ); |
332 |
|
warn "parser->_lookup_create = ", dump( $parser->{_lookup_create} ); |
333 |
|
|
334 |
$log->info("working on input '$input->{name}' in $input->{path} [type: $input->{type}] using $input_module lookup '$input->{lookup}'"); |
my $lookup; |
335 |
|
|
336 |
my $input_db = new WebPAC::Input( |
my $input_db = new WebPAC::Input( |
337 |
module => $input_module, |
module => $input_module, |
338 |
code_page => $config->{webpac}->{webpac_encoding}, |
encoding => $config->webpac('webpac_encoding'), |
339 |
limit => $limit || $input->{limit}, |
limit => $limit || $input->{limit}, |
340 |
offset => $offset, |
offset => $offset, |
341 |
lookup => $lookup, |
lookup_coderef => sub { |
342 |
|
my $rec = shift || return; |
343 |
|
$lookup->add( $rec ); |
344 |
|
}, |
345 |
recode => $input->{recode}, |
recode => $input->{recode}, |
346 |
stats => $stats, |
stats => $stats, |
347 |
|
modify_records => $input->{modify_records}, |
348 |
|
modify_file => $input->{modify_file}, |
349 |
); |
); |
350 |
$log->logdie("can't create input using $input_module") unless ($input); |
$log->logdie("can't create input using $input_module") unless ($input); |
351 |
|
|
352 |
|
if (defined( $input->{lookup} )) { |
353 |
|
$log->warn("$database/", $input->{name}, " has depriciated lookup definition, removing it..."); |
354 |
|
delete( $input->{lookup} ); |
355 |
|
} |
356 |
|
|
357 |
my $maxmfn = $input_db->open( |
my $maxmfn = $input_db->open( |
358 |
path => $input->{path}, |
path => $input->{path}, |
359 |
code_page => $input->{encoding}, # database encoding |
code_page => $input->{encoding}, # database encoding |
360 |
|
%{ $input }, |
361 |
); |
); |
362 |
|
|
363 |
my $n = new WebPAC::Normalize::XML( |
my $report_fh; |
364 |
# filter => { 'foo' => sub { shift } }, |
if ($stats || $validate) { |
365 |
db => $db, |
my $path = "out/report/" . $database . '-' . $input->{name} . '.txt'; |
366 |
lookup_regex => $lookup->regex, |
open($report_fh, '>', $path) || $log->logdie("can't open $path: $!"); |
367 |
lookup => $lookup, |
|
368 |
prefix => $input->{name}, |
print $report_fh "Report for database '$database' input '$input->{name}' records ", |
369 |
); |
$offset || 1, "-", $limit || $input->{limit} || $maxmfn, "\n\n"; |
370 |
|
$log->info("Generating report file $path"); |
371 |
|
} |
372 |
|
|
373 |
my $rules; |
my @norm_array = ref($input->{normalize}) eq 'ARRAY' ? |
374 |
my $normalize_path = $input->{normalize}->{path}; |
@{ $input->{normalize} } : ( $input->{normalize} ); |
375 |
|
|
376 |
if ($force_set) { |
if ($marc_normalize) { |
377 |
my $new_norm_path = $normalize_path; |
@norm_array = ( { |
378 |
$new_norm_path =~ s/\.xml$/.pl/; |
path => $marc_normalize, |
379 |
if (-e $new_norm_path) { |
output => $marc_output || 'out/marc/' . $database . '-' . $input->{name} . '.marc', |
380 |
$log->debug("--force-set replaced $normalize_path with $new_norm_path"); |
} ); |
|
$normalize_path = $new_norm_path; |
|
|
} else { |
|
|
$log->debug("--force-set failed on $new_norm_path, fallback to $normalize_path"); |
|
|
} |
|
381 |
} |
} |
382 |
|
|
383 |
if ($normalize_path =~ m/\.xml$/i) { |
foreach my $normalize (@norm_array) { |
|
$n->open( |
|
|
tag => $input->{normalize}->{tag}, |
|
|
xml_file => $normalize_path, |
|
|
); |
|
|
} elsif ($normalize_path =~ m/\.(?:yml|yaml)$/i) { |
|
|
$n->open_yaml( |
|
|
path => $normalize_path, |
|
|
tag => $input->{normalize}->{tag}, |
|
|
); |
|
|
} elsif ($normalize_path =~ m/\.(?:pl)$/i) { |
|
|
$n = undef; |
|
|
$log->info("using WebPAC::Normalize::Set to process $normalize_path"); |
|
|
$rules = read_file( $normalize_path ) or die "can't open $normalize_path: $!"; |
|
|
} |
|
384 |
|
|
385 |
foreach my $pos ( 0 ... $input_db->size ) { |
my $normalize_path = $normalize->{path} || $log->logdie("can't find normalize path in config"); |
386 |
|
|
387 |
my $row = $input_db->fetch || next; |
$log->logdie("Found '$normalize_path' as normalization file which isn't supported any more!") unless ( $normalize_path =~ m!\.pl$!i ); |
388 |
|
|
389 |
my $mfn = $row->{'000'}->[0]; |
my $rules = read_file( $normalize_path ) or die "can't open $normalize_path: $!"; |
390 |
|
|
391 |
if (! $mfn || $mfn !~ m#^\d+$#) { |
$log->info("Using $normalize_path for normalization..."); |
|
$log->warn("record $pos doesn't have valid MFN but '$mfn', using $pos"); |
|
|
$mfn = $pos; |
|
|
push @{ $row->{'000'} }, $pos; |
|
|
} |
|
392 |
|
|
393 |
|
my $marc = new WebPAC::Output::MARC( |
394 |
|
path => $normalize->{output}, |
395 |
|
lint => $marc_lint, |
396 |
|
dump => $marc_dump, |
397 |
|
) if ($normalize->{output}); |
398 |
|
|
399 |
if ($validate) { |
# reset position in database |
400 |
my @errors = $validate->validate_errors( $row ); |
$input_db->seek(1); |
|
$log->error( "MFN $mfn validation errors:\n", join("\n", @errors) ) if (@errors); |
|
|
} |
|
401 |
|
|
402 |
|
# generate name of config key for indexer (strip everything after -) |
403 |
my $ds; |
my $indexer_config = $use_indexer; |
404 |
if ($n) { |
$indexer_config =~ s/^(\w+)-?.*$/$1/g if ($indexer_config); |
405 |
$ds = $n->data_structure($row); |
|
406 |
} else { |
foreach my $pos ( 0 ... $input_db->size ) { |
407 |
$ds = WebPAC::Normalize::Set::data_structure( |
|
408 |
|
my $row = $input_db->fetch || next; |
409 |
|
|
410 |
|
my $mfn = $row->{'000'}->[0]; |
411 |
|
|
412 |
|
if (! $mfn || $mfn !~ m#^\d+$#) { |
413 |
|
$log->warn("record $pos doesn't have valid MFN but '$mfn', using $pos"); |
414 |
|
$mfn = $pos; |
415 |
|
push @{ $row->{'000'} }, $pos; |
416 |
|
} |
417 |
|
|
418 |
|
|
419 |
|
if ($validate) { |
420 |
|
if ( my $errors = $validate->validate_errors( $row, $input_db->dump ) ) { |
421 |
|
$log->error( "MFN $mfn validation error:\n", |
422 |
|
$validate->report_error( $errors ) |
423 |
|
); |
424 |
|
} |
425 |
|
} |
426 |
|
|
427 |
|
my $ds_config = dclone($db_config); |
428 |
|
|
429 |
|
# default values -> database key |
430 |
|
$ds_config->{_} = $database; |
431 |
|
|
432 |
|
# current mfn |
433 |
|
$ds_config->{_mfn} = $mfn; |
434 |
|
|
435 |
|
# attach current input |
436 |
|
$ds_config->{input} = $input; |
437 |
|
|
438 |
|
my $ds = WebPAC::Normalize::data_structure( |
439 |
row => $row, |
row => $row, |
440 |
rules => $rules, |
rules => $rules, |
441 |
lookup => $lookup->lookup_hash, |
lookup => $lookup ? $lookup->lookup_hash : undef, |
442 |
|
config => $ds_config, |
443 |
|
marc_encoding => 'utf-8', |
444 |
); |
); |
445 |
|
|
446 |
$db->save_ds( |
$db->save_ds( |
448 |
ds => $ds, |
ds => $ds, |
449 |
prefix => $input->{name}, |
prefix => $input->{name}, |
450 |
) if ($ds && !$stats); |
) if ($ds && !$stats); |
451 |
|
|
452 |
|
$indexer->add( |
453 |
|
id => $input->{name} . "/" . $mfn, |
454 |
|
ds => $ds, |
455 |
|
type => $config->get($indexer_config)->{type}, |
456 |
|
) if ($indexer && $ds); |
457 |
|
|
458 |
|
if ($marc) { |
459 |
|
my $i = 0; |
460 |
|
|
461 |
|
while (my $fields = WebPAC::Normalize::_get_marc_fields( fetch_next => 1 ) ) { |
462 |
|
$marc->add( |
463 |
|
id => $mfn . ( $i ? "/$i" : '' ), |
464 |
|
fields => $fields, |
465 |
|
leader => WebPAC::Normalize::marc_leader(), |
466 |
|
row => $row, |
467 |
|
); |
468 |
|
$i++; |
469 |
|
} |
470 |
|
|
471 |
|
$log->info("Created $i instances of MFN $mfn\n") if ($i > 1); |
472 |
|
} |
473 |
|
|
474 |
|
$total_rows++; |
475 |
} |
} |
476 |
|
|
477 |
$indexer->add( |
if ($validate) { |
478 |
id => $input->{name} . "/" . $mfn, |
my $errors = $validate->report; |
479 |
ds => $ds, |
if ($errors) { |
480 |
type => $config->{$use_indexer}->{type}, |
$log->info("validation errors:\n$errors\n" ); |
481 |
) if ($indexer); |
print $report_fh "$errors\n" if ($report_fh); |
482 |
|
} |
483 |
|
} |
484 |
|
|
485 |
$total_rows++; |
if ($stats) { |
486 |
} |
my $s = $input_db->stats; |
487 |
|
$log->info("statistics of fields usage:\n$s"); |
488 |
|
print $report_fh "Statistics of fields usage:\n$s" if ($report_fh); |
489 |
|
} |
490 |
|
|
491 |
|
# close MARC file |
492 |
|
$marc->finish if ($marc); |
493 |
|
|
494 |
$log->info("statistics of fields usage:\n", $input_db->stats) if ($stats); |
# close report |
495 |
|
close($report_fh) if ($report_fh) |
496 |
|
} |
497 |
|
|
498 |
}; |
} |
499 |
|
|
500 |
eval { $indexer->finish } if ($indexer && $indexer->can('finish')); |
eval { $indexer->finish } if ($indexer && $indexer->can('finish')); |
501 |
|
|
506 |
) |
) |
507 |
); |
); |
508 |
|
|
509 |
# |
|
510 |
# add Hyper Estraier links to other databases |
# end forked process |
511 |
# |
if ($parallel) { |
512 |
if (ref($db_config->{links}) eq 'ARRAY') { |
$log->info("parallel process $$ finished"); |
513 |
foreach my $link (@{ $db_config->{links} }) { |
exit(0); |
|
if ($use_indexer eq 'hyperestraier') { |
|
|
$log->info("adding link $database -> $link->{to} [$link->{credit}]"); |
|
|
$indexer->add_link( |
|
|
from => $database, |
|
|
to => $link->{to}, |
|
|
credit => $link->{credit}, |
|
|
); |
|
|
} else { |
|
|
$log->warn("NOT IMPLEMENTED WITH $use_indexer: adding link $database -> $link->{to} [$link->{credit}]"); |
|
|
} |
|
|
} |
|
514 |
} |
} |
515 |
|
|
516 |
} |
} |
517 |
|
|
518 |
|
if ($parallel) { |
519 |
|
# wait all children to finish |
520 |
|
sleep(1) while wait != -1; |
521 |
|
$log->info("all parallel processes finished"); |
522 |
|
} |
523 |
|
|
524 |
|
# |
525 |
|
# handle links or merge after indexing |
526 |
|
# |
527 |
|
|
528 |
|
if ($merge) { |
529 |
|
print $estcmd_fh 'sudo /etc/init.d/hyperestraier start',$/; |
530 |
|
close($estcmd_fh); |
531 |
|
chmod 0700, $estcmd_path || $log->warn("can't chmod 0700 $estcmd_path: $!"); |
532 |
|
system $estcmd_path; |
533 |
|
} else { |
534 |
|
foreach my $link (@links) { |
535 |
|
$log->logdie("coderef in link ", Dumper($link), " is ", ref($link), " and not CODE") unless (ref($link) eq 'CODE'); |
536 |
|
$link->(); |
537 |
|
} |
538 |
|
} |