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::Input 0.11; |
12 |
use WebPAC::Store 0.03; |
use WebPAC::Store 0.03; |
13 |
use WebPAC::Normalize::XML; |
use WebPAC::Normalize 0.11; |
|
use WebPAC::Normalize::Set; |
|
14 |
use WebPAC::Output::TT; |
use WebPAC::Output::TT; |
15 |
use YAML qw/LoadFile/; |
use WebPAC::Validate 0.06; |
16 |
|
use WebPAC::Output::MARC; |
17 |
|
use WebPAC::Config; |
18 |
use Getopt::Long; |
use Getopt::Long; |
19 |
use File::Path; |
use File::Path; |
20 |
use Time::HiRes qw/time/; |
use Time::HiRes qw/time/; |
21 |
use File::Slurp; |
use File::Slurp; |
22 |
|
use Data::Dump qw/dump/; |
23 |
|
use Storable qw/dclone/; |
24 |
|
|
25 |
|
use Proc::Queue size => 1; |
26 |
|
use POSIX ":sys_wait_h"; # imports WNOHANG |
27 |
|
|
28 |
=head1 NAME |
=head1 NAME |
29 |
|
|
47 |
|
|
48 |
remove database and Hyper Estraier index before indexing |
remove database and Hyper Estraier index before indexing |
49 |
|
|
50 |
=item --only=database_name |
=item --only=database_name/input_filter |
51 |
|
|
52 |
reindex just single database (legacy name is --one) |
reindex just single database (legacy name is --one) |
53 |
|
|
54 |
|
C</input_filter> is optional part which can be C<name> |
55 |
|
or C<type> from input |
56 |
|
|
57 |
=item --config conf/config.yml |
=item --config conf/config.yml |
58 |
|
|
59 |
path to YAML configuration file |
path to YAML configuration file |
60 |
|
|
61 |
=item --force-set |
=item --stats |
62 |
|
|
63 |
force conversion C<normalize->path> in C<config.yml> from |
disable indexing, modify_* in configuration and dump statistics about field |
64 |
C<.xml> to C<.pl> |
and subfield usage for each input |
65 |
|
|
66 |
=item --stats |
=item --validate path/to/validation_file |
67 |
|
|
68 |
|
turn on extra validation of imput records, see L<WebPAC::Validation> |
69 |
|
|
70 |
|
=item --marc-normalize conf/normalize/mapping.pl |
71 |
|
|
72 |
|
This option specifies normalisation file for MARC creation |
73 |
|
|
74 |
|
=item --marc-output out/marc/test.marc |
75 |
|
|
76 |
|
Optional path to output file |
77 |
|
|
78 |
|
=item --marc-lint |
79 |
|
|
80 |
|
By default turned on if C<--marc-normalize> is used. You can disable lint |
81 |
|
messages with C<--no-marc-lint>. |
82 |
|
|
83 |
|
=item --marc-dump |
84 |
|
|
85 |
|
Force dump or input and marc record for debugging. |
86 |
|
|
87 |
|
=item --parallel 4 |
88 |
|
|
89 |
dump statistics about used fields and subfields in each input |
Run databases in parallel (aproximatly same as number of processors in |
90 |
|
machine if you want to use full load) |
91 |
|
|
92 |
|
=item --only-links |
93 |
|
|
94 |
|
Create just links |
95 |
|
|
96 |
|
=item --merge |
97 |
|
|
98 |
|
Create merged index of databases which have links |
99 |
|
|
100 |
=back |
=back |
101 |
|
|
105 |
my $limit; |
my $limit; |
106 |
|
|
107 |
my $clean = 0; |
my $clean = 0; |
108 |
my $config = 'conf/config.yml'; |
my $config_path; |
109 |
my $debug = 0; |
my $debug = 0; |
110 |
my $only_db_name; |
my $only_filter; |
|
my $force_set = 0; |
|
111 |
my $stats = 0; |
my $stats = 0; |
112 |
|
my $validate_path; |
113 |
|
my ($marc_normalize, $marc_output); |
114 |
|
my $marc_lint = 1; |
115 |
|
my $marc_dump = 0; |
116 |
|
my $parallel = 0; |
117 |
|
my $only_links = 0; |
118 |
|
my $merge = 0; |
119 |
|
|
120 |
|
my $log = _new WebPAC::Common()->_get_logger(); |
121 |
|
|
122 |
GetOptions( |
GetOptions( |
123 |
"limit=i" => \$limit, |
"limit=i" => \$limit, |
124 |
"offset=i" => \$offset, |
"offset=i" => \$offset, |
125 |
"clean" => \$clean, |
"clean" => \$clean, |
126 |
"one=s" => \$only_db_name, |
"one=s" => \$only_filter, |
127 |
"only=s" => \$only_db_name, |
"only=s" => \$only_filter, |
128 |
"config" => \$config, |
"config" => \$config_path, |
129 |
"debug" => \$debug, |
"debug+" => \$debug, |
|
"force-set" => \$force_set, |
|
130 |
"stats" => \$stats, |
"stats" => \$stats, |
131 |
|
"validate=s" => \$validate_path, |
132 |
|
"marc-normalize=s" => \$marc_normalize, |
133 |
|
"marc-output=s" => \$marc_output, |
134 |
|
"marc-lint!" => \$marc_lint, |
135 |
|
"marc-dump!" => \$marc_dump, |
136 |
|
"parallel=i" => \$parallel, |
137 |
|
"only-links!" => \$only_links, |
138 |
|
"merge" => \$merge, |
139 |
); |
); |
140 |
|
|
141 |
$config = LoadFile($config); |
my $config = new WebPAC::Config( path => $config_path ); |
142 |
|
|
143 |
print "config = ",Dumper($config) if ($debug); |
#print "config = ",dump($config) if ($debug); |
144 |
|
|
145 |
die "no databases in config file!\n" unless ($config->{databases}); |
die "no databases in config file!\n" unless ($config->databases); |
146 |
|
|
147 |
my $log = _new WebPAC::Common()->_get_logger(); |
$log->info( "-" x 79 ); |
148 |
|
|
149 |
my $use_indexer = $config->{use_indexer} || 'hyperestraier'; |
|
150 |
$log->info("using $use_indexer indexing engine..."); |
my $estcmd_fh; |
151 |
|
my $estcmd_path = './estcmd-merge.sh'; |
152 |
|
if ($merge) { |
153 |
|
open($estcmd_fh, '>', $estcmd_path) || $log->logdie("can't open $estcmd_path: $!"); |
154 |
|
print $estcmd_fh 'cd /data/estraier/_node/ || exit 1',$/; |
155 |
|
print $estcmd_fh 'sudo /etc/init.d/hyperestraier stop',$/; |
156 |
|
$log->info("created merge batch file $estcmd_path"); |
157 |
|
} |
158 |
|
|
159 |
|
|
160 |
|
my $validate; |
161 |
|
$validate = new WebPAC::Validate( |
162 |
|
path => $validate_path, |
163 |
|
) if ($validate_path); |
164 |
|
|
165 |
|
|
166 |
|
my $use_indexer = $config->use_indexer; |
167 |
|
if ($stats) { |
168 |
|
$log->debug("option --stats disables update of indexing engine..."); |
169 |
|
$use_indexer = undef; |
170 |
|
} else { |
171 |
|
$log->info("using $use_indexer indexing engine..."); |
172 |
|
} |
173 |
|
|
174 |
|
# disable indexing when creating marc |
175 |
|
$use_indexer = undef if ($marc_normalize); |
176 |
|
|
177 |
|
# parse normalize files and create source files for lookup and normalization |
178 |
|
|
179 |
|
my $parser = new WebPAC::Parser( config => $config ); |
180 |
|
|
181 |
my $total_rows = 0; |
my $total_rows = 0; |
182 |
my $start_t = time(); |
my $start_t = time(); |
183 |
|
|
184 |
while (my ($database, $db_config) = each %{ $config->{databases} }) { |
my @links; |
185 |
|
|
186 |
|
if ($parallel) { |
187 |
|
$log->info("Using $parallel processes for speedup"); |
188 |
|
Proc::Queue::size($parallel); |
189 |
|
} |
190 |
|
|
191 |
|
while (my ($database, $db_config) = each %{ $config->databases }) { |
192 |
|
|
193 |
|
my ($only_database,$only_input) = split(m#/#, $only_filter) if ($only_filter); |
194 |
|
next if ($only_database && $database !~ m/$only_database/i); |
195 |
|
|
196 |
next if ($only_db_name && $database !~ m/$only_db_name/i); |
if ($parallel) { |
197 |
|
my $f=fork; |
198 |
|
if(defined ($f) and $f==0) { |
199 |
|
$log->info("Created processes $$ for speedup"); |
200 |
|
} else { |
201 |
|
next; |
202 |
|
} |
203 |
|
} |
204 |
|
|
205 |
my $indexer; |
my $indexer; |
206 |
|
if ($use_indexer) { |
207 |
|
|
208 |
my $indexer_config = $config->{$use_indexer} || $log->logdie("can't find '$use_indexer' part in confguration"); |
my $cfg_name = $use_indexer; |
209 |
$indexer_config->{database} = $database; |
$cfg_name =~ s/\-.*$//; |
|
$indexer_config->{clean} = $clean; |
|
|
$indexer_config->{label} = $db_config->{name}; |
|
|
|
|
|
if ($use_indexer eq 'hyperestraier') { |
|
|
|
|
|
# open Hyper Estraier database |
|
|
use WebPAC::Output::Estraier '0.10'; |
|
|
$indexer = new WebPAC::Output::Estraier( %{ $indexer_config } ); |
|
|
|
|
|
} elsif ($use_indexer eq 'kinosearch') { |
|
|
|
|
|
# open KinoSearch |
|
|
use WebPAC::Output::KinoSearch; |
|
|
$indexer_config->{clean} = 1 unless (-e $indexer_config->{index_path}); |
|
|
$indexer = new WebPAC::Output::KinoSearch( %{ $indexer_config } ); |
|
210 |
|
|
211 |
} else { |
my $indexer_config = $config->get( $cfg_name ) || $log->logdie("can't find '$cfg_name' part in confguration"); |
212 |
$log->logdie("unknown use_indexer: $use_indexer"); |
$indexer_config->{database} = $database; |
213 |
|
$indexer_config->{clean} = $clean; |
214 |
|
$indexer_config->{label} = $db_config->{name}; |
215 |
|
|
216 |
|
# force clean if database has links |
217 |
|
$indexer_config->{clean} = 1 if ($db_config->{links}); |
218 |
|
|
219 |
|
if ($use_indexer eq 'hyperestraier') { |
220 |
|
|
221 |
|
# open Hyper Estraier database |
222 |
|
use WebPAC::Output::Estraier '0.10'; |
223 |
|
$indexer = new WebPAC::Output::Estraier( %{ $indexer_config } ); |
224 |
|
|
225 |
|
} elsif ($use_indexer eq 'hyperestraier-native') { |
226 |
|
|
227 |
|
# open Hyper Estraier database |
228 |
|
use WebPAC::Output::EstraierNative; |
229 |
|
$indexer = new WebPAC::Output::EstraierNative( %{ $indexer_config } ); |
230 |
|
|
231 |
|
} elsif ($use_indexer eq 'kinosearch') { |
232 |
|
|
233 |
|
# open KinoSearch |
234 |
|
use WebPAC::Output::KinoSearch; |
235 |
|
$indexer_config->{clean} = 1 unless (-e $indexer_config->{index_path}); |
236 |
|
$indexer = new WebPAC::Output::KinoSearch( %{ $indexer_config } ); |
237 |
|
|
238 |
|
} else { |
239 |
|
$log->logdie("unknown use_indexer: $use_indexer"); |
240 |
|
} |
241 |
|
|
242 |
|
$log->logide("can't continue without valid indexer") unless ($indexer); |
243 |
|
} |
244 |
|
|
245 |
|
|
246 |
|
# |
247 |
|
# store Hyper Estraier links to other databases |
248 |
|
# |
249 |
|
if (ref($db_config->{links}) eq 'ARRAY' && $use_indexer) { |
250 |
|
foreach my $link (@{ $db_config->{links} }) { |
251 |
|
if ($use_indexer eq 'hyperestraier') { |
252 |
|
if ($merge) { |
253 |
|
print $estcmd_fh 'sudo -u www-data estcmd merge ' . $database . ' ' . $link->{to},$/; |
254 |
|
} else { |
255 |
|
$log->info("saving link $database -> $link->{to} [$link->{credit}]"); |
256 |
|
push @links, sub { |
257 |
|
$log->info("adding link $database -> $link->{to} [$link->{credit}]"); |
258 |
|
$indexer->add_link( |
259 |
|
from => $database, |
260 |
|
to => $link->{to}, |
261 |
|
credit => $link->{credit}, |
262 |
|
); |
263 |
|
}; |
264 |
|
} |
265 |
|
} else { |
266 |
|
$log->warn("NOT IMPLEMENTED WITH $use_indexer: adding link $database -> $link->{to} [$link->{credit}]"); |
267 |
|
} |
268 |
|
} |
269 |
} |
} |
270 |
|
next if ($only_links); |
271 |
|
|
|
$log->logide("can't continue without valid indexer") unless ($indexer); |
|
272 |
|
|
273 |
# |
# |
274 |
# now WebPAC::Store |
# now WebPAC::Store |
276 |
my $abs_path = abs_path($0); |
my $abs_path = abs_path($0); |
277 |
$abs_path =~ s#/[^/]*$#/#; |
$abs_path =~ s#/[^/]*$#/#; |
278 |
|
|
279 |
my $db_path = $config->{webpac}->{db_path} . '/' . $database; |
my $db_path = $config->get('webpac')->{db_path} . '/' . $database; |
280 |
|
|
281 |
if ($clean) { |
if ($clean) { |
282 |
$log->info("creating new database $database in $db_path"); |
$log->info("creating new database '$database' in $db_path"); |
283 |
rmtree( $db_path ) || $log->warn("can't remove $db_path: $!"); |
rmtree( $db_path ) || $log->warn("can't remove $db_path: $!"); |
284 |
} else { |
} else { |
285 |
$log->debug("working on $database in $db_path"); |
$log->info("working on database '$database' in $db_path"); |
286 |
} |
} |
287 |
|
|
288 |
my $db = new WebPAC::Store( |
my $db = new WebPAC::Store( |
305 |
$log->info("database $database doesn't have inputs defined"); |
$log->info("database $database doesn't have inputs defined"); |
306 |
} |
} |
307 |
|
|
|
my @supported_inputs = keys %{ $config->{webpac}->{inputs} }; |
|
|
|
|
308 |
foreach my $input (@inputs) { |
foreach my $input (@inputs) { |
309 |
|
|
310 |
|
next if ($only_input && ($input->{name} !~ m#$only_input#i && $input->{type} !~ m#$only_input#i)); |
311 |
|
|
312 |
my $type = lc($input->{type}); |
my $type = lc($input->{type}); |
313 |
|
|
314 |
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'))); |
315 |
|
|
316 |
|
my $input_module = $config->webpac('inputs')->{$type}; |
317 |
|
|
318 |
|
my @lookups = $parser->have_lookup_create($database, $input); |
319 |
|
|
320 |
my $lookup = new WebPAC::Lookup( |
$log->info("working on input '$input->{name}' in $input->{path} [type: $input->{type}] using $input_module", |
321 |
lookup_file => $input->{lookup}, |
@lookups ? "lookup ".join(", ", @lookups) : "" |
322 |
); |
); |
323 |
|
|
324 |
my $input_module = $config->{webpac}->{inputs}->{$type}; |
warn "lookups = ", dump( @lookups ); |
325 |
|
|
326 |
$log->info("working on input '$input->{path}' [$input->{type}] using $input_module lookup '$input->{lookup}'"); |
if ($stats) { |
327 |
|
# disable modification of records if --stats is in use |
328 |
|
delete($input->{modify_records}); |
329 |
|
delete($input->{modify_file}); |
330 |
|
} |
331 |
|
|
332 |
|
warn "parser->depends = ", dump( $parser->{depends} ); |
333 |
|
warn "depends on: ", dump( $parser->depends($database, $input->{name}) ); |
334 |
|
warn "lookup_create_rules = ", dump( $parser->lookup_create_rules($database, $input->{name}) ); |
335 |
|
warn "parser->_lookup_create = ", dump( $parser->{_lookup_create} ); |
336 |
|
|
337 |
|
my $lookup; |
338 |
|
|
339 |
my $input_db = new WebPAC::Input( |
my $input_db = new WebPAC::Input( |
340 |
module => $input_module, |
module => $input_module, |
341 |
code_page => $config->{webpac}->{webpac_encoding}, |
encoding => $config->webpac('webpac_encoding'), |
342 |
limit => $limit || $input->{limit}, |
limit => $limit || $input->{limit}, |
343 |
offset => $offset, |
offset => $offset, |
344 |
lookup => $lookup, |
lookup_coderef => sub { |
345 |
|
my $rec = shift || return; |
346 |
|
$lookup->add( $rec ); |
347 |
|
}, |
348 |
recode => $input->{recode}, |
recode => $input->{recode}, |
349 |
stats => $stats, |
stats => $stats, |
350 |
|
modify_records => $input->{modify_records}, |
351 |
|
modify_file => $input->{modify_file}, |
352 |
); |
); |
353 |
$log->logdie("can't create input using $input_module") unless ($input); |
$log->logdie("can't create input using $input_module") unless ($input); |
354 |
|
|
355 |
|
if (defined( $input->{lookup} )) { |
356 |
|
$log->warn("$database/", $input->{name}, " has depriciated lookup definition, removing it..."); |
357 |
|
delete( $input->{lookup} ); |
358 |
|
} |
359 |
|
|
360 |
my $maxmfn = $input_db->open( |
my $maxmfn = $input_db->open( |
361 |
path => $input->{path}, |
path => $input->{path}, |
362 |
code_page => $input->{encoding}, # database encoding |
code_page => $input->{encoding}, # database encoding |
363 |
|
%{ $input }, |
364 |
); |
); |
365 |
|
|
366 |
my $n = new WebPAC::Normalize::XML( |
my $report_fh; |
367 |
# filter => { 'foo' => sub { shift } }, |
if ($stats || $validate) { |
368 |
db => $db, |
my $path = "out/report/" . $database . '-' . $input->{name} . '.txt'; |
369 |
lookup_regex => $lookup->regex, |
open($report_fh, '>', $path) || $log->logdie("can't open $path: $!"); |
370 |
lookup => $lookup, |
|
371 |
prefix => $input->{name}, |
print $report_fh "Report for database '$database' input '$input->{name}' records ", |
372 |
); |
$offset || 1, "-", $limit || $input->{limit} || $maxmfn, "\n\n"; |
373 |
|
$log->info("Generating report file $path"); |
374 |
|
} |
375 |
|
|
376 |
my $rules; |
my @norm_array = ref($input->{normalize}) eq 'ARRAY' ? |
377 |
my $normalize_path = $input->{normalize}->{path}; |
@{ $input->{normalize} } : ( $input->{normalize} ); |
378 |
|
|
379 |
if ($force_set) { |
if ($marc_normalize) { |
380 |
my $new_norm_path = $normalize_path; |
@norm_array = ( { |
381 |
$new_norm_path =~ s/\.xml$/.pl/; |
path => $marc_normalize, |
382 |
if (-e $new_norm_path) { |
output => $marc_output || 'out/marc/' . $database . '-' . $input->{name} . '.marc', |
383 |
$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"); |
|
|
} |
|
384 |
} |
} |
385 |
|
|
386 |
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: $!"; |
|
|
} |
|
387 |
|
|
388 |
foreach my $pos ( 0 ... $input_db->size ) { |
my $normalize_path = $normalize->{path} || $log->logdie("can't find normalize path in config"); |
389 |
|
|
390 |
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 ); |
391 |
|
|
392 |
my $mfn = $row->{'000'}->[0]; |
my $rules = read_file( $normalize_path ) or die "can't open $normalize_path: $!"; |
393 |
|
|
394 |
if (! $mfn || $mfn !~ m#^\d+$#) { |
$log->info("Using $normalize_path for normalization..."); |
395 |
$log->warn("record $pos doesn't have valid MFN but '$mfn', using $pos"); |
|
396 |
$mfn = $pos; |
my $marc = new WebPAC::Output::MARC( |
397 |
push @{ $row->{'000'} }, $pos; |
path => $normalize->{output}, |
398 |
} |
lint => $marc_lint, |
399 |
|
dump => $marc_dump, |
400 |
|
) if ($normalize->{output}); |
401 |
|
|
402 |
|
# reset position in database |
403 |
|
$input_db->seek(1); |
404 |
|
|
405 |
|
# generate name of config key for indexer (strip everything after -) |
406 |
|
my $indexer_config = $use_indexer; |
407 |
|
$indexer_config =~ s/^(\w+)-?.*$/$1/g if ($indexer_config); |
408 |
|
|
409 |
|
foreach my $pos ( 0 ... $input_db->size ) { |
410 |
|
|
411 |
|
my $row = $input_db->fetch || next; |
412 |
|
|
413 |
|
my $mfn = $row->{'000'}->[0]; |
414 |
|
|
415 |
|
if (! $mfn || $mfn !~ m#^\d+$#) { |
416 |
|
$log->warn("record $pos doesn't have valid MFN but '$mfn', using $pos"); |
417 |
|
$mfn = $pos; |
418 |
|
push @{ $row->{'000'} }, $pos; |
419 |
|
} |
420 |
|
|
421 |
|
|
422 |
|
if ($validate) { |
423 |
|
if ( my $errors = $validate->validate_errors( $row, $input_db->dump ) ) { |
424 |
|
$log->error( "MFN $mfn validation error:\n", |
425 |
|
$validate->report_error( $errors ) |
426 |
|
); |
427 |
|
} |
428 |
|
} |
429 |
|
|
430 |
|
my $ds_config = dclone($db_config); |
431 |
|
|
432 |
my $ds = $n ? $n->data_structure($row) : |
# default values -> database key |
433 |
WebPAC::Normalize::Set::data_structure( |
$ds_config->{_} = $database; |
434 |
|
|
435 |
|
# current mfn |
436 |
|
$ds_config->{_mfn} = $mfn; |
437 |
|
|
438 |
|
# attach current input |
439 |
|
$ds_config->{input} = $input; |
440 |
|
|
441 |
|
my $ds = WebPAC::Normalize::data_structure( |
442 |
row => $row, |
row => $row, |
443 |
rules => $rules, |
rules => $rules, |
444 |
lookup => $lookup->lookup_hash, |
lookup => $lookup ? $lookup->lookup_hash : undef, |
445 |
|
config => $ds_config, |
446 |
|
marc_encoding => 'utf-8', |
447 |
); |
); |
448 |
|
|
449 |
$indexer->add( |
$db->save_ds( |
450 |
id => $input->{name} . "/" . $mfn, |
id => $mfn, |
451 |
ds => $ds, |
ds => $ds, |
452 |
type => $config->{$use_indexer}->{type}, |
prefix => $input->{name}, |
453 |
); |
) if ($ds && !$stats); |
454 |
|
|
455 |
|
$indexer->add( |
456 |
|
id => $input->{name} . "/" . $mfn, |
457 |
|
ds => $ds, |
458 |
|
type => $config->get($indexer_config)->{type}, |
459 |
|
) if ($indexer && $ds); |
460 |
|
|
461 |
|
if ($marc) { |
462 |
|
my $i = 0; |
463 |
|
|
464 |
|
while (my $fields = WebPAC::Normalize::_get_marc_fields( fetch_next => 1 ) ) { |
465 |
|
$marc->add( |
466 |
|
id => $mfn . ( $i ? "/$i" : '' ), |
467 |
|
fields => $fields, |
468 |
|
leader => WebPAC::Normalize::marc_leader(), |
469 |
|
row => $row, |
470 |
|
); |
471 |
|
$i++; |
472 |
|
} |
473 |
|
|
474 |
$total_rows++; |
$log->info("Created $i instances of MFN $mfn\n") if ($i > 1); |
475 |
} |
} |
476 |
|
|
477 |
|
$total_rows++; |
478 |
|
} |
479 |
|
|
480 |
|
if ($validate) { |
481 |
|
my $errors = $validate->report; |
482 |
|
if ($errors) { |
483 |
|
$log->info("validation errors:\n$errors\n" ); |
484 |
|
print $report_fh "$errors\n" if ($report_fh); |
485 |
|
} |
486 |
|
} |
487 |
|
|
488 |
$log->info("statistics of fields usage:\n", $input_db->stats) if ($stats); |
if ($stats) { |
489 |
|
my $s = $input_db->stats; |
490 |
|
$log->info("statistics of fields usage:\n$s"); |
491 |
|
print $report_fh "Statistics of fields usage:\n$s" if ($report_fh); |
492 |
|
} |
493 |
|
|
494 |
}; |
# close MARC file |
495 |
|
$marc->finish if ($marc); |
496 |
|
|
497 |
eval { $indexer->finish } if ($indexer->can('finish')); |
# close report |
498 |
|
close($report_fh) if ($report_fh) |
499 |
|
} |
500 |
|
|
501 |
|
} |
502 |
|
|
503 |
|
eval { $indexer->finish } if ($indexer && $indexer->can('finish')); |
504 |
|
|
505 |
my $dt = time() - $start_t; |
my $dt = time() - $start_t; |
506 |
$log->info("$total_rows records indexed in " . |
$log->info("$total_rows records ", $indexer ? "indexed " : "", |
507 |
sprintf("%.2f sec [%.2f rec/sec]", |
sprintf("in %.2f sec [%.2f rec/sec]", |
508 |
$dt, ($total_rows / $dt) |
$dt, ($total_rows / $dt) |
509 |
) |
) |
510 |
); |
); |
511 |
|
|
512 |
# |
|
513 |
# add Hyper Estraier links to other databases |
# end forked process |
514 |
# |
if ($parallel) { |
515 |
if (ref($db_config->{links}) eq 'ARRAY') { |
$log->info("parallel process $$ finished"); |
516 |
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}]"); |
|
|
} |
|
|
} |
|
517 |
} |
} |
518 |
|
|
519 |
} |
} |
520 |
|
|
521 |
|
if ($parallel) { |
522 |
|
# wait all children to finish |
523 |
|
sleep(1) while wait != -1; |
524 |
|
$log->info("all parallel processes finished"); |
525 |
|
} |
526 |
|
|
527 |
|
# |
528 |
|
# handle links or merge after indexing |
529 |
|
# |
530 |
|
|
531 |
|
if ($merge) { |
532 |
|
print $estcmd_fh 'sudo /etc/init.d/hyperestraier start',$/; |
533 |
|
close($estcmd_fh); |
534 |
|
chmod 0700, $estcmd_path || $log->warn("can't chmod 0700 $estcmd_path: $!"); |
535 |
|
system $estcmd_path; |
536 |
|
} else { |
537 |
|
foreach my $link (@links) { |
538 |
|
$log->logdie("coderef in link ", Dumper($link), " is ", ref($link), " and not CODE") unless (ref($link) eq 'CODE'); |
539 |
|
$link->(); |
540 |
|
} |
541 |
|
} |