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 0.08; |
11 |
use WebPAC::Input 0.03; |
use WebPAC::Input 0.16; |
12 |
use WebPAC::Store 0.03; |
use WebPAC::Store 0.14; |
13 |
use WebPAC::Normalize; |
use WebPAC::Normalize 0.22; |
14 |
use WebPAC::Output::TT; |
use WebPAC::Output::TT; |
15 |
use WebPAC::Validate; |
use WebPAC::Validate 0.11; |
16 |
use YAML qw/LoadFile/; |
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 MARC::Record 2.0; # need 2.0 for utf-8 encoding see marcpm.sf.net |
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 |
=head1 NAME |
30 |
|
|
32 |
|
|
33 |
B<this command will probably go away. Don't get used to it!> |
B<this command will probably go away. Don't get used to it!> |
34 |
|
|
35 |
Options: |
=head1 OPTIONS |
36 |
|
|
37 |
=over 4 |
=over 4 |
38 |
|
|
61 |
|
|
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 |
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 |
This option specifies normalisation file for MARC creation |
Force dump or input and marc record for debugging. |
92 |
|
|
93 |
=item --marc-output out/marc/test.marc |
=item --parallel 4 |
94 |
|
|
95 |
Optional path to output file |
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 |
=back |
107 |
|
|
111 |
my $limit; |
my $limit; |
112 |
|
|
113 |
my $clean = 0; |
my $clean = 0; |
114 |
my $config = 'conf/config.yml'; |
my $config_path; |
115 |
my $debug = 0; |
my $debug = 0; |
116 |
my $only_filter; |
my $only_filter; |
117 |
my $stats = 0; |
my $stats = 0; |
118 |
my $validate_path; |
my $validate_path; |
119 |
my ($marc_normalize, $marc_output); |
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( |
GetOptions( |
131 |
"limit=i" => \$limit, |
"limit=i" => \$limit, |
133 |
"clean" => \$clean, |
"clean" => \$clean, |
134 |
"one=s" => \$only_filter, |
"one=s" => \$only_filter, |
135 |
"only=s" => \$only_filter, |
"only=s" => \$only_filter, |
136 |
"config" => \$config, |
"config" => \$config_path, |
137 |
"debug" => \$debug, |
"debug+" => \$debug, |
138 |
"stats" => \$stats, |
"stats" => \$stats, |
139 |
"validate=s" => \$validate_path, |
"validate=s" => \$validate_path, |
140 |
"marc-normalize=s" => \$marc_normalize, |
"validate-delimiters=s" => \$validate_delimiters_path, |
141 |
"marc-output=s" => \$marc_output, |
"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 |
$config = LoadFile($config); |
$marc_generate = 0 if ( $validate_delimiters_path ); |
151 |
|
|
152 |
print "config = ",Dumper($config) if ($debug); |
pod2usage(-verbose => 2) if ($help); |
153 |
|
|
154 |
die "no databases in config file!\n" unless ($config->{databases}); |
my $config = new WebPAC::Config( path => $config_path ); |
155 |
|
|
156 |
|
#print "config = ",dump($config) if ($debug); |
157 |
|
|
158 |
|
die "no databases in config file!\n" unless ($config->databases); |
159 |
|
|
|
my $log = _new WebPAC::Common()->_get_logger(); |
|
160 |
$log->info( "-" x 79 ); |
$log->info( "-" x 79 ); |
161 |
|
|
162 |
|
my $log_file = 'log'; |
163 |
|
|
164 |
|
if (-e $log_file ) { # && -s $log_file > 5 * 1024 * 1024) { |
165 |
|
$log->info("moved old log with ", -s $log_file, " bytes to '${log_file}.old'"); |
166 |
|
rename $log_file, "${log_file}.old" || $log->logwarn("can't rename $log_file to ${log_file}.old: $!"); |
167 |
|
} |
168 |
|
|
169 |
|
my $estcmd_fh; |
170 |
|
my $estcmd_path = './estcmd-merge.sh'; |
171 |
|
if ($merge) { |
172 |
|
open($estcmd_fh, '>', $estcmd_path) || $log->logdie("can't open $estcmd_path: $!"); |
173 |
|
print $estcmd_fh 'cd /data/estraier/_node/ || exit 1',$/; |
174 |
|
print $estcmd_fh 'sudo /etc/init.d/hyperestraier stop',$/; |
175 |
|
$log->info("created merge batch file $estcmd_path"); |
176 |
|
} |
177 |
|
|
178 |
my $validate; |
my $validate; |
179 |
$validate = new WebPAC::Validate( |
$validate = new WebPAC::Validate( |
180 |
path => $validate_path, |
delimiters => $config->webpac('delimiters'), |
181 |
) if ($validate_path); |
) if ($validate_path || $validate_delimiters_path); |
182 |
|
|
183 |
my $use_indexer = $config->{use_indexer} || 'hyperestraier'; |
my $use_indexer = $config->use_indexer; |
184 |
|
$stats ||= $validate; |
185 |
if ($stats) { |
if ($stats) { |
186 |
$log->debug("option --stats disables update of indexing engine..."); |
$log->debug("disabled indexing for stats collection"); |
187 |
$use_indexer = undef; |
$use_indexer = undef; |
188 |
} else { |
} elsif ( $use_indexer ) { |
189 |
$log->info("using $use_indexer indexing engine..."); |
$log->info("using $use_indexer indexing engine..."); |
190 |
} |
} |
191 |
|
|
192 |
# disable indexing when creating marc |
# parse normalize files and create source files for lookup and normalization |
193 |
$use_indexer = undef if ($marc_normalize); |
|
194 |
|
my $parser = new WebPAC::Parser( config => $config ); |
195 |
|
|
196 |
my $total_rows = 0; |
my $total_rows = 0; |
197 |
my $start_t = time(); |
my $start_t = time(); |
198 |
|
|
199 |
my @links; |
my @links; |
|
my $indexer; |
|
200 |
|
|
201 |
while (my ($database, $db_config) = each %{ $config->{databases} }) { |
if ($parallel) { |
202 |
|
$log->info("Using $parallel processes for speedup"); |
203 |
|
Proc::Queue::size($parallel); |
204 |
|
} |
205 |
|
|
206 |
|
sub create_ds_config { |
207 |
|
my ($db_config, $database, $input, $mfn) = @_; |
208 |
|
my $c = dclone( $db_config ); |
209 |
|
$c->{_} = $database || $log->logconfess("need database"); |
210 |
|
$c->{_mfn} = $mfn || $log->logconfess("need mfn"); |
211 |
|
$c->{input} = $input || $log->logconfess("need input"); |
212 |
|
return $c; |
213 |
|
} |
214 |
|
|
215 |
|
foreach my $database ( sort keys %{ $config->databases } ) { |
216 |
|
my $db_config = $config->databases->{$database}; |
217 |
|
|
218 |
my ($only_database,$only_input) = split(m#/#, $only_filter) if ($only_filter); |
my ($only_database,$only_input) = split(m#/#, $only_filter) if ($only_filter); |
219 |
next if ($only_database && $database !~ m/$only_database/i); |
next if ($only_database && $database !~ m/$only_database/i); |
220 |
|
|
221 |
if ($use_indexer) { |
if ($parallel) { |
222 |
my $indexer_config = $config->{$use_indexer} || $log->logdie("can't find '$use_indexer' part in confguration"); |
my $f=fork; |
223 |
|
if(defined ($f) and $f==0) { |
224 |
|
$log->info("Created processes $$ for speedup"); |
225 |
|
} else { |
226 |
|
next; |
227 |
|
} |
228 |
|
} |
229 |
|
|
230 |
|
my $indexer; |
231 |
|
if ($use_indexer && $parser->have_rules( 'search', $database )) { |
232 |
|
|
233 |
|
my $cfg_name = $use_indexer; |
234 |
|
$cfg_name =~ s/\-.*$//; |
235 |
|
|
236 |
|
my $indexer_config = $config->get( $cfg_name ) || $log->logdie("can't find '$cfg_name' part in confguration"); |
237 |
$indexer_config->{database} = $database; |
$indexer_config->{database} = $database; |
238 |
$indexer_config->{clean} = $clean; |
$indexer_config->{clean} = $clean; |
239 |
$indexer_config->{label} = $db_config->{name}; |
$indexer_config->{label} = $db_config->{name}; |
240 |
|
|
241 |
|
# force clean if database has links |
242 |
|
$indexer_config->{clean} = 1 if ($db_config->{links}); |
243 |
|
|
244 |
if ($use_indexer eq 'hyperestraier') { |
if ($use_indexer eq 'hyperestraier') { |
245 |
|
|
246 |
# open Hyper Estraier database |
# open Hyper Estraier database |
247 |
use WebPAC::Output::Estraier '0.10'; |
require WebPAC::Output::Estraier; |
248 |
$indexer = new WebPAC::Output::Estraier( %{ $indexer_config } ); |
$indexer = new WebPAC::Output::Estraier( %{ $indexer_config } ); |
249 |
|
|
250 |
|
} elsif ($use_indexer eq 'hyperestraier-native') { |
251 |
|
|
252 |
|
# open Hyper Estraier database |
253 |
|
require WebPAC::Output::EstraierNative; |
254 |
|
$indexer = new WebPAC::Output::EstraierNative( %{ $indexer_config } ); |
255 |
|
|
256 |
} elsif ($use_indexer eq 'kinosearch') { |
} elsif ($use_indexer eq 'kinosearch') { |
257 |
|
|
258 |
# open KinoSearch |
# open KinoSearch |
259 |
use WebPAC::Output::KinoSearch; |
require WebPAC::Output::KinoSearch; |
260 |
$indexer_config->{clean} = 1 unless (-e $indexer_config->{index_path}); |
$indexer_config->{clean} = 1 unless (-e $indexer_config->{index_path}); |
261 |
$indexer = new WebPAC::Output::KinoSearch( %{ $indexer_config } ); |
$indexer = new WebPAC::Output::KinoSearch( %{ $indexer_config } ); |
262 |
|
|
264 |
$log->logdie("unknown use_indexer: $use_indexer"); |
$log->logdie("unknown use_indexer: $use_indexer"); |
265 |
} |
} |
266 |
|
|
267 |
$log->logide("can't continue without valid indexer") unless ($indexer); |
$log->logdie("can't continue without valid indexer") unless ($indexer); |
268 |
} |
} |
269 |
|
|
270 |
|
|
271 |
# |
# |
272 |
|
# store Hyper Estraier links to other databases |
273 |
|
# |
274 |
|
if (ref($db_config->{links}) eq 'ARRAY' && $use_indexer) { |
275 |
|
foreach my $link (@{ $db_config->{links} }) { |
276 |
|
if ($use_indexer eq 'hyperestraier') { |
277 |
|
if ($merge) { |
278 |
|
print $estcmd_fh 'sudo -u www-data estcmd merge ' . $database . ' ' . $link->{to},$/; |
279 |
|
} else { |
280 |
|
$log->info("saving link $database -> $link->{to} [$link->{credit}]"); |
281 |
|
push @links, sub { |
282 |
|
$log->info("adding link $database -> $link->{to} [$link->{credit}]"); |
283 |
|
$indexer->add_link( |
284 |
|
from => $database, |
285 |
|
to => $link->{to}, |
286 |
|
credit => $link->{credit}, |
287 |
|
); |
288 |
|
}; |
289 |
|
} |
290 |
|
} else { |
291 |
|
$log->warn("NOT IMPLEMENTED WITH $use_indexer: adding link $database -> $link->{to} [$link->{credit}]"); |
292 |
|
} |
293 |
|
} |
294 |
|
} |
295 |
|
next if ($only_links); |
296 |
|
|
297 |
|
|
298 |
|
# |
299 |
# now WebPAC::Store |
# now WebPAC::Store |
300 |
# |
# |
301 |
my $abs_path = abs_path($0); |
my $abs_path = abs_path($0); |
302 |
$abs_path =~ s#/[^/]*$#/#; |
$abs_path =~ s#/[^/]*$#/#; # |
303 |
|
|
304 |
my $db_path = $config->{webpac}->{db_path} . '/' . $database; |
my $db_path = $config->webpac('db_path'); |
305 |
|
|
306 |
if ($clean) { |
if ($clean) { |
307 |
$log->info("creating new database '$database' in $db_path"); |
$log->info("creating new database '$database' in $db_path"); |
310 |
$log->info("working on database '$database' in $db_path"); |
$log->info("working on database '$database' in $db_path"); |
311 |
} |
} |
312 |
|
|
313 |
my $db = new WebPAC::Store( |
my $store = new WebPAC::Store( |
314 |
path => $db_path, |
path => $db_path, |
|
database => $database, |
|
315 |
debug => $debug, |
debug => $debug, |
316 |
); |
); |
317 |
|
|
318 |
|
# |
319 |
|
# prepare output |
320 |
|
# |
321 |
|
my @outputs; |
322 |
|
if (defined( $db_config->{output} )) { |
323 |
|
my $module = $db_config->{output}->{module} || $log->logdie("need module in output section of $database"); |
324 |
|
$module = 'WebPAC::Output::' . $module unless $module =~ m/::/; |
325 |
|
$log->debug("loading output module $module"); |
326 |
|
eval "require $module"; |
327 |
|
my $out = new $module->new( $db_config->{output} ); |
328 |
|
$out->init; |
329 |
|
push @outputs, $out; |
330 |
|
} |
331 |
|
|
332 |
|
|
333 |
# |
# |
334 |
# now, iterate through input formats |
# now, iterate through input formats |
343 |
$log->info("database $database doesn't have inputs defined"); |
$log->info("database $database doesn't have inputs defined"); |
344 |
} |
} |
345 |
|
|
|
my @supported_inputs = keys %{ $config->{webpac}->{inputs} }; |
|
|
|
|
346 |
foreach my $input (@inputs) { |
foreach my $input (@inputs) { |
347 |
|
|
348 |
next if ($only_input && ($input->{name} !~ m#$only_input#i && $input->{type} !~ m#$only_input#i)); |
my $input_name = $input->{name} || $log->logdie("input without a name isn't valid: ",dump($input)); |
349 |
|
|
350 |
|
next if ($only_input && ($input_name !~ m#$only_input#i && $input->{type} !~ m#$only_input#i)); |
351 |
|
|
352 |
my $type = lc($input->{type}); |
my $type = lc($input->{type}); |
353 |
|
|
354 |
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'))); |
355 |
|
|
356 |
my $lookup = new WebPAC::Lookup( |
my $input_module = $config->webpac('inputs')->{$type}; |
|
lookup_file => $input->{lookup}, |
|
|
) if ($input->{lookup}); |
|
357 |
|
|
358 |
my $input_module = $config->{webpac}->{inputs}->{$type}; |
my @lookups = $parser->have_lookup_create($database, $input); |
359 |
|
|
360 |
$log->info("working on input '$input->{name}' in $input->{path} [type: $input->{type}] using $input_module", |
$log->info("working on input '$input_name' in $input->{path} [type: $input->{type}] using $input_module", |
361 |
$input->{lookup} ? "lookup '$input->{lookup}'" : "" |
@lookups ? " creating lookups: ".join(", ", @lookups) : "" |
362 |
); |
); |
363 |
|
|
364 |
|
if ($stats) { |
365 |
|
# disable modification of records if --stats is in use |
366 |
|
delete($input->{modify_records}); |
367 |
|
delete($input->{modify_file}); |
368 |
|
} |
369 |
|
|
370 |
my $input_db = new WebPAC::Input( |
my $input_db = new WebPAC::Input( |
371 |
module => $input_module, |
module => $input_module, |
372 |
code_page => $config->{webpac}->{webpac_encoding}, |
encoding => $config->webpac('webpac_encoding'), |
373 |
limit => $limit || $input->{limit}, |
limit => $limit || $input->{limit}, |
374 |
offset => $offset, |
offset => $offset, |
|
lookup => $lookup, |
|
375 |
recode => $input->{recode}, |
recode => $input->{recode}, |
376 |
stats => $stats, |
stats => $stats, |
377 |
|
modify_records => $input->{modify_records}, |
378 |
|
modify_file => $input->{modify_file}, |
379 |
|
input_config => $input, |
380 |
); |
); |
381 |
$log->logdie("can't create input using $input_module") unless ($input); |
$log->logdie("can't create input using $input_module") unless ($input); |
382 |
|
|
383 |
|
if (defined( $input->{lookup} )) { |
384 |
|
$log->warn("$database/$input_name has depriciated lookup definition, removing it..."); |
385 |
|
delete( $input->{lookup} ); |
386 |
|
} |
387 |
|
|
388 |
|
my $lookup_coderef; |
389 |
|
|
390 |
|
if (@lookups) { |
391 |
|
|
392 |
|
my $rules = $parser->lookup_create_rules($database, $input) || $log->logdie("no rules found for $database/$input"); |
393 |
|
|
394 |
|
$lookup_coderef = sub { |
395 |
|
my $rec = shift || die "need rec!"; |
396 |
|
my $mfn = $rec->{'000'}->[0] || die "need mfn in 000"; |
397 |
|
|
398 |
|
WebPAC::Normalize::data_structure( |
399 |
|
row => $rec, |
400 |
|
rules => $rules, |
401 |
|
config => create_ds_config( $db_config, $database, $input, $mfn ), |
402 |
|
); |
403 |
|
|
404 |
|
#warn "current lookup: ", dump(WebPAC::Normalize::_get_lookup()); |
405 |
|
}; |
406 |
|
|
407 |
|
WebPAC::Normalize::_set_lookup( undef ); |
408 |
|
|
409 |
|
$log->debug("created lookup_coderef using:\n$rules"); |
410 |
|
|
411 |
|
}; |
412 |
|
|
413 |
|
my $lookup_jar; |
414 |
|
|
415 |
my $maxmfn = $input_db->open( |
my $maxmfn = $input_db->open( |
416 |
path => $input->{path}, |
path => $input->{path}, |
417 |
code_page => $input->{encoding}, # database encoding |
code_page => $input->{encoding}, # database encoding |
418 |
|
lookup_coderef => $lookup_coderef, |
419 |
|
lookup => $lookup_jar, |
420 |
%{ $input }, |
%{ $input }, |
421 |
|
load_row => sub { |
422 |
|
my $a = shift; |
423 |
|
return $store->load_row( |
424 |
|
database => $database, |
425 |
|
input => $input_name, |
426 |
|
id => $a->{id}, |
427 |
|
); |
428 |
|
}, |
429 |
|
save_row => sub { |
430 |
|
my $a = shift; |
431 |
|
return $store->save_row( |
432 |
|
database => $database, |
433 |
|
input => $input_name, |
434 |
|
id => $a->{id}, |
435 |
|
row => $a->{row}, |
436 |
|
); |
437 |
|
}, |
438 |
|
|
439 |
); |
); |
440 |
|
|
441 |
my @norm_array = ref($input->{normalize}) eq 'ARRAY' ? |
my $lookup_data = WebPAC::Normalize::_get_lookup(); |
|
@{ $input->{normalize} } : ( $input->{normalize} ); |
|
442 |
|
|
443 |
if ($marc_normalize) { |
if (defined( $lookup_data->{$database}->{$input_name} )) { |
444 |
@norm_array = ( { |
$log->debug("created following lookups: ", sub { dump( $lookup_data ) } ); |
445 |
path => $marc_normalize, |
|
446 |
output => $marc_output || 'out/marc/' . $database . '-' . $input->{name} . '.marc', |
foreach my $key (keys %{ $lookup_data->{$database}->{$input_name} }) { |
447 |
} ); |
$store->save_lookup( |
448 |
|
database => $database, |
449 |
|
input => $input_name, |
450 |
|
key => $key, |
451 |
|
data => $lookup_data->{$database}->{$input_name}->{$key}, |
452 |
|
); |
453 |
|
} |
454 |
} |
} |
455 |
|
|
456 |
foreach my $normalize (@norm_array) { |
my $report_fh; |
457 |
|
if ($stats || $validate) { |
458 |
|
my $path = "out/report/${database}-${input_name}.txt"; |
459 |
|
open($report_fh, '>', $path) || $log->logdie("can't open $path: $!"); |
460 |
|
|
461 |
|
print $report_fh "Report for database '$database' input '$input_name' records ", |
462 |
|
$offset || 1, "-", $limit || $input->{limit} || $maxmfn, "\n\n"; |
463 |
|
$log->info("Generating report file $path"); |
464 |
|
|
465 |
|
if ( $validate ) { |
466 |
|
$validate->read_validate_file( $validate->fill_in( $validate_path, database => $database, input => $input_name ) ) if ( $validate_path ); |
467 |
|
$validate->read_validate_delimiters_file( $validate->fill_in( $validate_delimiters_path, database => $database, input => $input_name ) ) if ( $validate_delimiters_path ); |
468 |
|
} |
469 |
|
} |
470 |
|
|
471 |
my $normalize_path = $normalize->{path} || $log->logdie("can't find normalize path in config"); |
my $marc; |
472 |
|
if ($marc_generate && $parser->have_rules( 'marc', $database, $input_name )) { |
473 |
|
$marc = new WebPAC::Output::MARC( |
474 |
|
path => "out/marc/${database}-${input_name}.marc", |
475 |
|
lint => $marc_lint, |
476 |
|
dump => $marc_dump, |
477 |
|
); |
478 |
|
} |
479 |
|
|
480 |
$log->logdie("Found '$normalize_path' as normalization file which isn't supported any more!") unless ( $normalize_path =~ m!\.pl$!i ); |
my $rules = $parser->normalize_rules($database,$input_name) || $log->logdie("no normalize rules found for $database/$input_name"); |
481 |
|
$log->debug("parsed normalize rules:\n$rules"); |
482 |
|
|
483 |
my $rules = read_file( $normalize_path ) or die "can't open $normalize_path: $!"; |
# reset position in database |
484 |
|
$input_db->seek(1); |
485 |
|
|
486 |
$log->info("Using $normalize_path for normalization..."); |
# generate name of config key for indexer (strip everything after -) |
487 |
|
my $indexer_config = $use_indexer; |
488 |
|
$indexer_config =~ s/^(\w+)-?.*$/$1/g if ($indexer_config); |
489 |
|
|
490 |
|
my $lookup_hash; |
491 |
|
my $depends = $parser->depends($database,$input_name); |
492 |
|
|
493 |
|
if ($depends) { |
494 |
|
$log->debug("$database/$input_name depends on: ", dump($depends)) if ($depends); |
495 |
|
$log->logdie("parser->depends didn't return HASH") unless (ref($depends) eq 'HASH'); |
496 |
|
|
497 |
|
foreach my $db (keys %$depends) { |
498 |
|
foreach my $i (keys %{$depends->{$db}}) { |
499 |
|
foreach my $k (keys %{$depends->{$db}->{$i}}) { |
500 |
|
my $t = time(); |
501 |
|
$log->debug("loading lookup $db/$i"); |
502 |
|
$lookup_hash->{$db}->{$i}->{$k} = $store->load_lookup( |
503 |
|
database => $db, |
504 |
|
input => $i, |
505 |
|
key => $k, |
506 |
|
); |
507 |
|
$log->debug(sprintf("lookup $db/$i took %.2fs", time() - $t)); |
508 |
|
} |
509 |
|
} |
510 |
|
} |
511 |
|
|
512 |
my $marc_fh; |
$log->debug("lookup_hash = ", sub { dump( $lookup_hash ) }); |
513 |
if (my $path = $normalize->{output}) { |
} |
|
open($marc_fh, '>', $path) || |
|
|
$log->logdie("can't open MARC output $path: $!"); |
|
514 |
|
|
|
$log->info("Creating MARC export file $path\n"); |
|
|
} |
|
515 |
|
|
516 |
# reset position in database |
foreach my $pos ( 0 ... $input_db->size ) { |
|
$input_db->seek(1); |
|
517 |
|
|
518 |
foreach my $pos ( 0 ... $input_db->size ) { |
my $row = $input_db->fetch || next; |
519 |
|
|
520 |
my $row = $input_db->fetch || next; |
$total_rows++; |
521 |
|
|
522 |
my $mfn = $row->{'000'}->[0]; |
my $mfn = $row->{'000'}->[0]; |
523 |
|
|
524 |
if (! $mfn || $mfn !~ m#^\d+$#) { |
if (! $mfn || $mfn !~ m{^\d+$}) { |
525 |
$log->warn("record $pos doesn't have valid MFN but '$mfn', using $pos"); |
$log->warn("record $pos doesn't have valid MFN but '$mfn', using $pos"); |
526 |
$mfn = $pos; |
$mfn = $pos; |
527 |
push @{ $row->{'000'} }, $pos; |
push @{ $row->{'000'} }, $pos; |
528 |
} |
} |
529 |
|
|
530 |
|
|
531 |
if ($validate) { |
if ($validate) { |
532 |
my @errors = $validate->validate_errors( $row ); |
if ( my $errors = $validate->validate_rec( $row, $input_db->dump_ascii ) ) { |
533 |
$log->error( "MFN $mfn validation errors:\n", join("\n", @errors) ) if (@errors); |
$log->error( "MFN $mfn validation error:\n", |
534 |
|
$validate->report_error( $errors ) |
535 |
|
); |
536 |
} |
} |
537 |
|
next; # validation doesn't create any output |
538 |
|
} |
539 |
|
|
540 |
|
my $ds = WebPAC::Normalize::data_structure( |
541 |
my $ds = WebPAC::Normalize::data_structure( |
row => $row, |
542 |
row => $row, |
rules => $rules, |
543 |
rules => $rules, |
lookup => $lookup_hash, |
544 |
lookup => $lookup ? $lookup->lookup_hash : undef, |
config => create_ds_config( $db_config, $database, $input, $mfn ), |
545 |
marc_encoding => 'utf-8', |
marc_encoding => 'utf-8', |
546 |
); |
load_row_coderef => sub { |
547 |
|
my ($database,$input,$mfn) = @_; |
548 |
$db->save_ds( |
return $store->load_row( |
549 |
id => $mfn, |
database => $database, |
550 |
ds => $ds, |
input => $input, |
551 |
prefix => $input->{name}, |
id => $mfn, |
552 |
) if ($ds && !$stats); |
); |
553 |
|
}, |
554 |
$indexer->add( |
); |
555 |
id => $input->{name} . "/" . $mfn, |
|
556 |
ds => $ds, |
$log->debug("ds = ", sub { dump($ds) }) if ($ds); |
557 |
type => $config->{$use_indexer}->{type}, |
|
558 |
) if ($indexer && $ds); |
$store->save_ds( |
559 |
|
database => $database, |
560 |
if ($marc_fh) { |
input => $input_name, |
561 |
my $marc = new MARC::Record; |
id => $mfn, |
562 |
$marc->encoding( 'utf-8' ); |
ds => $ds, |
563 |
$marc->add_fields( WebPAC::Normalize::_get_marc_fields() ); |
) if ($ds && !$stats); |
564 |
print $marc_fh $marc->as_usmarc; |
|
565 |
|
$indexer->add( |
566 |
|
id => "${input_name}/${mfn}", |
567 |
|
ds => $ds, |
568 |
|
type => $config->get($indexer_config)->{type}, |
569 |
|
) if ($indexer && $ds); |
570 |
|
|
571 |
|
if ($marc) { |
572 |
|
my $i = 0; |
573 |
|
|
574 |
|
while (my $fields = WebPAC::Normalize::_get_marc_fields( fetch_next => 1 ) ) { |
575 |
|
$marc->add( |
576 |
|
id => $mfn . ( $i ? "/$i" : '' ), |
577 |
|
fields => $fields, |
578 |
|
leader => WebPAC::Normalize::_get_marc_leader(), |
579 |
|
row => $row, |
580 |
|
); |
581 |
|
$i++; |
582 |
} |
} |
583 |
|
|
584 |
$total_rows++; |
$log->info("Created $i instances of MFN $mfn\n") if ($i > 1); |
585 |
} |
} |
586 |
|
|
587 |
$log->info("statistics of fields usage:\n", $input_db->stats) if ($stats); |
foreach my $out ( @outputs ) { |
588 |
|
$out->add( $mfn, $ds ) if $out->can('add'); |
589 |
|
} |
590 |
|
|
591 |
|
} |
592 |
|
|
593 |
# close MARC file |
if ($validate) { |
594 |
close($marc_fh) if ($marc_fh); |
my $errors = $validate->report; |
595 |
|
if ($errors) { |
596 |
|
$log->info("validation errors:\n$errors\n" ); |
597 |
|
print $report_fh "$errors\n" if ($report_fh); |
598 |
|
} |
599 |
|
|
600 |
|
print $report_fh "\nAll possible subfields/delimiter templates:\n", $validate->delimiters_templates( report => 1, current_input => 1 ), "\n\n"; |
601 |
|
|
602 |
|
# must be last thing that touches $validate for this input |
603 |
|
$validate->reset; |
604 |
|
} |
605 |
|
|
606 |
|
if ($stats) { |
607 |
|
my $s = $input_db->stats; |
608 |
|
$log->info("statistics of fields usage:\n$s"); |
609 |
|
print $report_fh "Statistics of fields usage:\n$s" if ($report_fh); |
610 |
} |
} |
611 |
|
|
612 |
|
# close MARC file |
613 |
|
$marc->finish if ($marc); |
614 |
|
|
615 |
|
# close report |
616 |
|
close($report_fh) if ($report_fh); |
617 |
} |
} |
618 |
|
|
619 |
eval { $indexer->finish } if ($indexer && $indexer->can('finish')); |
eval { $indexer->finish } if ($indexer && $indexer->can('finish')); |
620 |
|
|
621 |
|
foreach my $out ( @outputs ) { |
622 |
|
$out->finish if $out->can('finish'); |
623 |
|
} |
624 |
|
|
625 |
my $dt = time() - $start_t; |
my $dt = time() - $start_t; |
626 |
$log->info("$total_rows records ", $indexer ? "indexed " : "", |
$log->info("$total_rows records ", $indexer ? "indexed " : "", |
627 |
sprintf("in %.2f sec [%.2f rec/sec]", |
sprintf("in %.2f sec [%.2f rec/sec]", |
629 |
) |
) |
630 |
); |
); |
631 |
|
|
632 |
# |
|
633 |
# add Hyper Estraier links to other databases |
# end forked process |
634 |
# |
if ($parallel) { |
635 |
if (ref($db_config->{links}) eq 'ARRAY' && $use_indexer) { |
$log->info("parallel process $$ finished"); |
636 |
foreach my $link (@{ $db_config->{links} }) { |
exit(0); |
|
if ($use_indexer eq 'hyperestraier') { |
|
|
$log->info("saving link $database -> $link->{to} [$link->{credit}]"); |
|
|
push @links, { |
|
|
from => $database, |
|
|
to => $link->{to}, |
|
|
credit => $link->{credit}, |
|
|
}; |
|
|
} else { |
|
|
$log->warn("NOT IMPLEMENTED WITH $use_indexer: adding link $database -> $link->{to} [$link->{credit}]"); |
|
|
} |
|
|
} |
|
637 |
} |
} |
638 |
|
|
639 |
} |
} |
640 |
|
|
641 |
foreach my $link (@links) { |
if ($parallel) { |
642 |
$log->info("adding link $link->{from} -> $link->{to} [$link->{credit}]"); |
# wait all children to finish |
643 |
$indexer->add_link( %{ $link } ); |
sleep(1) while wait != -1; |
644 |
|
$log->info("all parallel processes finished"); |
645 |
|
} |
646 |
|
|
647 |
|
# save new delimiters if needed |
648 |
|
$validate->save_delimiters_templates if ( $validate_delimiters_path ); |
649 |
|
|
650 |
|
# |
651 |
|
# handle links or merge after indexing |
652 |
|
# |
653 |
|
|
654 |
|
if ($merge) { |
655 |
|
print $estcmd_fh 'sudo /etc/init.d/hyperestraier start',$/; |
656 |
|
close($estcmd_fh); |
657 |
|
chmod 0700, $estcmd_path || $log->warn("can't chmod 0700 $estcmd_path: $!"); |
658 |
|
system $estcmd_path; |
659 |
|
} else { |
660 |
|
foreach my $link (@links) { |
661 |
|
$log->logdie("coderef in link ", Dumper($link), " is ", ref($link), " and not CODE") unless (ref($link) eq 'CODE'); |
662 |
|
$link->(); |
663 |
|
} |
664 |
} |
} |