3 |
use warnings; |
use warnings; |
4 |
use strict; |
use strict; |
5 |
|
|
6 |
|
use blib; |
7 |
|
|
8 |
use WebPAC::Common; |
use WebPAC::Common; |
9 |
use base qw/WebPAC::Common/; |
use base qw/WebPAC::Common/; |
10 |
use Text::Iconv; |
use Data::Dump qw/dump/; |
11 |
use Data::Dumper; |
use Encode qw/from_to/; |
12 |
|
|
13 |
=head1 NAME |
=head1 NAME |
14 |
|
|
16 |
|
|
17 |
=head1 VERSION |
=head1 VERSION |
18 |
|
|
19 |
Version 0.04 |
Version 0.18 |
20 |
|
|
21 |
=cut |
=cut |
22 |
|
|
23 |
our $VERSION = '0.04'; |
our $VERSION = '0.18'; |
24 |
|
|
25 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
26 |
|
|
39 |
|
|
40 |
Perhaps a little code snippet. |
Perhaps a little code snippet. |
41 |
|
|
42 |
use WebPAC::Input; |
use WebPAC::Input; |
43 |
|
|
44 |
my $db = WebPAC::Input->new( |
my $db = WebPAC::Input->new( |
45 |
module => 'WebPAC::Input::ISIS', |
module => 'WebPAC::Input::ISIS', |
46 |
config => $config, |
); |
|
lookup => $lookup_obj, |
|
|
low_mem => 1, |
|
|
); |
|
47 |
|
|
48 |
$db->open('/path/to/database'); |
$db->open( path => '/path/to/database' ); |
49 |
print "database size: ",$db->size,"\n"; |
print "database size: ",$db->size,"\n"; |
50 |
while (my $rec = $db->fetch) { |
while (my $rec = $db->fetch) { |
51 |
# do something with $rec |
# do something with $rec |
61 |
|
|
62 |
my $db = new WebPAC::Input( |
my $db = new WebPAC::Input( |
63 |
module => 'WebPAC::Input::MARC', |
module => 'WebPAC::Input::MARC', |
64 |
code_page => 'ISO-8859-2', |
encoding => 'ISO-8859-2', |
|
low_mem => 1, |
|
65 |
recode => 'char pairs', |
recode => 'char pairs', |
66 |
no_progress_bar => 1, |
no_progress_bar => 1, |
67 |
); |
); |
68 |
|
|
69 |
C<module> is low-level file format module. See L<WebPAC::Input::Isis> and |
C<module> is low-level file format module. See L<WebPAC::Input::ISIS> and |
70 |
L<WebPAC::Input::MARC>. |
L<WebPAC::Input::MARC>. |
71 |
|
|
72 |
Optional parametar C<code_page> specify application code page (which will be |
Optional parametar C<encoding> specify application code page (which will be |
73 |
used internally). This should probably be your terminal encoding, and by |
used internally). This should probably be your terminal encoding, and by |
74 |
default, it C<ISO-8859-2>. |
default, it C<ISO-8859-2>. |
75 |
|
|
|
Default is not to use C<low_mem> options (see L<MEMORY USAGE> below). |
|
|
|
|
76 |
C<recode> is optional string constisting of character or words pairs that |
C<recode> is optional string constisting of character or words pairs that |
77 |
should be replaced in input stream. |
should be replaced in input stream. |
78 |
|
|
90 |
|
|
91 |
my $log = $self->_get_logger; |
my $log = $self->_get_logger; |
92 |
|
|
93 |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
$log->logconfess("code_page argument is not suppored any more. change it to encoding") if ($self->{lookup}); |
94 |
my $module = $self->{module}; |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if ($self->{lookup}); |
95 |
$module =~ s#::#/#g; |
$log->logconfess("low_mem argument is not suppored any more. rewrite it to load_row and save_row") if ($self->{low_mem}); |
|
$module .= '.pm'; |
|
|
$log->debug("require low-level module $self->{module} from $module"); |
|
|
|
|
|
require $module; |
|
|
#eval $self->{module} .'->import'; |
|
|
|
|
|
# check if required subclasses are implemented |
|
|
foreach my $subclass (qw/open_db fetch_rec init/) { |
|
|
my $n = $self->{module} . '::' . $subclass; |
|
|
if (! defined &{ $n }) { |
|
|
my $missing = "missing $subclass in $self->{module}"; |
|
|
$self->{$subclass} = sub { $log->logwarn($missing) }; |
|
|
} else { |
|
|
$self->{$subclass} = \&{ $n }; |
|
|
} |
|
|
} |
|
|
|
|
|
if ($self->{init}) { |
|
|
$log->debug("calling init"); |
|
|
$self->{init}->($self, @_); |
|
|
} |
|
96 |
|
|
97 |
$self->{'code_page'} ||= 'ISO-8859-2'; |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
98 |
|
my $module_path = $self->{module}; |
99 |
# running with low_mem flag? well, use DBM::Deep then. |
$module_path =~ s#::#/#g; |
100 |
if ($self->{'low_mem'}) { |
$module_path .= '.pm'; |
101 |
$log->info("running with low_mem which impacts performance (<32 Mb memory usage)"); |
$log->debug("require low-level module $self->{module} from $module_path"); |
|
|
|
|
my $db_file = "data.db"; |
|
|
|
|
|
if (-e $db_file) { |
|
|
unlink $db_file or $log->logdie("can't remove '$db_file' from last run"); |
|
|
$log->debug("removed '$db_file' from last run"); |
|
|
} |
|
|
|
|
|
require DBM::Deep; |
|
102 |
|
|
103 |
my $db = new DBM::Deep $db_file; |
require $module_path; |
104 |
|
|
105 |
$log->logdie("DBM::Deep error: $!") unless ($db); |
$self->{'encoding'} ||= 'ISO-8859-2'; |
|
|
|
|
if ($db->error()) { |
|
|
$log->logdie("can't open '$db_file' under low_mem: ",$db->error()); |
|
|
} else { |
|
|
$log->debug("using file '$db_file' for DBM::Deep"); |
|
|
} |
|
|
|
|
|
$self->{'db'} = $db; |
|
|
} |
|
106 |
|
|
107 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
108 |
} |
} |
111 |
|
|
112 |
This function will read whole database in memory and produce lookups. |
This function will read whole database in memory and produce lookups. |
113 |
|
|
114 |
|
my $store; # simple in-memory hash |
115 |
|
|
116 |
$input->open( |
$input->open( |
117 |
path => '/path/to/database/file', |
path => '/path/to/database/file', |
118 |
code_page => '852', |
code_page => 'cp852', |
119 |
limit => 500, |
limit => 500, |
120 |
offset => 6000, |
offset => 6000, |
121 |
lookup => $lookup_obj, |
stats => 1, |
122 |
|
lookup_coderef => sub { |
123 |
|
my $rec = shift; |
124 |
|
# store lookups |
125 |
|
}, |
126 |
|
modify_records => { |
127 |
|
900 => { '^a' => { ' : ' => '^b' } }, |
128 |
|
901 => { '*' => { '^b' => ' ; ' } }, |
129 |
|
}, |
130 |
|
modify_file => 'conf/modify/mapping.map', |
131 |
|
save_row => sub { |
132 |
|
my $a = shift; |
133 |
|
$store->{ $a->{id} } = $a->{row}; |
134 |
|
}, |
135 |
|
load_row => sub { |
136 |
|
my $a = shift; |
137 |
|
return defined($store->{ $a->{id} }) && |
138 |
|
$store->{ $a->{id} }; |
139 |
|
}, |
140 |
|
|
141 |
); |
); |
142 |
|
|
143 |
By default, C<code_page> is assumed to be C<852>. |
By default, C<code_page> is assumed to be C<cp852>. |
144 |
|
|
145 |
C<offset> is optional parametar to position at some offset before reading from database. |
C<offset> is optional parametar to position at some offset before reading from database. |
146 |
|
|
147 |
C<limit> is optional parametar to read just C<limit> records from database |
C<limit> is optional parametar to read just C<limit> records from database |
148 |
|
|
149 |
|
C<stats> create optional report about usage of fields and subfields |
150 |
|
|
151 |
|
C<lookup_coderef> is closure to called to save data into lookups |
152 |
|
|
153 |
|
C<modify_records> specify mapping from subfields to delimiters or from |
154 |
|
delimiters to subfields, as well as oprations on fields (if subfield is |
155 |
|
defined as C<*>. |
156 |
|
|
157 |
|
C<modify_file> is alternative for C<modify_records> above which preserves order and offers |
158 |
|
(hopefully) simplier sintax than YAML or perl (see L</modify_file_regex>). This option |
159 |
|
overrides C<modify_records> if both exists for same input. |
160 |
|
|
161 |
|
C<save_row> and C<load_row> are low-level implementation of store engine. Calling convention |
162 |
|
is documented in example above. |
163 |
|
|
164 |
Returns size of database, regardless of C<offset> and C<limit> |
Returns size of database, regardless of C<offset> and C<limit> |
165 |
parametars, see also C<size>. |
parametars, see also C<size>. |
166 |
|
|
172 |
|
|
173 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
174 |
|
|
175 |
|
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_coderef") if ($arg->{lookup}); |
176 |
|
$log->logconfess("lookup_coderef must be CODE, not ",ref($arg->{lookup_coderef})) |
177 |
|
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
178 |
|
|
179 |
|
$log->debug( $arg->{lookup_coderef} ? '' : 'not ', "using lookup_coderef"); |
180 |
|
|
181 |
$log->logcroak("need path") if (! $arg->{'path'}); |
$log->logcroak("need path") if (! $arg->{'path'}); |
182 |
my $code_page = $arg->{'code_page'} || '852'; |
my $code_page = $arg->{'code_page'} || 'cp852'; |
183 |
|
|
184 |
# store data in object |
# store data in object |
185 |
$self->{'input_code_page'} = $code_page; |
$self->{'input_code_page'} = $code_page; |
187 |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
188 |
} |
} |
189 |
|
|
190 |
# create Text::Iconv object |
if ($arg->{load_row} || $arg->{save_row}) { |
191 |
$self->{iconv} = Text::Iconv->new($code_page,$self->{'code_page'}); |
$log->logconfess("save_row and load_row must be defined in pair and be CODE") unless ( |
192 |
|
ref($arg->{load_row}) eq 'CODE' && |
193 |
|
ref($arg->{save_row}) eq 'CODE' |
194 |
|
); |
195 |
|
$self->{load_row} = $arg->{load_row}; |
196 |
|
$self->{save_row} = $arg->{save_row}; |
197 |
|
$log->debug("using load_row and save_row instead of in-memory hash"); |
198 |
|
} |
199 |
|
|
200 |
my $filter_ref; |
my $filter_ref; |
201 |
|
my $recode_regex; |
202 |
|
my $recode_map; |
203 |
|
|
204 |
if ($self->{recode}) { |
if ($self->{recode}) { |
205 |
my @r = split(/\s/, $self->{recode}); |
my @r = split(/\s/, $self->{recode}); |
206 |
if ($#r % 2 != 1) { |
if ($#r % 2 != 1) { |
207 |
$log->logwarn("recode needs even number of elements (some number of valid pairs)"); |
$log->logwarn("recode needs even number of elements (some number of valid pairs)"); |
208 |
} else { |
} else { |
|
my $recode; |
|
209 |
while (@r) { |
while (@r) { |
210 |
my $from = shift @r; |
my $from = shift @r; |
211 |
my $to = shift @r; |
my $to = shift @r; |
212 |
$recode->{$from} = $to; |
$recode_map->{$from} = $to; |
213 |
} |
} |
214 |
|
|
215 |
my $regex = join '|' => keys %{ $recode }; |
$recode_regex = join '|' => keys %{ $recode_map }; |
|
|
|
|
$log->debug("using recode regex: $regex"); |
|
|
|
|
|
$filter_ref = sub { |
|
|
my $t = shift; |
|
|
$t =~ s/($regex)/$recode->{$1}/g; |
|
|
return $t; |
|
|
}; |
|
216 |
|
|
217 |
|
$log->debug("using recode regex: $recode_regex"); |
218 |
} |
} |
219 |
|
|
220 |
} |
} |
221 |
|
|
222 |
my ($db, $size) = $self->{open_db}->( $self, |
my $rec_regex; |
223 |
|
if (my $p = $arg->{modify_file}) { |
224 |
|
$log->debug("using modify_file $p"); |
225 |
|
$rec_regex = $self->modify_file_regexps( $p ); |
226 |
|
} elsif (my $h = $arg->{modify_records}) { |
227 |
|
$log->debug("using modify_records ", sub { dump( $h ) }); |
228 |
|
$rec_regex = $self->modify_record_regexps(%{ $h }); |
229 |
|
} |
230 |
|
$log->debug("rec_regex: ", sub { dump($rec_regex) }) if ($rec_regex); |
231 |
|
|
232 |
|
my $class = $self->{module} || $log->logconfess("can't get low-level module name!"); |
233 |
|
|
234 |
|
my $ll_db = $class->new( |
235 |
path => $arg->{path}, |
path => $arg->{path}, |
236 |
filter => $filter_ref, |
# filter => sub { |
237 |
|
# my ($l,$f_nr) = @_; |
238 |
|
# return unless defined($l); |
239 |
|
# from_to($l, $code_page, $self->{'encoding'}); |
240 |
|
# $l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
241 |
|
# return $l; |
242 |
|
# }, |
243 |
|
%{ $arg }, |
244 |
); |
); |
245 |
|
|
246 |
unless ($db) { |
unless (defined($ll_db)) { |
247 |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
248 |
return; |
return; |
249 |
} |
} |
250 |
|
|
251 |
|
my $size = $ll_db->size; |
252 |
|
|
253 |
unless ($size) { |
unless ($size) { |
254 |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
255 |
return; |
return; |
259 |
my $to_rec = $size; |
my $to_rec = $size; |
260 |
|
|
261 |
if (my $s = $self->{offset}) { |
if (my $s = $self->{offset}) { |
262 |
$log->info("skipping to MFN $s"); |
$log->debug("skipping to MFN $s"); |
263 |
$from_rec = $s; |
$from_rec = $s; |
264 |
} else { |
} else { |
265 |
$self->{offset} = $from_rec; |
$self->{offset} = $from_rec; |
274 |
# store size for later |
# store size for later |
275 |
$self->{size} = ($to_rec - $from_rec) ? ($to_rec - $from_rec + 1) : 0; |
$self->{size} = ($to_rec - $from_rec) ? ($to_rec - $from_rec + 1) : 0; |
276 |
|
|
277 |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{code_page}"); |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{encoding}", $self->{stats} ? ' [stats]' : ''); |
278 |
|
|
279 |
# read database |
# read database |
280 |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
281 |
|
|
282 |
$log->debug("position: $pos\n"); |
$log->debug("position: $pos\n"); |
283 |
|
|
284 |
my $rec = $self->{fetch_rec}->($self, $db, $pos ); |
my $rec = $ll_db->fetch_rec($pos, sub { |
285 |
|
my ($l,$f_nr,$debug) = @_; |
286 |
|
# return unless defined($l); |
287 |
|
# return $l unless ($rec_regex && $f_nr); |
288 |
|
|
289 |
|
return unless ( defined($l) && defined($f_nr) ); |
290 |
|
|
291 |
|
warn "-=> $f_nr ## |$l|\n" if ($debug); |
292 |
|
$log->debug("-=> $f_nr ## $l"); |
293 |
|
|
294 |
|
# codepage conversion and recode_regex |
295 |
|
from_to($l, $code_page, $self->{'encoding'}); |
296 |
|
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
297 |
|
|
298 |
|
# apply regexps |
299 |
|
if ($rec_regex && defined($rec_regex->{$f_nr})) { |
300 |
|
$log->logconfess("regexps->{$f_nr} must be ARRAY") if (ref($rec_regex->{$f_nr}) ne 'ARRAY'); |
301 |
|
my $c = 0; |
302 |
|
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
303 |
|
my $old_l = $l; |
304 |
|
$log->logconfess("expected regex in ", dump( $r )) unless defined($r->{regex}); |
305 |
|
eval '$l =~ ' . $r->{regex}; |
306 |
|
if ($old_l ne $l) { |
307 |
|
my $d = "|$old_l| -> |$l| "; # . $r->{regex}; |
308 |
|
$d .= ' +' . $r->{line} . ' ' . $r->{file} if defined($r->{line}); |
309 |
|
$d .= ' ' . $r->{debug} if defined($r->{debug}); |
310 |
|
$log->debug("MODIFY $d"); |
311 |
|
warn "*** $d\n" if ($debug); |
312 |
|
|
313 |
|
} |
314 |
|
$log->error("error applying regex: $r") if ($@); |
315 |
|
} |
316 |
|
} |
317 |
|
|
318 |
|
$log->debug("<=- $f_nr ## |$l|"); |
319 |
|
warn "<=- $f_nr ## $l\n" if ($debug); |
320 |
|
return $l; |
321 |
|
}); |
322 |
|
|
323 |
$log->debug(sub { Dumper($rec) }); |
$log->debug(sub { dump($rec) }); |
324 |
|
|
325 |
if (! $rec) { |
if (! $rec) { |
326 |
$log->warn("record $pos empty? skipping..."); |
$log->warn("record $pos empty? skipping..."); |
328 |
} |
} |
329 |
|
|
330 |
# store |
# store |
331 |
if ($self->{low_mem}) { |
if ($self->{save_row}) { |
332 |
$self->{db}->put($pos, $rec); |
$self->{save_row}->({ |
333 |
|
id => $pos, |
334 |
|
row => $rec, |
335 |
|
}); |
336 |
} else { |
} else { |
337 |
$self->{data}->{$pos} = $rec; |
$self->{data}->{$pos} = $rec; |
338 |
} |
} |
339 |
|
|
340 |
# create lookup |
# create lookup |
341 |
$self->{'lookup'}->add( $rec ) if ($rec && $self->{'lookup'}); |
$arg->{'lookup_coderef'}->( $rec ) if ($rec && $arg->{'lookup_coderef'}); |
342 |
|
|
343 |
|
# update counters for statistics |
344 |
|
if ($self->{stats}) { |
345 |
|
|
346 |
|
# fetch clean record with regexpes applied for statistics |
347 |
|
my $rec = $ll_db->fetch_rec($pos); |
348 |
|
|
349 |
|
foreach my $fld (keys %{ $rec }) { |
350 |
|
$self->{_stats}->{fld}->{ $fld }++; |
351 |
|
|
352 |
|
$log->logdie("invalid record fild $fld, not ARRAY") |
353 |
|
unless (ref($rec->{ $fld }) eq 'ARRAY'); |
354 |
|
|
355 |
|
foreach my $row (@{ $rec->{$fld} }) { |
356 |
|
|
357 |
|
if (ref($row) eq 'HASH') { |
358 |
|
|
359 |
|
foreach my $sf (keys %{ $row }) { |
360 |
|
next if ($sf eq 'subfields'); |
361 |
|
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{count}++; |
362 |
|
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{repeatable}++ |
363 |
|
if (ref($row->{$sf}) eq 'ARRAY'); |
364 |
|
} |
365 |
|
|
366 |
|
} else { |
367 |
|
$self->{_stats}->{repeatable}->{ $fld }++; |
368 |
|
} |
369 |
|
} |
370 |
|
} |
371 |
|
} |
372 |
|
|
373 |
$self->progress_bar($pos,$to_rec) unless ($self->{no_progress_bar}); |
$self->progress_bar($pos,$to_rec) unless ($self->{no_progress_bar}); |
374 |
|
|
381 |
$self->{max_pos} = $to_rec; |
$self->{max_pos} = $to_rec; |
382 |
$log->debug("max_pos: $to_rec"); |
$log->debug("max_pos: $to_rec"); |
383 |
|
|
384 |
|
# save for dump |
385 |
|
$self->{ll_db} = $ll_db; |
386 |
|
|
387 |
return $size; |
return $size; |
388 |
} |
} |
389 |
|
|
423 |
|
|
424 |
my $rec; |
my $rec; |
425 |
|
|
426 |
if ($self->{low_mem}) { |
if ($self->{load_row}) { |
427 |
$rec = $self->{db}->get($mfn); |
$rec = $self->{load_row}->({ id => $mfn }); |
428 |
} else { |
} else { |
429 |
$rec = $self->{data}->{$mfn}; |
$rec = $self->{data}->{$mfn}; |
430 |
} |
} |
479 |
|
|
480 |
sub seek { |
sub seek { |
481 |
my $self = shift; |
my $self = shift; |
482 |
my $pos = shift || return; |
my $pos = shift; |
483 |
|
|
484 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
485 |
|
|
486 |
|
$log->logconfess("called without pos") unless defined($pos); |
487 |
|
|
488 |
if ($pos < 1) { |
if ($pos < 1) { |
489 |
$log->warn("seek before first record"); |
$log->warn("seek before first record"); |
490 |
$pos = 1; |
$pos = 1; |
496 |
return $self->{pos} = (($pos - 1) || -1); |
return $self->{pos} = (($pos - 1) || -1); |
497 |
} |
} |
498 |
|
|
499 |
|
=head2 stats |
500 |
|
|
501 |
|
Dump statistics about field and subfield usage |
502 |
|
|
503 |
|
print $input->stats; |
504 |
|
|
505 |
|
=cut |
506 |
|
|
507 |
|
sub stats { |
508 |
|
my $self = shift; |
509 |
|
|
510 |
=head1 MEMORY USAGE |
my $log = $self->_get_logger(); |
511 |
|
|
512 |
C<low_mem> options is double-edged sword. If enabled, WebPAC |
my $s = $self->{_stats}; |
513 |
will run on memory constraint machines (which doesn't have enough |
if (! $s) { |
514 |
physical RAM to create memory structure for whole source database). |
$log->warn("called stats, but there is no statistics collected"); |
515 |
|
return; |
516 |
If your machine has 512Mb or more of RAM and database is around 10000 records, |
} |
517 |
memory shouldn't be an issue. If you don't have enough physical RAM, you |
|
518 |
might consider using virtual memory (if your operating system is handling it |
my $max_fld = 0; |
519 |
well, like on FreeBSD or Linux) instead of dropping to L<DBM::Deep> to handle |
|
520 |
parsed structure of ISIS database (this is what C<low_mem> option does). |
my $out = join("\n", |
521 |
|
map { |
522 |
Hitting swap at end of reading source database is probably o.k. However, |
my $f = $_ || die "no field"; |
523 |
hitting swap before 90% will dramatically decrease performance and you will |
my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}"; |
524 |
be better off with C<low_mem> and using rest of availble memory for |
$max_fld = $v if ($v > $max_fld); |
525 |
operating system disk cache (Linux is particuallary good about this). |
|
526 |
However, every access to database record will require disk access, so |
my $o = sprintf("%4s %d ~", $f, $v); |
527 |
generation phase will be slower 10-100 times. |
|
528 |
|
if (defined($s->{sf}->{$f})) { |
529 |
Parsed structures are essential - you just have option to trade RAM memory |
map { |
530 |
(which is fast) for disk space (which is slow). Be sure to have planty of |
$o .= sprintf(" %s:%d%s", $_, |
531 |
disk space if you are using C<low_mem> and thus L<DBM::Deep>. |
$s->{sf}->{$f}->{$_}->{count}, |
532 |
|
$s->{sf}->{$f}->{$_}->{repeatable} ? '*' : '', |
533 |
However, when WebPAC is running on desktop machines (or laptops :-), it's |
); |
534 |
highly undesireable for system to start swapping. Using C<low_mem> option can |
} sort keys %{ $s->{sf}->{$f} }; |
535 |
reduce WecPAC memory usage to around 64Mb for same database with lookup |
} |
536 |
fields and sorted indexes which stay in RAM. Performance will suffer, but |
|
537 |
memory usage will really be minimal. It might be also more confortable to |
if (my $v_r = $s->{repeatable}->{$f}) { |
538 |
run WebPAC reniced on those machines. |
$o .= " ($v_r)" if ($v_r != $v); |
539 |
|
} |
540 |
|
|
541 |
|
$o; |
542 |
|
} sort { $a cmp $b } keys %{ $s->{fld} } |
543 |
|
); |
544 |
|
|
545 |
|
$log->debug( sub { dump($s) } ); |
546 |
|
|
547 |
|
return $out; |
548 |
|
} |
549 |
|
|
550 |
|
=head2 dump_ascii |
551 |
|
|
552 |
|
Display humanly readable dump of record |
553 |
|
|
554 |
|
=cut |
555 |
|
|
556 |
|
sub dump_ascii { |
557 |
|
my $self = shift; |
558 |
|
|
559 |
|
return unless $self->{ll_db}; |
560 |
|
|
561 |
|
if ($self->{ll_db}->can('dump_rec')) { |
562 |
|
return $self->{ll_db}->dump_ascii( $self->{pos} ); |
563 |
|
} else { |
564 |
|
return dump( $self->{ll_db}->fetch_rec( $self->{pos} ) ); |
565 |
|
} |
566 |
|
} |
567 |
|
|
568 |
|
=head2 _get_regex |
569 |
|
|
570 |
|
Helper function called which create regexps to be execute on code. |
571 |
|
|
572 |
|
_get_regex( 900, 'regex:[0-9]+' ,'numbers' ); |
573 |
|
_get_regex( 900, '^b', ' : ^b' ); |
574 |
|
|
575 |
|
It supports perl regexps with C<regex:> prefix to from value and has |
576 |
|
additional logic to skip empty subfields. |
577 |
|
|
578 |
|
=cut |
579 |
|
|
580 |
|
sub _get_regex { |
581 |
|
my ($sf,$from,$to) = @_; |
582 |
|
|
583 |
|
# protect / |
584 |
|
$from =~ s!/!\\/!gs; |
585 |
|
$to =~ s!/!\\/!gs; |
586 |
|
|
587 |
|
if ($from =~ m/^regex:(.+)$/) { |
588 |
|
$from = $1; |
589 |
|
} else { |
590 |
|
$from = '\Q' . $from . '\E'; |
591 |
|
} |
592 |
|
if ($sf =~ /^\^/) { |
593 |
|
my $need_subfield_data = '*'; # no |
594 |
|
# if from is also subfield, require some data in between |
595 |
|
# to correctly skip empty subfields |
596 |
|
$need_subfield_data = '+' if ($from =~ m/^\\Q\^/); |
597 |
|
return |
598 |
|
's/\Q'. $sf .'\E([^\^]' . $need_subfield_data . '?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
599 |
|
} else { |
600 |
|
return |
601 |
|
's/'. $from .'/'. $to .'/g'; |
602 |
|
} |
603 |
|
} |
604 |
|
|
605 |
|
|
606 |
|
=head2 modify_record_regexps |
607 |
|
|
608 |
|
Generate hash with regexpes to be applied using L<filter>. |
609 |
|
|
610 |
|
my $regexpes = $input->modify_record_regexps( |
611 |
|
900 => { '^a' => { ' : ' => '^b' } }, |
612 |
|
901 => { '*' => { '^b' => ' ; ' } }, |
613 |
|
); |
614 |
|
|
615 |
|
=cut |
616 |
|
|
617 |
|
sub modify_record_regexps { |
618 |
|
my $self = shift; |
619 |
|
my $modify_record = {@_}; |
620 |
|
|
621 |
|
my $regexpes; |
622 |
|
|
623 |
|
my $log = $self->_get_logger(); |
624 |
|
|
625 |
|
foreach my $f (keys %$modify_record) { |
626 |
|
$log->debug("field: $f"); |
627 |
|
|
628 |
|
foreach my $sf (keys %{ $modify_record->{$f} }) { |
629 |
|
$log->debug("subfield: $sf"); |
630 |
|
|
631 |
|
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
632 |
|
my $to = $modify_record->{$f}->{$sf}->{$from}; |
633 |
|
#die "no field?" unless defined($to); |
634 |
|
my $d = "|$from| -> |$to|"; |
635 |
|
$log->debug("transform: $d"); |
636 |
|
|
637 |
|
my $regex = _get_regex($sf,$from,$to); |
638 |
|
push @{ $regexpes->{$f} }, { regex => $regex, debug => $d }; |
639 |
|
$log->debug("regex: $regex"); |
640 |
|
} |
641 |
|
} |
642 |
|
} |
643 |
|
|
644 |
|
return $regexpes; |
645 |
|
} |
646 |
|
|
647 |
|
=head2 modify_file_regexps |
648 |
|
|
649 |
|
Generate hash with regexpes to be applied using L<filter> from |
650 |
|
pseudo hash/yaml format for regex mappings. |
651 |
|
|
652 |
|
It should be obvious: |
653 |
|
|
654 |
|
200 |
655 |
|
'^a' |
656 |
|
' : ' => '^e' |
657 |
|
' = ' => '^d' |
658 |
|
|
659 |
|
In field I<200> find C<'^a'> and then C<' : '>, and replace it with C<'^e'>. |
660 |
|
In field I<200> find C<'^a'> and then C<' = '>, and replace it with C<'^d'>. |
661 |
|
|
662 |
|
my $regexpes = $input->modify_file_regexps( 'conf/modify/common.pl' ); |
663 |
|
|
664 |
|
On undef path it will just return. |
665 |
|
|
666 |
|
=cut |
667 |
|
|
668 |
|
sub modify_file_regexps { |
669 |
|
my $self = shift; |
670 |
|
|
671 |
|
my $modify_path = shift || return; |
672 |
|
|
673 |
|
my $log = $self->_get_logger(); |
674 |
|
|
675 |
|
my $regexpes; |
676 |
|
|
677 |
|
CORE::open(my $fh, $modify_path) || $log->logdie("can't open modify file $modify_path: $!"); |
678 |
|
|
679 |
|
my ($f,$sf); |
680 |
|
|
681 |
|
while(<$fh>) { |
682 |
|
chomp; |
683 |
|
next if (/^#/ || /^\s*$/); |
684 |
|
|
685 |
|
if (/^\s*(\d+)\s*$/) { |
686 |
|
$f = $1; |
687 |
|
$log->debug("field: $f"); |
688 |
|
next; |
689 |
|
} elsif (/^\s*'([^']*)'\s*$/) { |
690 |
|
$sf = $1; |
691 |
|
$log->die("can't define subfiled before field in: $_") unless ($f); |
692 |
|
$log->debug("subfield: $sf"); |
693 |
|
} elsif (/^\s*'([^']*)'\s*=>\s*'([^']*)'\s*$/) { |
694 |
|
my ($from,$to) = ($1, $2); |
695 |
|
|
696 |
|
$log->debug("transform: |$from| -> |$to|"); |
697 |
|
|
698 |
|
my $regex = _get_regex($sf,$from,$to); |
699 |
|
push @{ $regexpes->{$f} }, { |
700 |
|
regex => $regex, |
701 |
|
file => $modify_path, |
702 |
|
line => $., |
703 |
|
}; |
704 |
|
$log->debug("regex: $regex"); |
705 |
|
} |
706 |
|
} |
707 |
|
|
708 |
|
return $regexpes; |
709 |
|
} |
710 |
|
|
711 |
=head1 AUTHOR |
=head1 AUTHOR |
712 |
|
|
714 |
|
|
715 |
=head1 COPYRIGHT & LICENSE |
=head1 COPYRIGHT & LICENSE |
716 |
|
|
717 |
Copyright 2005 Dobrica Pavlinusic, All Rights Reserved. |
Copyright 2005-2006 Dobrica Pavlinusic, All Rights Reserved. |
718 |
|
|
719 |
This program is free software; you can redistribute it and/or modify it |
This program is free software; you can redistribute it and/or modify it |
720 |
under the same terms as Perl itself. |
under the same terms as Perl itself. |