1 |
package WebPAC::Input; |
2 |
|
3 |
use warnings; |
4 |
use strict; |
5 |
|
6 |
use blib; |
7 |
|
8 |
use WebPAC::Common; |
9 |
use base qw/WebPAC::Common/; |
10 |
use Data::Dumper; |
11 |
use Encode qw/from_to/; |
12 |
|
13 |
=head1 NAME |
14 |
|
15 |
WebPAC::Input - read different file formats into WebPAC |
16 |
|
17 |
=head1 VERSION |
18 |
|
19 |
Version 0.11 |
20 |
|
21 |
=cut |
22 |
|
23 |
our $VERSION = '0.11'; |
24 |
|
25 |
=head1 SYNOPSIS |
26 |
|
27 |
This module implements input as database which have fixed and known |
28 |
I<size> while indexing and single unique numeric identifier for database |
29 |
position ranging from 1 to I<size>. |
30 |
|
31 |
Simply, something that is indexed by unmber from 1 .. I<size>. |
32 |
|
33 |
Examples of such databases are CDS/ISIS files, MARC files, lines in |
34 |
text file, and so on. |
35 |
|
36 |
Specific file formats are implemented using low-level interface modules, |
37 |
located in C<WebPAC::Input::*> namespace which export C<open_db>, |
38 |
C<fetch_rec> and optional C<init> functions. |
39 |
|
40 |
Perhaps a little code snippet. |
41 |
|
42 |
use WebPAC::Input; |
43 |
|
44 |
my $db = WebPAC::Input->new( |
45 |
module => 'WebPAC::Input::ISIS', |
46 |
low_mem => 1, |
47 |
); |
48 |
|
49 |
$db->open( path => '/path/to/database' ); |
50 |
print "database size: ",$db->size,"\n"; |
51 |
while (my $rec = $db->fetch) { |
52 |
# do something with $rec |
53 |
} |
54 |
|
55 |
|
56 |
|
57 |
=head1 FUNCTIONS |
58 |
|
59 |
=head2 new |
60 |
|
61 |
Create new input database object. |
62 |
|
63 |
my $db = new WebPAC::Input( |
64 |
module => 'WebPAC::Input::MARC', |
65 |
encoding => 'ISO-8859-2', |
66 |
low_mem => 1, |
67 |
recode => 'char pairs', |
68 |
no_progress_bar => 1, |
69 |
); |
70 |
|
71 |
C<module> is low-level file format module. See L<WebPAC::Input::ISIS> and |
72 |
L<WebPAC::Input::MARC>. |
73 |
|
74 |
Optional parametar C<encoding> specify application code page (which will be |
75 |
used internally). This should probably be your terminal encoding, and by |
76 |
default, it C<ISO-8859-2>. |
77 |
|
78 |
Default is not to use C<low_mem> options (see L<MEMORY USAGE> below). |
79 |
|
80 |
C<recode> is optional string constisting of character or words pairs that |
81 |
should be replaced in input stream. |
82 |
|
83 |
C<no_progress_bar> disables progress bar output on C<STDOUT> |
84 |
|
85 |
This function will also call low-level C<init> if it exists with same |
86 |
parametars. |
87 |
|
88 |
=cut |
89 |
|
90 |
sub new { |
91 |
my $class = shift; |
92 |
my $self = {@_}; |
93 |
bless($self, $class); |
94 |
|
95 |
my $log = $self->_get_logger; |
96 |
|
97 |
$log->logconfess("code_page argument is not suppored any more. change it to encoding") if ($self->{lookup}); |
98 |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if ($self->{lookup}); |
99 |
|
100 |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
101 |
my $module = $self->{module}; |
102 |
$module =~ s#::#/#g; |
103 |
$module .= '.pm'; |
104 |
$log->debug("require low-level module $self->{module} from $module"); |
105 |
|
106 |
require $module; |
107 |
#eval $self->{module} .'->import'; |
108 |
|
109 |
# check if required subclasses are implemented |
110 |
foreach my $subclass (qw/open_db fetch_rec init/) { |
111 |
my $n = $self->{module} . '::' . $subclass; |
112 |
if (! defined &{ $n }) { |
113 |
my $missing = "missing $subclass in $self->{module}"; |
114 |
$self->{$subclass} = sub { $log->logwarn($missing) }; |
115 |
} else { |
116 |
$self->{$subclass} = \&{ $n }; |
117 |
} |
118 |
} |
119 |
|
120 |
if ($self->{init}) { |
121 |
$log->debug("calling init"); |
122 |
$self->{init}->($self, @_); |
123 |
} |
124 |
|
125 |
$self->{'encoding'} ||= 'ISO-8859-2'; |
126 |
|
127 |
# running with low_mem flag? well, use DBM::Deep then. |
128 |
if ($self->{'low_mem'}) { |
129 |
$log->info("running with low_mem which impacts performance (<32 Mb memory usage)"); |
130 |
|
131 |
my $db_file = "data.db"; |
132 |
|
133 |
if (-e $db_file) { |
134 |
unlink $db_file or $log->logdie("can't remove '$db_file' from last run"); |
135 |
$log->debug("removed '$db_file' from last run"); |
136 |
} |
137 |
|
138 |
require DBM::Deep; |
139 |
|
140 |
my $db = new DBM::Deep $db_file; |
141 |
|
142 |
$log->logdie("DBM::Deep error: $!") unless ($db); |
143 |
|
144 |
if ($db->error()) { |
145 |
$log->logdie("can't open '$db_file' under low_mem: ",$db->error()); |
146 |
} else { |
147 |
$log->debug("using file '$db_file' for DBM::Deep"); |
148 |
} |
149 |
|
150 |
$self->{'db'} = $db; |
151 |
} |
152 |
|
153 |
$self ? return $self : return undef; |
154 |
} |
155 |
|
156 |
=head2 open |
157 |
|
158 |
This function will read whole database in memory and produce lookups. |
159 |
|
160 |
$input->open( |
161 |
path => '/path/to/database/file', |
162 |
code_page => 'cp852', |
163 |
limit => 500, |
164 |
offset => 6000, |
165 |
lookup => $lookup_obj, |
166 |
stats => 1, |
167 |
lookup_ref => sub { |
168 |
my ($k,$v) = @_; |
169 |
# store lookup $k => $v |
170 |
}, |
171 |
modify_records => { |
172 |
900 => { '^a' => { ' : ' => '^b' } }, |
173 |
901 => { '*' => { '^b' => ' ; ' } }, |
174 |
}, |
175 |
); |
176 |
|
177 |
By default, C<code_page> is assumed to be C<cp852>. |
178 |
|
179 |
C<offset> is optional parametar to position at some offset before reading from database. |
180 |
|
181 |
C<limit> is optional parametar to read just C<limit> records from database |
182 |
|
183 |
C<stats> create optional report about usage of fields and subfields |
184 |
|
185 |
C<lookup_coderef> is closure to call when adding C<< key => 'value' >> combinations to |
186 |
lookup. |
187 |
|
188 |
C<modify_records> specify mapping from subfields to delimiters or from |
189 |
delimiters to subfields, as well as oprations on fields (if subfield is |
190 |
defined as C<*>. |
191 |
|
192 |
Returns size of database, regardless of C<offset> and C<limit> |
193 |
parametars, see also C<size>. |
194 |
|
195 |
=cut |
196 |
|
197 |
sub open { |
198 |
my $self = shift; |
199 |
my $arg = {@_}; |
200 |
|
201 |
my $log = $self->_get_logger(); |
202 |
|
203 |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_coderef") if ($arg->{lookup}); |
204 |
$log->logconfess("lookup_coderef must be CODE, not ",ref($arg->{lookup_coderef})) |
205 |
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
206 |
|
207 |
$log->logcroak("need path") if (! $arg->{'path'}); |
208 |
my $code_page = $arg->{'code_page'} || 'cp852'; |
209 |
|
210 |
# store data in object |
211 |
$self->{'input_code_page'} = $code_page; |
212 |
foreach my $v (qw/path offset limit/) { |
213 |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
214 |
} |
215 |
|
216 |
my $filter_ref; |
217 |
my $recode_regex; |
218 |
my $recode_map; |
219 |
|
220 |
if ($self->{recode}) { |
221 |
my @r = split(/\s/, $self->{recode}); |
222 |
if ($#r % 2 != 1) { |
223 |
$log->logwarn("recode needs even number of elements (some number of valid pairs)"); |
224 |
} else { |
225 |
while (@r) { |
226 |
my $from = shift @r; |
227 |
my $to = shift @r; |
228 |
$recode_map->{$from} = $to; |
229 |
} |
230 |
|
231 |
$recode_regex = join '|' => keys %{ $recode_map }; |
232 |
|
233 |
$log->debug("using recode regex: $recode_regex"); |
234 |
} |
235 |
|
236 |
} |
237 |
|
238 |
my $rec_regex = $self->modify_record_regexps(%{ $arg->{modify_records} }); |
239 |
$log->debug("rec_regex: ", Dumper($rec_regex)); |
240 |
|
241 |
my ($db, $size) = $self->{open_db}->( $self, |
242 |
path => $arg->{path}, |
243 |
# filter => sub { |
244 |
# my ($l,$f_nr) = @_; |
245 |
# return unless defined($l); |
246 |
# from_to($l, $code_page, $self->{'encoding'}); |
247 |
# $l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
248 |
# return $l; |
249 |
# }, |
250 |
%{ $arg }, |
251 |
); |
252 |
|
253 |
unless (defined($db)) { |
254 |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
255 |
return; |
256 |
} |
257 |
|
258 |
unless ($size) { |
259 |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
260 |
return; |
261 |
} |
262 |
|
263 |
my $from_rec = 1; |
264 |
my $to_rec = $size; |
265 |
|
266 |
if (my $s = $self->{offset}) { |
267 |
$log->debug("skipping to MFN $s"); |
268 |
$from_rec = $s; |
269 |
} else { |
270 |
$self->{offset} = $from_rec; |
271 |
} |
272 |
|
273 |
if ($self->{limit}) { |
274 |
$log->debug("limiting to ",$self->{limit}," records"); |
275 |
$to_rec = $from_rec + $self->{limit} - 1; |
276 |
$to_rec = $size if ($to_rec > $size); |
277 |
} |
278 |
|
279 |
# store size for later |
280 |
$self->{size} = ($to_rec - $from_rec) ? ($to_rec - $from_rec + 1) : 0; |
281 |
|
282 |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{encoding}", $self->{stats} ? ' [stats]' : ''); |
283 |
|
284 |
# read database |
285 |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
286 |
|
287 |
$log->debug("position: $pos\n"); |
288 |
|
289 |
my $rec = $self->{fetch_rec}->($self, $db, $pos, sub { |
290 |
my ($l,$f_nr) = @_; |
291 |
# return unless defined($l); |
292 |
# return $l unless ($rec_regex && $f_nr); |
293 |
|
294 |
$log->debug("-=> $f_nr ## $l"); |
295 |
|
296 |
# codepage conversion and recode_regex |
297 |
from_to($l, $code_page, $self->{'encoding'}); |
298 |
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
299 |
|
300 |
# apply regexps |
301 |
if ($rec_regex && defined($rec_regex->{$f_nr})) { |
302 |
$log->logconfess("regexps->{$f_nr} must be ARRAY") if (ref($rec_regex->{$f_nr}) ne 'ARRAY'); |
303 |
my $c = 0; |
304 |
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
305 |
my $old_l = $l; |
306 |
eval '$l =~ ' . $r; |
307 |
if ($old_l ne $l) { |
308 |
$log->debug("REGEX on $f_nr eval \$l =~ $r\n## old l: [$old_l]\n## new l: [$l]"); |
309 |
} |
310 |
$log->error("error applying regex: $r") if ($@); |
311 |
} |
312 |
} |
313 |
|
314 |
$log->debug("<=- $f_nr ## $l"); |
315 |
return $l; |
316 |
}); |
317 |
|
318 |
$log->debug(sub { Dumper($rec) }); |
319 |
|
320 |
if (! $rec) { |
321 |
$log->warn("record $pos empty? skipping..."); |
322 |
next; |
323 |
} |
324 |
|
325 |
# store |
326 |
if ($self->{low_mem}) { |
327 |
$self->{db}->put($pos, $rec); |
328 |
} else { |
329 |
$self->{data}->{$pos} = $rec; |
330 |
} |
331 |
|
332 |
# create lookup |
333 |
$arg->{'lookup_coderef'}->( $rec ) if ($rec && $arg->{'lookup_coderef'}); |
334 |
|
335 |
# update counters for statistics |
336 |
if ($self->{stats}) { |
337 |
|
338 |
# fetch clean record with regexpes applied for statistics |
339 |
my $rec = $self->{fetch_rec}->($self, $db, $pos); |
340 |
|
341 |
foreach my $fld (keys %{ $rec }) { |
342 |
$self->{_stats}->{fld}->{ $fld }++; |
343 |
|
344 |
$log->logdie("invalid record fild $fld, not ARRAY") |
345 |
unless (ref($rec->{ $fld }) eq 'ARRAY'); |
346 |
|
347 |
foreach my $row (@{ $rec->{$fld} }) { |
348 |
|
349 |
if (ref($row) eq 'HASH') { |
350 |
|
351 |
foreach my $sf (keys %{ $row }) { |
352 |
next if ($sf eq 'subfields'); |
353 |
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{count}++; |
354 |
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{repeatable}++ |
355 |
if (ref($row->{$sf}) eq 'ARRAY'); |
356 |
} |
357 |
|
358 |
} else { |
359 |
$self->{_stats}->{repeatable}->{ $fld }++; |
360 |
} |
361 |
} |
362 |
} |
363 |
} |
364 |
|
365 |
$self->progress_bar($pos,$to_rec) unless ($self->{no_progress_bar}); |
366 |
|
367 |
} |
368 |
|
369 |
$self->{pos} = -1; |
370 |
$self->{last_pcnt} = 0; |
371 |
|
372 |
# store max mfn and return it. |
373 |
$self->{max_pos} = $to_rec; |
374 |
$log->debug("max_pos: $to_rec"); |
375 |
|
376 |
return $size; |
377 |
} |
378 |
|
379 |
=head2 fetch |
380 |
|
381 |
Fetch next record from database. It will also displays progress bar. |
382 |
|
383 |
my $rec = $isis->fetch; |
384 |
|
385 |
Record from this function should probably go to C<data_structure> for |
386 |
normalisation. |
387 |
|
388 |
=cut |
389 |
|
390 |
sub fetch { |
391 |
my $self = shift; |
392 |
|
393 |
my $log = $self->_get_logger(); |
394 |
|
395 |
$log->logconfess("it seems that you didn't load database!") unless ($self->{pos}); |
396 |
|
397 |
if ($self->{pos} == -1) { |
398 |
$self->{pos} = $self->{offset}; |
399 |
} else { |
400 |
$self->{pos}++; |
401 |
} |
402 |
|
403 |
my $mfn = $self->{pos}; |
404 |
|
405 |
if ($mfn > $self->{max_pos}) { |
406 |
$self->{pos} = $self->{max_pos}; |
407 |
$log->debug("at EOF"); |
408 |
return; |
409 |
} |
410 |
|
411 |
$self->progress_bar($mfn,$self->{max_pos}) unless ($self->{no_progress_bar}); |
412 |
|
413 |
my $rec; |
414 |
|
415 |
if ($self->{low_mem}) { |
416 |
$rec = $self->{db}->get($mfn); |
417 |
} else { |
418 |
$rec = $self->{data}->{$mfn}; |
419 |
} |
420 |
|
421 |
$rec ||= 0E0; |
422 |
} |
423 |
|
424 |
=head2 pos |
425 |
|
426 |
Returns current record number (MFN). |
427 |
|
428 |
print $isis->pos; |
429 |
|
430 |
First record in database has position 1. |
431 |
|
432 |
=cut |
433 |
|
434 |
sub pos { |
435 |
my $self = shift; |
436 |
return $self->{pos}; |
437 |
} |
438 |
|
439 |
|
440 |
=head2 size |
441 |
|
442 |
Returns number of records in database |
443 |
|
444 |
print $isis->size; |
445 |
|
446 |
Result from this function can be used to loop through all records |
447 |
|
448 |
foreach my $mfn ( 1 ... $isis->size ) { ... } |
449 |
|
450 |
because it takes into account C<offset> and C<limit>. |
451 |
|
452 |
=cut |
453 |
|
454 |
sub size { |
455 |
my $self = shift; |
456 |
return $self->{size}; |
457 |
} |
458 |
|
459 |
=head2 seek |
460 |
|
461 |
Seek to specified MFN in file. |
462 |
|
463 |
$isis->seek(42); |
464 |
|
465 |
First record in database has position 1. |
466 |
|
467 |
=cut |
468 |
|
469 |
sub seek { |
470 |
my $self = shift; |
471 |
my $pos = shift || return; |
472 |
|
473 |
my $log = $self->_get_logger(); |
474 |
|
475 |
if ($pos < 1) { |
476 |
$log->warn("seek before first record"); |
477 |
$pos = 1; |
478 |
} elsif ($pos > $self->{max_pos}) { |
479 |
$log->warn("seek beyond last record"); |
480 |
$pos = $self->{max_pos}; |
481 |
} |
482 |
|
483 |
return $self->{pos} = (($pos - 1) || -1); |
484 |
} |
485 |
|
486 |
=head2 stats |
487 |
|
488 |
Dump statistics about field and subfield usage |
489 |
|
490 |
print $input->stats; |
491 |
|
492 |
=cut |
493 |
|
494 |
sub stats { |
495 |
my $self = shift; |
496 |
|
497 |
my $log = $self->_get_logger(); |
498 |
|
499 |
my $s = $self->{_stats}; |
500 |
if (! $s) { |
501 |
$log->warn("called stats, but there is no statistics collected"); |
502 |
return; |
503 |
} |
504 |
|
505 |
my $max_fld = 0; |
506 |
|
507 |
my $out = join("\n", |
508 |
map { |
509 |
my $f = $_ || die "no field"; |
510 |
my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}"; |
511 |
$max_fld = $v if ($v > $max_fld); |
512 |
|
513 |
my $o = sprintf("%4s %d ~", $f, $v); |
514 |
|
515 |
if (defined($s->{sf}->{$f})) { |
516 |
map { |
517 |
$o .= sprintf(" %s:%d%s", $_, |
518 |
$s->{sf}->{$f}->{$_}->{count}, |
519 |
$s->{sf}->{$f}->{$_}->{repeatable} ? '*' : '', |
520 |
); |
521 |
} sort keys %{ $s->{sf}->{$f} }; |
522 |
} |
523 |
|
524 |
if (my $v_r = $s->{repeatable}->{$f}) { |
525 |
$o .= " ($v_r)" if ($v_r != $v); |
526 |
} |
527 |
|
528 |
$o; |
529 |
} sort { $a cmp $b } keys %{ $s->{fld} } |
530 |
); |
531 |
|
532 |
$log->debug( sub { Dumper($s) } ); |
533 |
|
534 |
return $out; |
535 |
} |
536 |
|
537 |
=head2 modify_record_regexps |
538 |
|
539 |
Generate hash with regexpes to be applied using L<filter>. |
540 |
|
541 |
my $regexpes = $input->modify_record_regexps( |
542 |
900 => { '^a' => { ' : ' => '^b' } }, |
543 |
901 => { '*' => { '^b' => ' ; ' } }, |
544 |
); |
545 |
|
546 |
=cut |
547 |
|
548 |
sub modify_record_regexps { |
549 |
my $self = shift; |
550 |
my $modify_record = {@_}; |
551 |
|
552 |
my $regexpes; |
553 |
|
554 |
my $log = $self->_get_logger(); |
555 |
|
556 |
foreach my $f (keys %$modify_record) { |
557 |
$log->debug("field: $f"); |
558 |
|
559 |
foreach my $sf (keys %{ $modify_record->{$f} }) { |
560 |
$log->debug("subfield: $sf"); |
561 |
|
562 |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
563 |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
564 |
#die "no field?" unless defined($to); |
565 |
$log->debug("transform: |$from| -> |$to|"); |
566 |
|
567 |
if ($sf =~ /^\^/) { |
568 |
my $regex = |
569 |
's/\Q'. $sf .'\E(.*?)\Q'. $from .'\E(.*?)/'. $sf .'$1'. $to .'$2/g'; |
570 |
push @{ $regexpes->{$f} }, $regex; |
571 |
$log->debug(">>>>> $regex [sf]"); |
572 |
} else { |
573 |
my $regex = |
574 |
's/\Q'. $from .'\E/'. $to .'/g'; |
575 |
push @{ $regexpes->{$f} }, $regex; |
576 |
$log->debug(">>>>> $regex [global]"); |
577 |
} |
578 |
|
579 |
} |
580 |
} |
581 |
} |
582 |
|
583 |
return $regexpes; |
584 |
} |
585 |
|
586 |
=head1 MEMORY USAGE |
587 |
|
588 |
C<low_mem> options is double-edged sword. If enabled, WebPAC |
589 |
will run on memory constraint machines (which doesn't have enough |
590 |
physical RAM to create memory structure for whole source database). |
591 |
|
592 |
If your machine has 512Mb or more of RAM and database is around 10000 records, |
593 |
memory shouldn't be an issue. If you don't have enough physical RAM, you |
594 |
might consider using virtual memory (if your operating system is handling it |
595 |
well, like on FreeBSD or Linux) instead of dropping to L<DBM::Deep> to handle |
596 |
parsed structure of ISIS database (this is what C<low_mem> option does). |
597 |
|
598 |
Hitting swap at end of reading source database is probably o.k. However, |
599 |
hitting swap before 90% will dramatically decrease performance and you will |
600 |
be better off with C<low_mem> and using rest of availble memory for |
601 |
operating system disk cache (Linux is particuallary good about this). |
602 |
However, every access to database record will require disk access, so |
603 |
generation phase will be slower 10-100 times. |
604 |
|
605 |
Parsed structures are essential - you just have option to trade RAM memory |
606 |
(which is fast) for disk space (which is slow). Be sure to have planty of |
607 |
disk space if you are using C<low_mem> and thus L<DBM::Deep>. |
608 |
|
609 |
However, when WebPAC is running on desktop machines (or laptops :-), it's |
610 |
highly undesireable for system to start swapping. Using C<low_mem> option can |
611 |
reduce WecPAC memory usage to around 64Mb for same database with lookup |
612 |
fields and sorted indexes which stay in RAM. Performance will suffer, but |
613 |
memory usage will really be minimal. It might be also more confortable to |
614 |
run WebPAC reniced on those machines. |
615 |
|
616 |
|
617 |
=head1 AUTHOR |
618 |
|
619 |
Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >> |
620 |
|
621 |
=head1 COPYRIGHT & LICENSE |
622 |
|
623 |
Copyright 2005-2006 Dobrica Pavlinusic, All Rights Reserved. |
624 |
|
625 |
This program is free software; you can redistribute it and/or modify it |
626 |
under the same terms as Perl itself. |
627 |
|
628 |
=cut |
629 |
|
630 |
1; # End of WebPAC::Input |