9 use base qw/WebPAC::Common/;
11 use Encode qw/from_to/;
15 WebPAC::Input - read different file formats into WebPAC
23 our $VERSION = '0.11';
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>.
31 Simply, something that is indexed by unmber from 1 .. I<size>.
33 Examples of such databases are CDS/ISIS files, MARC files, lines in
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.
40 Perhaps a little code snippet.
44 my $db = WebPAC::Input->new(
45 module => 'WebPAC::Input::ISIS',
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
61 Create new input database object.
63 my $db = new WebPAC::Input(
64 module => 'WebPAC::Input::MARC',
65 encoding => 'ISO-8859-2',
67 recode => 'char pairs',
71 C<module> is low-level file format module. See L<WebPAC::Input::ISIS> and
72 L<WebPAC::Input::MARC>.
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>.
78 Default is not to use C<low_mem> options (see L<MEMORY USAGE> below).
80 C<recode> is optional string constisting of character or words pairs that
81 should be replaced in input stream.
83 C<no_progress_bar> disables progress bar output on C<STDOUT>
85 This function will also call low-level C<init> if it exists with same
95 my $log = $self->_get_logger;
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});
100 $log->logconfess("specify low-level file format module") unless ($self->{module});
101 my $module = $self->{module};
104 $log->debug("require low-level module $self->{module} from $module");
107 #eval $self->{module} .'->import';
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) };
116 $self->{$subclass} = \&{ $n };
121 $log->debug("calling init");
122 $self->{init}->($self, @_);
125 $self->{'encoding'} ||= 'ISO-8859-2';
127 # running with low_mem flag? well, use DBM::Deep then.
128 if ($self->{'low_mem'}) {
129 $log->info("running with low_mem which impacts performance (<32 Mb memory usage)");
131 my $db_file = "data.db";
134 unlink $db_file or $log->logdie("can't remove '$db_file' from last run");
135 $log->debug("removed '$db_file' from last run");
140 my $db = new DBM::Deep $db_file;
142 $log->logdie("DBM::Deep error: $!") unless ($db);
145 $log->logdie("can't open '$db_file' under low_mem: ",$db->error());
147 $log->debug("using file '$db_file' for DBM::Deep");
153 $self ? return $self : return undef;
158 This function will read whole database in memory and produce lookups.
161 path => '/path/to/database/file',
162 code_page => 'cp852',
165 lookup => $lookup_obj,
169 # store lookup $k => $v
172 900 => { '^a' => { ' : ' => '^b' } },
173 901 => { '*' => { '^b' => ' ; ' } },
177 By default, C<code_page> is assumed to be C<cp852>.
179 C<offset> is optional parametar to position at some offset before reading from database.
181 C<limit> is optional parametar to read just C<limit> records from database
183 C<stats> create optional report about usage of fields and subfields
185 C<lookup_coderef> is closure to call when adding C<< key => 'value' >> combinations to
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
192 Returns size of database, regardless of C<offset> and C<limit>
193 parametars, see also C<size>.
201 my $log = $self->_get_logger();
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');
207 $log->logcroak("need path") if (! $arg->{'path'});
208 my $code_page = $arg->{'code_page'} || 'cp852';
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});
220 if ($self->{recode}) {
221 my @r = split(/\s/, $self->{recode});
223 $log->logwarn("recode needs even number of elements (some number of valid pairs)");
228 $recode_map->{$from} = $to;
231 $recode_regex = join '|' => keys %{ $recode_map };
233 $log->debug("using recode regex: $recode_regex");
238 my $rec_regex = $self->modify_record_regexps(%{ $arg->{modify_records} });
239 $log->debug("rec_regex: ", Dumper($rec_regex));
241 my ($db, $size) = $self->{open_db}->( $self,
242 path => $arg->{path},
244 # my ($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 && $recode_map);
253 unless (defined($db)) {
254 $log->logwarn("can't open database $arg->{path}, skipping...");
259 $log->logwarn("no records in database $arg->{path}, skipping...");
266 if (my $s = $self->{offset}) {
267 $log->debug("skipping to MFN $s");
270 $self->{offset} = $from_rec;
273 if ($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);
279 # store size for later
280 $self->{size} = ($to_rec - $from_rec) ? ($to_rec - $from_rec + 1) : 0;
282 $log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{encoding}", $self->{stats} ? ' [stats]' : '');
285 for (my $pos = $from_rec; $pos <= $to_rec; $pos++) {
287 $log->debug("position: $pos\n");
289 my $rec = $self->{fetch_rec}->($self, $db, $pos, sub {
291 # return unless defined($l);
292 # return $l unless ($rec_regex && $f_nr);
294 $log->debug("-=> $f_nr ## $l");
296 # codepage conversion and recode_regex
297 # from_to($l, $code_page, $self->{'encoding'});
298 from_to($l, $code_page, 'utf-8');
299 $l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map);
302 if ($rec_regex && defined($rec_regex->{$f_nr})) {
303 $log->logconfess("regexps->{$f_nr} must be ARRAY") if (ref($rec_regex->{$f_nr}) ne 'ARRAY');
305 foreach my $r (@{ $rec_regex->{$f_nr} }) {
306 #$log->debug("\$l = $l\neval \$l =~ $r");
308 $log->error("error applying regex: $r") if ($@);
312 $log->debug("<=- $f_nr ## $l");
316 $log->debug(sub { Dumper($rec) });
319 $log->warn("record $pos empty? skipping...");
324 if ($self->{low_mem}) {
325 $self->{db}->put($pos, $rec);
327 $self->{data}->{$pos} = $rec;
331 $arg->{'lookup_coderef'}->( $rec ) if ($rec && $arg->{'lookup_coderef'});
333 # update counters for statistics
334 if ($self->{stats}) {
336 # fetch clean record with regexpes applied for statistics
337 my $rec = $self->{fetch_rec}->($self, $db, $pos);
339 foreach my $fld (keys %{ $rec }) {
340 $self->{_stats}->{fld}->{ $fld }++;
342 $log->logdie("invalid record fild $fld, not ARRAY")
343 unless (ref($rec->{ $fld }) eq 'ARRAY');
345 foreach my $row (@{ $rec->{$fld} }) {
347 if (ref($row) eq 'HASH') {
349 foreach my $sf (keys %{ $row }) {
350 next if ($sf eq 'subfields');
351 $self->{_stats}->{sf}->{ $fld }->{ $sf }->{count}++;
352 $self->{_stats}->{sf}->{ $fld }->{ $sf }->{repeatable}++
353 if (ref($row->{$sf}) eq 'ARRAY');
357 $self->{_stats}->{repeatable}->{ $fld }++;
363 $self->progress_bar($pos,$to_rec) unless ($self->{no_progress_bar});
368 $self->{last_pcnt} = 0;
370 # store max mfn and return it.
371 $self->{max_pos} = $to_rec;
372 $log->debug("max_pos: $to_rec");
379 Fetch next record from database. It will also displays progress bar.
381 my $rec = $isis->fetch;
383 Record from this function should probably go to C<data_structure> for
391 my $log = $self->_get_logger();
393 $log->logconfess("it seems that you didn't load database!") unless ($self->{pos});
395 if ($self->{pos} == -1) {
396 $self->{pos} = $self->{offset};
401 my $mfn = $self->{pos};
403 if ($mfn > $self->{max_pos}) {
404 $self->{pos} = $self->{max_pos};
405 $log->debug("at EOF");
409 $self->progress_bar($mfn,$self->{max_pos}) unless ($self->{no_progress_bar});
413 if ($self->{low_mem}) {
414 $rec = $self->{db}->get($mfn);
416 $rec = $self->{data}->{$mfn};
424 Returns current record number (MFN).
428 First record in database has position 1.
440 Returns number of records in database
444 Result from this function can be used to loop through all records
446 foreach my $mfn ( 1 ... $isis->size ) { ... }
448 because it takes into account C<offset> and C<limit>.
454 return $self->{size};
459 Seek to specified MFN in file.
463 First record in database has position 1.
469 my $pos = shift || return;
471 my $log = $self->_get_logger();
474 $log->warn("seek before first record");
476 } elsif ($pos > $self->{max_pos}) {
477 $log->warn("seek beyond last record");
478 $pos = $self->{max_pos};
481 return $self->{pos} = (($pos - 1) || -1);
486 Dump statistics about field and subfield usage
495 my $log = $self->_get_logger();
497 my $s = $self->{_stats};
499 $log->warn("called stats, but there is no statistics collected");
507 my $f = $_ || die "no field";
508 my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}";
509 $max_fld = $v if ($v > $max_fld);
511 my $o = sprintf("%4s %d ~", $f, $v);
513 if (defined($s->{sf}->{$f})) {
515 $o .= sprintf(" %s:%d%s", $_,
516 $s->{sf}->{$f}->{$_}->{count},
517 $s->{sf}->{$f}->{$_}->{repeatable} ? '*' : '',
519 } sort keys %{ $s->{sf}->{$f} };
522 if (my $v_r = $s->{repeatable}->{$f}) {
523 $o .= " ($v_r)" if ($v_r != $v);
527 } sort { $a cmp $b } keys %{ $s->{fld} }
530 $log->debug( sub { Dumper($s) } );
535 =head2 modify_record_regexps
537 Generate hash with regexpes to be applied using L<filter>.
539 my $regexpes = $input->modify_record_regexps(
540 900 => { '^a' => { ' : ' => '^b' } },
541 901 => { '*' => { '^b' => ' ; ' } },
546 sub modify_record_regexps {
548 my $modify_record = {@_};
552 foreach my $f (keys %$modify_record) {
554 foreach my $sf (keys %{ $modify_record->{$f} }) {
555 warn "---- sf: $sf\n";
556 foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) {
557 my $to = $modify_record->{$f}->{$sf}->{$from};
558 #die "no field?" unless defined($to);
559 warn "----- transform: |$from| -> |$to|\n";
563 's/\Q'. $sf .'\E([^\^]+)\Q'. $from .'\E([^\^]+)/'. $sf .'$1'. $to .'$2/g';
564 push @{ $regexpes->{$f} }, $regex;
565 warn ">>>>> $regex [sf]\n";
568 's/\Q'. $from .'\E/'. $to .'/g';
569 push @{ $regexpes->{$f} }, $regex;
570 warn ">>>>> $regex [global]\n";
582 C<low_mem> options is double-edged sword. If enabled, WebPAC
583 will run on memory constraint machines (which doesn't have enough
584 physical RAM to create memory structure for whole source database).
586 If your machine has 512Mb or more of RAM and database is around 10000 records,
587 memory shouldn't be an issue. If you don't have enough physical RAM, you
588 might consider using virtual memory (if your operating system is handling it
589 well, like on FreeBSD or Linux) instead of dropping to L<DBM::Deep> to handle
590 parsed structure of ISIS database (this is what C<low_mem> option does).
592 Hitting swap at end of reading source database is probably o.k. However,
593 hitting swap before 90% will dramatically decrease performance and you will
594 be better off with C<low_mem> and using rest of availble memory for
595 operating system disk cache (Linux is particuallary good about this).
596 However, every access to database record will require disk access, so
597 generation phase will be slower 10-100 times.
599 Parsed structures are essential - you just have option to trade RAM memory
600 (which is fast) for disk space (which is slow). Be sure to have planty of
601 disk space if you are using C<low_mem> and thus L<DBM::Deep>.
603 However, when WebPAC is running on desktop machines (or laptops :-), it's
604 highly undesireable for system to start swapping. Using C<low_mem> option can
605 reduce WecPAC memory usage to around 64Mb for same database with lookup
606 fields and sorted indexes which stay in RAM. Performance will suffer, but
607 memory usage will really be minimal. It might be also more confortable to
608 run WebPAC reniced on those machines.
613 Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >>
615 =head1 COPYRIGHT & LICENSE
617 Copyright 2005-2006 Dobrica Pavlinusic, All Rights Reserved.
619 This program is free software; you can redistribute it and/or modify it
620 under the same terms as Perl itself.
624 1; # End of WebPAC::Input