7 |
|
|
8 |
use WebPAC::Common; |
use WebPAC::Common; |
9 |
use base qw/WebPAC::Common/; |
use base qw/WebPAC::Common/; |
10 |
use Data::Dumper; |
use Data::Dump qw/dump/; |
11 |
use Encode qw/from_to/; |
use Encode qw/from_to/; |
12 |
|
|
13 |
=head1 NAME |
=head1 NAME |
16 |
|
|
17 |
=head1 VERSION |
=head1 VERSION |
18 |
|
|
19 |
Version 0.13 |
Version 0.18 |
20 |
|
|
21 |
=cut |
=cut |
22 |
|
|
23 |
our $VERSION = '0.13'; |
our $VERSION = '0.18'; |
24 |
|
|
25 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
26 |
|
|
43 |
|
|
44 |
my $db = WebPAC::Input->new( |
my $db = WebPAC::Input->new( |
45 |
module => 'WebPAC::Input::ISIS', |
module => 'WebPAC::Input::ISIS', |
|
low_mem => 1, |
|
46 |
); |
); |
47 |
|
|
48 |
$db->open( path => '/path/to/database' ); |
$db->open( path => '/path/to/database' ); |
62 |
my $db = new WebPAC::Input( |
my $db = new WebPAC::Input( |
63 |
module => 'WebPAC::Input::MARC', |
module => 'WebPAC::Input::MARC', |
64 |
encoding => '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 |
); |
); |
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 |
|
|
92 |
|
|
93 |
$log->logconfess("code_page argument is not suppored any more. change it to encoding") if ($self->{lookup}); |
$log->logconfess("code_page argument is not suppored any more. change it to encoding") if ($self->{lookup}); |
94 |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if ($self->{lookup}); |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if ($self->{lookup}); |
95 |
|
$log->logconfess("low_mem argument is not suppored any more. rewrite it to load_row and save_row") if ($self->{low_mem}); |
96 |
|
|
97 |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
98 |
my $module_path = $self->{module}; |
my $module_path = $self->{module}; |
102 |
|
|
103 |
require $module_path; |
require $module_path; |
104 |
|
|
|
# check if required subclasses are implemented |
|
|
foreach my $subclass (qw/open_db fetch_rec init dump_rec/) { |
|
|
# FIXME |
|
|
} |
|
|
|
|
105 |
$self->{'encoding'} ||= 'ISO-8859-2'; |
$self->{'encoding'} ||= 'ISO-8859-2'; |
106 |
|
|
|
# running with low_mem flag? well, use DBM::Deep then. |
|
|
if ($self->{'low_mem'}) { |
|
|
$log->info("running with low_mem which impacts performance (<32 Mb memory usage)"); |
|
|
|
|
|
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; |
|
|
|
|
|
my $db = new DBM::Deep $db_file; |
|
|
|
|
|
$log->logdie("DBM::Deep error: $!") unless ($db); |
|
|
|
|
|
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; |
|
|
} |
|
|
|
|
107 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
108 |
} |
} |
109 |
|
|
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 => 'cp852', |
code_page => 'cp852', |
128 |
901 => { '*' => { '^b' => ' ; ' } }, |
901 => { '*' => { '^b' => ' ; ' } }, |
129 |
}, |
}, |
130 |
modify_file => 'conf/modify/mapping.map', |
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<cp852>. |
By default, C<code_page> is assumed to be C<cp852>. |
158 |
(hopefully) simplier sintax than YAML or perl (see L</modify_file_regex>). This option |
(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. |
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 |
|
|
187 |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
188 |
} |
} |
189 |
|
|
190 |
|
if ($arg->{load_row} || $arg->{save_row}) { |
191 |
|
$log->logconfess("save_row and load_row must be defined in pair and be CODE") unless ( |
192 |
|
ref($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; |
my $recode_regex; |
202 |
my $recode_map; |
my $recode_map; |
224 |
$log->debug("using modify_file $p"); |
$log->debug("using modify_file $p"); |
225 |
$rec_regex = $self->modify_file_regexps( $p ); |
$rec_regex = $self->modify_file_regexps( $p ); |
226 |
} elsif (my $h = $arg->{modify_records}) { |
} elsif (my $h = $arg->{modify_records}) { |
227 |
$log->debug("using modify_records ", Dumper( $h )); |
$log->debug("using modify_records ", sub { dump( $h ) }); |
228 |
$rec_regex = $self->modify_record_regexps(%{ $h }); |
$rec_regex = $self->modify_record_regexps(%{ $h }); |
229 |
} |
} |
230 |
$log->debug("rec_regex: ", Dumper($rec_regex)) if ($rec_regex); |
$log->debug("rec_regex: ", sub { dump($rec_regex) }) if ($rec_regex); |
231 |
|
|
232 |
my $class = $self->{module} || $log->logconfess("can't get low-level module name!"); |
my $class = $self->{module} || $log->logconfess("can't get low-level module name!"); |
233 |
|
|
282 |
$log->debug("position: $pos\n"); |
$log->debug("position: $pos\n"); |
283 |
|
|
284 |
my $rec = $ll_db->fetch_rec($pos, sub { |
my $rec = $ll_db->fetch_rec($pos, sub { |
285 |
my ($l,$f_nr) = @_; |
my ($l,$f_nr,$debug) = @_; |
286 |
# return unless defined($l); |
# return unless defined($l); |
287 |
# return $l unless ($rec_regex && $f_nr); |
# 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"); |
$log->debug("-=> $f_nr ## $l"); |
293 |
|
|
294 |
# codepage conversion and recode_regex |
# codepage conversion and recode_regex |
301 |
my $c = 0; |
my $c = 0; |
302 |
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
303 |
my $old_l = $l; |
my $old_l = $l; |
304 |
eval '$l =~ ' . $r; |
$log->logconfess("expected regex in ", dump( $r )) unless defined($r->{regex}); |
305 |
|
eval '$l =~ ' . $r->{regex}; |
306 |
if ($old_l ne $l) { |
if ($old_l ne $l) { |
307 |
$log->debug("REGEX on $f_nr eval \$l =~ $r\n## old l: [$old_l]\n## new l: [$l]"); |
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: $r") if ($@); |
$log->error("error applying regex: $r") if ($@); |
315 |
} |
} |
316 |
} |
} |
317 |
|
|
318 |
$log->debug("<=- $f_nr ## $l"); |
$log->debug("<=- $f_nr ## |$l|"); |
319 |
|
warn "<=- $f_nr ## $l\n" if ($debug); |
320 |
return $l; |
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 |
} |
} |
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; |
542 |
} sort { $a cmp $b } keys %{ $s->{fld} } |
} sort { $a cmp $b } keys %{ $s->{fld} } |
543 |
); |
); |
544 |
|
|
545 |
$log->debug( sub { Dumper($s) } ); |
$log->debug( sub { dump($s) } ); |
546 |
|
|
547 |
return $out; |
return $out; |
548 |
} |
} |
549 |
|
|
550 |
=head2 dump |
=head2 dump_ascii |
551 |
|
|
552 |
Display humanly readable dump of record |
Display humanly readable dump of record |
553 |
|
|
554 |
=cut |
=cut |
555 |
|
|
556 |
sub dump { |
sub dump_ascii { |
557 |
my $self = shift; |
my $self = shift; |
558 |
|
|
559 |
return $self->{dump_rec}->($self, $self->{pos}); |
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 modify_record_regexps |
=head2 _get_regex |
569 |
|
|
570 |
Generate hash with regexpes to be applied using l<filter>. |
Helper function called which create regexps to be execute on code. |
571 |
|
|
572 |
my $regexpes = $input->modify_record_regexps( |
_get_regex( 900, 'regex:[0-9]+' ,'numbers' ); |
573 |
900 => { '^a' => { ' : ' => '^b' } }, |
_get_regex( 900, '^b', ' : ^b' ); |
574 |
901 => { '*' => { '^b' => ' ; ' } }, |
|
575 |
); |
It supports perl regexps with C<regex:> prefix to from value and has |
576 |
|
additional logic to skip empty subfields. |
577 |
|
|
578 |
=cut |
=cut |
579 |
|
|
580 |
sub _get_regex { |
sub _get_regex { |
581 |
my ($sf,$from,$to) = @_; |
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 =~ /^\^/) { |
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 |
return |
598 |
's/\Q'. $sf .'\E([^\^]*?)\Q'. $from .'\E([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
's/\Q'. $sf .'\E([^\^]' . $need_subfield_data . '?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
599 |
} else { |
} else { |
600 |
return |
return |
601 |
's/\Q'. $from .'\E/'. $to .'/g'; |
'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 { |
sub modify_record_regexps { |
618 |
my $self = shift; |
my $self = shift; |
619 |
my $modify_record = {@_}; |
my $modify_record = {@_}; |
631 |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
632 |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
633 |
#die "no field?" unless defined($to); |
#die "no field?" unless defined($to); |
634 |
$log->debug("transform: |$from| -> |$to|"); |
my $d = "|$from| -> |$to|"; |
635 |
|
$log->debug("transform: $d"); |
636 |
|
|
637 |
my $regex = _get_regex($sf,$from,$to); |
my $regex = _get_regex($sf,$from,$to); |
638 |
push @{ $regexpes->{$f} }, $regex; |
push @{ $regexpes->{$f} }, { regex => $regex, debug => $d }; |
639 |
$log->debug("regex: $regex"); |
$log->debug("regex: $regex"); |
640 |
} |
} |
641 |
} |
} |
646 |
|
|
647 |
=head2 modify_file_regexps |
=head2 modify_file_regexps |
648 |
|
|
649 |
Generate hash with regexpes to be applied using l<filter> from |
Generate hash with regexpes to be applied using L<filter> from |
650 |
pseudo hash/yaml format for regex mappings. |
pseudo hash/yaml format for regex mappings. |
651 |
|
|
652 |
It should be obvious: |
It should be obvious: |
696 |
$log->debug("transform: |$from| -> |$to|"); |
$log->debug("transform: |$from| -> |$to|"); |
697 |
|
|
698 |
my $regex = _get_regex($sf,$from,$to); |
my $regex = _get_regex($sf,$from,$to); |
699 |
push @{ $regexpes->{$f} }, $regex; |
push @{ $regexpes->{$f} }, { |
700 |
|
regex => $regex, |
701 |
|
file => $modify_path, |
702 |
|
line => $., |
703 |
|
}; |
704 |
$log->debug("regex: $regex"); |
$log->debug("regex: $regex"); |
705 |
} |
} |
706 |
} |
} |
708 |
return $regexpes; |
return $regexpes; |
709 |
} |
} |
710 |
|
|
|
=head1 MEMORY USAGE |
|
|
|
|
|
C<low_mem> options is double-edged sword. If enabled, WebPAC |
|
|
will run on memory constraint machines (which doesn't have enough |
|
|
physical RAM to create memory structure for whole source database). |
|
|
|
|
|
If your machine has 512Mb or more of RAM and database is around 10000 records, |
|
|
memory shouldn't be an issue. If you don't have enough physical RAM, you |
|
|
might consider using virtual memory (if your operating system is handling it |
|
|
well, like on FreeBSD or Linux) instead of dropping to L<DBM::Deep> to handle |
|
|
parsed structure of ISIS database (this is what C<low_mem> option does). |
|
|
|
|
|
Hitting swap at end of reading source database is probably o.k. However, |
|
|
hitting swap before 90% will dramatically decrease performance and you will |
|
|
be better off with C<low_mem> and using rest of availble memory for |
|
|
operating system disk cache (Linux is particuallary good about this). |
|
|
However, every access to database record will require disk access, so |
|
|
generation phase will be slower 10-100 times. |
|
|
|
|
|
Parsed structures are essential - you just have option to trade RAM memory |
|
|
(which is fast) for disk space (which is slow). Be sure to have planty of |
|
|
disk space if you are using C<low_mem> and thus L<DBM::Deep>. |
|
|
|
|
|
However, when WebPAC is running on desktop machines (or laptops :-), it's |
|
|
highly undesireable for system to start swapping. Using C<low_mem> option can |
|
|
reduce WecPAC memory usage to around 64Mb for same database with lookup |
|
|
fields and sorted indexes which stay in RAM. Performance will suffer, but |
|
|
memory usage will really be minimal. It might be also more confortable to |
|
|
run WebPAC reniced on those machines. |
|
|
|
|
|
|
|
711 |
=head1 AUTHOR |
=head1 AUTHOR |
712 |
|
|
713 |
Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >> |
Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >> |