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 0.17 |
20 |
|
|
21 |
=cut |
=cut |
22 |
|
|
23 |
our $VERSION = '0.13'; |
our $VERSION = '0.17'; |
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 |
|
|
107 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
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 |
|
|
276 |
|
|
277 |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{encoding}", $self->{stats} ? ' [stats]' : ''); |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{encoding}", $self->{stats} ? ' [stats]' : ''); |
278 |
|
|
|
# turn on low_mem for databases with more than 100000 records! |
|
|
if (! $self->{low_mem} && $size > 100000) { |
|
|
$log->warn("Using on-disk storage instead of memory for input data. This will affect performance."); |
|
|
$self->{low_mem}++; |
|
|
} |
|
|
|
|
|
# running with low_mem flag? well, use DBM::Deep then. |
|
|
if ($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; |
|
|
} |
|
|
|
|
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 = $ll_db->fetch_rec($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 |
|
warn "-=> $f_nr ## $l\n" if ($debug); |
290 |
$log->debug("-=> $f_nr ## $l"); |
$log->debug("-=> $f_nr ## $l"); |
291 |
|
|
292 |
# codepage conversion and recode_regex |
# codepage conversion and recode_regex |
302 |
eval '$l =~ ' . $r; |
eval '$l =~ ' . $r; |
303 |
if ($old_l ne $l) { |
if ($old_l ne $l) { |
304 |
$log->debug("REGEX on $f_nr eval \$l =~ $r\n## old l: [$old_l]\n## new l: [$l]"); |
$log->debug("REGEX on $f_nr eval \$l =~ $r\n## old l: [$old_l]\n## new l: [$l]"); |
305 |
|
warn "*** $r |$old_l| -> |$l|\n" if ($debug); |
306 |
} |
} |
307 |
$log->error("error applying regex: $r") if ($@); |
$log->error("error applying regex: $r") if ($@); |
308 |
} |
} |
309 |
} |
} |
310 |
|
|
311 |
$log->debug("<=- $f_nr ## $l"); |
$log->debug("<=- $f_nr ## $l"); |
312 |
|
warn "<=- $f_nr ## $l\n" if ($debug); |
313 |
return $l; |
return $l; |
314 |
}); |
}); |
315 |
|
|
316 |
$log->debug(sub { Dumper($rec) }); |
$log->debug(sub { dump($rec) }); |
317 |
|
|
318 |
if (! $rec) { |
if (! $rec) { |
319 |
$log->warn("record $pos empty? skipping..."); |
$log->warn("record $pos empty? skipping..."); |
321 |
} |
} |
322 |
|
|
323 |
# store |
# store |
324 |
if ($self->{low_mem}) { |
if ($self->{save_row}) { |
325 |
$self->{db}->put($pos, $rec); |
$self->{save_row}->({ |
326 |
|
id => $pos, |
327 |
|
row => $rec, |
328 |
|
}); |
329 |
} else { |
} else { |
330 |
$self->{data}->{$pos} = $rec; |
$self->{data}->{$pos} = $rec; |
331 |
} |
} |
416 |
|
|
417 |
my $rec; |
my $rec; |
418 |
|
|
419 |
if ($self->{low_mem}) { |
if ($self->{load_row}) { |
420 |
$rec = $self->{db}->get($mfn); |
$rec = $self->{load_row}->({ id => $mfn }); |
421 |
} else { |
} else { |
422 |
$rec = $self->{data}->{$mfn}; |
$rec = $self->{data}->{$mfn}; |
423 |
} |
} |
472 |
|
|
473 |
sub seek { |
sub seek { |
474 |
my $self = shift; |
my $self = shift; |
475 |
my $pos = shift || return; |
my $pos = shift; |
476 |
|
|
477 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
478 |
|
|
479 |
|
$log->logconfess("called without pos") unless defined($pos); |
480 |
|
|
481 |
if ($pos < 1) { |
if ($pos < 1) { |
482 |
$log->warn("seek before first record"); |
$log->warn("seek before first record"); |
483 |
$pos = 1; |
$pos = 1; |
535 |
} sort { $a cmp $b } keys %{ $s->{fld} } |
} sort { $a cmp $b } keys %{ $s->{fld} } |
536 |
); |
); |
537 |
|
|
538 |
$log->debug( sub { Dumper($s) } ); |
$log->debug( sub { dump($s) } ); |
539 |
|
|
540 |
return $out; |
return $out; |
541 |
} |
} |
542 |
|
|
543 |
=head2 dump |
=head2 dump_ascii |
544 |
|
|
545 |
Display humanly readable dump of record |
Display humanly readable dump of record |
546 |
|
|
547 |
=cut |
=cut |
548 |
|
|
549 |
sub dump { |
sub dump_ascii { |
550 |
my $self = shift; |
my $self = shift; |
551 |
|
|
552 |
return $self->{ll_db}->dump_rec( $self->{pos} ); |
return unless $self->{ll_db}; |
553 |
|
|
554 |
|
if ($self->{ll_db}->can('dump_rec')) { |
555 |
|
return $self->{ll_db}->dump_ascii( $self->{pos} ); |
556 |
|
} else { |
557 |
|
return dump( $self->{ll_db}->fetch_rec( $self->{pos} ) ); |
558 |
|
} |
559 |
} |
} |
560 |
|
|
561 |
=head2 modify_record_regexps |
=head2 modify_record_regexps |
562 |
|
|
563 |
Generate hash with regexpes to be applied using l<filter>. |
Generate hash with regexpes to be applied using L<filter>. |
564 |
|
|
565 |
my $regexpes = $input->modify_record_regexps( |
my $regexpes = $input->modify_record_regexps( |
566 |
900 => { '^a' => { ' : ' => '^b' } }, |
900 => { '^a' => { ' : ' => '^b' } }, |
571 |
|
|
572 |
sub _get_regex { |
sub _get_regex { |
573 |
my ($sf,$from,$to) = @_; |
my ($sf,$from,$to) = @_; |
574 |
|
|
575 |
|
# protect / |
576 |
|
$from =~ s!/!\\/!gs; |
577 |
|
$to =~ s!/!\\/!gs; |
578 |
|
|
579 |
|
if ($from =~ m/^regex:(.+)$/) { |
580 |
|
$from = $1; |
581 |
|
} else { |
582 |
|
$from = '\Q' . $from . '\E'; |
583 |
|
} |
584 |
if ($sf =~ /^\^/) { |
if ($sf =~ /^\^/) { |
585 |
return |
return |
586 |
's/\Q'. $sf .'\E([^\^]*?)\Q'. $from .'\E([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
's/\Q'. $sf .'\E([^\^]*?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
587 |
} else { |
} else { |
588 |
return |
return |
589 |
's/\Q'. $from .'\E/'. $to .'/g'; |
's/'. $from .'/'. $to .'/g'; |
590 |
} |
} |
591 |
} |
} |
592 |
|
|
621 |
|
|
622 |
=head2 modify_file_regexps |
=head2 modify_file_regexps |
623 |
|
|
624 |
Generate hash with regexpes to be applied using l<filter> from |
Generate hash with regexpes to be applied using L<filter> from |
625 |
pseudo hash/yaml format for regex mappings. |
pseudo hash/yaml format for regex mappings. |
626 |
|
|
627 |
It should be obvious: |
It should be obvious: |
679 |
return $regexpes; |
return $regexpes; |
680 |
} |
} |
681 |
|
|
|
=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. |
|
|
|
|
|
|
|
682 |
=head1 AUTHOR |
=head1 AUTHOR |
683 |
|
|
684 |
Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >> |
Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >> |