16 |
|
|
17 |
=head1 VERSION |
=head1 VERSION |
18 |
|
|
19 |
Version 0.11 |
Version 0.14 |
20 |
|
|
21 |
=cut |
=cut |
22 |
|
|
23 |
our $VERSION = '0.11'; |
our $VERSION = '0.14'; |
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 = $self->{module}; |
my $module_path = $self->{module}; |
99 |
$module =~ s#::#/#g; |
$module_path =~ s#::#/#g; |
100 |
$module .= '.pm'; |
$module_path .= '.pm'; |
101 |
$log->debug("require low-level module $self->{module} from $module"); |
$log->debug("require low-level module $self->{module} from $module_path"); |
102 |
|
|
103 |
require $module; |
require $module_path; |
|
#eval $self->{module} .'->import'; |
|
104 |
|
|
105 |
# check if required subclasses are implemented |
# check if required subclasses are implemented |
106 |
foreach my $subclass (qw/open_db fetch_rec init/) { |
foreach my $subclass (qw/open_db fetch_rec init dump_rec/) { |
107 |
my $n = $self->{module} . '::' . $subclass; |
# FIXME |
|
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, @_); |
|
108 |
} |
} |
109 |
|
|
110 |
$self->{'encoding'} ||= 'ISO-8859-2'; |
$self->{'encoding'} ||= 'ISO-8859-2'; |
111 |
|
|
|
# running with low_mem flag? well, use DBM::Deep then. |
|
|
if ($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; |
|
|
} |
|
|
|
|
112 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
113 |
} |
} |
114 |
|
|
116 |
|
|
117 |
This function will read whole database in memory and produce lookups. |
This function will read whole database in memory and produce lookups. |
118 |
|
|
119 |
|
my $store; # simple in-memory hash |
120 |
|
|
121 |
$input->open( |
$input->open( |
122 |
path => '/path/to/database/file', |
path => '/path/to/database/file', |
123 |
code_page => 'cp852', |
code_page => 'cp852', |
124 |
limit => 500, |
limit => 500, |
125 |
offset => 6000, |
offset => 6000, |
|
lookup => $lookup_obj, |
|
126 |
stats => 1, |
stats => 1, |
127 |
lookup_ref => sub { |
lookup_coderef => sub { |
128 |
my ($k,$v) = @_; |
my $rec = shift; |
129 |
# store lookup $k => $v |
# store lookups |
130 |
}, |
}, |
131 |
modify_records => { |
modify_records => { |
132 |
900 => { '^a' => { ' : ' => '^b' } }, |
900 => { '^a' => { ' : ' => '^b' } }, |
133 |
901 => { '*' => { '^b' => ' ; ' } }, |
901 => { '*' => { '^b' => ' ; ' } }, |
134 |
}, |
}, |
135 |
|
modify_file => 'conf/modify/mapping.map', |
136 |
|
save_row => sub { |
137 |
|
my $a = shift; |
138 |
|
$store->{ $a->{id} } = $a->{row}; |
139 |
|
}, |
140 |
|
load_row => sub { |
141 |
|
my $a = shift; |
142 |
|
return defined($store->{ $a->{id} }) && |
143 |
|
$store->{ $a->{id} }; |
144 |
|
}, |
145 |
|
|
146 |
); |
); |
147 |
|
|
148 |
By default, C<code_page> is assumed to be C<cp852>. |
By default, C<code_page> is assumed to be C<cp852>. |
153 |
|
|
154 |
C<stats> create optional report about usage of fields and subfields |
C<stats> create optional report about usage of fields and subfields |
155 |
|
|
156 |
C<lookup_coderef> is closure to call when adding C<< key => 'value' >> combinations to |
C<lookup_coderef> is closure to called to save data into lookups |
|
lookup. |
|
157 |
|
|
158 |
C<modify_records> specify mapping from subfields to delimiters or from |
C<modify_records> specify mapping from subfields to delimiters or from |
159 |
delimiters to subfields, as well as oprations on fields (if subfield is |
delimiters to subfields, as well as oprations on fields (if subfield is |
160 |
defined as C<*>. |
defined as C<*>. |
161 |
|
|
162 |
|
C<modify_file> is alternative for C<modify_records> above which preserves order and offers |
163 |
|
(hopefully) simplier sintax than YAML or perl (see L</modify_file_regex>). This option |
164 |
|
overrides C<modify_records> if both exists for same input. |
165 |
|
|
166 |
|
C<save_row> and C<load_row> are low-level implementation of store engine. Calling convention |
167 |
|
is documented in example above. |
168 |
|
|
169 |
Returns size of database, regardless of C<offset> and C<limit> |
Returns size of database, regardless of C<offset> and C<limit> |
170 |
parametars, see also C<size>. |
parametars, see also C<size>. |
171 |
|
|
181 |
$log->logconfess("lookup_coderef must be CODE, not ",ref($arg->{lookup_coderef})) |
$log->logconfess("lookup_coderef must be CODE, not ",ref($arg->{lookup_coderef})) |
182 |
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
183 |
|
|
184 |
|
$log->debug( $arg->{lookup_coderef} ? '' : 'not ', "using lookup_coderef"); |
185 |
|
|
186 |
$log->logcroak("need path") if (! $arg->{'path'}); |
$log->logcroak("need path") if (! $arg->{'path'}); |
187 |
my $code_page = $arg->{'code_page'} || 'cp852'; |
my $code_page = $arg->{'code_page'} || 'cp852'; |
188 |
|
|
192 |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
193 |
} |
} |
194 |
|
|
195 |
|
if ($arg->{load_row} || $arg->{save_row}) { |
196 |
|
$log->logconfess("save_row and load_row must be defined in pair and be CODE") unless ( |
197 |
|
ref($arg->{load_row}) eq 'CODE' && |
198 |
|
ref($arg->{save_row}) eq 'CODE' |
199 |
|
); |
200 |
|
$self->{load_row} = $arg->{load_row}; |
201 |
|
$self->{save_row} = $arg->{save_row}; |
202 |
|
$log->debug("using load_row and save_row instead of in-memory hash"); |
203 |
|
} |
204 |
|
|
205 |
my $filter_ref; |
my $filter_ref; |
206 |
my $recode_regex; |
my $recode_regex; |
207 |
my $recode_map; |
my $recode_map; |
224 |
|
|
225 |
} |
} |
226 |
|
|
227 |
my $rec_regex = $self->modify_record_regexps(%{ $arg->{modify_records} }); |
my $rec_regex; |
228 |
$log->debug("rec_regex: ", Dumper($rec_regex)); |
if (my $p = $arg->{modify_file}) { |
229 |
|
$log->debug("using modify_file $p"); |
230 |
|
$rec_regex = $self->modify_file_regexps( $p ); |
231 |
|
} elsif (my $h = $arg->{modify_records}) { |
232 |
|
$log->debug("using modify_records ", Dumper( $h )); |
233 |
|
$rec_regex = $self->modify_record_regexps(%{ $h }); |
234 |
|
} |
235 |
|
$log->debug("rec_regex: ", Dumper($rec_regex)) if ($rec_regex); |
236 |
|
|
237 |
|
my $class = $self->{module} || $log->logconfess("can't get low-level module name!"); |
238 |
|
|
239 |
my ($db, $size) = $self->{open_db}->( $self, |
my $ll_db = $class->new( |
240 |
path => $arg->{path}, |
path => $arg->{path}, |
241 |
# filter => sub { |
# filter => sub { |
242 |
# my ($l,$f_nr) = @_; |
# my ($l,$f_nr) = @_; |
248 |
%{ $arg }, |
%{ $arg }, |
249 |
); |
); |
250 |
|
|
251 |
unless (defined($db)) { |
unless (defined($ll_db)) { |
252 |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
253 |
return; |
return; |
254 |
} |
} |
255 |
|
|
256 |
|
my $size = $ll_db->size; |
257 |
|
|
258 |
unless ($size) { |
unless ($size) { |
259 |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
260 |
return; |
return; |
286 |
|
|
287 |
$log->debug("position: $pos\n"); |
$log->debug("position: $pos\n"); |
288 |
|
|
289 |
my $rec = $self->{fetch_rec}->($self, $db, $pos, sub { |
my $rec = $ll_db->fetch_rec($pos, sub { |
290 |
my ($l,$f_nr) = @_; |
my ($l,$f_nr) = @_; |
291 |
# return unless defined($l); |
# return unless defined($l); |
292 |
# return $l unless ($rec_regex && $f_nr); |
# return $l unless ($rec_regex && $f_nr); |
294 |
$log->debug("-=> $f_nr ## $l"); |
$log->debug("-=> $f_nr ## $l"); |
295 |
|
|
296 |
# codepage conversion and recode_regex |
# codepage conversion and recode_regex |
297 |
# from_to($l, $code_page, $self->{'encoding'}); |
from_to($l, $code_page, $self->{'encoding'}); |
|
from_to($l, $code_page, 'utf-8'); |
|
298 |
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
299 |
|
|
300 |
# apply regexps |
# apply regexps |
302 |
$log->logconfess("regexps->{$f_nr} must be ARRAY") if (ref($rec_regex->{$f_nr}) ne 'ARRAY'); |
$log->logconfess("regexps->{$f_nr} must be ARRAY") if (ref($rec_regex->{$f_nr}) ne 'ARRAY'); |
303 |
my $c = 0; |
my $c = 0; |
304 |
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
305 |
#$log->debug("\$l = $l\neval \$l =~ $r"); |
my $old_l = $l; |
306 |
eval '$l =~ ' . $r; |
eval '$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: $r") if ($@); |
$log->error("error applying regex: $r") if ($@); |
311 |
} |
} |
312 |
} |
} |
323 |
} |
} |
324 |
|
|
325 |
# store |
# store |
326 |
if ($self->{low_mem}) { |
if ($self->{save_row}) { |
327 |
$self->{db}->put($pos, $rec); |
$self->{save_row}->({ |
328 |
|
id => $pos, |
329 |
|
row => $rec, |
330 |
|
}); |
331 |
} else { |
} else { |
332 |
$self->{data}->{$pos} = $rec; |
$self->{data}->{$pos} = $rec; |
333 |
} |
} |
339 |
if ($self->{stats}) { |
if ($self->{stats}) { |
340 |
|
|
341 |
# fetch clean record with regexpes applied for statistics |
# fetch clean record with regexpes applied for statistics |
342 |
my $rec = $self->{fetch_rec}->($self, $db, $pos); |
my $rec = $ll_db->fetch_rec($pos); |
343 |
|
|
344 |
foreach my $fld (keys %{ $rec }) { |
foreach my $fld (keys %{ $rec }) { |
345 |
$self->{_stats}->{fld}->{ $fld }++; |
$self->{_stats}->{fld}->{ $fld }++; |
376 |
$self->{max_pos} = $to_rec; |
$self->{max_pos} = $to_rec; |
377 |
$log->debug("max_pos: $to_rec"); |
$log->debug("max_pos: $to_rec"); |
378 |
|
|
379 |
|
# save for dump |
380 |
|
$self->{ll_db} = $ll_db; |
381 |
|
|
382 |
return $size; |
return $size; |
383 |
} |
} |
384 |
|
|
418 |
|
|
419 |
my $rec; |
my $rec; |
420 |
|
|
421 |
if ($self->{low_mem}) { |
if ($self->{load_row}) { |
422 |
$rec = $self->{db}->get($mfn); |
$rec = $self->{load_row}->({ id => $mfn }); |
423 |
} else { |
} else { |
424 |
$rec = $self->{data}->{$mfn}; |
$rec = $self->{data}->{$mfn}; |
425 |
} |
} |
540 |
return $out; |
return $out; |
541 |
} |
} |
542 |
|
|
543 |
|
=head2 dump |
544 |
|
|
545 |
|
Display humanly readable dump of record |
546 |
|
|
547 |
|
=cut |
548 |
|
|
549 |
|
sub dump { |
550 |
|
my $self = shift; |
551 |
|
|
552 |
|
return $self->{ll_db}->dump_rec( $self->{pos} ); |
553 |
|
|
554 |
|
} |
555 |
|
|
556 |
=head2 modify_record_regexps |
=head2 modify_record_regexps |
557 |
|
|
558 |
Generate hash with regexpes to be applied using L<filter>. |
Generate hash with regexpes to be applied using l<filter>. |
559 |
|
|
560 |
my $regexpes = $input->modify_record_regexps( |
my $regexpes = $input->modify_record_regexps( |
561 |
900 => { '^a' => { ' : ' => '^b' } }, |
900 => { '^a' => { ' : ' => '^b' } }, |
564 |
|
|
565 |
=cut |
=cut |
566 |
|
|
567 |
|
sub _get_regex { |
568 |
|
my ($sf,$from,$to) = @_; |
569 |
|
if ($sf =~ /^\^/) { |
570 |
|
return |
571 |
|
's/\Q'. $sf .'\E([^\^]*?)\Q'. $from .'\E([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
572 |
|
} else { |
573 |
|
return |
574 |
|
's/\Q'. $from .'\E/'. $to .'/g'; |
575 |
|
} |
576 |
|
} |
577 |
|
|
578 |
sub modify_record_regexps { |
sub modify_record_regexps { |
579 |
my $self = shift; |
my $self = shift; |
580 |
my $modify_record = {@_}; |
my $modify_record = {@_}; |
581 |
|
|
582 |
my $regexpes; |
my $regexpes; |
583 |
|
|
584 |
|
my $log = $self->_get_logger(); |
585 |
|
|
586 |
foreach my $f (keys %$modify_record) { |
foreach my $f (keys %$modify_record) { |
587 |
warn "--- f: $f\n"; |
$log->debug("field: $f"); |
588 |
|
|
589 |
foreach my $sf (keys %{ $modify_record->{$f} }) { |
foreach my $sf (keys %{ $modify_record->{$f} }) { |
590 |
warn "---- sf: $sf\n"; |
$log->debug("subfield: $sf"); |
591 |
|
|
592 |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
593 |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
594 |
#die "no field?" unless defined($to); |
#die "no field?" unless defined($to); |
595 |
warn "----- transform: |$from| -> |$to|\n"; |
$log->debug("transform: |$from| -> |$to|"); |
|
|
|
|
if ($sf =~ /^\^/) { |
|
|
my $regex = |
|
|
's/\Q'. $sf .'\E([^\^]+)\Q'. $from .'\E([^\^]+)/'. $sf .'$1'. $to .'$2/g'; |
|
|
push @{ $regexpes->{$f} }, $regex; |
|
|
warn ">>>>> $regex [sf]\n"; |
|
|
} else { |
|
|
my $regex = |
|
|
's/\Q'. $from .'\E/'. $to .'/g'; |
|
|
push @{ $regexpes->{$f} }, $regex; |
|
|
warn ">>>>> $regex [global]\n"; |
|
|
} |
|
596 |
|
|
597 |
|
my $regex = _get_regex($sf,$from,$to); |
598 |
|
push @{ $regexpes->{$f} }, $regex; |
599 |
|
$log->debug("regex: $regex"); |
600 |
} |
} |
601 |
} |
} |
602 |
} |
} |
604 |
return $regexpes; |
return $regexpes; |
605 |
} |
} |
606 |
|
|
607 |
=head1 MEMORY USAGE |
=head2 modify_file_regexps |
608 |
|
|
609 |
|
Generate hash with regexpes to be applied using l<filter> from |
610 |
|
pseudo hash/yaml format for regex mappings. |
611 |
|
|
612 |
|
It should be obvious: |
613 |
|
|
614 |
C<low_mem> options is double-edged sword. If enabled, WebPAC |
200 |
615 |
will run on memory constraint machines (which doesn't have enough |
'^a' |
616 |
physical RAM to create memory structure for whole source database). |
' : ' => '^e' |
617 |
|
' = ' => '^d' |
|
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. |
|
618 |
|
|
619 |
|
In field I<200> find C<'^a'> and then C<' : '>, and replace it with C<'^e'>. |
620 |
|
In field I<200> find C<'^a'> and then C<' = '>, and replace it with C<'^d'>. |
621 |
|
|
622 |
|
my $regexpes = $input->modify_file_regexps( 'conf/modify/common.pl' ); |
623 |
|
|
624 |
|
On undef path it will just return. |
625 |
|
|
626 |
|
=cut |
627 |
|
|
628 |
|
sub modify_file_regexps { |
629 |
|
my $self = shift; |
630 |
|
|
631 |
|
my $modify_path = shift || return; |
632 |
|
|
633 |
|
my $log = $self->_get_logger(); |
634 |
|
|
635 |
|
my $regexpes; |
636 |
|
|
637 |
|
CORE::open(my $fh, $modify_path) || $log->logdie("can't open modify file $modify_path: $!"); |
638 |
|
|
639 |
|
my ($f,$sf); |
640 |
|
|
641 |
|
while(<$fh>) { |
642 |
|
chomp; |
643 |
|
next if (/^#/ || /^\s*$/); |
644 |
|
|
645 |
|
if (/^\s*(\d+)\s*$/) { |
646 |
|
$f = $1; |
647 |
|
$log->debug("field: $f"); |
648 |
|
next; |
649 |
|
} elsif (/^\s*'([^']*)'\s*$/) { |
650 |
|
$sf = $1; |
651 |
|
$log->die("can't define subfiled before field in: $_") unless ($f); |
652 |
|
$log->debug("subfield: $sf"); |
653 |
|
} elsif (/^\s*'([^']*)'\s*=>\s*'([^']*)'\s*$/) { |
654 |
|
my ($from,$to) = ($1, $2); |
655 |
|
|
656 |
|
$log->debug("transform: |$from| -> |$to|"); |
657 |
|
|
658 |
|
my $regex = _get_regex($sf,$from,$to); |
659 |
|
push @{ $regexpes->{$f} }, $regex; |
660 |
|
$log->debug("regex: $regex"); |
661 |
|
} |
662 |
|
} |
663 |
|
|
664 |
|
return $regexpes; |
665 |
|
} |
666 |
|
|
667 |
=head1 AUTHOR |
=head1 AUTHOR |
668 |
|
|