7 |
|
|
8 |
use WebPAC::Common; |
use WebPAC::Common; |
9 |
use base qw/WebPAC::Common/; |
use base qw/WebPAC::Common/; |
10 |
use Text::Iconv; |
use Data::Dump qw/dump/; |
11 |
|
use Encode qw/from_to/; |
12 |
|
|
13 |
=head1 NAME |
=head1 NAME |
14 |
|
|
15 |
WebPAC::Input - core module for input file format |
WebPAC::Input - read different file formats into WebPAC |
16 |
|
|
17 |
=head1 VERSION |
=head1 VERSION |
18 |
|
|
19 |
Version 0.02 |
Version 0.17 |
20 |
|
|
21 |
=cut |
=cut |
22 |
|
|
23 |
our $VERSION = '0.02'; |
our $VERSION = '0.17'; |
24 |
|
|
25 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
26 |
|
|
27 |
This module is used as base class for all database specific modules |
This module implements input as database which have fixed and known |
28 |
(basically, files which have one handle, fixed size while indexing and some |
I<size> while indexing and single unique numeric identifier for database |
29 |
kind of numeric idefinirier which goes from 1 to filesize). |
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. |
Perhaps a little code snippet. |
41 |
|
|
42 |
use WebPAC::Input; |
use WebPAC::Input; |
43 |
|
|
44 |
|
my $db = WebPAC::Input->new( |
45 |
|
module => 'WebPAC::Input::ISIS', |
46 |
|
); |
47 |
|
|
48 |
|
$db->open( path => '/path/to/database' ); |
49 |
|
print "database size: ",$db->size,"\n"; |
50 |
|
while (my $rec = $db->fetch) { |
51 |
|
# do something with $rec |
52 |
|
} |
53 |
|
|
54 |
|
|
|
my $db = WebPAC::Input->new( |
|
|
format => 'NULL', |
|
|
config => $config, |
|
|
lookup => $lookup_obj, |
|
|
low_mem => 1, |
|
|
); |
|
|
|
|
|
$db->open('/path/to/database'); |
|
|
print "database size: ",$db->size,"\n"; |
|
|
while (my $row = $db->fetch) { |
|
|
... |
|
|
} |
|
55 |
|
|
56 |
=head1 FUNCTIONS |
=head1 FUNCTIONS |
57 |
|
|
60 |
Create new input database object. |
Create new input database object. |
61 |
|
|
62 |
my $db = new WebPAC::Input( |
my $db = new WebPAC::Input( |
63 |
format => 'NULL' |
module => 'WebPAC::Input::MARC', |
64 |
code_page => 'ISO-8859-2', |
encoding => 'ISO-8859-2', |
65 |
low_mem => 1, |
recode => 'char pairs', |
66 |
|
no_progress_bar => 1, |
67 |
); |
); |
68 |
|
|
69 |
Optional parametar C<code_page> specify application code page (which will be |
C<module> is low-level file format module. See L<WebPAC::Input::ISIS> and |
70 |
|
L<WebPAC::Input::MARC>. |
71 |
|
|
72 |
|
Optional parametar C<encoding> specify application code page (which will be |
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 |
|
|
76 |
Default is not to use C<low_mem> options (see L<MEMORY USAGE> below). |
C<recode> is optional string constisting of character or words pairs that |
77 |
|
should be replaced in input stream. |
78 |
|
|
79 |
|
C<no_progress_bar> disables progress bar output on C<STDOUT> |
80 |
|
|
81 |
This function will also call low-level C<init> if it exists with same |
This function will also call low-level C<init> if it exists with same |
82 |
parametars. |
parametars. |
90 |
|
|
91 |
my $log = $self->_get_logger; |
my $log = $self->_get_logger; |
92 |
|
|
93 |
# check if required subclasses are implemented |
$log->logconfess("code_page argument is not suppored any more. change it to encoding") if ($self->{lookup}); |
94 |
foreach my $subclass (qw/open_db fetch_rec/) { |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if ($self->{lookup}); |
95 |
$log->logdie("missing implementation of $subclass") unless ($self->SUPER::can($subclass)); |
$log->logconfess("low_mem argument is not suppored any more. rewrite it to load_row and save_row") if ($self->{low_mem}); |
96 |
} |
|
97 |
|
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
98 |
|
my $module_path = $self->{module}; |
99 |
|
$module_path =~ s#::#/#g; |
100 |
|
$module_path .= '.pm'; |
101 |
|
$log->debug("require low-level module $self->{module} from $module_path"); |
102 |
|
|
103 |
if ($self->can('init')) { |
require $module_path; |
|
$log->debug("calling init"); |
|
|
$self->init(@_); |
|
|
} |
|
|
|
|
|
$self->{'code_page'} ||= 'ISO-8859-2'; |
|
104 |
|
|
105 |
# running with low_mem flag? well, use DBM::Deep then. |
$self->{'encoding'} ||= 'ISO-8859-2'; |
|
if ($self->{'low_mem'}) { |
|
|
$log->info("running with low_mem which impacts performance (<32 Mb memory usage)"); |
|
106 |
|
|
107 |
my $db_file = "data.db"; |
$self ? return $self : return undef; |
108 |
|
} |
109 |
|
|
110 |
if (-e $db_file) { |
=head2 open |
|
unlink $db_file or $log->logdie("can't remove '$db_file' from last run"); |
|
|
$log->debug("removed '$db_file' from last run"); |
|
|
} |
|
111 |
|
|
112 |
require DBM::Deep; |
This function will read whole database in memory and produce lookups. |
113 |
|
|
114 |
my $db = new DBM::Deep $db_file; |
my $store; # simple in-memory hash |
115 |
|
|
116 |
$log->logdie("DBM::Deep error: $!") unless ($db); |
$input->open( |
117 |
|
path => '/path/to/database/file', |
118 |
|
code_page => 'cp852', |
119 |
|
limit => 500, |
120 |
|
offset => 6000, |
121 |
|
stats => 1, |
122 |
|
lookup_coderef => sub { |
123 |
|
my $rec = shift; |
124 |
|
# store lookups |
125 |
|
}, |
126 |
|
modify_records => { |
127 |
|
900 => { '^a' => { ' : ' => '^b' } }, |
128 |
|
901 => { '*' => { '^b' => ' ; ' } }, |
129 |
|
}, |
130 |
|
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->{ $a->{id} }; |
139 |
|
}, |
140 |
|
|
141 |
if ($db->error()) { |
); |
|
$log->logdie("can't open '$db_file' under low_mem: ",$db->error()); |
|
|
} else { |
|
|
$log->debug("using file '$db_file' for DBM::Deep"); |
|
|
} |
|
142 |
|
|
143 |
$self->{'db'} = $db; |
By default, C<code_page> is assumed to be C<cp852>. |
|
} |
|
144 |
|
|
145 |
$self ? return $self : return undef; |
C<offset> is optional parametar to position at some offset before reading from database. |
|
} |
|
146 |
|
|
147 |
=head2 open |
C<limit> is optional parametar to read just C<limit> records from database |
148 |
|
|
149 |
This function will read whole database in memory and produce lookups. |
C<stats> create optional report about usage of fields and subfields |
150 |
|
|
151 |
$isis->open( |
C<lookup_coderef> is closure to called to save data into lookups |
|
path => '/path/to/database/file', |
|
|
code_page => '852', |
|
|
limit_mfn => 500, |
|
|
start_mfn => 6000, |
|
|
lookup => $lookup_obj, |
|
|
); |
|
152 |
|
|
153 |
By default, C<code_page> is assumed to be C<852>. |
C<modify_records> specify mapping from subfields to delimiters or from |
154 |
|
delimiters to subfields, as well as oprations on fields (if subfield is |
155 |
|
defined as C<*>. |
156 |
|
|
157 |
If optional parametar C<start_mfn> is set, this will be first MFN to read |
C<modify_file> is alternative for C<modify_records> above which preserves order and offers |
158 |
from database (so you can skip beginning of your database if you need to). |
(hopefully) simplier sintax than YAML or perl (see L</modify_file_regex>). This option |
159 |
|
overrides C<modify_records> if both exists for same input. |
160 |
|
|
161 |
If optional parametar C<limit_mfn> is set, it will read just 500 records |
C<save_row> and C<load_row> are low-level implementation of store engine. Calling convention |
162 |
from database in example above. |
is documented in example above. |
163 |
|
|
164 |
Returns size of database, regardless of C<start_mfn> and C<limit_mfn> |
Returns size of database, regardless of C<offset> and C<limit> |
165 |
parametars, see also C<$isis->size>. |
parametars, see also C<size>. |
166 |
|
|
167 |
=cut |
=cut |
168 |
|
|
172 |
|
|
173 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
174 |
|
|
175 |
|
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_coderef") if ($arg->{lookup}); |
176 |
|
$log->logconfess("lookup_coderef must be CODE, not ",ref($arg->{lookup_coderef})) |
177 |
|
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
178 |
|
|
179 |
|
$log->debug( $arg->{lookup_coderef} ? '' : 'not ', "using lookup_coderef"); |
180 |
|
|
181 |
$log->logcroak("need path") if (! $arg->{'path'}); |
$log->logcroak("need path") if (! $arg->{'path'}); |
182 |
my $code_page = $arg->{'code_page'} || '852'; |
my $code_page = $arg->{'code_page'} || 'cp852'; |
183 |
|
|
184 |
# store data in object |
# store data in object |
185 |
$self->{'code_page'} = $code_page; |
$self->{'input_code_page'} = $code_page; |
186 |
foreach my $v (qw/path start_mfn limit_mfn/) { |
foreach my $v (qw/path offset limit/) { |
187 |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
188 |
} |
} |
189 |
|
|
190 |
# create Text::Iconv object |
if ($arg->{load_row} || $arg->{save_row}) { |
191 |
$self->{iconv} = Text::Iconv->new($code_page,$self->{'code_page'}); |
$log->logconfess("save_row and load_row must be defined in pair and be CODE") unless ( |
192 |
|
ref($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; |
201 |
|
my $recode_regex; |
202 |
|
my $recode_map; |
203 |
|
|
204 |
|
if ($self->{recode}) { |
205 |
|
my @r = split(/\s/, $self->{recode}); |
206 |
|
if ($#r % 2 != 1) { |
207 |
|
$log->logwarn("recode needs even number of elements (some number of valid pairs)"); |
208 |
|
} else { |
209 |
|
while (@r) { |
210 |
|
my $from = shift @r; |
211 |
|
my $to = shift @r; |
212 |
|
$recode_map->{$from} = $to; |
213 |
|
} |
214 |
|
|
215 |
|
$recode_regex = join '|' => keys %{ $recode_map }; |
216 |
|
|
217 |
|
$log->debug("using recode regex: $recode_regex"); |
218 |
|
} |
219 |
|
|
220 |
|
} |
221 |
|
|
222 |
|
my $rec_regex; |
223 |
|
if (my $p = $arg->{modify_file}) { |
224 |
|
$log->debug("using modify_file $p"); |
225 |
|
$rec_regex = $self->modify_file_regexps( $p ); |
226 |
|
} elsif (my $h = $arg->{modify_records}) { |
227 |
|
$log->debug("using modify_records ", sub { dump( $h ) }); |
228 |
|
$rec_regex = $self->modify_record_regexps(%{ $h }); |
229 |
|
} |
230 |
|
$log->debug("rec_regex: ", sub { dump($rec_regex) }) if ($rec_regex); |
231 |
|
|
232 |
|
my $class = $self->{module} || $log->logconfess("can't get low-level module name!"); |
233 |
|
|
234 |
my ($db, $size) = $self->open_db( |
my $ll_db = $class->new( |
235 |
path => $arg->{path}, |
path => $arg->{path}, |
236 |
|
# filter => sub { |
237 |
|
# my ($l,$f_nr) = @_; |
238 |
|
# return unless defined($l); |
239 |
|
# from_to($l, $code_page, $self->{'encoding'}); |
240 |
|
# $l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
241 |
|
# return $l; |
242 |
|
# }, |
243 |
|
%{ $arg }, |
244 |
); |
); |
245 |
|
|
246 |
unless ($db) { |
unless (defined($ll_db)) { |
247 |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
248 |
return; |
return; |
249 |
} |
} |
250 |
|
|
251 |
|
my $size = $ll_db->size; |
252 |
|
|
253 |
unless ($size) { |
unless ($size) { |
254 |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
255 |
return; |
return; |
256 |
} |
} |
257 |
|
|
258 |
my $startmfn = 1; |
my $from_rec = 1; |
259 |
my $maxmfn = $size; |
my $to_rec = $size; |
260 |
|
|
261 |
if (my $s = $self->{start_mfn}) { |
if (my $s = $self->{offset}) { |
262 |
$log->info("skipping to MFN $s"); |
$log->debug("skipping to MFN $s"); |
263 |
$startmfn = $s; |
$from_rec = $s; |
264 |
} else { |
} else { |
265 |
$self->{start_mfn} = $startmfn; |
$self->{offset} = $from_rec; |
266 |
} |
} |
267 |
|
|
268 |
if ($self->{limit_mfn}) { |
if ($self->{limit}) { |
269 |
$log->info("limiting to ",$self->{limit_mfn}," records"); |
$log->debug("limiting to ",$self->{limit}," records"); |
270 |
$maxmfn = $startmfn + $self->{limit_mfn} - 1; |
$to_rec = $from_rec + $self->{limit} - 1; |
271 |
$maxmfn = $size if ($maxmfn > $size); |
$to_rec = $size if ($to_rec > $size); |
272 |
} |
} |
273 |
|
|
274 |
# store size for later |
# store size for later |
275 |
$self->{size} = ($maxmfn - $startmfn) ? ($maxmfn - $startmfn + 1) : 0; |
$self->{size} = ($to_rec - $from_rec) ? ($to_rec - $from_rec + 1) : 0; |
276 |
|
|
277 |
$log->info("processing $self->{size} records in $code_page, convert to $self->{code_page}"); |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{encoding}", $self->{stats} ? ' [stats]' : ''); |
278 |
|
|
279 |
# read database |
# read database |
280 |
for (my $mfn = $startmfn; $mfn <= $maxmfn; $mfn++) { |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
281 |
|
|
282 |
|
$log->debug("position: $pos\n"); |
283 |
|
|
284 |
$log->debug("mfn: $mfn\n"); |
my $rec = $ll_db->fetch_rec($pos, sub { |
285 |
|
my ($l,$f_nr) = @_; |
286 |
|
# return unless defined($l); |
287 |
|
# return $l unless ($rec_regex && $f_nr); |
288 |
|
|
289 |
|
$log->debug("-=> $f_nr ## $l"); |
290 |
|
|
291 |
|
# codepage conversion and recode_regex |
292 |
|
from_to($l, $code_page, $self->{'encoding'}); |
293 |
|
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
294 |
|
|
295 |
|
# apply regexps |
296 |
|
if ($rec_regex && defined($rec_regex->{$f_nr})) { |
297 |
|
$log->logconfess("regexps->{$f_nr} must be ARRAY") if (ref($rec_regex->{$f_nr}) ne 'ARRAY'); |
298 |
|
my $c = 0; |
299 |
|
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
300 |
|
my $old_l = $l; |
301 |
|
eval '$l =~ ' . $r; |
302 |
|
if ($old_l ne $l) { |
303 |
|
$log->debug("REGEX on $f_nr eval \$l =~ $r\n## old l: [$old_l]\n## new l: [$l]"); |
304 |
|
} |
305 |
|
$log->error("error applying regex: $r") if ($@); |
306 |
|
} |
307 |
|
} |
308 |
|
|
309 |
|
$log->debug("<=- $f_nr ## $l"); |
310 |
|
return $l; |
311 |
|
}); |
312 |
|
|
313 |
my $rec = $self->fetch_rec( $db, $mfn ); |
$log->debug(sub { dump($rec) }); |
314 |
|
|
315 |
if (! $rec) { |
if (! $rec) { |
316 |
$log->warn("record $mfn empty? skipping..."); |
$log->warn("record $pos empty? skipping..."); |
317 |
next; |
next; |
318 |
} |
} |
319 |
|
|
320 |
# store |
# store |
321 |
if ($self->{'low_mem'}) { |
if ($self->{save_row}) { |
322 |
$self->{'db'}->put($mfn, $rec); |
$self->{save_row}->({ |
323 |
|
id => $pos, |
324 |
|
row => $rec, |
325 |
|
}); |
326 |
} else { |
} else { |
327 |
$self->{'data'}->{$mfn} = $rec; |
$self->{data}->{$pos} = $rec; |
328 |
} |
} |
329 |
|
|
330 |
# create lookup |
# create lookup |
331 |
$self->{'lookup'}->add( $rec ) if ($rec && $self->{'lookup'}); |
$arg->{'lookup_coderef'}->( $rec ) if ($rec && $arg->{'lookup_coderef'}); |
332 |
|
|
333 |
$self->progress_bar($mfn,$maxmfn); |
# update counters for statistics |
334 |
|
if ($self->{stats}) { |
335 |
|
|
336 |
} |
# fetch clean record with regexpes applied for statistics |
337 |
|
my $rec = $ll_db->fetch_rec($pos); |
338 |
|
|
339 |
$self->{'current_mfn'} = -1; |
foreach my $fld (keys %{ $rec }) { |
340 |
$self->{'last_pcnt'} = 0; |
$self->{_stats}->{fld}->{ $fld }++; |
341 |
|
|
342 |
|
$log->logdie("invalid record fild $fld, not ARRAY") |
343 |
|
unless (ref($rec->{ $fld }) eq 'ARRAY'); |
344 |
|
|
345 |
|
foreach my $row (@{ $rec->{$fld} }) { |
346 |
|
|
347 |
|
if (ref($row) eq 'HASH') { |
348 |
|
|
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'); |
354 |
|
} |
355 |
|
|
356 |
|
} else { |
357 |
|
$self->{_stats}->{repeatable}->{ $fld }++; |
358 |
|
} |
359 |
|
} |
360 |
|
} |
361 |
|
} |
362 |
|
|
363 |
|
$self->progress_bar($pos,$to_rec) unless ($self->{no_progress_bar}); |
364 |
|
|
365 |
|
} |
366 |
|
|
367 |
$log->debug("max mfn: $maxmfn"); |
$self->{pos} = -1; |
368 |
|
$self->{last_pcnt} = 0; |
369 |
|
|
370 |
# store max mfn and return it. |
# store max mfn and return it. |
371 |
$self->{'max_mfn'} = $maxmfn; |
$self->{max_pos} = $to_rec; |
372 |
|
$log->debug("max_pos: $to_rec"); |
373 |
|
|
374 |
|
# save for dump |
375 |
|
$self->{ll_db} = $ll_db; |
376 |
|
|
377 |
return $size; |
return $size; |
378 |
} |
} |
393 |
|
|
394 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
395 |
|
|
396 |
$log->logconfess("it seems that you didn't load database!") unless ($self->{'current_mfn'}); |
$log->logconfess("it seems that you didn't load database!") unless ($self->{pos}); |
397 |
|
|
398 |
if ($self->{'current_mfn'} == -1) { |
if ($self->{pos} == -1) { |
399 |
$self->{'current_mfn'} = $self->{'start_mfn'}; |
$self->{pos} = $self->{offset}; |
400 |
} else { |
} else { |
401 |
$self->{'current_mfn'}++; |
$self->{pos}++; |
402 |
} |
} |
403 |
|
|
404 |
my $mfn = $self->{'current_mfn'}; |
my $mfn = $self->{pos}; |
405 |
|
|
406 |
if ($mfn > $self->{'max_mfn'}) { |
if ($mfn > $self->{max_pos}) { |
407 |
$self->{'current_mfn'} = $self->{'max_mfn'}; |
$self->{pos} = $self->{max_pos}; |
408 |
$log->debug("at EOF"); |
$log->debug("at EOF"); |
409 |
return; |
return; |
410 |
} |
} |
411 |
|
|
412 |
$self->progress_bar($mfn,$self->{'max_mfn'}); |
$self->progress_bar($mfn,$self->{max_pos}) unless ($self->{no_progress_bar}); |
413 |
|
|
414 |
my $rec; |
my $rec; |
415 |
|
|
416 |
if ($self->{'low_mem'}) { |
if ($self->{load_row}) { |
417 |
$rec = $self->{'db'}->get($mfn); |
$rec = $self->{load_row}->({ id => $mfn }); |
418 |
} else { |
} else { |
419 |
$rec = $self->{'data'}->{$mfn}; |
$rec = $self->{data}->{$mfn}; |
420 |
} |
} |
421 |
|
|
422 |
$rec ||= 0E0; |
$rec ||= 0E0; |
434 |
|
|
435 |
sub pos { |
sub pos { |
436 |
my $self = shift; |
my $self = shift; |
437 |
return $self->{'current_mfn'}; |
return $self->{pos}; |
438 |
} |
} |
439 |
|
|
440 |
|
|
448 |
|
|
449 |
foreach my $mfn ( 1 ... $isis->size ) { ... } |
foreach my $mfn ( 1 ... $isis->size ) { ... } |
450 |
|
|
451 |
because it takes into account C<start_mfn> and C<limit_mfn>. |
because it takes into account C<offset> and C<limit>. |
452 |
|
|
453 |
=cut |
=cut |
454 |
|
|
455 |
sub size { |
sub size { |
456 |
my $self = shift; |
my $self = shift; |
457 |
return $self->{'size'}; |
return $self->{size}; |
458 |
} |
} |
459 |
|
|
460 |
=head2 seek |
=head2 seek |
469 |
|
|
470 |
sub seek { |
sub seek { |
471 |
my $self = shift; |
my $self = shift; |
472 |
my $pos = shift || return; |
my $pos = shift; |
473 |
|
|
474 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
475 |
|
|
476 |
|
$log->confess("called without pos") unless defined($pos); |
477 |
|
|
478 |
if ($pos < 1) { |
if ($pos < 1) { |
479 |
$log->warn("seek before first record"); |
$log->warn("seek before first record"); |
480 |
$pos = 1; |
$pos = 1; |
481 |
} elsif ($pos > $self->{'max_mfn'}) { |
} elsif ($pos > $self->{max_pos}) { |
482 |
$log->warn("seek beyond last record"); |
$log->warn("seek beyond last record"); |
483 |
$pos = $self->{'max_mfn'}; |
$pos = $self->{max_pos}; |
484 |
|
} |
485 |
|
|
486 |
|
return $self->{pos} = (($pos - 1) || -1); |
487 |
|
} |
488 |
|
|
489 |
|
=head2 stats |
490 |
|
|
491 |
|
Dump statistics about field and subfield usage |
492 |
|
|
493 |
|
print $input->stats; |
494 |
|
|
495 |
|
=cut |
496 |
|
|
497 |
|
sub stats { |
498 |
|
my $self = shift; |
499 |
|
|
500 |
|
my $log = $self->_get_logger(); |
501 |
|
|
502 |
|
my $s = $self->{_stats}; |
503 |
|
if (! $s) { |
504 |
|
$log->warn("called stats, but there is no statistics collected"); |
505 |
|
return; |
506 |
|
} |
507 |
|
|
508 |
|
my $max_fld = 0; |
509 |
|
|
510 |
|
my $out = join("\n", |
511 |
|
map { |
512 |
|
my $f = $_ || die "no field"; |
513 |
|
my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}"; |
514 |
|
$max_fld = $v if ($v > $max_fld); |
515 |
|
|
516 |
|
my $o = sprintf("%4s %d ~", $f, $v); |
517 |
|
|
518 |
|
if (defined($s->{sf}->{$f})) { |
519 |
|
map { |
520 |
|
$o .= sprintf(" %s:%d%s", $_, |
521 |
|
$s->{sf}->{$f}->{$_}->{count}, |
522 |
|
$s->{sf}->{$f}->{$_}->{repeatable} ? '*' : '', |
523 |
|
); |
524 |
|
} sort keys %{ $s->{sf}->{$f} }; |
525 |
|
} |
526 |
|
|
527 |
|
if (my $v_r = $s->{repeatable}->{$f}) { |
528 |
|
$o .= " ($v_r)" if ($v_r != $v); |
529 |
|
} |
530 |
|
|
531 |
|
$o; |
532 |
|
} sort { $a cmp $b } keys %{ $s->{fld} } |
533 |
|
); |
534 |
|
|
535 |
|
$log->debug( sub { dump($s) } ); |
536 |
|
|
537 |
|
return $out; |
538 |
|
} |
539 |
|
|
540 |
|
=head2 dump_ascii |
541 |
|
|
542 |
|
Display humanly readable dump of record |
543 |
|
|
544 |
|
=cut |
545 |
|
|
546 |
|
sub dump_ascii { |
547 |
|
my $self = shift; |
548 |
|
|
549 |
|
return unless $self->{ll_db}; |
550 |
|
|
551 |
|
if ($self->{ll_db}->can('dump_rec')) { |
552 |
|
return $self->{ll_db}->dump_ascii( $self->{pos} ); |
553 |
|
} else { |
554 |
|
return dump( $self->{ll_db}->fetch_rec( $self->{pos} ) ); |
555 |
|
} |
556 |
|
} |
557 |
|
|
558 |
|
=head2 modify_record_regexps |
559 |
|
|
560 |
|
Generate hash with regexpes to be applied using l<filter>. |
561 |
|
|
562 |
|
my $regexpes = $input->modify_record_regexps( |
563 |
|
900 => { '^a' => { ' : ' => '^b' } }, |
564 |
|
901 => { '*' => { '^b' => ' ; ' } }, |
565 |
|
); |
566 |
|
|
567 |
|
=cut |
568 |
|
|
569 |
|
sub _get_regex { |
570 |
|
my ($sf,$from,$to) = @_; |
571 |
|
if ($from =~ m/^regex:(.+)$/) { |
572 |
|
$from = $1; |
573 |
|
} else { |
574 |
|
$from = '\Q' . $from . '\E'; |
575 |
} |
} |
576 |
|
if ($sf =~ /^\^/) { |
577 |
|
return |
578 |
|
's/\Q'. $sf .'\E([^\^]*?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
579 |
|
} else { |
580 |
|
return |
581 |
|
's/'. $from .'/'. $to .'/g'; |
582 |
|
} |
583 |
|
} |
584 |
|
|
585 |
|
sub modify_record_regexps { |
586 |
|
my $self = shift; |
587 |
|
my $modify_record = {@_}; |
588 |
|
|
589 |
|
my $regexpes; |
590 |
|
|
591 |
|
my $log = $self->_get_logger(); |
592 |
|
|
593 |
return $self->{'current_mfn'} = (($pos - 1) || -1); |
foreach my $f (keys %$modify_record) { |
594 |
|
$log->debug("field: $f"); |
595 |
|
|
596 |
|
foreach my $sf (keys %{ $modify_record->{$f} }) { |
597 |
|
$log->debug("subfield: $sf"); |
598 |
|
|
599 |
|
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
600 |
|
my $to = $modify_record->{$f}->{$sf}->{$from}; |
601 |
|
#die "no field?" unless defined($to); |
602 |
|
$log->debug("transform: |$from| -> |$to|"); |
603 |
|
|
604 |
|
my $regex = _get_regex($sf,$from,$to); |
605 |
|
push @{ $regexpes->{$f} }, $regex; |
606 |
|
$log->debug("regex: $regex"); |
607 |
|
} |
608 |
|
} |
609 |
|
} |
610 |
|
|
611 |
|
return $regexpes; |
612 |
} |
} |
613 |
|
|
614 |
|
=head2 modify_file_regexps |
615 |
|
|
616 |
|
Generate hash with regexpes to be applied using l<filter> from |
617 |
|
pseudo hash/yaml format for regex mappings. |
618 |
|
|
619 |
=head1 MEMORY USAGE |
It should be obvious: |
620 |
|
|
621 |
C<low_mem> options is double-edged sword. If enabled, WebPAC |
200 |
622 |
will run on memory constraint machines (which doesn't have enough |
'^a' |
623 |
physical RAM to create memory structure for whole source database). |
' : ' => '^e' |
624 |
|
' = ' => '^d' |
625 |
|
|
626 |
If your machine has 512Mb or more of RAM and database is around 10000 records, |
In field I<200> find C<'^a'> and then C<' : '>, and replace it with C<'^e'>. |
627 |
memory shouldn't be an issue. If you don't have enough physical RAM, you |
In field I<200> find C<'^a'> and then C<' = '>, and replace it with C<'^d'>. |
|
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). |
|
628 |
|
|
629 |
Hitting swap at end of reading source database is probably o.k. However, |
my $regexpes = $input->modify_file_regexps( 'conf/modify/common.pl' ); |
|
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. |
|
630 |
|
|
631 |
Parsed structures are essential - you just have option to trade RAM memory |
On undef path it will just return. |
|
(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>. |
|
632 |
|
|
633 |
However, when WebPAC is running on desktop machines (or laptops :-), it's |
=cut |
|
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. |
|
634 |
|
|
635 |
|
sub modify_file_regexps { |
636 |
|
my $self = shift; |
637 |
|
|
638 |
|
my $modify_path = shift || return; |
639 |
|
|
640 |
|
my $log = $self->_get_logger(); |
641 |
|
|
642 |
|
my $regexpes; |
643 |
|
|
644 |
|
CORE::open(my $fh, $modify_path) || $log->logdie("can't open modify file $modify_path: $!"); |
645 |
|
|
646 |
|
my ($f,$sf); |
647 |
|
|
648 |
|
while(<$fh>) { |
649 |
|
chomp; |
650 |
|
next if (/^#/ || /^\s*$/); |
651 |
|
|
652 |
|
if (/^\s*(\d+)\s*$/) { |
653 |
|
$f = $1; |
654 |
|
$log->debug("field: $f"); |
655 |
|
next; |
656 |
|
} elsif (/^\s*'([^']*)'\s*$/) { |
657 |
|
$sf = $1; |
658 |
|
$log->die("can't define subfiled before field in: $_") unless ($f); |
659 |
|
$log->debug("subfield: $sf"); |
660 |
|
} elsif (/^\s*'([^']*)'\s*=>\s*'([^']*)'\s*$/) { |
661 |
|
my ($from,$to) = ($1, $2); |
662 |
|
|
663 |
|
$log->debug("transform: |$from| -> |$to|"); |
664 |
|
|
665 |
|
my $regex = _get_regex($sf,$from,$to); |
666 |
|
push @{ $regexpes->{$f} }, $regex; |
667 |
|
$log->debug("regex: $regex"); |
668 |
|
} |
669 |
|
} |
670 |
|
|
671 |
|
return $regexpes; |
672 |
|
} |
673 |
|
|
674 |
=head1 AUTHOR |
=head1 AUTHOR |
675 |
|
|
677 |
|
|
678 |
=head1 COPYRIGHT & LICENSE |
=head1 COPYRIGHT & LICENSE |
679 |
|
|
680 |
Copyright 2005 Dobrica Pavlinusic, All Rights Reserved. |
Copyright 2005-2006 Dobrica Pavlinusic, All Rights Reserved. |
681 |
|
|
682 |
This program is free software; you can redistribute it and/or modify it |
This program is free software; you can redistribute it and/or modify it |
683 |
under the same terms as Perl itself. |
under the same terms as Perl itself. |