| 1 |
package WebPAC::Input; |
| 2 |
|
| 3 |
use warnings; |
| 4 |
use strict; |
| 5 |
|
| 6 |
use WebPAC::Common; |
| 7 |
use base qw/WebPAC::Common/; |
| 8 |
use Text::Iconv; |
| 9 |
use Data::Dumper; |
| 10 |
|
| 11 |
=head1 NAME |
| 12 |
|
| 13 |
WebPAC::Input - read different file formats into WebPAC |
| 14 |
|
| 15 |
=head1 VERSION |
| 16 |
|
| 17 |
Version 0.04 |
| 18 |
|
| 19 |
=cut |
| 20 |
|
| 21 |
our $VERSION = '0.04'; |
| 22 |
|
| 23 |
=head1 SYNOPSIS |
| 24 |
|
| 25 |
This module implements input as database which have fixed and known |
| 26 |
I<size> while indexing and single unique numeric identifier for database |
| 27 |
position ranging from 1 to I<size>. |
| 28 |
|
| 29 |
Simply, something that is indexed by unmber from 1 .. I<size>. |
| 30 |
|
| 31 |
Examples of such databases are CDS/ISIS files, MARC files, lines in |
| 32 |
text file, and so on. |
| 33 |
|
| 34 |
Specific file formats are implemented using low-level interface modules, |
| 35 |
located in C<WebPAC::Input::*> namespace which export C<open_db>, |
| 36 |
C<fetch_rec> and optional C<init> functions. |
| 37 |
|
| 38 |
Perhaps a little code snippet. |
| 39 |
|
| 40 |
use WebPAC::Input; |
| 41 |
|
| 42 |
my $db = WebPAC::Input->new( |
| 43 |
module => 'WebPAC::Input::ISIS', |
| 44 |
config => $config, |
| 45 |
lookup => $lookup_obj, |
| 46 |
low_mem => 1, |
| 47 |
); |
| 48 |
|
| 49 |
$db->open('/path/to/database'); |
| 50 |
print "database size: ",$db->size,"\n"; |
| 51 |
while (my $rec = $db->fetch) { |
| 52 |
# do something with $rec |
| 53 |
} |
| 54 |
|
| 55 |
|
| 56 |
|
| 57 |
=head1 FUNCTIONS |
| 58 |
|
| 59 |
=head2 new |
| 60 |
|
| 61 |
Create new input database object. |
| 62 |
|
| 63 |
my $db = new WebPAC::Input( |
| 64 |
module => 'WebPAC::Input::MARC', |
| 65 |
code_page => 'ISO-8859-2', |
| 66 |
low_mem => 1, |
| 67 |
recode => 'char pairs', |
| 68 |
); |
| 69 |
|
| 70 |
C<module> is low-level file format module. See L<WebPAC::Input::Isis> and |
| 71 |
L<WebPAC::Input::MARC>. |
| 72 |
|
| 73 |
Optional parametar C<code_page> specify application code page (which will be |
| 74 |
used internally). This should probably be your terminal encoding, and by |
| 75 |
default, it C<ISO-8859-2>. |
| 76 |
|
| 77 |
Default is not to use C<low_mem> options (see L<MEMORY USAGE> below). |
| 78 |
|
| 79 |
This function will also call low-level C<init> if it exists with same |
| 80 |
parametars. |
| 81 |
|
| 82 |
=cut |
| 83 |
|
| 84 |
sub new { |
| 85 |
my $class = shift; |
| 86 |
my $self = {@_}; |
| 87 |
bless($self, $class); |
| 88 |
|
| 89 |
my $log = $self->_get_logger; |
| 90 |
|
| 91 |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
| 92 |
my $module = $self->{module}; |
| 93 |
$module =~ s#::#/#g; |
| 94 |
$module .= '.pm'; |
| 95 |
$log->debug("require low-level module $self->{module} from $module"); |
| 96 |
|
| 97 |
require $module; |
| 98 |
#eval $self->{module} .'->import'; |
| 99 |
|
| 100 |
# check if required subclasses are implemented |
| 101 |
foreach my $subclass (qw/open_db fetch_rec init/) { |
| 102 |
my $n = $self->{module} . '::' . $subclass; |
| 103 |
if (! defined &{ $n }) { |
| 104 |
my $missing = "missing $subclass in $self->{module}"; |
| 105 |
$self->{$subclass} = sub { $log->logwarn($missing) }; |
| 106 |
} else { |
| 107 |
$self->{$subclass} = \&{ $n }; |
| 108 |
} |
| 109 |
} |
| 110 |
|
| 111 |
if ($self->{init}) { |
| 112 |
$log->debug("calling init"); |
| 113 |
$self->{init}->($self, @_); |
| 114 |
} |
| 115 |
|
| 116 |
$self->{'code_page'} ||= 'ISO-8859-2'; |
| 117 |
|
| 118 |
# running with low_mem flag? well, use DBM::Deep then. |
| 119 |
if ($self->{'low_mem'}) { |
| 120 |
$log->info("running with low_mem which impacts performance (<32 Mb memory usage)"); |
| 121 |
|
| 122 |
my $db_file = "data.db"; |
| 123 |
|
| 124 |
if (-e $db_file) { |
| 125 |
unlink $db_file or $log->logdie("can't remove '$db_file' from last run"); |
| 126 |
$log->debug("removed '$db_file' from last run"); |
| 127 |
} |
| 128 |
|
| 129 |
require DBM::Deep; |
| 130 |
|
| 131 |
my $db = new DBM::Deep $db_file; |
| 132 |
|
| 133 |
$log->logdie("DBM::Deep error: $!") unless ($db); |
| 134 |
|
| 135 |
if ($db->error()) { |
| 136 |
$log->logdie("can't open '$db_file' under low_mem: ",$db->error()); |
| 137 |
} else { |
| 138 |
$log->debug("using file '$db_file' for DBM::Deep"); |
| 139 |
} |
| 140 |
|
| 141 |
$self->{'db'} = $db; |
| 142 |
} |
| 143 |
|
| 144 |
$self ? return $self : return undef; |
| 145 |
} |
| 146 |
|
| 147 |
=head2 open |
| 148 |
|
| 149 |
This function will read whole database in memory and produce lookups. |
| 150 |
|
| 151 |
$input->open( |
| 152 |
path => '/path/to/database/file', |
| 153 |
code_page => '852', |
| 154 |
limit => 500, |
| 155 |
offset => 6000, |
| 156 |
lookup => $lookup_obj, |
| 157 |
); |
| 158 |
|
| 159 |
By default, C<code_page> is assumed to be C<852>. |
| 160 |
|
| 161 |
C<offset> is optional parametar to position at some offset before reading from database. |
| 162 |
|
| 163 |
C<limit> is optional parametar to read just C<limit> records from database |
| 164 |
|
| 165 |
Returns size of database, regardless of C<offset> and C<limit> |
| 166 |
parametars, see also C<size>. |
| 167 |
|
| 168 |
=cut |
| 169 |
|
| 170 |
sub open { |
| 171 |
my $self = shift; |
| 172 |
my $arg = {@_}; |
| 173 |
|
| 174 |
my $log = $self->_get_logger(); |
| 175 |
|
| 176 |
$log->logcroak("need path") if (! $arg->{'path'}); |
| 177 |
my $code_page = $arg->{'code_page'} || '852'; |
| 178 |
|
| 179 |
# store data in object |
| 180 |
$self->{'input_code_page'} = $code_page; |
| 181 |
foreach my $v (qw/path offset limit/) { |
| 182 |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
| 183 |
} |
| 184 |
|
| 185 |
# create Text::Iconv object |
| 186 |
$self->{iconv} = Text::Iconv->new($code_page,$self->{'code_page'}); |
| 187 |
|
| 188 |
my $filter_ref; |
| 189 |
|
| 190 |
if ($self->{recode}) { |
| 191 |
my @r = split(/\s/, $self->{recode}); |
| 192 |
if ($#r % 2 != 1) { |
| 193 |
$log->logwarn("recode needs even number of elements (some number of valid pairs)"); |
| 194 |
} else { |
| 195 |
my $recode; |
| 196 |
while (@r) { |
| 197 |
my $from = shift @r; |
| 198 |
my $to = shift @r; |
| 199 |
$recode->{$from} = $to; |
| 200 |
} |
| 201 |
|
| 202 |
my $regex = join '|' => keys %{ $recode }; |
| 203 |
|
| 204 |
$log->debug("using recode regex: $regex"); |
| 205 |
|
| 206 |
$filter_ref = sub { |
| 207 |
my $t = shift; |
| 208 |
$t =~ s/($regex)/$recode->{$1}/g; |
| 209 |
return $t; |
| 210 |
}; |
| 211 |
|
| 212 |
} |
| 213 |
|
| 214 |
} |
| 215 |
|
| 216 |
my ($db, $size) = $self->{open_db}->( $self, |
| 217 |
path => $arg->{path}, |
| 218 |
filter => $filter_ref, |
| 219 |
); |
| 220 |
|
| 221 |
unless ($db) { |
| 222 |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
| 223 |
return; |
| 224 |
} |
| 225 |
|
| 226 |
unless ($size) { |
| 227 |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
| 228 |
return; |
| 229 |
} |
| 230 |
|
| 231 |
my $from_rec = 1; |
| 232 |
my $to_rec = $size; |
| 233 |
|
| 234 |
if (my $s = $self->{offset}) { |
| 235 |
$log->info("skipping to MFN $s"); |
| 236 |
$from_rec = $s; |
| 237 |
} else { |
| 238 |
$self->{offset} = $from_rec; |
| 239 |
} |
| 240 |
|
| 241 |
if ($self->{limit}) { |
| 242 |
$log->debug("limiting to ",$self->{limit}," records"); |
| 243 |
$to_rec = $from_rec + $self->{limit} - 1; |
| 244 |
$to_rec = $size if ($to_rec > $size); |
| 245 |
} |
| 246 |
|
| 247 |
# store size for later |
| 248 |
$self->{size} = ($to_rec - $from_rec) ? ($to_rec - $from_rec + 1) : 0; |
| 249 |
|
| 250 |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{code_page}"); |
| 251 |
|
| 252 |
# read database |
| 253 |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
| 254 |
|
| 255 |
$log->debug("position: $pos\n"); |
| 256 |
|
| 257 |
my $rec = $self->{fetch_rec}->($self, $db, $pos ); |
| 258 |
|
| 259 |
$log->debug(sub { Dumper($rec) }); |
| 260 |
|
| 261 |
if (! $rec) { |
| 262 |
$log->warn("record $pos empty? skipping..."); |
| 263 |
next; |
| 264 |
} |
| 265 |
|
| 266 |
# store |
| 267 |
if ($self->{low_mem}) { |
| 268 |
$self->{db}->put($pos, $rec); |
| 269 |
} else { |
| 270 |
$self->{data}->{$pos} = $rec; |
| 271 |
} |
| 272 |
|
| 273 |
# create lookup |
| 274 |
$self->{'lookup'}->add( $rec ) if ($rec && $self->{'lookup'}); |
| 275 |
|
| 276 |
$self->progress_bar($pos,$to_rec); |
| 277 |
|
| 278 |
} |
| 279 |
|
| 280 |
$self->{pos} = -1; |
| 281 |
$self->{last_pcnt} = 0; |
| 282 |
|
| 283 |
# store max mfn and return it. |
| 284 |
$self->{max_pos} = $to_rec; |
| 285 |
$log->debug("max_pos: $to_rec"); |
| 286 |
|
| 287 |
return $size; |
| 288 |
} |
| 289 |
|
| 290 |
=head2 fetch |
| 291 |
|
| 292 |
Fetch next record from database. It will also displays progress bar. |
| 293 |
|
| 294 |
my $rec = $isis->fetch; |
| 295 |
|
| 296 |
Record from this function should probably go to C<data_structure> for |
| 297 |
normalisation. |
| 298 |
|
| 299 |
=cut |
| 300 |
|
| 301 |
sub fetch { |
| 302 |
my $self = shift; |
| 303 |
|
| 304 |
my $log = $self->_get_logger(); |
| 305 |
|
| 306 |
$log->logconfess("it seems that you didn't load database!") unless ($self->{pos}); |
| 307 |
|
| 308 |
if ($self->{pos} == -1) { |
| 309 |
$self->{pos} = $self->{offset}; |
| 310 |
} else { |
| 311 |
$self->{pos}++; |
| 312 |
} |
| 313 |
|
| 314 |
my $mfn = $self->{pos}; |
| 315 |
|
| 316 |
if ($mfn > $self->{max_pos}) { |
| 317 |
$self->{pos} = $self->{max_pos}; |
| 318 |
$log->debug("at EOF"); |
| 319 |
return; |
| 320 |
} |
| 321 |
|
| 322 |
$self->progress_bar($mfn,$self->{max_pos}); |
| 323 |
|
| 324 |
my $rec; |
| 325 |
|
| 326 |
if ($self->{low_mem}) { |
| 327 |
$rec = $self->{db}->get($mfn); |
| 328 |
} else { |
| 329 |
$rec = $self->{data}->{$mfn}; |
| 330 |
} |
| 331 |
|
| 332 |
$rec ||= 0E0; |
| 333 |
} |
| 334 |
|
| 335 |
=head2 pos |
| 336 |
|
| 337 |
Returns current record number (MFN). |
| 338 |
|
| 339 |
print $isis->pos; |
| 340 |
|
| 341 |
First record in database has position 1. |
| 342 |
|
| 343 |
=cut |
| 344 |
|
| 345 |
sub pos { |
| 346 |
my $self = shift; |
| 347 |
return $self->{pos}; |
| 348 |
} |
| 349 |
|
| 350 |
|
| 351 |
=head2 size |
| 352 |
|
| 353 |
Returns number of records in database |
| 354 |
|
| 355 |
print $isis->size; |
| 356 |
|
| 357 |
Result from this function can be used to loop through all records |
| 358 |
|
| 359 |
foreach my $mfn ( 1 ... $isis->size ) { ... } |
| 360 |
|
| 361 |
because it takes into account C<offset> and C<limit>. |
| 362 |
|
| 363 |
=cut |
| 364 |
|
| 365 |
sub size { |
| 366 |
my $self = shift; |
| 367 |
return $self->{size}; |
| 368 |
} |
| 369 |
|
| 370 |
=head2 seek |
| 371 |
|
| 372 |
Seek to specified MFN in file. |
| 373 |
|
| 374 |
$isis->seek(42); |
| 375 |
|
| 376 |
First record in database has position 1. |
| 377 |
|
| 378 |
=cut |
| 379 |
|
| 380 |
sub seek { |
| 381 |
my $self = shift; |
| 382 |
my $pos = shift || return; |
| 383 |
|
| 384 |
my $log = $self->_get_logger(); |
| 385 |
|
| 386 |
if ($pos < 1) { |
| 387 |
$log->warn("seek before first record"); |
| 388 |
$pos = 1; |
| 389 |
} elsif ($pos > $self->{max_pos}) { |
| 390 |
$log->warn("seek beyond last record"); |
| 391 |
$pos = $self->{max_pos}; |
| 392 |
} |
| 393 |
|
| 394 |
return $self->{pos} = (($pos - 1) || -1); |
| 395 |
} |
| 396 |
|
| 397 |
|
| 398 |
=head1 MEMORY USAGE |
| 399 |
|
| 400 |
C<low_mem> options is double-edged sword. If enabled, WebPAC |
| 401 |
will run on memory constraint machines (which doesn't have enough |
| 402 |
physical RAM to create memory structure for whole source database). |
| 403 |
|
| 404 |
If your machine has 512Mb or more of RAM and database is around 10000 records, |
| 405 |
memory shouldn't be an issue. If you don't have enough physical RAM, you |
| 406 |
might consider using virtual memory (if your operating system is handling it |
| 407 |
well, like on FreeBSD or Linux) instead of dropping to L<DBM::Deep> to handle |
| 408 |
parsed structure of ISIS database (this is what C<low_mem> option does). |
| 409 |
|
| 410 |
Hitting swap at end of reading source database is probably o.k. However, |
| 411 |
hitting swap before 90% will dramatically decrease performance and you will |
| 412 |
be better off with C<low_mem> and using rest of availble memory for |
| 413 |
operating system disk cache (Linux is particuallary good about this). |
| 414 |
However, every access to database record will require disk access, so |
| 415 |
generation phase will be slower 10-100 times. |
| 416 |
|
| 417 |
Parsed structures are essential - you just have option to trade RAM memory |
| 418 |
(which is fast) for disk space (which is slow). Be sure to have planty of |
| 419 |
disk space if you are using C<low_mem> and thus L<DBM::Deep>. |
| 420 |
|
| 421 |
However, when WebPAC is running on desktop machines (or laptops :-), it's |
| 422 |
highly undesireable for system to start swapping. Using C<low_mem> option can |
| 423 |
reduce WecPAC memory usage to around 64Mb for same database with lookup |
| 424 |
fields and sorted indexes which stay in RAM. Performance will suffer, but |
| 425 |
memory usage will really be minimal. It might be also more confortable to |
| 426 |
run WebPAC reniced on those machines. |
| 427 |
|
| 428 |
|
| 429 |
=head1 AUTHOR |
| 430 |
|
| 431 |
Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >> |
| 432 |
|
| 433 |
=head1 COPYRIGHT & LICENSE |
| 434 |
|
| 435 |
Copyright 2005 Dobrica Pavlinusic, All Rights Reserved. |
| 436 |
|
| 437 |
This program is free software; you can redistribute it and/or modify it |
| 438 |
under the same terms as Perl itself. |
| 439 |
|
| 440 |
=cut |
| 441 |
|
| 442 |
1; # End of WebPAC::Input |
Parent Directory
| 
Revision Log