| 3 |
use warnings; |
use warnings; |
| 4 |
use strict; |
use strict; |
| 5 |
|
|
| 6 |
use blib; |
use lib 'lib'; |
| 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 Data::Dumper; |
use Encode qw/decode from_to/; |
| 12 |
|
use YAML; |
| 13 |
|
|
| 14 |
=head1 NAME |
=head1 NAME |
| 15 |
|
|
| 16 |
WebPAC::Input - read different file formats into WebPAC |
WebPAC::Input - read different file formats into WebPAC |
| 17 |
|
|
|
=head1 VERSION |
|
|
|
|
|
Version 0.05 |
|
|
|
|
| 18 |
=cut |
=cut |
| 19 |
|
|
| 20 |
our $VERSION = '0.05'; |
our $VERSION = '0.19'; |
| 21 |
|
|
| 22 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
| 23 |
|
|
| 36 |
|
|
| 37 |
Perhaps a little code snippet. |
Perhaps a little code snippet. |
| 38 |
|
|
| 39 |
use WebPAC::Input; |
use WebPAC::Input; |
| 40 |
|
|
| 41 |
my $db = WebPAC::Input->new( |
my $db = WebPAC::Input->new( |
| 42 |
module => 'WebPAC::Input::ISIS', |
module => 'WebPAC::Input::ISIS', |
| 43 |
config => $config, |
); |
|
lookup => $lookup_obj, |
|
|
low_mem => 1, |
|
|
); |
|
| 44 |
|
|
| 45 |
$db->open('/path/to/database'); |
$db->open( path => '/path/to/database' ); |
| 46 |
print "database size: ",$db->size,"\n"; |
print "database size: ",$db->size,"\n"; |
| 47 |
while (my $rec = $db->fetch) { |
while (my $rec = $db->fetch) { |
| 48 |
# do something with $rec |
# do something with $rec |
| 58 |
|
|
| 59 |
my $db = new WebPAC::Input( |
my $db = new WebPAC::Input( |
| 60 |
module => 'WebPAC::Input::MARC', |
module => 'WebPAC::Input::MARC', |
|
code_page => 'ISO-8859-2', |
|
|
low_mem => 1, |
|
| 61 |
recode => 'char pairs', |
recode => 'char pairs', |
| 62 |
no_progress_bar => 1, |
no_progress_bar => 1, |
| 63 |
|
input_config => { |
| 64 |
|
mapping => [ 'foo', 'bar', 'baz' ], |
| 65 |
|
}, |
| 66 |
); |
); |
| 67 |
|
|
| 68 |
C<module> is low-level file format module. See L<WebPAC::Input::Isis> and |
C<module> is low-level file format module. See L<WebPAC::Input::ISIS> and |
| 69 |
L<WebPAC::Input::MARC>. |
L<WebPAC::Input::MARC>. |
| 70 |
|
|
|
Optional parametar C<code_page> specify application code page (which will be |
|
|
used internally). This should probably be your terminal encoding, and by |
|
|
default, it C<ISO-8859-2>. |
|
|
|
|
|
Default is not to use C<low_mem> options (see L<MEMORY USAGE> below). |
|
|
|
|
| 71 |
C<recode> is optional string constisting of character or words pairs that |
C<recode> is optional string constisting of character or words pairs that |
| 72 |
should be replaced in input stream. |
should be replaced in input stream. |
| 73 |
|
|
| 85 |
|
|
| 86 |
my $log = $self->_get_logger; |
my $log = $self->_get_logger; |
| 87 |
|
|
| 88 |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
$log->logconfess("code_page argument is not suppored any more.") if $self->{code_page}; |
| 89 |
my $module = $self->{module}; |
$log->logconfess("encoding argument is not suppored any more.") if $self->{encoding}; |
| 90 |
$module =~ s#::#/#g; |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if $self->{lookup}; |
| 91 |
$module .= '.pm'; |
$log->logconfess("low_mem argument is not suppored any more. rewrite it to load_row and save_row") if $self->{low_mem}; |
|
$log->debug("require low-level module $self->{module} from $module"); |
|
|
|
|
|
require $module; |
|
|
#eval $self->{module} .'->import'; |
|
|
|
|
|
# check if required subclasses are implemented |
|
|
foreach my $subclass (qw/open_db fetch_rec init/) { |
|
|
my $n = $self->{module} . '::' . $subclass; |
|
|
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, @_); |
|
|
} |
|
|
|
|
|
$self->{'code_page'} ||= 'ISO-8859-2'; |
|
|
|
|
|
# running with low_mem flag? well, use DBM::Deep then. |
|
|
if ($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; |
|
| 92 |
|
|
| 93 |
$log->logdie("DBM::Deep error: $!") unless ($db); |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
| 94 |
|
my $module_path = $self->{module}; |
| 95 |
if ($db->error()) { |
$module_path =~ s#::#/#g; |
| 96 |
$log->logdie("can't open '$db_file' under low_mem: ",$db->error()); |
$module_path .= '.pm'; |
| 97 |
} else { |
$log->debug("require low-level module $self->{module} from $module_path"); |
|
$log->debug("using file '$db_file' for DBM::Deep"); |
|
|
} |
|
| 98 |
|
|
| 99 |
$self->{'db'} = $db; |
require $module_path; |
|
} |
|
| 100 |
|
|
| 101 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
| 102 |
} |
} |
| 105 |
|
|
| 106 |
This function will read whole database in memory and produce lookups. |
This function will read whole database in memory and produce lookups. |
| 107 |
|
|
| 108 |
|
my $store; # simple in-memory hash |
| 109 |
|
|
| 110 |
$input->open( |
$input->open( |
| 111 |
path => '/path/to/database/file', |
path => '/path/to/database/file', |
| 112 |
code_page => '852', |
input_encoding => 'cp852', |
| 113 |
|
strict_encoding => 0, |
| 114 |
limit => 500, |
limit => 500, |
| 115 |
offset => 6000, |
offset => 6000, |
|
lookup => $lookup_obj, |
|
| 116 |
stats => 1, |
stats => 1, |
| 117 |
|
lookup_coderef => sub { |
| 118 |
|
my $rec = shift; |
| 119 |
|
# store lookups |
| 120 |
|
}, |
| 121 |
|
modify_records => { |
| 122 |
|
900 => { '^a' => { ' : ' => '^b' } }, |
| 123 |
|
901 => { '*' => { '^b' => ' ; ' } }, |
| 124 |
|
}, |
| 125 |
|
modify_file => 'conf/modify/mapping.map', |
| 126 |
|
save_row => sub { |
| 127 |
|
my $a = shift; |
| 128 |
|
$store->{ $a->{id} } = $a->{row}; |
| 129 |
|
}, |
| 130 |
|
load_row => sub { |
| 131 |
|
my $a = shift; |
| 132 |
|
return defined($store->{ $a->{id} }) && |
| 133 |
|
$store->{ $a->{id} }; |
| 134 |
|
}, |
| 135 |
|
|
| 136 |
); |
); |
| 137 |
|
|
| 138 |
By default, C<code_page> is assumed to be C<852>. |
By default, C<input_encoding> is assumed to be C<cp852>. |
| 139 |
|
|
| 140 |
C<offset> is optional parametar to position at some offset before reading from database. |
C<offset> is optional parametar to skip records at beginning. |
| 141 |
|
|
| 142 |
C<limit> is optional parametar to read just C<limit> records from database |
C<limit> is optional parametar to read just C<limit> records from database |
| 143 |
|
|
| 144 |
C<stats> create optional report about usage of fields and subfields |
C<stats> create optional report about usage of fields and subfields |
| 145 |
|
|
| 146 |
|
C<lookup_coderef> is closure to called to save data into lookups |
| 147 |
|
|
| 148 |
|
C<modify_records> specify mapping from subfields to delimiters or from |
| 149 |
|
delimiters to subfields, as well as oprations on fields (if subfield is |
| 150 |
|
defined as C<*>. |
| 151 |
|
|
| 152 |
|
C<modify_file> is alternative for C<modify_records> above which preserves order and offers |
| 153 |
|
(hopefully) simplier sintax than YAML or perl (see L</modify_file_regex>). This option |
| 154 |
|
overrides C<modify_records> if both exists for same input. |
| 155 |
|
|
| 156 |
|
C<save_row> and C<load_row> are low-level implementation of store engine. Calling convention |
| 157 |
|
is documented in example above. |
| 158 |
|
|
| 159 |
|
C<strict_encoding> should really default to 1, but it doesn't for now. |
| 160 |
|
|
| 161 |
Returns size of database, regardless of C<offset> and C<limit> |
Returns size of database, regardless of C<offset> and C<limit> |
| 162 |
parametars, see also C<size>. |
parametars, see also C<size>. |
| 163 |
|
|
| 168 |
my $arg = {@_}; |
my $arg = {@_}; |
| 169 |
|
|
| 170 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
| 171 |
|
$log->debug( "arguments: ",dump( $arg )); |
| 172 |
|
|
| 173 |
|
$log->logconfess("encoding argument is not suppored any more.") if $self->{encoding}; |
| 174 |
|
$log->logconfess("code_page argument is not suppored any more.") if $self->{code_page}; |
| 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 $input_encoding = $arg->{'input_encoding'} || $self->{'input_encoding'} || 'cp852'; |
| 183 |
|
|
| 184 |
# store data in object |
# store data in object |
| 185 |
$self->{'input_code_page'} = $code_page; |
$self->{$_} = $arg->{$_} foreach grep { defined $arg->{$_} } qw(path offset limit); |
|
foreach my $v (qw/path offset limit/) { |
|
|
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
|
|
} |
|
| 186 |
|
|
| 187 |
# create Text::Iconv object |
if ($arg->{load_row} || $arg->{save_row}) { |
| 188 |
$self->{iconv} = Text::Iconv->new($code_page,$self->{'code_page'}); |
$log->logconfess("save_row and load_row must be defined in pair and be CODE") unless ( |
| 189 |
|
ref($arg->{load_row}) eq 'CODE' && |
| 190 |
|
ref($arg->{save_row}) eq 'CODE' |
| 191 |
|
); |
| 192 |
|
$self->{load_row} = $arg->{load_row}; |
| 193 |
|
$self->{save_row} = $arg->{save_row}; |
| 194 |
|
$log->debug("using load_row and save_row instead of in-memory hash"); |
| 195 |
|
} |
| 196 |
|
|
| 197 |
my $filter_ref; |
my $filter_ref; |
| 198 |
|
my $recode_regex; |
| 199 |
|
my $recode_map; |
| 200 |
|
|
| 201 |
if ($self->{recode}) { |
if ($self->{recode}) { |
| 202 |
my @r = split(/\s/, $self->{recode}); |
my @r = split(/\s/, $self->{recode}); |
| 203 |
if ($#r % 2 != 1) { |
if ($#r % 2 != 1) { |
| 204 |
$log->logwarn("recode needs even number of elements (some number of valid pairs)"); |
$log->logwarn("recode needs even number of elements (some number of valid pairs)"); |
| 205 |
} else { |
} else { |
|
my $recode; |
|
| 206 |
while (@r) { |
while (@r) { |
| 207 |
my $from = shift @r; |
my $from = shift @r; |
| 208 |
my $to = shift @r; |
my $to = shift @r; |
| 209 |
$recode->{$from} = $to; |
$recode_map->{$from} = $to; |
| 210 |
} |
} |
| 211 |
|
|
| 212 |
my $regex = join '|' => keys %{ $recode }; |
$recode_regex = join '|' => keys %{ $recode_map }; |
|
|
|
|
$log->debug("using recode regex: $regex"); |
|
|
|
|
|
$filter_ref = sub { |
|
|
my $t = shift; |
|
|
$t =~ s/($regex)/$recode->{$1}/g; |
|
|
return $t; |
|
|
}; |
|
| 213 |
|
|
| 214 |
|
$log->debug("using recode regex: $recode_regex"); |
| 215 |
} |
} |
| 216 |
|
|
| 217 |
} |
} |
| 218 |
|
|
| 219 |
my ($db, $size) = $self->{open_db}->( $self, |
my $rec_regex; |
| 220 |
|
if (my $p = $arg->{modify_file}) { |
| 221 |
|
$log->debug("using modify_file $p"); |
| 222 |
|
$rec_regex = $self->modify_file_regexps( $p ); |
| 223 |
|
} elsif (my $h = $arg->{modify_records}) { |
| 224 |
|
$log->debug("using modify_records ", sub { dump( $h ) }); |
| 225 |
|
$rec_regex = $self->modify_record_regexps(%{ $h }); |
| 226 |
|
} |
| 227 |
|
$log->debug("rec_regex: ", sub { dump($rec_regex) }) if ($rec_regex); |
| 228 |
|
|
| 229 |
|
my $class = $self->{module} || $log->logconfess("can't get low-level module name!"); |
| 230 |
|
|
| 231 |
|
$arg->{$_} = $self->{$_} foreach qw(offset limit); |
| 232 |
|
|
| 233 |
|
my $ll_db = $class->new( |
| 234 |
path => $arg->{path}, |
path => $arg->{path}, |
| 235 |
filter => $filter_ref, |
input_config => $arg->{input_config} || $self->{input_config}, |
| 236 |
|
# filter => sub { |
| 237 |
|
# my ($l,$f_nr) = @_; |
| 238 |
|
# return unless defined($l); |
| 239 |
|
# $l = decode($input_encoding, $l); |
| 240 |
|
# $l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
| 241 |
|
# return $l; |
| 242 |
|
# }, |
| 243 |
|
%{ $arg }, |
| 244 |
); |
); |
| 245 |
|
|
| 246 |
unless (defined($db)) { |
# save for dump and input_module |
| 247 |
|
$self->{ll_db} = $ll_db; |
| 248 |
|
|
| 249 |
|
unless (defined($ll_db)) { |
| 250 |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
| 251 |
return; |
return; |
| 252 |
} |
} |
| 253 |
|
|
| 254 |
|
my $size = $ll_db->size; |
| 255 |
|
|
| 256 |
unless ($size) { |
unless ($size) { |
| 257 |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
| 258 |
return; |
return; |
| 262 |
my $to_rec = $size; |
my $to_rec = $size; |
| 263 |
|
|
| 264 |
if (my $s = $self->{offset}) { |
if (my $s = $self->{offset}) { |
| 265 |
$log->info("skipping to MFN $s"); |
$log->debug("offset $s records"); |
| 266 |
$from_rec = $s; |
$from_rec = $s + 1; |
| 267 |
} else { |
} else { |
| 268 |
$self->{offset} = $from_rec; |
$self->{offset} = $from_rec - 1; |
| 269 |
} |
} |
| 270 |
|
|
| 271 |
if ($self->{limit}) { |
if ($self->{limit}) { |
| 274 |
$to_rec = $size if ($to_rec > $size); |
$to_rec = $size if ($to_rec > $size); |
| 275 |
} |
} |
| 276 |
|
|
| 277 |
# store size for later |
my $strict_encoding = $arg->{strict_encoding} || $self->{strict_encoding}; ## FIXME should be 1 really |
| 278 |
$self->{size} = ($to_rec - $from_rec) ? ($to_rec - $from_rec + 1) : 0; |
|
| 279 |
|
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec]", |
| 280 |
|
" encoding $input_encoding ", $strict_encoding ? ' [strict]' : '', |
| 281 |
|
$self->{stats} ? ' [stats]' : '', |
| 282 |
|
); |
| 283 |
|
|
| 284 |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{code_page}", $self->{stats} ? ' [stats]' : ''); |
$self->{size} = 0; |
| 285 |
|
|
| 286 |
# read database |
# read database |
| 287 |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
| 288 |
|
|
| 289 |
$log->debug("position: $pos\n"); |
$log->debug("position: $pos\n"); |
| 290 |
|
|
| 291 |
my $rec = $self->{fetch_rec}->($self, $db, $pos ); |
$self->{size}++; # XXX I could move this more down if I didn't want empty records... |
| 292 |
|
|
| 293 |
|
my $rec = $ll_db->fetch_rec($pos, sub { |
| 294 |
|
my ($l,$f_nr,$debug) = @_; |
| 295 |
|
# return unless defined($l); |
| 296 |
|
# return $l unless ($rec_regex && $f_nr); |
| 297 |
|
|
| 298 |
|
return unless ( defined($l) && defined($f_nr) ); |
| 299 |
|
|
| 300 |
|
warn "-=> $f_nr ## |$l|\n" if ($debug); |
| 301 |
|
$log->debug("-=> $f_nr ## $l"); |
| 302 |
|
|
| 303 |
|
# codepage conversion and recode_regex |
| 304 |
|
$l = decode($input_encoding, $l, 1); |
| 305 |
|
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
| 306 |
|
|
| 307 |
|
# apply regexps |
| 308 |
|
if ($rec_regex && defined($rec_regex->{$f_nr})) { |
| 309 |
|
$log->logconfess("regexps->{$f_nr} must be ARRAY") if (ref($rec_regex->{$f_nr}) ne 'ARRAY'); |
| 310 |
|
my $c = 0; |
| 311 |
|
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
| 312 |
|
my $old_l = $l; |
| 313 |
|
$log->logconfess("expected regex in ", dump( $r )) unless defined($r->{regex}); |
| 314 |
|
eval '$l =~ ' . $r->{regex}; |
| 315 |
|
if ($old_l ne $l) { |
| 316 |
|
my $d = "|$old_l| -> |$l| "; # . $r->{regex}; |
| 317 |
|
$d .= ' +' . $r->{line} . ' ' . $r->{file} if defined($r->{line}); |
| 318 |
|
$d .= ' ' . $r->{debug} if defined($r->{debug}); |
| 319 |
|
$log->debug("MODIFY $d"); |
| 320 |
|
warn "*** $d\n" if ($debug); |
| 321 |
|
|
| 322 |
|
} |
| 323 |
|
$log->error("error applying regex: ",dump($r), $@) if $@; |
| 324 |
|
} |
| 325 |
|
} |
| 326 |
|
|
| 327 |
$log->debug(sub { Dumper($rec) }); |
$log->debug("<=- $f_nr ## |$l|"); |
| 328 |
|
warn "<=- $f_nr ## $l\n" if ($debug); |
| 329 |
|
return $l; |
| 330 |
|
}); |
| 331 |
|
|
| 332 |
|
$log->debug(sub { dump($rec) }); |
| 333 |
|
|
| 334 |
if (! $rec) { |
if (! $rec) { |
| 335 |
$log->warn("record $pos empty? skipping..."); |
$log->warn("record $pos empty? skipping..."); |
| 337 |
} |
} |
| 338 |
|
|
| 339 |
# store |
# store |
| 340 |
if ($self->{low_mem}) { |
if ($self->{save_row}) { |
| 341 |
$self->{db}->put($pos, $rec); |
$self->{save_row}->({ |
| 342 |
|
id => $pos, |
| 343 |
|
row => $rec, |
| 344 |
|
}); |
| 345 |
} else { |
} else { |
| 346 |
$self->{data}->{$pos} = $rec; |
$self->{data}->{$pos} = $rec; |
| 347 |
} |
} |
| 348 |
|
|
| 349 |
# create lookup |
# create lookup |
| 350 |
$self->{'lookup'}->add( $rec ) if ($rec && $self->{'lookup'}); |
$arg->{'lookup_coderef'}->( $rec ) if ($rec && $arg->{'lookup_coderef'}); |
| 351 |
|
|
| 352 |
# update counters for statistics |
# update counters for statistics |
| 353 |
if ($self->{stats}) { |
if ($self->{stats}) { |
| 354 |
map { |
|
| 355 |
my $fld = $_; |
# fetch clean record with regexpes applied for statistics |
| 356 |
|
my $rec = $ll_db->fetch_rec($pos); |
| 357 |
|
|
| 358 |
|
foreach my $fld (keys %{ $rec }) { |
| 359 |
$self->{_stats}->{fld}->{ $fld }++; |
$self->{_stats}->{fld}->{ $fld }++; |
| 360 |
if (ref($rec->{ $fld }) eq 'ARRAY') { |
|
| 361 |
map { |
#$log->logdie("invalid record fild $fld, not ARRAY") |
| 362 |
if (ref($_) eq 'HASH') { |
next unless (ref($rec->{ $fld }) eq 'ARRAY'); |
| 363 |
map { |
|
| 364 |
$self->{_stats}->{sf}->{ $fld }->{ $_ }++; |
foreach my $row (@{ $rec->{$fld} }) { |
| 365 |
} keys %{ $_ }; |
|
| 366 |
} else { |
if (ref($row) eq 'HASH') { |
| 367 |
$self->{_stats}->{repeatable}->{ $fld }++; |
|
| 368 |
|
foreach my $sf (keys %{ $row }) { |
| 369 |
|
next if ($sf eq 'subfields'); |
| 370 |
|
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{count}++; |
| 371 |
|
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{repeatable}++ |
| 372 |
|
if (ref($row->{$sf}) eq 'ARRAY'); |
| 373 |
} |
} |
| 374 |
} @{ $rec->{$fld} }; |
|
| 375 |
|
} else { |
| 376 |
|
$self->{_stats}->{repeatable}->{ $fld }++; |
| 377 |
|
} |
| 378 |
} |
} |
| 379 |
} keys %{ $rec }; |
} |
| 380 |
} |
} |
| 381 |
|
|
| 382 |
$self->progress_bar($pos,$to_rec) unless ($self->{no_progress_bar}); |
$self->progress_bar($pos,$to_rec) unless ($self->{no_progress_bar}); |
| 393 |
return $size; |
return $size; |
| 394 |
} |
} |
| 395 |
|
|
| 396 |
|
sub input_module { $_[0]->{ll_db} } |
| 397 |
|
|
| 398 |
=head2 fetch |
=head2 fetch |
| 399 |
|
|
| 400 |
Fetch next record from database. It will also displays progress bar. |
Fetch next record from database. It will also displays progress bar. |
| 414 |
$log->logconfess("it seems that you didn't load database!") unless ($self->{pos}); |
$log->logconfess("it seems that you didn't load database!") unless ($self->{pos}); |
| 415 |
|
|
| 416 |
if ($self->{pos} == -1) { |
if ($self->{pos} == -1) { |
| 417 |
$self->{pos} = $self->{offset}; |
$self->{pos} = $self->{offset} + 1; |
| 418 |
} else { |
} else { |
| 419 |
$self->{pos}++; |
$self->{pos}++; |
| 420 |
} |
} |
| 431 |
|
|
| 432 |
my $rec; |
my $rec; |
| 433 |
|
|
| 434 |
if ($self->{low_mem}) { |
if ($self->{load_row}) { |
| 435 |
$rec = $self->{db}->get($mfn); |
$rec = $self->{load_row}->({ id => $mfn }); |
| 436 |
} else { |
} else { |
| 437 |
$rec = $self->{data}->{$mfn}; |
$rec = $self->{data}->{$mfn}; |
| 438 |
} |
} |
| 472 |
|
|
| 473 |
sub size { |
sub size { |
| 474 |
my $self = shift; |
my $self = shift; |
| 475 |
return $self->{size}; |
return $self->{size}; # FIXME this is buggy if open is called multiple times! |
| 476 |
} |
} |
| 477 |
|
|
| 478 |
=head2 seek |
=head2 seek |
| 487 |
|
|
| 488 |
sub seek { |
sub seek { |
| 489 |
my $self = shift; |
my $self = shift; |
| 490 |
my $pos = shift || return; |
my $pos = shift; |
| 491 |
|
|
| 492 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
| 493 |
|
|
| 494 |
|
$log->logconfess("called without pos") unless defined($pos); |
| 495 |
|
|
| 496 |
if ($pos < 1) { |
if ($pos < 1) { |
| 497 |
$log->warn("seek before first record"); |
$log->warn("seek before first record"); |
| 498 |
$pos = 1; |
$pos = 1; |
| 527 |
|
|
| 528 |
my $out = join("\n", |
my $out = join("\n", |
| 529 |
map { |
map { |
| 530 |
my $f = $_ || die "no field"; |
my $f = $_; |
| 531 |
|
die "no field in ", dump( $s->{fld} ) unless defined( $f ); |
| 532 |
my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}"; |
my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}"; |
| 533 |
$max_fld = $v if ($v > $max_fld); |
$max_fld = $v if ($v > $max_fld); |
| 534 |
|
|
| 535 |
my $o = sprintf("%4d %d ~", $f, $v); |
my $o = sprintf("%4s %d ~", $f, $v); |
| 536 |
|
|
| 537 |
if (defined($s->{sf}->{$f})) { |
if (defined($s->{sf}->{$f})) { |
| 538 |
|
my @subfields = keys %{ $s->{sf}->{$f} }; |
| 539 |
map { |
map { |
| 540 |
$o .= sprintf(" %s:%d", $_, $s->{sf}->{$f}->{$_}); |
$o .= sprintf(" %s:%d%s", $_, |
| 541 |
} sort keys %{ $s->{sf}->{$f} }; |
$s->{sf}->{$f}->{$_}->{count}, |
| 542 |
|
$s->{sf}->{$f}->{$_}->{repeatable} ? '*' : '', |
| 543 |
|
); |
| 544 |
|
} ( |
| 545 |
|
# first indicators and other special subfields |
| 546 |
|
sort( grep { length($_) > 1 } @subfields ), |
| 547 |
|
# then subfileds (single char) |
| 548 |
|
sort( grep { length($_) == 1 } @subfields ), |
| 549 |
|
); |
| 550 |
} |
} |
| 551 |
|
|
| 552 |
if (my $v_r = $s->{repeatable}->{$f}) { |
if (my $v_r = $s->{repeatable}->{$f}) { |
| 554 |
} |
} |
| 555 |
|
|
| 556 |
$o; |
$o; |
| 557 |
} sort { $a <=> $b } keys %{ $s->{fld} } |
} sort { |
| 558 |
|
if ( $a =~ m/^\d+$/ && $b =~ m/^\d+$/ ) { |
| 559 |
|
$a <=> $b |
| 560 |
|
} else { |
| 561 |
|
$a cmp $b |
| 562 |
|
} |
| 563 |
|
} keys %{ $s->{fld} } |
| 564 |
); |
); |
| 565 |
|
|
| 566 |
$log->debug( sub { Dumper($s) } ); |
$log->debug( sub { dump($s) } ); |
| 567 |
|
|
| 568 |
|
my $path = 'var/stats.yml'; |
| 569 |
|
YAML::DumpFile( $path, $s ); |
| 570 |
|
$log->info( 'created ', $path, ' with ', -s $path, ' bytes' ); |
| 571 |
|
|
| 572 |
return $out; |
return $out; |
| 573 |
} |
} |
| 574 |
|
|
| 575 |
=head1 MEMORY USAGE |
=head2 dump_ascii |
| 576 |
|
|
| 577 |
|
Display humanly readable dump of record |
| 578 |
|
|
| 579 |
|
=cut |
| 580 |
|
|
| 581 |
|
sub dump_ascii { |
| 582 |
|
my $self = shift; |
| 583 |
|
|
| 584 |
|
return unless $self->{ll_db}; |
| 585 |
|
|
| 586 |
|
if ($self->{ll_db}->can('dump_ascii')) { |
| 587 |
|
return $self->{ll_db}->dump_ascii( $self->{pos} ); |
| 588 |
|
} else { |
| 589 |
|
return dump( $self->{ll_db}->fetch_rec( $self->{pos} ) ); |
| 590 |
|
} |
| 591 |
|
} |
| 592 |
|
|
| 593 |
|
=head2 _get_regex |
| 594 |
|
|
| 595 |
|
Helper function called which create regexps to be execute on code. |
| 596 |
|
|
| 597 |
|
_get_regex( 900, 'regex:[0-9]+' ,'numbers' ); |
| 598 |
|
_get_regex( 900, '^b', ' : ^b' ); |
| 599 |
|
|
| 600 |
|
It supports perl regexps with C<regex:> prefix to from value and has |
| 601 |
|
additional logic to skip empty subfields. |
| 602 |
|
|
| 603 |
|
=cut |
| 604 |
|
|
| 605 |
|
sub _get_regex { |
| 606 |
|
my ($sf,$from,$to) = @_; |
| 607 |
|
|
| 608 |
|
# protect / |
| 609 |
|
$from =~ s!/!\\/!gs; |
| 610 |
|
$to =~ s!/!\\/!gs; |
| 611 |
|
|
| 612 |
|
if ($from =~ m/^regex:(.+)$/) { |
| 613 |
|
$from = $1; |
| 614 |
|
} else { |
| 615 |
|
$from = '\Q' . $from . '\E'; |
| 616 |
|
} |
| 617 |
|
if ($sf =~ /^\^/) { |
| 618 |
|
my $need_subfield_data = '*'; # no |
| 619 |
|
# if from is also subfield, require some data in between |
| 620 |
|
# to correctly skip empty subfields |
| 621 |
|
$need_subfield_data = '+' if ($from =~ m/^\\Q\^/); |
| 622 |
|
return |
| 623 |
|
's/\Q'. $sf .'\E([^\^]' . $need_subfield_data . '?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
| 624 |
|
} else { |
| 625 |
|
return |
| 626 |
|
's/'. $from .'/'. $to .'/g'; |
| 627 |
|
} |
| 628 |
|
} |
| 629 |
|
|
| 630 |
|
|
| 631 |
|
=head2 modify_record_regexps |
| 632 |
|
|
| 633 |
|
Generate hash with regexpes to be applied using L<filter>. |
| 634 |
|
|
| 635 |
|
my $regexpes = $input->modify_record_regexps( |
| 636 |
|
900 => { '^a' => { ' : ' => '^b' } }, |
| 637 |
|
901 => { '*' => { '^b' => ' ; ' } }, |
| 638 |
|
); |
| 639 |
|
|
| 640 |
|
=cut |
| 641 |
|
|
| 642 |
|
sub modify_record_regexps { |
| 643 |
|
my $self = shift; |
| 644 |
|
my $modify_record = {@_}; |
| 645 |
|
|
| 646 |
|
my $regexpes; |
| 647 |
|
|
| 648 |
|
my $log = $self->_get_logger(); |
| 649 |
|
|
| 650 |
|
foreach my $f (keys %$modify_record) { |
| 651 |
|
$log->debug("field: $f"); |
| 652 |
|
|
| 653 |
C<low_mem> options is double-edged sword. If enabled, WebPAC |
foreach my $sf (keys %{ $modify_record->{$f} }) { |
| 654 |
will run on memory constraint machines (which doesn't have enough |
$log->debug("subfield: $sf"); |
|
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. |
|
| 655 |
|
|
| 656 |
|
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
| 657 |
|
my $to = $modify_record->{$f}->{$sf}->{$from}; |
| 658 |
|
#die "no field?" unless defined($to); |
| 659 |
|
my $d = "|$from| -> |$to|"; |
| 660 |
|
$log->debug("transform: $d"); |
| 661 |
|
|
| 662 |
|
my $regex = _get_regex($sf,$from,$to); |
| 663 |
|
push @{ $regexpes->{$f} }, { regex => $regex, debug => $d }; |
| 664 |
|
$log->debug("regex: $regex"); |
| 665 |
|
} |
| 666 |
|
} |
| 667 |
|
} |
| 668 |
|
|
| 669 |
|
return $regexpes; |
| 670 |
|
} |
| 671 |
|
|
| 672 |
|
=head2 modify_file_regexps |
| 673 |
|
|
| 674 |
|
Generate hash with regexpes to be applied using L<filter> from |
| 675 |
|
pseudo hash/yaml format for regex mappings. |
| 676 |
|
|
| 677 |
|
It should be obvious: |
| 678 |
|
|
| 679 |
|
200 |
| 680 |
|
'^a' |
| 681 |
|
' : ' => '^e' |
| 682 |
|
' = ' => '^d' |
| 683 |
|
|
| 684 |
|
In field I<200> find C<'^a'> and then C<' : '>, and replace it with C<'^e'>. |
| 685 |
|
In field I<200> find C<'^a'> and then C<' = '>, and replace it with C<'^d'>. |
| 686 |
|
|
| 687 |
|
my $regexpes = $input->modify_file_regexps( 'conf/modify/common.pl' ); |
| 688 |
|
|
| 689 |
|
On undef path it will just return. |
| 690 |
|
|
| 691 |
|
=cut |
| 692 |
|
|
| 693 |
|
sub modify_file_regexps { |
| 694 |
|
my $self = shift; |
| 695 |
|
|
| 696 |
|
my $modify_path = shift || return; |
| 697 |
|
|
| 698 |
|
my $log = $self->_get_logger(); |
| 699 |
|
|
| 700 |
|
my $regexpes; |
| 701 |
|
|
| 702 |
|
CORE::open(my $fh, $modify_path) || $log->logdie("can't open modify file $modify_path: $!"); |
| 703 |
|
|
| 704 |
|
my ($f,$sf); |
| 705 |
|
|
| 706 |
|
while(<$fh>) { |
| 707 |
|
chomp; |
| 708 |
|
next if (/^#/ || /^\s*$/); |
| 709 |
|
|
| 710 |
|
if (/^\s*(\d+)\s*$/) { |
| 711 |
|
$f = $1; |
| 712 |
|
$log->debug("field: $f"); |
| 713 |
|
next; |
| 714 |
|
} elsif (/^\s*'([^']*)'\s*$/) { |
| 715 |
|
$sf = $1; |
| 716 |
|
$log->die("can't define subfiled before field in: $_") unless ($f); |
| 717 |
|
$log->debug("subfield: $sf"); |
| 718 |
|
} elsif (/^\s*'([^']*)'\s*=>\s*'([^']*)'\s*$/) { |
| 719 |
|
my ($from,$to) = ($1, $2); |
| 720 |
|
|
| 721 |
|
$log->debug("transform: |$from| -> |$to|"); |
| 722 |
|
|
| 723 |
|
my $regex = _get_regex($sf,$from,$to); |
| 724 |
|
push @{ $regexpes->{$f} }, { |
| 725 |
|
regex => $regex, |
| 726 |
|
file => $modify_path, |
| 727 |
|
line => $., |
| 728 |
|
}; |
| 729 |
|
$log->debug("regex: $regex"); |
| 730 |
|
} else { |
| 731 |
|
die "can't parse: $_"; |
| 732 |
|
} |
| 733 |
|
} |
| 734 |
|
|
| 735 |
|
return $regexpes; |
| 736 |
|
} |
| 737 |
|
|
| 738 |
=head1 AUTHOR |
=head1 AUTHOR |
| 739 |
|
|
| 741 |
|
|
| 742 |
=head1 COPYRIGHT & LICENSE |
=head1 COPYRIGHT & LICENSE |
| 743 |
|
|
| 744 |
Copyright 2005 Dobrica Pavlinusic, All Rights Reserved. |
Copyright 2005-2006 Dobrica Pavlinusic, All Rights Reserved. |
| 745 |
|
|
| 746 |
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 |
| 747 |
under the same terms as Perl itself. |
under the same terms as Perl itself. |
Parent Directory
| 
Revision Log
| 
Patch