| 1 | #################################################################### | ||
| 2 | # Copyright (C) 2004 Jörg Tiedemann <joerg@stp.ling.uu.se> | ||
| 3 | # | ||
| 4 | # This program is free software; you can redistribute it and/or modify | ||
| 5 | # it under the terms of the GNU General Public License as published by | ||
| 6 | # the Free Software Foundation; either version 2 of the License, or | ||
| 7 | # (at your option) any later version. | ||
| 8 | # | ||
| 9 | # This program is distributed in the hope that it will be useful, | ||
| 10 | # but WITHOUT ANY WARRANTY; without even the implied warranty of | ||
| 11 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | ||
| 12 | # GNU General Public License for more details. | ||
| 13 | # | ||
| 14 | # You should have received a copy of the GNU General Public License | ||
| 15 | # along with this program; if not, write to the Free Software | ||
| 16 | # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA | ||
| 17 | # | ||
| 18 | # Uplug::IO::DBM | ||
| 19 | # | ||
| 20 | # simple module for storing uplug-data in DBM-files | ||
| 21 | # | ||
| 22 | ##################################################################### | ||
| 23 | # $Author$ | ||
| 24 | # $Id$ | ||
| 25 | |||
| 26 | |||
| 27 | |||
| 28 | package Uplug::IO::DBM; | ||
| 29 | |||
| 30 | # use utf8; | ||
| 31 | # use bytes; | ||
| 32 | use strict; | ||
| 33 | use vars qw(@ISA $DEFAULTMODE $DEFAULTCACHESIZE); | ||
| 34 | use Uplug::IO; | ||
| 35 | use Uplug::Data; | ||
| 36 | use Uplug::Encoding; | ||
| 37 | use Data::Dumper; | ||
| 38 | |||
| 39 | @ISA = qw( Uplug::IO ); | ||
| 40 | |||
| 41 | BEGIN { @AnyDBM_File::ISA=qw(DB_File GDBM_File SDBM_File NDBM_File ODBM_File) } | ||
| 42 | # BEGIN { @AnyDBM_File::ISA=qw(SDBM_File GDBM_File_ NDBM_File ODBM_File) } | ||
| 43 | use AnyDBM_File; | ||
| 44 | use POSIX; | ||
| 45 | |||
| 46 | #--------------------------------------------------------------------------- | ||
| 47 | # simple caching: the first selected items get chached | ||
| 48 | # nothing will be added if the cache is full until the DBM is opened again | ||
| 49 | |||
| 50 | $DEFAULTCACHESIZE=200000; # max number of items cached by GetKeyData | ||
| 51 | #$DEFAULTCACHESIZE=0; | ||
| 52 | $DEFAULTMODE=0644; # default file mode for tie | ||
| 53 | |||
| 54 | |||
| 55 | |||
| 56 | sub open{ | ||
| 57 | my $self = shift; | ||
| 58 | $self->{AccessMode} = shift; | ||
| 59 | my $HeaderHash = $_[0]; | ||
| 60 | if (not $self->{AccessMode}){$self->{AccessMode}='read';} | ||
| 61 | if ($self->{AccessMode} eq 'write'){ | ||
| 62 | if (defined $HeaderHash->{'write_mode'}){ | ||
| 63 | $self->{AccessMode}=$HeaderHash->{'write_mode'}; | ||
| 64 | } | ||
| 65 | } | ||
| 66 | |||
| 67 | $self->{CACHED}=0; | ||
| 68 | $self->{CACHESIZE}=$DEFAULTCACHESIZE; | ||
| 69 | $self->SUPER::init($HeaderHash); | ||
| 70 | |||
| 71 | if ((not defined $self->{StreamOptions}->{FileName}) and | ||
| 72 | (defined $self->{StreamOptions}->{file})){ | ||
| 73 | $self->{StreamOptions}->{FileName}=$self->{StreamOptions}->{file}; | ||
| 74 | } | ||
| 75 | $self->{FileName}=$self->{StreamOptions}->{FileName}; | ||
| 76 | $self->{StreamOptions}->{file}=$self->{StreamOptions}->{FileName}; | ||
| 77 | |||
| 78 | |||
| 79 | if (defined $self->{FileName}){ | ||
| 80 | $self->{DBMhash}={}; | ||
| 81 | if (not defined $self->{StreamOptions}->{'file mode'}){ | ||
| 82 | $self->{StreamOptions}->{'file mode'}=$Uplug::IO::DBM::DEFAULTMODE; | ||
| 83 | } | ||
| 84 | if (-e $self->{'FileName'}){ | ||
| 85 | if ($self->{AccessMode} eq 'write'){ | ||
| 86 | warn "Cannot open $self->{'FileName'} for writing!"; | ||
| 87 | return 0; | ||
| 88 | } | ||
| 89 | if ($self->{AccessMode} eq 'overwrite'){ | ||
| 90 | unlink $self->{'FileName'}; | ||
| 91 | } | ||
| 92 | } | ||
| 93 | if ($self->{AccessMode} eq 'read'){ | ||
| 94 | if (not $self->{DBH}=tie %{$self->{DBMhash}}, | ||
| 95 | 'AnyDBM_File', | ||
| 96 | $self->{'FileName'}, | ||
| 97 | O_RDONLY, | ||
| 98 | $self->{StreamOptions}->{'file mode'}){ | ||
| 99 | return 0; | ||
| 100 | } | ||
| 101 | } | ||
| 102 | elsif (not $self->{DBH}=tie %{$self->{DBMhash}}, | ||
| 103 | 'AnyDBM_File', | ||
| 104 | $self->{'FileName'}, | ||
| 105 | O_RDWR|O_CREAT, | ||
| 106 | $self->{StreamOptions}->{'file mode'}){ | ||
| 107 | return 0; | ||
| 108 | } | ||
| 109 | |||
| 110 | |||
| 111 | ## JT 2005-03-14: | ||
| 112 | |||
| 113 | if ($]>=5.006){ | ||
| 114 | my $mod=qw( Encode ); | ||
| 115 | eval "require $mod"; | ||
| 116 | if ($@){warn $@;exit;} | ||
| 117 | # $self->{DBH}->filter_store_key(sub {Encode::_utf8_off($_)}); | ||
| 118 | # $self->{DBH}->filter_store_value(sub{Encode::_utf8_off($_)}); | ||
| 119 | # $self->{DBH}->filter_fetch_key(sub{Encode::_utf8_off($_)}); | ||
| 120 | # $self->{DBH}->filter_fetch_value(sub {Encode::_utf8_off($_)}); | ||
| 121 | $self->{DBH}->filter_store_key(sub {$_=Encode::encode('utf8',$_)}); | ||
| 122 | $self->{DBH}->filter_store_value(sub{$_=Encode::encode('utf8',$_)}); | ||
| 123 | $self->{DBH}->filter_fetch_key(sub{$_=Encode::decode('utf8',$_)}); | ||
| 124 | $self->{DBH}->filter_fetch_value(sub {$_=Encode::decode('utf8',$_)}); | ||
| 125 | } | ||
| 126 | } | ||
| 127 | if (defined $HeaderHash->{'key'}){ | ||
| 128 | if (ref($HeaderHash->{'key'}) eq 'ARRAY'){ | ||
| 129 | my @arr=@{$HeaderHash->{'key'}}; | ||
| 130 | $self->{'key'}=[]; | ||
| 131 | @{$self->{'key'}}=@arr; | ||
| 132 | } | ||
| 133 | else{ | ||
| 134 | if ($HeaderHash->{'key'}=~/^\((.*)\)$/){ | ||
| 135 | my @arr=split(/\s*,\s*/,$1); | ||
| 136 | $self->{'key'}=[]; | ||
| 137 | @{$self->{'key'}}=@arr; | ||
| 138 | } | ||
| 139 | else{ | ||
| 140 | $self->{'key'}=[]; | ||
| 141 | @{$self->{'key'}}=($HeaderHash->{'key'}); | ||
| 142 | } | ||
| 143 | } | ||
| 144 | } | ||
| 145 | my %HeaderOptions=('FileName' => $self->{FileName}.'.head'); | ||
| 146 | $self->{HeaderStream}=Uplug::IO::Text->new; | ||
| 147 | $self->{HeaderStream}->open($self->{AccessMode},\%HeaderOptions); | ||
| 148 | $self->addheader($self->{HeaderStream}->header); | ||
| 149 | return $self->SUPER::open($self->{AccessMode},$HeaderHash); | ||
| 150 | } | ||
| 151 | |||
| 152 | |||
| 153 | |||
| 154 | sub close{ | ||
| 155 | my $self=shift; | ||
| 156 | $self->{HeaderStream}->close; | ||
| 157 | untie %{$self->{DBMhash}}; | ||
| 158 | return 1; | ||
| 159 | } | ||
| 160 | |||
| 161 | |||
| 162 | sub read{ | ||
| 163 | my ($self,$TreeData)=@_; | ||
| 164 | |||
| 165 | $TreeData->init; | ||
| 166 | |||
| 167 | my $data={}; | ||
| 168 | my ($key,$val)=each %{$self->{DBMhash}}; | ||
| 169 | |||
| 170 | if ((not defined $key) and (not defined $val)){ | ||
| 171 | return 0; | ||
| 172 | } | ||
| 173 | |||
| 174 | my @keyval=split(/\x00/,$key); | ||
| 175 | %{$data}=split(/\x00/,$val); | ||
| 176 | if ($val=~/\|/){ | ||
| 177 | foreach (keys %{$data}){ | ||
| 178 | if ($data->{$_}=~/\|/){ | ||
| 179 | my @arr=split(/\|/,$data->{$_}); | ||
| 180 | $data->{$_}=undef; | ||
| 181 | @{$data->{$_}}=@arr; | ||
| 182 | } | ||
| 183 | } | ||
| 184 | } | ||
| 185 | if (defined $self->{key}){ | ||
| 186 | foreach (@{$self->{'key'}}){ | ||
| 187 | $data->{$_}=shift(@keyval); | ||
| 188 | } | ||
| 189 | } | ||
| 190 | %{$data}=&expandData($data); | ||
| 191 | $TreeData->setData($data); | ||
| 192 | return 1; | ||
| 193 | } | ||
| 194 | |||
| 195 | |||
| 196 | |||
| 197 | sub write{ | ||
| 198 | my ($self,$TreeData)=@_; | ||
| 199 | |||
| 200 | $self->SUPER::write($TreeData); | ||
| 201 | my $data=$TreeData->data; | ||
| 202 | if (ref($data) ne 'HASH'){return;} | ||
| 203 | |||
| 204 | my %dat=%{$data}; | ||
| 205 | my $key; | ||
| 206 | my @fields=(); | ||
| 207 | if (defined $self->{key}){ | ||
| 208 | foreach (@{$self->{'key'}}){ | ||
| 209 | push (@fields,$dat{$_}); | ||
| 210 | delete $dat{$_}; | ||
| 211 | } | ||
| 212 | } | ||
| 213 | $key=join ("\x00",@fields); | ||
| 214 | if (not $key){ | ||
| 215 | $self->{'LastNr'}++; | ||
| 216 | $key=$self->{'LastNr'}; | ||
| 217 | } | ||
| 218 | |||
| 219 | my %TmpData=$self->GetKeyValue($key); | ||
| 220 | if (keys %TmpData){ | ||
| 221 | my $NewData=&JoinComplexHashs(\%TmpData,\%dat); | ||
| 222 | %dat=%{$NewData}; | ||
| 223 | } | ||
| 224 | foreach (keys %dat){if ($dat{$_} eq ''){$dat{$_}=' ';}} | ||
| 225 | |||
| 226 | %dat=&dumpData(\%dat); | ||
| 227 | my $d=join ("\x00",%dat); | ||
| 228 | |||
| 229 | eval { $self->{DBMhash}->{$key}=$d; }; | ||
| 230 | return 1; | ||
| 231 | } | ||
| 232 | |||
| 233 | |||
| 234 | sub count{ | ||
| 235 | my $self=shift; | ||
| 236 | my $records=keys %{$self->{DBMhash}}; | ||
| 237 | return $records; | ||
| 238 | } | ||
| 239 | |||
| 240 | |||
| 241 | |||
| 242 | #----------------------------------------------------------------------------- | ||
| 243 | |||
| 244 | sub select{ | ||
| 245 | my $self=shift; | ||
| 246 | my ($data, | ||
| 247 | $Query, | ||
| 248 | $ListOfAttr, | ||
| 249 | $CmpOperator)=@_; | ||
| 250 | |||
| 251 | if (not ref($data)){return;} # no return data object found | ||
| 252 | if (ref($Query) ne 'HASH'){return;} # query pattern is no hash | ||
| 253 | my %Pattern=%{$Query}; # save query in a new hash | ||
| 254 | |||
| 255 | my @keys; | ||
| 256 | if (defined $self->{'key'}){ # hash-key fields | ||
| 257 | @keys=@{$self->{'key'}}; | ||
| 258 | } | ||
| 259 | if (not @keys){ # if no hash keys specified: | ||
| 260 | return $self->SUPER::select($data, # use standard select | ||
| 261 | $Query, # (read sequentially through | ||
| 262 | $ListOfAttr, # the database --> slow!) | ||
| 263 | $CmpOperator); | ||
| 264 | } | ||
| 265 | my $key; | ||
| 266 | my $keycomplete=1; # hash-key is complete | ||
| 267 | my @fields=(); | ||
| 268 | foreach my $k (@keys){ # check all key-fields | ||
| 269 | if (not defined $Pattern{$k}){ # if one of them is not | ||
| 270 | $keycomplete=0; # specified --> reset flag | ||
| 271 | push (@fields,'.*'); # and create a wild-card-key | ||
| 272 | } | ||
| 273 | else{ | ||
| 274 | push(@fields,$Pattern{$k}); # otherwise: save key-field | ||
| 275 | delete $Pattern{$k}; # and delete pattern | ||
| 276 | } | ||
| 277 | } | ||
| 278 | $key=join ("\x00",@fields); # join all keys together | ||
| 279 | |||
| 280 | #------------------------------------------------------------------------- | ||
| 281 | my %hash=(); | ||
| 282 | if ($keycomplete){ # if the key is complete: | ||
| 283 | %hash=$self->GetKeyData($key); # * get data from hash | ||
| 284 | } | ||
| 285 | else{ # key is not complete! | ||
| 286 | $key=qr/^$key\$/; # make regular expression | ||
| 287 | my ($k,$v); | ||
| 288 | while (($k,$v)=$self->getNext()){ # run through all keys | ||
| 289 | if ($k=~/$key/){ # and try to match the key! | ||
| 290 | %hash=$self->GetKeyData($k); # - get the data | ||
| 291 | } | ||
| 292 | if (keys %hash){last;} | ||
| 293 | } | ||
| 294 | } | ||
| 295 | #------------------------------------------------------------------------- | ||
| 296 | if (not keys %hash){return 0;} # - key not found -> return | ||
| 297 | if (not &MatchQuery(\%Pattern,\%hash)){ # - other attr don't match | ||
| 298 | return 0; | ||
| 299 | } | ||
| 300 | $data->setData(\%hash); | ||
| 301 | return 1; | ||
| 302 | } | ||
| 303 | |||
| 304 | #----------------------------------------------------------------------------- | ||
| 305 | |||
| 306 | |||
| 307 | sub getNext{ | ||
| 308 | my $self=shift; | ||
| 309 | return each %{$self->{DBMhash}}; | ||
| 310 | } | ||
| 311 | |||
| 312 | sub getValue{ | ||
| 313 | return $_[0]->{DBMhash}->{$_[1]}; | ||
| 314 | } | ||
| 315 | |||
| 316 | #------------------------------------------------------------------- | ||
| 317 | # match a query pattern with a data structure | ||
| 318 | # ... this is much too simple ... | ||
| 319 | # ... should be improved in many ways | ||
| 320 | # | ||
| 321 | |||
| 322 | sub MatchQuery{ | ||
| 323 | my ($query,$data)=@_; | ||
| 324 | |||
| 325 | if (ref($query) ne ref($data)){return 0;} | ||
| 326 | if (not ref($query)){ | ||
| 327 | return $data=~/^$query$/; | ||
| 328 | } | ||
| 329 | if (ref($query) eq 'ARRAY'){ | ||
| 330 | foreach my $a (@{$query}){ | ||
| 331 | if (not &MatchQuery($query->[$a],$data->[$a])){ | ||
| 332 | return 0; | ||
| 333 | } | ||
| 334 | } | ||
| 335 | } | ||
| 336 | foreach my $a (keys %{$query}){ | ||
| 337 | if (not &MatchQuery($query->{$a},$data->{$a})){ | ||
| 338 | return 0; | ||
| 339 | } | ||
| 340 | } | ||
| 341 | return 1; | ||
| 342 | } | ||
| 343 | |||
| 344 | |||
| 345 | sub readheader{ | ||
| 346 | my $self=shift; | ||
| 347 | $self->{HeaderStream}->readheader; | ||
| 348 | $self->addheader($self->{HeaderStream}->header); | ||
| 349 | } | ||
| 350 | |||
| 351 | sub writeheader{ | ||
| 352 | my $self=shift; | ||
| 353 | $self->{HeaderStream}->addheader($self->header); | ||
| 354 | $self->{HeaderStream}->writeheader; | ||
| 355 | } | ||
| 356 | |||
| 357 | |||
| 358 | |||
| 359 | |||
| 360 | sub GetKeyData{ | ||
| 361 | my ($self,$key)=@_; | ||
| 362 | |||
| 363 | #----------------------------------------------- | ||
| 364 | if (defined $self->{DBMcache}->{$key}){ | ||
| 365 | # print STDERR "found $key in cache ($self->{CACHED})!\n"; | ||
| 366 | return %{$self->{DBMcache}->{$key}}; | ||
| 367 | } | ||
| 368 | #----------------------------------------------- | ||
| 369 | |||
| 370 | if (defined $self->{DBMhash}->{$key}){ | ||
| 371 | my $dat=$self->{DBMhash}->{$key}; | ||
| 372 | my @keyval=split(/\x00/,$key); | ||
| 373 | my %FoundData=split(/\x00/,$dat); | ||
| 374 | &expandData(\%FoundData); | ||
| 375 | foreach (@{$self->{'key'}}){ | ||
| 376 | $FoundData{$_}=shift(@keyval); | ||
| 377 | } | ||
| 378 | #----------------------------------------------- | ||
| 379 | if ($self->{CACHED}<$self->{CACHESIZE}){ | ||
| 380 | %{$self->{DBMcache}->{$key}}=%FoundData; | ||
| 381 | $self->{CACHED}++; | ||
| 382 | } | ||
| 383 | #----------------------------------------------- | ||
| 384 | return %FoundData; | ||
| 385 | } | ||
| 386 | #----------------------------------------------- | ||
| 387 | # JT 14-11-2004: cache even not-found keys | ||
| 388 | if ($self->{CACHED}<$self->{CACHESIZE}){ | ||
| 389 | %{$self->{DBMcache}->{$key}}=(); | ||
| 390 | $self->{CACHED}++; | ||
| 391 | } | ||
| 392 | #----------------------------------------------- | ||
| 393 | return (); | ||
| 394 | } | ||
| 395 | |||
| 396 | sub GetKeyValue{ | ||
| 397 | my ($self,$key)=@_; | ||
| 398 | if (defined $self->{DBMhash}->{$key}){ | ||
| 399 | my $dat=$self->{DBMhash}->{$key}; | ||
| 400 | my %FoundData=split(/\x00/,$dat); | ||
| 401 | &expandData(\%FoundData); | ||
| 402 | return %FoundData; | ||
| 403 | } | ||
| 404 | return (); | ||
| 405 | } | ||
| 406 | |||
| 407 | |||
| 408 | |||
| 409 | #################################################### | ||
| 410 | # put 2 hash-structures together | ||
| 411 | |||
| 412 | sub JoinComplexHashs{ | ||
| 413 | |||
| 414 | my ($data1,$data2)=@_; | ||
| 415 | if (not defined $data2->{freq}){$data2->{freq}=1;} | ||
| 416 | if (not defined $data1->{freq}){$data1->{freq}=1;} | ||
| 417 | $data1->{freq}+=$data2->{freq}; | ||
| 418 | foreach my $key (keys %{$data2}){ | ||
| 419 | if ($key eq 'freq'){next;} | ||
| 420 | if (defined $data1->{$key}){ | ||
| 421 | if ((not ref($data2->{$key})) and | ||
| 422 | ($data1->{$key} eq $data2->{$key})){ | ||
| 423 | next; | ||
| 424 | } | ||
| 425 | if (not ref($data1->{$key})){ | ||
| 426 | my $old=$data1->{$key}; | ||
| 427 | $data1->{$key}=[]; | ||
| 428 | $data1->{$key}->[0]=$old; | ||
| 429 | } | ||
| 430 | if (not ref($data2->{$key})){ | ||
| 431 | my $old=$data2->{$key}; | ||
| 432 | $data2->{$key}=[]; | ||
| 433 | $data2->{$key}->[0]=$old; | ||
| 434 | } | ||
| 435 | # if (ref($data1->{$key}) ne ref($data2->{$key})){ | ||
| 436 | # my $old=$data1->{$key}; | ||
| 437 | # my $new=$data2->{$key}; | ||
| 438 | # $data1->{$key}=[]; | ||
| 439 | # $data1->{$key}->[0]=$old; | ||
| 440 | # $Data::Dumper::Terse=1; | ||
| 441 | # $Data::Dumper::Purity=1; | ||
| 442 | # my $ValString=Dumper($data2->{$key}); | ||
| 443 | # $data1->{$key}->[1]=$ValString; | ||
| 444 | # return; | ||
| 445 | # } | ||
| 446 | if (ref($data2->{$key}) eq 'ARRAY'){ | ||
| 447 | push (@{$data1->{$key}},@{$data2->{$key}}); | ||
| 448 | } | ||
| 449 | if (ref($data2->{$key}) eq 'HASH'){ | ||
| 450 | foreach (keys %{$data2->{$key}}){ | ||
| 451 | $data1->{$key}->{$_}=$data2->{$key}->{$_}; | ||
| 452 | } | ||
| 453 | } | ||
| 454 | # $Data::Dumper::Terse=1; | ||
| 455 | # $Data::Dumper::Purity=1; | ||
| 456 | # my $DataString=Dumper($data1->{$key}); | ||
| 457 | # delete $data1->{$key}; | ||
| 458 | # $data1->{$key}=eval $DataString; | ||
| 459 | } | ||
| 460 | else{ | ||
| 461 | if (ref($data2->{$key})){ | ||
| 462 | $Data::Dumper::Terse=1; | ||
| 463 | $Data::Dumper::Purity=1; | ||
| 464 | my $ValString=Dumper($data2->{$key}); | ||
| 465 | $data1->{$key}=eval $ValString; | ||
| 466 | } | ||
| 467 | else{ | ||
| 468 | $data1->{$key}=$data2->{$key}; | ||
| 469 | } | ||
| 470 | } | ||
| 471 | } | ||
| 472 | return $data1; | ||
| 473 | } | ||
| 474 | |||
| 475 | |||
| 476 | ####################################### | ||
| 477 | |||
| 478 | sub expandData{ | ||
| 479 | my ($data)=@_; | ||
| 480 | my %FlatHash=%{$data}; | ||
| 481 | my $VAR1; | ||
| 482 | foreach (keys %FlatHash){ | ||
| 483 | if ($FlatHash{$_}=~/^\s*\$VAR1\s*\=/){ | ||
| 484 | $data->{$_}=eval $FlatHash{$_}; | ||
| 485 | } | ||
| 486 | } | ||
| 487 | return %{$data}; | ||
| 488 | } | ||
| 489 | |||
| 490 | sub dumpData{ | ||
| 491 | my ($data)=@_; | ||
| 492 | my %FlatHash; | ||
| 493 | my $VAR1; | ||
| 494 | foreach (keys %{$data}){ | ||
| 495 | if (ref($data->{$_})){ | ||
| 496 | $Data::Dumper::Indent=0; | ||
| 497 | $Data::Dumper::Terse=0; | ||
| 498 | $Data::Dumper::Purity=1; | ||
| 499 | $FlatHash{$_}=Dumper($data->{$_}); | ||
| 500 | } | ||
| 501 | else{ | ||
| 502 | $FlatHash{$_}=$data->{$_}; | ||
| 503 | } | ||
| 504 | } | ||
| 505 | return %FlatHash; | ||
| 506 | } | ||
| 507 | |||
| 508 | |||
| 509 | ####################################### | ||
| 510 | |||
| 511 | 1; |