[537] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
| 3 | # AZList.pm --
|
---|
| 4 | # A component of the Greenstone digital library software
|
---|
| 5 | # from the New Zealand Digital Library Project at the
|
---|
| 6 | # University of Waikato, New Zealand.
|
---|
| 7 | #
|
---|
| 8 | # Copyright (C) 1999 New Zealand Digital Library Project
|
---|
| 9 | #
|
---|
| 10 | # This program is free software; you can redistribute it and/or modify
|
---|
| 11 | # it under the terms of the GNU General Public License as published by
|
---|
| 12 | # the Free Software Foundation; either version 2 of the License, or
|
---|
| 13 | # (at your option) any later version.
|
---|
| 14 | #
|
---|
| 15 | # This program is distributed in the hope that it will be useful,
|
---|
| 16 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 17 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 18 | # GNU General Public License for more details.
|
---|
| 19 | #
|
---|
| 20 | # You should have received a copy of the GNU General Public License
|
---|
| 21 | # along with this program; if not, write to the Free Software
|
---|
| 22 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 23 | #
|
---|
| 24 | ###########################################################################
|
---|
| 25 |
|
---|
[256] | 26 | # classifier plugin for sorting alphabetically
|
---|
| 27 |
|
---|
| 28 | package AZList;
|
---|
| 29 |
|
---|
[1483] | 30 | use BasClas;
|
---|
[385] | 31 | use sorttools;
|
---|
| 32 |
|
---|
[10253] | 33 | use strict;
|
---|
| 34 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
| 35 |
|
---|
[1483] | 36 | sub BEGIN {
|
---|
[10253] | 37 | @AZList::ISA = ('BasClas');
|
---|
[1483] | 38 | }
|
---|
| 39 |
|
---|
[6806] | 40 |
|
---|
[4759] | 41 | my $arguments =
|
---|
| 42 | [ { 'name' => "metadata",
|
---|
[6940] | 43 | 'desc' => "{AZList.metadata}",
|
---|
[4759] | 44 | 'type' => "metadata",
|
---|
| 45 | 'reqd' => "yes" } ,
|
---|
| 46 | { 'name' => "removeprefix",
|
---|
[9064] | 47 | 'desc' => "{BasClas.removeprefix}",
|
---|
[6408] | 48 | 'type' => "regexp",
|
---|
[4759] | 49 | 'deft' => "",
|
---|
[4840] | 50 | 'reqd' => "no" } ,
|
---|
| 51 | { 'name' => "removesuffix",
|
---|
[9064] | 52 | 'desc' => "{BasClas.removesuffix}",
|
---|
[6408] | 53 | 'type' => "regexp",
|
---|
[4840] | 54 | 'deft' => "",
|
---|
| 55 | 'reqd' => "no" }
|
---|
| 56 | ];
|
---|
[3540] | 57 |
|
---|
| 58 | my $options = { 'name' => "AZList",
|
---|
[5645] | 59 | 'desc' => "{AZList.desc}",
|
---|
[6408] | 60 | 'abstract' => "no",
|
---|
| 61 | 'inherits' => "yes",
|
---|
[4759] | 62 | 'args' => $arguments };
|
---|
[3540] | 63 |
|
---|
[1839] | 64 |
|
---|
[256] | 65 | sub new {
|
---|
[10218] | 66 | my ($class) = shift (@_);
|
---|
| 67 | my ($classifierslist,$inputargs,$hashArgOptLists) = @_;
|
---|
| 68 | push(@$classifierslist, $class);
|
---|
[256] | 69 |
|
---|
[10218] | 70 | if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
|
---|
| 71 | if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
|
---|
| 72 |
|
---|
[12891] | 73 | my $self = new BasClas($classifierslist, $inputargs, $hashArgOptLists);
|
---|
[10218] | 74 |
|
---|
[10253] | 75 | if ($self->{'info_only'}) {
|
---|
| 76 | # don't worry about any options etc
|
---|
| 77 | return bless $self, $class;
|
---|
| 78 | }
|
---|
| 79 |
|
---|
[10218] | 80 | if (!$self->{"metadata"}) {
|
---|
[6984] | 81 | print STDERR "AZList Error: required option -metadata not supplied \n";
|
---|
[4873] | 82 | $self->print_txt_usage(""); # Use default resource bundle
|
---|
[6961] | 83 |
|
---|
[6984] | 84 | die "AZList Error: required option -metadata not supplied\n";
|
---|
[1483] | 85 | }
|
---|
[10218] | 86 |
|
---|
| 87 | # Manually set $self parameters.
|
---|
| 88 | $self->{'list'} = {};
|
---|
[6961] | 89 |
|
---|
[10993] | 90 | # Transfer value from Auto Parsing to the variable name that used in previous GreenStone.
|
---|
[10218] | 91 | my @meta_list = split(/,/, $self->{"metadata"});
|
---|
[6961] | 92 | $self->{'meta_list'} = \@meta_list;
|
---|
[10218] | 93 |
|
---|
| 94 | $self->{'buttonname'} = $self->generate_title_from_metadata($self->{'metadata'}) unless ($self->{'buttonname'});
|
---|
| 95 |
|
---|
| 96 | # Further setup
|
---|
| 97 | if (defined($self->{"removeprefix"}) && $self->{"removeprefix"}) {
|
---|
| 98 | $self->{"removeprefix"} =~ s/^\^//; # don't need a leading ^
|
---|
[2954] | 99 | }
|
---|
[10218] | 100 | if (defined($self->{"removesuffix"}) && $self->{"removesuffix"}) {
|
---|
| 101 | $self->{"removesuffix"} =~ s/\$$//; # don't need a trailing $
|
---|
[4840] | 102 | }
|
---|
[1483] | 103 |
|
---|
[10218] | 104 | # Clean out the unused keys
|
---|
| 105 | delete $self->{"metadata"}; # Delete this key
|
---|
| 106 |
|
---|
| 107 | if($self->{"removeprefix"} eq "") {delete $self->{"removeprefix"};}
|
---|
| 108 | if($self->{"removesuffix"} eq "") {delete $self->{"removesuffix"};}
|
---|
| 109 |
|
---|
[1483] | 110 | return bless $self, $class;
|
---|
[256] | 111 | }
|
---|
| 112 |
|
---|
| 113 | sub init {
|
---|
| 114 | my $self = shift (@_);
|
---|
| 115 |
|
---|
| 116 | $self->{'list'} = {};
|
---|
| 117 | }
|
---|
| 118 |
|
---|
| 119 | sub classify {
|
---|
| 120 | my $self = shift (@_);
|
---|
| 121 | my ($doc_obj) = @_;
|
---|
| 122 |
|
---|
| 123 | my $doc_OID = $doc_obj->get_OID();
|
---|
[6961] | 124 | my $outhandle = $self->{'outhandle'};
|
---|
[256] | 125 |
|
---|
[6961] | 126 | my $metavalue;
|
---|
| 127 | my $metaname;
|
---|
| 128 | # should we extend this to use all available metadata not just the first?
|
---|
| 129 | if (!defined $self->{'meta_list'}) {
|
---|
| 130 | # just in case
|
---|
| 131 | return;
|
---|
| 132 | }
|
---|
[2954] | 133 |
|
---|
[6961] | 134 | # find the first available metadata
|
---|
[10253] | 135 | foreach my $m (@{$self->{'meta_list'}}) {
|
---|
[6961] | 136 | $metavalue = $doc_obj->get_metadata_element($doc_obj->get_top_section(), $m);
|
---|
| 137 | $metaname = $m;
|
---|
| 138 | last if defined $metavalue;
|
---|
| 139 | }
|
---|
| 140 |
|
---|
| 141 | #if we haven't found a metavalue here, then the doc shouldn't be included
|
---|
| 142 | if (!defined $metavalue || $metavalue eq "") {
|
---|
| 143 | print $outhandle "WARNING: AZList: $doc_OID metadata is empty - not classifying\n";
|
---|
| 144 | return;
|
---|
[256] | 145 | }
|
---|
[6961] | 146 |
|
---|
| 147 | if (defined($self->{'removeprefix'}) &&
|
---|
| 148 | length($self->{'removeprefix'})) {
|
---|
| 149 | $metavalue =~ s/^$self->{'removeprefix'}//;
|
---|
| 150 | }
|
---|
| 151 | if (defined($self->{'removesuffix'}) &&
|
---|
| 152 | length($self->{'removesuffix'})) {
|
---|
| 153 | $metavalue =~ s/$self->{'removesuffix'}$//;
|
---|
| 154 | }
|
---|
| 155 |
|
---|
[10630] | 156 |
|
---|
| 157 | $metavalue = &sorttools::format_metadata_for_sorting($metaname, $metavalue, $doc_obj) unless $self->{'no_metadata_formatting'};
|
---|
[6961] | 158 |
|
---|
| 159 | if (defined $self->{'list'}->{$doc_OID}) {
|
---|
| 160 | print $outhandle "WARNING: AZList::classify called multiple times for $doc_OID\n";
|
---|
| 161 | }
|
---|
| 162 | if ($metavalue) {
|
---|
| 163 | $self->{'list'}->{$doc_OID} = $metavalue;
|
---|
| 164 | } else {
|
---|
| 165 | # the formatting has made it empty
|
---|
| 166 | my $outhandle = $self->{'outhandle'};
|
---|
| 167 | print $outhandle "WARNING: AZList: $doc_OID metadata has become empty - not classifying\n";
|
---|
| 168 | }
|
---|
| 169 |
|
---|
[256] | 170 | }
|
---|
| 171 |
|
---|
[837] | 172 | sub alpha_numeric_cmp
|
---|
| 173 | {
|
---|
| 174 | my ($self,$a,$b) = @_;
|
---|
| 175 |
|
---|
| 176 | my $title_a = $self->{'list'}->{$a};
|
---|
| 177 | my $title_b = $self->{'list'}->{$b};
|
---|
| 178 |
|
---|
| 179 | if ($title_a =~ m/^(\d+(\.\d+)?)/)
|
---|
| 180 | {
|
---|
| 181 | my $val_a = $1;
|
---|
| 182 | if ($title_b =~ m/^(\d+(\.\d+)?)/)
|
---|
| 183 | {
|
---|
| 184 | my $val_b = $1;
|
---|
| 185 | if ($val_a != $val_b)
|
---|
| 186 | {
|
---|
| 187 | return ($val_a <=> $val_b);
|
---|
| 188 | }
|
---|
| 189 | }
|
---|
| 190 | }
|
---|
| 191 |
|
---|
| 192 | return ($title_a cmp $title_b);
|
---|
| 193 | }
|
---|
| 194 |
|
---|
[316] | 195 | sub get_classify_info {
|
---|
[256] | 196 | my $self = shift (@_);
|
---|
| 197 |
|
---|
[837] | 198 | my @classlist
|
---|
| 199 | = sort { $self->alpha_numeric_cmp($a,$b) } keys %{$self->{'list'}};
|
---|
[256] | 200 |
|
---|
[409] | 201 | return $self->splitlist (\@classlist);
|
---|
[256] | 202 | }
|
---|
| 203 |
|
---|
[316] | 204 | sub get_entry {
|
---|
| 205 | my $self = shift (@_);
|
---|
[677] | 206 | my ($title, $childtype, $thistype) = @_;
|
---|
[316] | 207 |
|
---|
| 208 | # organise into classification structure
|
---|
[677] | 209 | my %classifyinfo = ('childtype'=>$childtype,
|
---|
[316] | 210 | 'Title'=>$title,
|
---|
| 211 | 'contains'=>[]);
|
---|
[677] | 212 | $classifyinfo{'thistype'} = $thistype
|
---|
| 213 | if defined $thistype && $thistype =~ /\w/;
|
---|
[316] | 214 |
|
---|
| 215 | return \%classifyinfo;
|
---|
| 216 | }
|
---|
| 217 |
|
---|
[256] | 218 | # splitlist takes an ordered list of classifications (@$classlistref) and splits it
|
---|
[409] | 219 | # up into alphabetical sub-sections.
|
---|
[256] | 220 | sub splitlist {
|
---|
| 221 | my $self = shift (@_);
|
---|
| 222 | my ($classlistref) = @_;
|
---|
| 223 | my $classhash = {};
|
---|
| 224 |
|
---|
[409] | 225 | # top level
|
---|
[677] | 226 | my $childtype = "HList";
|
---|
[741] | 227 | if (scalar (@$classlistref) <= 39) {$childtype = "VList";}
|
---|
[10218] | 228 | my $classifyinfo = $self->get_entry ($self->{'buttonname'}, $childtype, "Invisible");
|
---|
[409] | 229 |
|
---|
[741] | 230 | # don't need to do any splitting if there are less than 39 (max + min -1) classifications
|
---|
| 231 | if ((scalar @$classlistref) <= 39) {
|
---|
[6806] | 232 | foreach my $subOID (@$classlistref) {
|
---|
[409] | 233 | push (@{$classifyinfo->{'contains'}}, {'OID'=>$subOID});
|
---|
| 234 | }
|
---|
| 235 | return $classifyinfo;
|
---|
| 236 | }
|
---|
| 237 |
|
---|
[256] | 238 | # first split up the list into separate A-Z and 0-9 classifications
|
---|
[6806] | 239 | foreach my $classification (@$classlistref) {
|
---|
[256] | 240 | my $title = $self->{'list'}->{$classification};
|
---|
[837] | 241 |
|
---|
[708] | 242 | $title =~ s/^(&.{1,6};|<[^>]>|[^a-zA-Z0-9])//g; # remove any unwanted stuff
|
---|
[3109] | 243 | # only need first char for classification
|
---|
| 244 | $title =~ m/^(.)/; $title=$1;
|
---|
[256] | 245 | $title =~ tr/[a-z]/[A-Z]/;
|
---|
| 246 | if ($title =~ /^[0-9]$/) {$title = '0-9';}
|
---|
| 247 | elsif ($title !~ /^[A-Z]$/) {
|
---|
[1483] | 248 | my $outhandle = $self->{'outhandle'};
|
---|
| 249 | print $outhandle "AZList: WARNING $classification has badly formatted title ($title)\n";
|
---|
[256] | 250 | }
|
---|
| 251 | $classhash->{$title} = [] unless defined $classhash->{$title};
|
---|
| 252 | push (@{$classhash->{$title}}, $classification);
|
---|
| 253 | }
|
---|
| 254 | $classhash = $self->compactlist ($classhash);
|
---|
| 255 |
|
---|
| 256 | my @tmparr = ();
|
---|
[6806] | 257 | foreach my $subsection (sort keys (%$classhash)) {
|
---|
[256] | 258 | push (@tmparr, $subsection);
|
---|
| 259 | }
|
---|
[8866] | 260 | #if there is only one entry here, we suppress the buckets
|
---|
| 261 | if ((scalar @tmparr) == 1) {
|
---|
[9322] | 262 | $classifyinfo->{'childtype'} = "VList";
|
---|
[9204] | 263 | foreach my $OID (@{$classhash->{$tmparr[0]}}) {
|
---|
[8866] | 264 | push (@{$classifyinfo->{'contains'}}, {'OID'=>$OID});
|
---|
| 265 | }
|
---|
| 266 | return $classifyinfo;
|
---|
| 267 | }
|
---|
[256] | 268 |
|
---|
| 269 | # if there's a 0-9 section it will have been sorted to the beginning
|
---|
| 270 | # but we want it at the end
|
---|
| 271 | if ($tmparr[0] eq '0-9') {
|
---|
| 272 | shift @tmparr;
|
---|
| 273 | push (@tmparr, '0-9');
|
---|
| 274 | }
|
---|
| 275 |
|
---|
[6806] | 276 | foreach my $subclass (@tmparr) {
|
---|
[677] | 277 | my $tempclassify = $self->get_entry($subclass, "VList");
|
---|
[6806] | 278 | foreach my $subsubOID (@{$classhash->{$subclass}}) {
|
---|
[409] | 279 | push (@{$tempclassify->{'contains'}}, {'OID'=>$subsubOID});
|
---|
| 280 | }
|
---|
| 281 | push (@{$classifyinfo->{'contains'}}, $tempclassify);
|
---|
[256] | 282 | }
|
---|
| 283 |
|
---|
[409] | 284 | return $classifyinfo;
|
---|
[256] | 285 | }
|
---|
| 286 |
|
---|
| 287 | sub compactlist {
|
---|
| 288 | my $self = shift (@_);
|
---|
| 289 | my ($classhashref) = @_;
|
---|
| 290 | my $compactedhash = {};
|
---|
| 291 | my @currentOIDs = ();
|
---|
[6806] | 292 | my $currentfirstletter = ""; # start of working bin
|
---|
| 293 | my $currentlastletter = ""; # end of working bin
|
---|
| 294 | my $lastkey = ""; # the name of the last completed key
|
---|
[256] | 295 |
|
---|
| 296 | # minimum and maximum documents to be displayed per page.
|
---|
| 297 | # the actual maximum will be max + (min-1).
|
---|
| 298 | # the smallest sub-section is a single letter at present
|
---|
| 299 | # so in this case there may be many times max documents
|
---|
| 300 | # displayed on a page.
|
---|
| 301 | my $min = 10;
|
---|
| 302 | my $max = 30;
|
---|
| 303 |
|
---|
[6806] | 304 | foreach my $subsection (sort keys %$classhashref) {
|
---|
[256] | 305 | if ($subsection eq '0-9') {
|
---|
[6806] | 306 | # leave this bin as-is... copy it straight across
|
---|
[256] | 307 | @{$compactedhash->{$subsection}} = @{$classhashref->{$subsection}};
|
---|
| 308 | next;
|
---|
| 309 | }
|
---|
| 310 | $currentfirstletter = $subsection if $currentfirstletter eq "";
|
---|
| 311 | if ((scalar (@currentOIDs) < $min) ||
|
---|
| 312 | ((scalar (@currentOIDs) + scalar (@{$classhashref->{$subsection}})) <= $max)) {
|
---|
[6806] | 313 | # add this letter to the bin and continue
|
---|
[256] | 314 | push (@currentOIDs, @{$classhashref->{$subsection}});
|
---|
| 315 | $currentlastletter = $subsection;
|
---|
| 316 | } else {
|
---|
[6806] | 317 | # too many or too few for a separate bin
|
---|
[256] | 318 | if ($currentfirstletter eq $currentlastletter) {
|
---|
| 319 | @{$compactedhash->{$currentfirstletter}} = @currentOIDs;
|
---|
| 320 | $lastkey = $currentfirstletter;
|
---|
| 321 | } else {
|
---|
| 322 | @{$compactedhash->{"$currentfirstletter-$currentlastletter"}} = @currentOIDs;
|
---|
| 323 | $lastkey = "$currentfirstletter-$currentlastletter";
|
---|
| 324 | }
|
---|
| 325 | if (scalar (@{$classhashref->{$subsection}}) >= $max) {
|
---|
[6806] | 326 | # this key is now complete. Start a new one
|
---|
[256] | 327 | $compactedhash->{$subsection} = $classhashref->{$subsection};
|
---|
| 328 | @currentOIDs = ();
|
---|
| 329 | $currentfirstletter = "";
|
---|
[677] | 330 | $lastkey = $subsection;
|
---|
[256] | 331 | } else {
|
---|
| 332 | @currentOIDs = @{$classhashref->{$subsection}};
|
---|
| 333 | $currentfirstletter = $subsection;
|
---|
| 334 | $currentlastletter = $subsection;
|
---|
| 335 | }
|
---|
| 336 | }
|
---|
| 337 | }
|
---|
| 338 |
|
---|
| 339 | # add final OIDs to last sub-classification if there aren't many otherwise
|
---|
| 340 | # add final sub-classification
|
---|
[5646] | 341 | # BUG FIX: don't add anything if there are no currentOIDs (thanks to Don Gourley)
|
---|
[6806] | 342 | if (! scalar(@currentOIDs)) {return $compactedhash;}
|
---|
| 343 |
|
---|
| 344 | if (scalar (@currentOIDs) < $min) {
|
---|
| 345 | my ($newkey) = $lastkey =~ /^(.)/;
|
---|
| 346 | @currentOIDs = (@{$compactedhash->{$lastkey}}, @currentOIDs);
|
---|
| 347 | delete $compactedhash->{$lastkey};
|
---|
| 348 | @{$compactedhash->{"$newkey-$currentlastletter"}} = @currentOIDs;
|
---|
| 349 | } else {
|
---|
| 350 | if ($currentfirstletter eq $currentlastletter) {
|
---|
| 351 | @{$compactedhash->{$currentfirstletter}} = @currentOIDs;
|
---|
[5646] | 352 | }
|
---|
| 353 | else {
|
---|
[6806] | 354 | @{$compactedhash->{"$currentfirstletter-$currentlastletter"}} =
|
---|
| 355 | @currentOIDs;
|
---|
[5646] | 356 | }
|
---|
[256] | 357 | }
|
---|
| 358 |
|
---|
| 359 | return $compactedhash;
|
---|
| 360 | }
|
---|
| 361 |
|
---|
| 362 | 1;
|
---|