/[webpac2]/trunk/lib/WebPAC/Normalize.pm
This is repository of my old source code which isn't updated any more. Go to git.rot13.org for current projects!
ViewVC logotype

Diff of /trunk/lib/WebPAC/Normalize.pm

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 64 by dpavlin, Tue Nov 15 16:56:44 2005 UTC revision 544 by dpavlin, Thu Jun 29 21:52:51 2006 UTC
# Line 1  Line 1 
1  package WebPAC::Normalize;  package WebPAC::Normalize;
2    use Exporter 'import';
3    @EXPORT = qw/
4            _set_rec _set_lookup
5            _get_ds _clean_ds
6    
7            tag search display
8            marc21
9    
10            rec1 rec2 rec
11            regex prefix suffix surround
12            first lookup join_with
13    /;
14    
15  use warnings;  use warnings;
16  use strict;  use strict;
17  use base 'WebPAC::Common';  
18    #use base qw/WebPAC::Common/;
19  use Data::Dumper;  use Data::Dumper;
20    use Encode qw/from_to/;
21    
22  =head1 NAME  =head1 NAME
23    
24  WebPAC::Normalize - data mungling for normalisation  WebPAC::Normalize - describe normalisaton rules using sets
25    
26  =head1 VERSION  =head1 VERSION
27    
28  Version 0.01  Version 0.06
29    
30  =cut  =cut
31    
32  our $VERSION = '0.01';  our $VERSION = '0.06';
33    
34  =head1 SYNOPSIS  =head1 SYNOPSIS
35    
36  This package contains code that mungle data to produce normalized format.  This module uses C<conf/normalize/*.pl> files to perform normalisation
37    from input records using perl functions which are specialized for set
38  It contains several assumptions:  processing.
39    
40  =over  Sets are implemented as arrays, and normalisation file is valid perl, which
41    means that you check it's validity before running WebPAC using
42  =item *  C<perl -c normalize.pl>.
43    
44  format of fields is defined using C<v123^a> notation for repeatable fields  Normalisation can generate multiple output normalized data. For now, supported output
45  or C<s123^a> for single (or first) value, where C<123> is field number and  types (on the left side of definition) are: C<tag>, C<display>, C<search> and
46  C<a> is subfield.  C<marc21>.
   
 =item *  
   
 source data records (C<$rec>) have unique identifiers in field C<000>  
   
 =item *  
   
 optional C<eval{length('v123^a') == 3}> tag at B<beginning of format> will be  
 perl code that is evaluated before producing output (value of field will be  
 interpolated before that)  
47    
48  =item *  =head1 FUNCTIONS
   
 optional C<filter{filter_name}> at B<begining of format> will apply perl  
 code defined as code ref on format after field substitution to producing  
 output  
49    
50  =item *  Functions which start with C<_> are private and used by WebPAC internally.
51    All other functions are available for use within normalisation rules.
52    
53  optional C<lookup{...}> will be then performed. See C<WebPAC::Lookups>.  =head2 data_structure
54    
55  =item *  Return data structure
56    
57  at end, optional C<format>s rules are resolved. Format rules are similar to    my $ds = WebPAC::Normalize::data_structure(
58  C<sprintf> and can also contain C<lookup{...}> which is performed after          lookup => $lookup->lookup_hash,
59  values are inserted in format.          row => $row,
60            rules => $normalize_pl_config,
61            marc_encoding => 'utf-8',
62      );
63    
64  =back  Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all
65    other are optional.
66    
67  This also describes order in which transformations are applied (eval,  This function will B<die> if normalizastion can't be evaled.
 filter, lookup, format) which is important to undestand when deciding how to  
 solve your data mungling and normalisation process.  
68    
69    Since this function isn't exported you have to call it with
70    C<WebPAC::Normalize::data_structure>.
71    
72    =cut
73    
74    sub data_structure {
75            my $arg = {@_};
76    
77  =head1 FUNCTIONS          die "need row argument" unless ($arg->{row});
78            die "need normalisation argument" unless ($arg->{rules});
79    
80  =head2 new          no strict 'subs';
81            _set_lookup( $arg->{lookup} );
82            _set_rec( $arg->{row} );
83            _clean_ds( %{ $arg } );
84            eval "$arg->{rules}";
85            die "error evaling $arg->{rules}: $@\n" if ($@);
86    
87  Create new normalisation object          return _get_ds();
88    }
89    
90    my $n = new WebPAC::Normalize::Something(  =head2 _set_rec
         filter => {  
                 'filter_name_1' => sub {  
                         # filter code  
                         return length($_);  
                 }, ...  
         },  
         db => $db_obj,  
         lookup_regex => $lookup->regex,  
         lookup => $lookup_obj,  
   );  
91    
92  Parametar C<filter> defines user supplied snippets of perl code which can  Set current record hash
 be use with C<filter{...}> notation.  
93    
94  Recommended parametar C<lookup_regex> is used to enable parsing of lookups    _set_rec( $rec );
 in structures. If you pass this parametar, you must also pass C<lookup>  
 which is C<WebPAC::Lookup> object.  
95    
96  =cut  =cut
97    
98  sub new {  my $rec;
         my $class = shift;  
         my $self = {@_};  
         bless($self, $class);  
99    
100          my $r = $self->{'lookup_regex'} ? 1 : 0;  sub _set_rec {
101          my $l = $self->{'lookup'} ? 1 : 0;          $rec = shift or die "no record hash";
   
         my $log = $self->_get_logger();  
   
         # those two must be in pair  
         if ( ($r & $l) != ($r || $l) ) {  
                 my $log = $self->_get_logger();  
                 $log->logdie("lookup_regex and lookup must be in pair");  
         }  
   
         $log->logdie("lookup must be WebPAC::Lookup object") if ($self->{'lookup'} && ! $self->{'lookup'}->isa('WebPAC::Lookup'));  
   
         $self ? return $self : return undef;  
102  }  }
103    
104    =head2 _get_ds
105    
106  =head2 data_structure  Return hash formatted as data structure
   
 Create in-memory data structure which represents normalized layout from  
 C<conf/normalize/*.xml>.  
   
 This structures are used to produce output.  
   
  my @ds = $webpac->data_structure($rec);  
107    
108  B<Note: historical oddity follows>    my $ds = _get_ds();
   
 This method will also set C<< $webpac->{'currnet_filename'} >> if there is  
 C<< <filename> >> tag and C<< $webpac->{'headline'} >> if there is  
 C<< <headline> >> tag.  
109    
110  =cut  =cut
111    
112  sub data_structure {  my $out;
113          my $self = shift;  my $marc21;
114    my $marc_encoding;
         my $log = $self->_get_logger();  
   
         my $rec = shift;  
         $log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o);  
115    
116          my $cache_file;  sub _get_ds {
117            return $out;
118          if ($self->{'db'}) {  }
                 my @ds = $self->{'db'}->load_ds($rec);  
                 $log->debug("load_ds( rec = ", sub { Dumper($rec) }, ") = ", sub { Dumper(@ds) });  
                 return @ds if ($#ds > 0);  
                 $log->debug("cache miss, creating");  
         }  
   
         undef $self->{'currnet_filename'};  
         undef $self->{'headline'};  
119    
120          my @sorted_tags;  =head2 _clean_ds
         if ($self->{tags_by_order}) {  
                 @sorted_tags = @{$self->{tags_by_order}};  
         } else {  
                 @sorted_tags = sort { $self->_sort_by_order } keys %{$self->{'import_xml'}->{'indexer'}};  
                 $self->{tags_by_order} = \@sorted_tags;  
         }  
121    
122          my @ds;  Clean data structure hash for next record
123    
124          $log->debug("tags: ",sub { join(", ",@sorted_tags) });    _clean_ds();
125    
126          foreach my $field (@sorted_tags) {  =cut
127    
128                  my $row;  sub _clean_ds {
129            my $a = {@_};
130            $out = undef;
131            $marc21 = undef;
132            $marc_encoding = $a->{marc_encoding};
133    }
134    
135  #print "field $field [",$self->{'tag'},"] = ",Dumper($self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}});  =head2 _set_lookup
136    
137                  foreach my $tag (@{$self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}}) {  Set current lookup hash
                         my $format;  
138    
139                          $log->logdie("expected tag HASH and got $tag") unless (ref($tag) eq 'HASH');    _set_lookup( $lookup );
                         $format = $tag->{'value'} || $tag->{'content'};  
140    
141                          $log->debug("format: $format");  =cut
142    
143                          my @v;  my $lookup;
                         if ($self->{'lookup_regex'} && $format =~ $self->{'lookup_regex'}) {  
                                 @v = $self->fill_in_to_arr($rec,$format);  
                         } else {  
                                 @v = $self->parse_to_arr($rec,$format);  
                         }  
                         next if (! @v);  
144    
145                          if ($tag->{'sort'}) {  sub _set_lookup {
146                                  @v = $self->sort_arr(@v);          $lookup = shift;
147                          }  }
148    
149                          # use format?  =head2 _get_marc21_fields
                         if ($tag->{'format_name'}) {  
                                 @v = map { $self->apply_format($tag->{'format_name'},$tag->{'format_delimiter'},$_) } @v;  
                         }  
150    
151                          if ($field eq 'filename') {  Get all fields defined by calls to C<marc21>
                                 $self->{'current_filename'} = join('',@v);  
                                 $log->debug("filename: ",$self->{'current_filename'});  
                         } elsif ($field eq 'headline') {  
                                 $self->{'headline'} .= join('',@v);  
                                 $log->debug("headline: ",$self->{'headline'});  
                                 next; # don't return headline in data_structure!  
                         }  
152    
153                          # delimiter will join repeatable fields          $marc->add_fields( WebPAC::Normalize:_get_marc21_fields() );
                         if ($tag->{'delimiter'}) {  
                                 @v = ( join($tag->{'delimiter'}, @v) );  
                         }  
154    
                         # default types  
                         my @types = qw(display swish);  
                         # override by type attribute  
                         @types = ( $tag->{'type'} ) if ($tag->{'type'});  
   
                         foreach my $type (@types) {  
                                 # append to previous line?  
                                 $log->debug("type: $type ",sub { join(" ",@v) }, $row->{'append'} || 'no append');  
                                 if ($tag->{'append'}) {  
   
                                         # I will delimit appended part with  
                                         # delimiter (or ,)  
                                         my $d = $tag->{'delimiter'};  
                                         # default delimiter  
                                         $d ||= " ";  
   
                                         my $last = pop @{$row->{$type}};  
                                         $d = "" if (! $last);  
                                         $last .= $d . join($d, @v);  
                                         push @{$row->{$type}}, $last;  
   
                                 } else {  
                                         push @{$row->{$type}}, @v;  
                                 }  
                         }  
155    
156    
157                  }  We are using I<magic> which detect repeatable fields only from
158    sequence of field/subfield data generated by normalization.
159    
160                  if ($row) {  Repeatable field is created if there is second occurence of same subfield or
161                          $row->{'tag'} = $field;  if any of indicators are different. This is sane for most cases except for
162    non-repeatable fields with repeatable subfields.
163    
164                          # TODO: name_sigular, name_plural  B<TODO>: implement exceptions to magic
                         my $name = $self->{'import_xml'}->{'indexer'}->{$field}->{'name'};  
                         $row->{'name'} = $name ? $self->_x($name) : $field;  
   
                         # post-sort all values in field  
                         if ($self->{'import_xml'}->{'indexer'}->{$field}->{'sort'}) {  
                                 $log->warn("sort at field tag not implemented");  
                         }  
165    
166                          push @ds, $row;  =cut
167    
168                          $log->debug("row $field: ",sub { Dumper($row) });  sub _get_marc21_fields {
169            my @m;
170            my $last;
171            foreach my $row (@{ $marc21 }) {
172                    if ($last &&
173                            $last->[0] eq $row->[0] &&              # check if field is same
174                            $last->[1] eq $row->[1] &&              # check for i1
175                            $last->[2] eq $row->[2] &&              # and for i2
176                            $last->[3] ne $row->[3]                 # and subfield is different
177                    ) {
178                            push @$last, ( $row->[3] , $row->[4] );
179                            warn "## ++ added $row->[0] ^$row->[3] to $last->[0]\n";
180                            next;
181                    } elsif ($last) {
182                            push @m, $last;
183                  }                  }
184    
185                    $last = $row;
186          }          }
187    
188          $log->logdie("there is no current_filename defined! Do you have filename tag in conf/normalize/?.xml") unless ($self->{'current_filename'});          push @m, $last if ($last);
   
         $self->{'db'}->save_ds(  
                 ds => \@ds,  
                 current_filename => $self->{'current_filename'},  
                 headline => $self->{'headline'},  
         ) if ($self->{'db'});  
   
         $log->debug("ds: ", sub { Dumper(@ds) });  
   
         return @ds;  
189    
190            return @m;
191  }  }
192    
193  =head2 parse  =head1 Functions to create C<data_structure>
   
 Perform smart parsing of string, skipping delimiters for fields which aren't  
 defined. It can also eval code in format starting with C<eval{...}> and  
 return output or nothing depending on eval code.  
194    
195   my $text = $webpac->parse($rec,'eval{"v901^a" eq "Deskriptor"}descriptor: v250^a', $i);  Those functions generally have to first in your normalization file.
   
 =cut  
196    
197  sub parse {  =head2 tag
         my $self = shift;  
198    
199          my ($rec, $format_utf8, $i) = @_;  Define new tag for I<search> and I<display>.
200    
201          return if (! $format_utf8);    tag('Title', rec('200','a') );
202    
         my $log = $self->_get_logger();  
203    
204          $log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o);  =cut
   
         $i = 0 if (! $i);  
   
         my $format = $self->_x($format_utf8) || $log->logconfess("can't convert '$format_utf8' from UTF-8 to ",$self->{'code_page'});  
   
         my @out;  
   
         $log->debug("format: $format");  
   
         my $eval_code;  
         # remove eval{...} from beginning  
         $eval_code = $1 if ($format =~ s/^eval{([^}]+)}//s);  
   
         my $filter_name;  
         # remove filter{...} from beginning  
         $filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s);  
   
         my $prefix;  
         my $all_found=0;  
205    
206          while ($format =~ s/^(.*?)(v|s)(\d+)(?:\^(\w))?//s) {  sub tag {
207            my $name = shift or die "tag needs name as first argument";
208            my @o = grep { defined($_) && $_ ne '' } @_;
209            return unless (@o);
210            $out->{$name}->{tag} = $name;
211            $out->{$name}->{search} = \@o;
212            $out->{$name}->{display} = \@o;
213    }
214    
215                  my $del = $1 || '';  =head2 display
                 $prefix ||= $del if ($all_found == 0);  
216    
217                  # repeatable index  Define tag just for I<display>
                 my $r = $i;  
                 $r = 0 if (lc("$2") eq 's');  
218    
219                  my $found = 0;    @v = display('Title', rec('200','a') );
                 my $tmp = $self->get_data(\$rec,$3,$4,$r,\$found);  
220    
221                  if ($found) {  =cut
                         push @out, $del;  
                         push @out, $tmp;  
                         $all_found += $found;  
                 }  
         }  
222    
223          return if (! $all_found);  sub display {
224            my $name = shift or die "display needs name as first argument";
225            my @o = grep { defined($_) && $_ ne '' } @_;
226            return unless (@o);
227            $out->{$name}->{tag} = $name;
228            $out->{$name}->{display} = \@o;
229    }
230    
231          my $out = join('',@out);  =head2 search
232    
233          if ($out) {  Prepare values just for I<search>
                 # add rest of format (suffix)  
                 $out .= $format;  
234    
235                  # add prefix if not there    @v = search('Title', rec('200','a') );
                 $out = $prefix . $out if ($out !~ m/^\Q$prefix\E/);  
236    
237                  $log->debug("result: $out");  =cut
         }  
   
         if ($eval_code) {  
                 my $eval = $self->fill_in($rec,$eval_code,$i) || return;  
                 $log->debug("about to eval{$eval} format: $out");  
                 return if (! $self->_eval($eval));  
         }  
           
         if ($filter_name && $self->{'filter'}->{$filter_name}) {  
                 $log->debug("about to filter{$filter_name} format: $out");  
                 $out = $self->{'filter'}->{$filter_name}->($out);  
                 return unless(defined($out));  
                 $log->debug("filter result: $out");  
         }  
238    
239          return $out;  sub search {
240            my $name = shift or die "search needs name as first argument";
241            my @o = grep { defined($_) && $_ ne '' } @_;
242            return unless (@o);
243            $out->{$name}->{tag} = $name;
244            $out->{$name}->{search} = \@o;
245  }  }
246    
247  =head2 parse_to_arr  =head2 marc21
248    
249  Similar to C<parse>, but returns array of all repeatable fields  Save value for MARC field
250    
251   my @arr = $webpac->parse_to_arr($rec,'v250^a');    marc21('900','a', rec('200','a') );
252    
253  =cut  =cut
254    
255  sub parse_to_arr {  sub marc21 {
256          my $self = shift;          my $f = shift or die "marc21 needs field";
257            die "marc21 field must be numer" unless ($f =~ /^\d+$/);
         my ($rec, $format_utf8) = @_;  
258    
259          my $log = $self->_get_logger();          my $sf = shift or die "marc21 needs subfield";
260    
261          $log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o);          foreach (@_) {
262          return if (! $format_utf8);                  my $v = $_;             # make var read-write for Encode
263                    next unless (defined($v) && $v !~ /^\s*$/);
264          my $i = 0;                  from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding);
265          my @arr;                  push @{ $marc21 }, [ $f, ' ', ' ', $sf => $v ];
   
         while (my $v = $self->parse($rec,$format_utf8,$i++)) {  
                 push @arr, $v;  
266          }          }
   
         $log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr);  
   
         return @arr;  
267  }  }
268    
269    =head1 Functions to extract data from input
270    
271  =head2 fill_in  This function should be used inside functions to create C<data_structure> described
272    above.
 Workhourse of all: takes record from in-memory structure of database and  
 strings with placeholders and returns string or array of with substituted  
 values from record.  
273    
274   my $text = $webpac->fill_in($rec,'v250^a');  =head2 rec1
275    
276  Optional argument is ordinal number for repeatable fields. By default,  Return all values in some field
 it's assume to be first repeatable field (fields are perl array, so first  
 element is 0).  
 Following example will read second value from repeatable field.  
277    
278   my $text = $webpac->fill_in($rec,'Title: v250^a',1);    @v = rec1('200')
279    
280  This function B<does not> perform parsing of format to inteligenty skip  TODO: order of values is probably same as in source data, need to investigate that
 delimiters before fields which aren't used.  
   
 This method will automatically decode UTF-8 string to local code page  
 if needed.  
281    
282  =cut  =cut
283    
284  sub fill_in {  sub rec1 {
285          my $self = shift;          my $f = shift;
286            return unless (defined($rec) && defined($rec->{$f}));
287          my $log = $self->_get_logger();          if (ref($rec->{$f}) eq 'ARRAY') {
288                    return map {
289          my $rec = shift || $log->logconfess("need data record");                          if (ref($_) eq 'HASH') {
290          my $format = shift || $log->logconfess("need format to parse");                                  values %{$_};
         # iteration (for repeatable fields)  
         my $i = shift || 0;  
   
         $log->logdie("infitite loop in format $format") if ($i > ($self->{'max_mfn'} || 9999));  
   
         # FIXME remove for speedup?  
         $log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o);  
   
         if (utf8::is_utf8($format)) {  
                 $format = $self->_x($format);  
         }  
   
         my $found = 0;  
   
         my $eval_code;  
         # remove eval{...} from beginning  
         $eval_code = $1 if ($format =~ s/^eval{([^}]+)}//s);  
   
         my $filter_name;  
         # remove filter{...} from beginning  
         $filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s);  
   
         # do actual replacement of placeholders  
         # repeatable fields  
         $format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found)/ges;  
         # non-repeatable fields  
         $format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found)/ges;  
   
         if ($found) {  
                 $log->debug("format: $format");  
                 if ($eval_code) {  
                         my $eval = $self->fill_in($rec,$eval_code,$i);  
                         return if (! $self->_eval($eval));  
                 }  
                 if ($filter_name && $self->{'filter'}->{$filter_name}) {  
                         $log->debug("filter '$filter_name' for $format");  
                         $format = $self->{'filter'}->{$filter_name}->($format);  
                         return unless(defined($format));  
                         $log->debug("filter result: $format");  
                 }  
                 # do we have lookups?  
                 if ($self->{'lookup'}) {  
                         if ($self->{'lookup'}->can('lookup')) {  
                                 return $self->{'lookup'}->lookup($format);  
291                          } else {                          } else {
292                                  $log->warn("Have lookup object but can't invoke lookup method");                                  $_;
293                          }                          }
294                  } else {                  } @{ $rec->{$f} };
295                          return $format;          } elsif( defined($rec->{$f}) ) {
296                  }                  return $rec->{$f};
         } else {  
                 return;  
297          }          }
298  }  }
299    
300    =head2 rec2
301    
302  =head2 fill_in_to_arr  Return all values in specific field and subfield
303    
304  Similar to C<fill_in>, but returns array of all repeatable fields. Usable    @v = rec2('200','a')
 for fields which have lookups, so they shouldn't be parsed but rather  
 C<fill_id>ed.  
   
  my @arr = $webpac->fill_in_to_arr($rec,'[v900];;[v250^a]');  
305    
306  =cut  =cut
307    
308  sub fill_in_to_arr {  sub rec2 {
309          my $self = shift;          my $f = shift;
310            return unless (defined($rec && $rec->{$f}));
311            my $sf = shift;
312            return map { $_->{$sf} } grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} };
313    }
314    
315          my ($rec, $format_utf8) = @_;  =head2 rec
316    
317          my $log = $self->_get_logger();  syntaxtic sugar for
318    
319          $log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o);    @v = rec('200')
320          return if (! $format_utf8);    @v = rec('200','a')
321    
322          my $i = 0;  =cut
         my @arr;  
323    
324          while (my @v = $self->fill_in($rec,$format_utf8,$i++)) {  sub rec {
325                  push @arr, @v;          if ($#_ == 0) {
326                    return rec1(@_);
327            } elsif ($#_ == 1) {
328                    return rec2(@_);
329          }          }
   
         $log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr);  
   
         return @arr;  
330  }  }
331    
332    =head2 regex
333    
334  =head2 get_data  Apply regex to some or all values
   
 Returns value from record.  
   
  my $text = $self->get_data(\$rec,$f,$sf,$i,\$found);  
   
 Arguments are:  
 record reference C<$rec>,  
 field C<$f>,  
 optional subfiled C<$sf>,  
 index for repeatable values C<$i>.  
335    
336  Optinal variable C<$found> will be incremeted if there    @v = regex( 's/foo/bar/g', @v );
 is field.  
   
 Returns value or empty string.  
337    
338  =cut  =cut
339    
340  sub get_data {  sub regex {
341          my $self = shift;          my $r = shift;
342            my @out;
343          my ($rec,$f,$sf,$i,$found) = @_;          #warn "r: $r\n",Dumper(\@_);
344            foreach my $t (@_) {
345          if ($$rec->{$f}) {                  next unless ($t);
346                  return '' if (! $$rec->{$f}->[$i]);                  eval "\$t =~ $r";
347                  no strict 'refs';                  push @out, $t if ($t && $t ne '');
                 if ($sf && $$rec->{$f}->[$i]->{$sf}) {  
                         $$found++ if (defined($$found));  
                         return $$rec->{$f}->[$i]->{$sf};  
                 } elsif (! $sf && $$rec->{$f}->[$i]) {  
                         $$found++ if (defined($$found));  
                         # it still might have subfield, just  
                         # not specified, so we'll dump all  
                         if ($$rec->{$f}->[$i] =~ /HASH/o) {  
                                 my $out;  
                                 foreach my $k (keys %{$$rec->{$f}->[$i]}) {  
                                         $out .= $$rec->{$f}->[$i]->{$k}." ";  
                                 }  
                                 return $out;  
                         } else {  
                                 return $$rec->{$f}->[$i];  
                         }  
                 } else {  
                         return '';  
                 }  
         } else {  
                 return '';  
348          }          }
349            return @out;
350  }  }
351    
352    =head2 prefix
353    
354  =head2 apply_format  Prefix all values with a string
   
 Apply format specified in tag with C<format_name="name"> and  
 C<format_delimiter=";;">.  
355    
356   my $text = $webpac->apply_format($format_name,$format_delimiter,$data);    @v = prefix( 'my_', @v );
   
 Formats can contain C<lookup{...}> if you need them.  
357    
358  =cut  =cut
359    
360  sub apply_format {  sub prefix {
361          my $self = shift;          my $p = shift or die "prefix needs string as first argument";
362            return map { $p . $_ } grep { defined($_) } @_;
363          my ($name,$delimiter,$data) = @_;  }
   
         my $log = $self->_get_logger();  
   
         if (! $self->{'import_xml'}->{'format'}->{$name}) {  
                 $log->warn("<format name=\"$name\"> is not defined in ",$self->{'import_xml_file'});  
                 return $data;  
         }  
   
         $log->warn("no delimiter for format $name") if (! $delimiter);  
364    
365          my $format = $self->_x($self->{'import_xml'}->{'format'}->{$name}->{'content'}) || $log->logdie("can't find format '$name'");  =head2 suffix
366    
367          my @data = split(/\Q$delimiter\E/, $data);  suffix all values with a string
368    
369          my $out = sprintf($format, @data);    @v = suffix( '_my', @v );
         $log->debug("using format $name [$format] on $data to produce: $out");  
370    
371          if ($self->{'lookup_regex'} && $out =~ $self->{'lookup_regex'}) {  =cut
                 return $self->{'lookup'}->lookup($out);  
         } else {  
                 return $out;  
         }  
372    
373    sub suffix {
374            my $s = shift or die "suffix needs string as first argument";
375            return map { $_ . $s } grep { defined($_) } @_;
376  }  }
377    
378  =head2 sort_arr  =head2 surround
379    
380  Sort array ignoring case and html in data  surround all values with a two strings
381    
382   my @sorted = $webpac->sort_arr(@unsorted);    @v = surround( 'prefix_', '_suffix', @v );
383    
384  =cut  =cut
385    
386  sub sort_arr {  sub surround {
387          my $self = shift;          my $p = shift or die "surround need prefix as first argument";
388            my $s = shift or die "surround needs suffix as second argument";
389          my $log = $self->_get_logger();          return map { $p . $_ . $s } grep { defined($_) } @_;
   
         # FIXME add Schwartzian Transformation?  
   
         my @sorted = sort {  
                 $a =~ s#<[^>]+/*>##;  
                 $b =~ s#<[^>]+/*>##;  
                 lc($b) cmp lc($a)  
         } @_;  
         $log->debug("sorted values: ",sub { join(", ",@sorted) });  
   
         return @sorted;  
390  }  }
391    
392    =head2 first
393    
394  =head1 INTERNAL METHODS  Return first element
   
 =head2 _sort_by_order  
395    
396  Sort xml tags data structure accoding to C<order=""> attribute.    $v = first( @v );
397    
398  =cut  =cut
399    
400  sub _sort_by_order {  sub first {
401          my $self = shift;          my $r = shift;
402            return $r;
         my $va = $self->{'import_xml'}->{'indexer'}->{$a}->{'order'} ||  
                 $self->{'import_xml'}->{'indexer'}->{$a};  
         my $vb = $self->{'import_xml'}->{'indexer'}->{$b}->{'order'} ||  
                 $self->{'import_xml'}->{'indexer'}->{$b};  
   
         return $va <=> $vb;  
403  }  }
404    
405  =head2 _x  =head2 lookup
406    
407  Convert strings from C<conf/normalize/*.xml> encoding into application  Consult lookup hashes for some value
 specific encoding (optinally specified using C<code_page> to C<new>  
 constructor).  
408    
409   my $text = $n->_x('normalize text string');    @v = lookup( $v );
410      @v = lookup( @v );
 This is a stub so that other modules doesn't have to implement it.  
411    
412  =cut  =cut
413    
414  sub _x {  sub lookup {
415          my $self = shift;          my $k = shift or return;
416          return shift;          return unless (defined($lookup->{$k}));
417            if (ref($lookup->{$k}) eq 'ARRAY') {
418                    return @{ $lookup->{$k} };
419            } else {
420                    return $lookup->{$k};
421            }
422  }  }
423    
424    =head2 join_with
425    
426  =head1 AUTHOR  Joins walues with some delimiter
   
 Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >>  
427    
428  =head1 COPYRIGHT & LICENSE    $v = join_with(", ", @v);
   
 Copyright 2005 Dobrica Pavlinusic, All Rights Reserved.  
   
 This program is free software; you can redistribute it and/or modify it  
 under the same terms as Perl itself.  
429    
430  =cut  =cut
431    
432  1; # End of WebPAC::DB  sub join_with {
433            my $d = shift;
434            return join($d, grep { defined($_) && $_ ne '' } @_);
435    }
436    
437    # END
438    1;

Legend:
Removed from v.64  
changed lines
  Added in v.544

  ViewVC Help
Powered by ViewVC 1.1.26