/[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 372 by dpavlin, Sun Jan 8 21:50:34 2006 UTC revision 540 by dpavlin, Thu Jun 29 15:29:41 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 blib;  
18  use WebPAC::Common;  #use base qw/WebPAC::Common/;
 use base 'WebPAC::Common';  
19  use Data::Dumper;  use Data::Dumper;
20    
21  =head1 NAME  =head1 NAME
22    
23  WebPAC::Normalize - data mungling for normalisation  WebPAC::Normalize - describe normalisaton rules using sets
24    
25  =head1 VERSION  =head1 VERSION
26    
27  Version 0.08  Version 0.05
28    
29  =cut  =cut
30    
31  our $VERSION = '0.08';  our $VERSION = '0.05';
32    
33  =head1 SYNOPSIS  =head1 SYNOPSIS
34    
35  This package contains code that mungle data to produce normalized format.  This module uses C<conf/normalize/*.pl> files to perform normalisation
36    from input records using perl functions which are specialized for set
37  It contains several assumptions:  processing.
38    
39  =over  Sets are implemented as arrays, and normalisation file is valid perl, which
40    means that you check it's validity before running WebPAC using
41  =item *  C<perl -c normalize.pl>.
42    
43  format of fields is defined using C<v123^a> notation for repeatable fields  Normalisation can generate multiple output normalized data. For now, supported output
44  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
45  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)  
   
 =item *  
   
 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  
   
 There is one built-in filter called C<regex> which can be use like this:  
   
   filter{regex(s/foo/bar/)}  
   
 =item *  
   
 optional C<lookup{...}> will be then performed. See C<WebPAC::Lookups>.  
   
 =item *  
   
 at end, optional C<format>s rules are resolved. Format rules are similar to  
 C<sprintf> and can also contain C<lookup{...}> which is performed after  
 values are inserted in format.  
   
 =back  
   
 This also describes order in which transformations are applied (eval,  
 filter, lookup, format) which is important to undestand when deciding how to  
 solve your data mungling and normalisation process.  
   
   
   
46    
47  =head1 FUNCTIONS  =head1 FUNCTIONS
48    
49  =head2 new  Functions which start with C<_> are private and used by WebPAC internally.
50    All other functions are available for use within normalisation rules.
51    
52    =head2 data_structure
53    
54  Create new normalisation object  Return data structure
55    
56    my $n = new WebPAC::Normalize::Something(    my $ds = WebPAC::Normalize::data_structure(
57          filter => {          lookup => $lookup->lookup_hash,
58                  'filter_name_1' => sub {          row => $row,
59                          # filter code          rules => $normalize_pl_config,
                         return length($_);  
                 }, ...  
         },  
         db => $db_obj,  
         lookup_regex => $lookup->regex,  
         lookup => $lookup_obj,  
         prefix => 'foobar',  
60    );    );
61    
62  Parametar C<filter> defines user supplied snippets of perl code which can  Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all
63  be use with C<filter{...}> notation.  other are optional.
64    
65  C<prefix> is used to form filename for database record (to support multiple  This function will B<die> if normalizastion can't be evaled.
 source files which are joined in one database).  
66    
67  Recommended parametar C<lookup_regex> is used to enable parsing of lookups  Since this function isn't exported you have to call it with
68  in structures. If you pass this parametar, you must also pass C<lookup>  C<WebPAC::Normalize::data_structure>.
 which is C<WebPAC::Lookup> object.  
69    
70  =cut  =cut
71    
72  sub new {  sub data_structure {
73          my $class = shift;          my $arg = {@_};
         my $self = {@_};  
         bless($self, $class);  
   
         my $r = $self->{'lookup_regex'} ? 1 : 0;  
         my $l = $self->{'lookup'} ? 1 : 0;  
   
         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");  
         }  
74    
75          $log->logdie("lookup must be WebPAC::Lookup object") if ($self->{'lookup'} && ! $self->{'lookup'}->isa('WebPAC::Lookup'));          die "need row argument" unless ($arg->{row});
76            die "need normalisation argument" unless ($arg->{rules});
77    
78          $log->warn("no prefix defined. please check that!") unless ($self->{'prefix'});          no strict 'subs';
79            _set_lookup( $arg->{lookup} );
80            _set_rec( $arg->{row} );
81            _clean_ds();
82    
83          $log->debug("using lookup regex: ", $self->{lookup_regex}) if ($r && $l);          eval "$arg->{rules}";
84            die "error evaling $arg->{rules}: $@\n" if ($@);
85    
86          if (! $self->{filter} || ! $self->{filter}->{regex}) {          return _get_ds();
                 $log->debug("adding built-in filter regex");  
                 $self->{filter}->{regex} = sub {  
                         my ($val, $regex) = @_;  
                         eval "\$val =~ $regex";  
                         return $val;  
                 };  
         }  
   
         $self ? return $self : return undef;  
87  }  }
88    
89    =head2 _set_rec
90    
91  =head2 data_structure  Set current record hash
92    
93  Create in-memory data structure which represents normalized layout from    _set_rec( $rec );
 C<conf/normalize/*.xml>.  
   
 This structures are used to produce output.  
   
  my $ds = $webpac->data_structure($rec);  
94    
95  =cut  =cut
96    
97  sub data_structure {  my $rec;
         my $self = shift;  
   
         my $log = $self->_get_logger();  
   
         my $rec = shift;  
         $log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o);  
98    
99          $log->debug("data_structure rec = ", sub { Dumper($rec) });  sub _set_rec {
100            $rec = shift or die "no record hash";
101          $log->logdie("need unique ID (mfn) in field 000 of record " . Dumper($rec) ) unless (defined($rec->{'000'}));  }
   
         my $id = $rec->{'000'}->[0] || $log->logdie("field 000 isn't array!");  
   
         my $cache_file;  
   
         if ($self->{'db'}) {  
                 my $ds = $self->{'db'}->load_ds( id => $id, prefix => $self->{prefix} );  
                 $log->debug("load_ds( rec = ", sub { Dumper($rec) }, ") = ", sub { Dumper($ds) });  
                 return $ds if ($ds);  
                 $log->debug("cache miss, creating");  
         }  
   
         my @sorted_tags;  
         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;  
         }  
   
         my $ds;  
   
         $log->debug("tags: ",sub { join(", ",@sorted_tags) });  
   
         foreach my $field (@sorted_tags) {  
   
                 my $row;  
102    
103  #print "field $field [",$self->{'tag'},"] = ",Dumper($self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}});  =head2 _get_ds
104    
105                  foreach my $tag (@{$self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}}) {  Return hash formatted as data structure
                         my $format;  
106    
107                          $log->logdie("expected tag HASH and got $tag") unless (ref($tag) eq 'HASH');    my $ds = _get_ds();
                         $format = $tag->{'value'} || $tag->{'content'};  
108    
109                          my @v;  =cut
                         if ($self->{'lookup_regex'} && $format =~ $self->{'lookup_regex'}) {  
                                 @v = $self->_rec_to_arr($rec,$format,'fill_in');  
                         } else {  
                                 @v = $self->_rec_to_arr($rec,$format,'parse');  
                         }  
                         if (! @v) {  
                                 $log->debug("$field <",$self->{tag},"> format: $format no values");  
 #                               next;  
                         } else {  
                                 $log->debug("$field <",$self->{tag},"> format: $format values: ", join(",", @v));  
                         }  
   
                         if ($tag->{'sort'}) {  
                                 @v = $self->sort_arr(@v);  
                         }  
   
                         # use format?  
                         if ($tag->{'format_name'}) {  
                                 @v = map { $self->apply_format($tag->{'format_name'},$tag->{'format_delimiter'},$_) } @v;  
                         }  
110    
111                          # delimiter will join repeatable fields  my $out;
112                          if ($tag->{'delimiter'}) {  my $marc21;
                                 @v = ( join($tag->{'delimiter'}, @v) );  
                         }  
   
                         # default types  
                         my @types = qw(display search);  
                         # override by type attribute  
                         @types = ( $tag->{'type'} ) if ($tag->{'type'});  
   
                         foreach my $type (@types) {  
                                 # append to previous line?  
                                 $log->debug("tag $field / $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;  
                                 }  
                         }  
113    
114    sub _get_ds {
115            return $out;
116    }
117    
118                  }  =head2 _clean_ds
119    
120                  if ($row) {  Clean data structure hash for next record
                         $row->{'tag'} = $field;  
121    
122                          # TODO: name_sigular, name_plural    _clean_ds();
                         my $name = $self->{'import_xml'}->{'indexer'}->{$field}->{'name'};  
                         my $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");  
                         }  
123    
124                          $ds->{$row_name} = $row;  =cut
125    
126                          $log->debug("row $field: ",sub { Dumper($row) });  sub _clean_ds {
127                  }          $out = undef;
128            $marc21 = undef;
129    }
130    
131          }  =head2 _set_lookup
132    
133          $self->{'db'}->save_ds(  Set current lookup hash
                 id => $id,  
                 ds => $ds,  
                 prefix => $self->{prefix},  
         ) if ($self->{'db'});  
134    
135          $log->debug("ds: ", sub { Dumper($ds) });    _set_lookup( $lookup );
136    
137          $log->logconfess("data structure returned is not array any more!") if wantarray;  =cut
138    
139          return $ds;  my $lookup;
140    
141    sub _set_lookup {
142            $lookup = shift;
143  }  }
144    
145  =head2 parse  =head2 _get_marc21_fields
   
 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.  
   
  my $text = $webpac->parse($rec,'eval{"v901^a" eq "Deskriptor"}descriptor: v250^a', $i);  
   
 Filters are implemented here. While simple form of filters looks like this:  
   
   filter{name_of_filter}  
146    
147  but, filters can also have variable number of parametars like this:  Get all fields defined by calls to C<marc21>
148    
149    filter{name_of_filter(param,param,param)}          $marc->add_fields( WebPAC::Normalize:_get_marc21_fields() );
150    
151  =cut  =cut
152    
153  my $warn_once;  sub _get_marc21_fields {
154            return @{$marc21};
155  sub parse {  }
         my $self = shift;  
156    
157          my ($rec, $format_utf8, $i, $rec_size) = @_;  =head1 Functions to create C<data_structure>
158    
159          return if (! $format_utf8);  Those functions generally have to first in your normalization file.
160    
161          my $log = $self->_get_logger();  =head2 tag
162    
163          $log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o);  Define new tag for I<search> and I<display>.
164    
165          $i = 0 if (! $i);    tag('Title', rec('200','a') );
166    
         my $format = $self->_x($format_utf8) || $log->logconfess("can't convert '$format_utf8' from UTF-8 to ",$self->{'code_page'});  
167    
168          my @out;  =cut
169    
170          $log->debug("format: $format [$i]");  sub tag {
171            my $name = shift or die "tag needs name as first argument";
172            my @o = grep { defined($_) && $_ ne '' } @_;
173            return unless (@o);
174            $out->{$name}->{tag} = $name;
175            $out->{$name}->{search} = \@o;
176            $out->{$name}->{display} = \@o;
177    }
178    
179          my $eval_code;  =head2 display
         # remove eval{...} from beginning  
         $eval_code = $1 if ($format =~ s/^eval{([^}]+)}//s);  
180    
181          my $filter_name;  Define tag just for I<display>
         # remove filter{...} from beginning  
         $filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s);  
182    
183          # did we found any (att all) field from format in row?    @v = display('Title', rec('200','a') );
         my $found_any;  
         # prefix before first field which we preserve it $found_any  
         my $prefix;  
184    
185          my $f_step = 1;  =cut
186    
187          while ($format =~ s/^(.*?)(v|s)(\d+)(?:\^(\w))?//s) {  sub display {
188            my $name = shift or die "display needs name as first argument";
189            my @o = grep { defined($_) && $_ ne '' } @_;
190            return unless (@o);
191            $out->{$name}->{tag} = $name;
192            $out->{$name}->{display} = \@o;
193    }
194    
195                  my $del = $1 || '';  =head2 search
                 $prefix = $del if ($f_step == 1);  
196    
197                  my $fld_type = lc($2);  Prepare values just for I<search>
198    
199                  # repeatable index    @v = search('Title', rec('200','a') );
                 my $r = $i;  
                 if ($fld_type eq 's') {  
                         if ($found_any->{'v'}) {  
                                 $r = 0;  
                         } else {  
                                 return;  
                         }  
                 }  
200    
201                  my $found = 0;  =cut
                 my $tmp = $self->get_data(\$rec,$3,$4,$r,\$found,$rec_size);  
202    
203                  if ($found) {  sub search {
204                          $found_any->{$fld_type} += $found;          my $name = shift or die "search needs name as first argument";
205            my @o = grep { defined($_) && $_ ne '' } @_;
206            return unless (@o);
207            $out->{$name}->{tag} = $name;
208            $out->{$name}->{search} = \@o;
209    }
210    
211                          # we will skip delimiter before first occurence of field!  =head2 marc21
                         push @out, $del unless($found_any->{$fld_type} == 1);  
                         push @out, $tmp;  
                 }  
                 $f_step++;  
         }  
212    
213          # test if any fields found?  Save value for MARC field
         return if (! $found_any->{'v'} && ! $found_any->{'s'});  
214    
215          my $out = join('',@out);    marc21('900','a', rec('200','a') );
216    
217          if ($out) {  =cut
                 # add rest of format (suffix)  
                 $out .= $format;  
218    
219                  # add prefix if not there  sub marc21 {
220                  $out = $prefix . $out if ($out !~ m/^\Q$prefix\E/);          my $f = shift or die "marc21 needs field";
221            die "marc21 field must be numer" unless ($f =~ /^\d+$/);
222    
223                  $log->debug("result: $out");          my $sf = shift or die "marc21 needs subfield";
         }  
224    
225          if ($eval_code) {          foreach my $v (@_) {
226                  my $eval = $self->fill_in($rec,$eval_code,$i) || return;                  push @{ $marc21 }, [ $f, ' ', ' ', $sf => $v ];
                 $log->debug("about to eval{$eval} format: $out");  
                 return if (! $self->_eval($eval));  
         }  
           
         if ($filter_name) {  
                 my @filter_args;  
                 if ($filter_name =~ s/(\w+)\((.*)\)/$1/) {  
                         @filter_args = split(/,/, $2);  
                 }  
                 if ($self->{'filter'}->{$filter_name}) {  
                         $log->debug("about to filter{$filter_name} format: $out with arguments: ", join(",", @filter_args));  
                         unshift @filter_args, $out;  
                         $out = $self->{'filter'}->{$filter_name}->(@filter_args);  
                         return unless(defined($out));  
                         $log->debug("filter result: $out");  
                 } elsif (! $warn_once->{$filter_name}) {  
                         $log->warn("trying to use undefined filter $filter_name");  
                         $warn_once->{$filter_name}++;  
                 }  
227          }          }
   
         return $out;  
228  }  }
229    
230  =head2 fill_in  =head1 Functions to extract data from input
   
 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.  
   
  my $text = $webpac->fill_in($rec,'v250^a');  
231    
232  Optional argument is ordinal number for repeatable fields. By default,  This function should be used inside functions to create C<data_structure> described
233  it's assume to be first repeatable field (fields are perl array, so first  above.
 element is 0).  
 Following example will read second value from repeatable field.  
234    
235   my $text = $webpac->fill_in($rec,'Title: v250^a',1);  =head2 rec1
236    
237  This function B<does not> perform parsing of format to inteligenty skip  Return all values in some field
 delimiters before fields which aren't used.  
238    
239  This method will automatically decode UTF-8 string to local code page    @v = rec1('200')
 if needed.  
240    
241  There is optional parametar C<$record_size> which can be used to get sizes of  TODO: order of values is probably same as in source data, need to investigate that
 all C<field^subfield> combinations in this format.  
   
  my $text = $webpac->fill_in($rec,'got: v900^a v900^x',0,\$rec_size);  
242    
243  =cut  =cut
244    
245  sub fill_in {  sub rec1 {
246          my $self = shift;          my $f = shift;
247            return unless (defined($rec) && defined($rec->{$f}));
248          my $log = $self->_get_logger();          if (ref($rec->{$f}) eq 'ARRAY') {
249                    return map {
250          my ($rec,$format,$i,$rec_size) = @_;                          if (ref($_) eq 'HASH') {
251                                    values %{$_};
         $log->logconfess("need data record") unless ($rec);  
         $log->logconfess("need format to parse") unless($format);  
   
         # iteration (for repeatable fields)  
         $i ||= 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 $just_single = 1;  
   
         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  
         if ($format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found,$rec_size)/ges) {  
                 $just_single = 0;  
         }  
   
         # non-repeatable fields  
         if ($format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found,$rec_size)/ges) {  
                 return if ($i > 0 && $just_single);  
         }  
   
         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')) {  
                                 my @lookup = $self->{lookup}->lookup($format);  
                                 $log->debug("lookup $format", join(", ", @lookup));  
                                 return @lookup;  
252                          } else {                          } else {
253                                  $log->warn("Have lookup object but can't invoke lookup method");                                  $_;
254                          }                          }
255                  } else {                  } @{ $rec->{$f} };
256                          return $format;          } elsif( defined($rec->{$f}) ) {
257                  }                  return $rec->{$f};
         } else {  
                 return;  
258          }          }
259  }  }
260    
261    =head2 rec2
262    
263  =head2 _rec_to_arr  Return all values in specific field and subfield
   
 Similar to C<parse> and C<fill_in>, but returns array of all repeatable fields. Usable  
 for fields which have lookups, so they shouldn't be parsed but rather  
 C<paste>d or C<fill_id>ed. Last argument is name of operation: C<paste> or C<fill_in>.  
264    
265   my @arr = $webpac->fill_in_to_arr($rec,'[v900];;[v250^a]','paste');    @v = rec2('200','a')
266    
267  =cut  =cut
268    
269  sub _rec_to_arr {  sub rec2 {
270          my $self = shift;          my $f = shift;
271            return unless (defined($rec && $rec->{$f}));
272            my $sf = shift;
273            return map { $_->{$sf} } grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} };
274    }
275    
276          my ($rec, $format_utf8, $code) = @_;  =head2 rec
277    
278          my $log = $self->_get_logger();  syntaxtic sugar for
279    
280          $log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o);    @v = rec('200')
281          return if (! $format_utf8);    @v = rec('200','a')
282    
283          my $i = 0;  =cut
         my $max = 0;  
         my @arr;  
         my $rec_size = {};  
284    
285          while ($i <= $max) {  sub rec {
286                  my $v = $self->$code($rec,$format_utf8,$i++,\$rec_size) || next;          if ($#_ == 0) {
287                  push @arr, $v;                  return rec1(@_);
288                  if ($rec_size) {          } elsif ($#_ == 1) {
289                          foreach my $f (keys %{ $rec_size }) {                  return rec2(@_);
                                 $max = $rec_size->{$f} if ($rec_size->{$f} > $max);  
                         }  
                         warn "max set to $max, rec_size = ", Dumper($rec_size);  
                         undef $rec_size;  
                 }  
290          }          }
   
         $log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr);  
   
         return @arr;  
291  }  }
292    
293    =head2 regex
294    
295  =head2 get_data  Apply regex to some or all values
   
 Returns value from record.  
   
  my $text = $self->get_data(\$rec,$f,$sf,$i,\$found,\$rec_size);  
   
 Required arguments are:  
   
 =over 8  
   
 =item C<$rec>  
   
 record reference  
   
 =item C<$f>  
   
 field  
   
 =item C<$sf>  
   
 optional subfield  
   
 =item C<$i>  
296    
297  index offset for repeatable values ( 0 ... $rec_size->{'400^a'} )    @v = regex( 's/foo/bar/g', @v );
   
 =item C<$found>  
   
 optional variable that will be incremeted if preset  
   
 =item C<$rec_size>  
   
 hash to hold maximum occurances of C<field^subfield> combinations  
 (which can be accessed using keys in same format)  
   
 =back  
   
 Returns value or empty string, updates C<$found> and C<rec_size>  
 if present.  
298    
299  =cut  =cut
300    
301  sub get_data {  sub regex {
302          my $self = shift;          my $r = shift;
303            my @out;
304          my ($rec,$f,$sf,$i,$found,$cache) = @_;          #warn "r: $r\n",Dumper(\@_);
305            foreach my $t (@_) {
306          return '' unless ($$rec->{$f} && ref($$rec->{$f}) eq 'ARRAY');                  next unless ($t);
307                    eval "\$t =~ $r";
308          if (defined($$cache)) {                  push @out, $t if ($t && $t ne '');
                 $$cache->{ $f . ( $sf ? '^' . $sf : '' ) } ||= scalar @{ $$rec->{$f} };  
         }  
   
         return '' unless ($$rec->{$f}->[$i]);  
   
         {  
                 no strict 'refs';  
                 if (defined($sf)) {  
                         $$found++ if (defined($$found) && $$rec->{$f}->[$i]->{$sf});  
                         return $$rec->{$f}->[$i]->{$sf};  
                 } else {  
                         $$found++ if (defined($$found));  
                         # it still might have subfields, just  
                         # not specified, so we'll dump some debug info  
                         if ($$rec->{$f}->[$i] =~ /HASH/o) {  
                                 my $out;  
                                 foreach my $k (keys %{$$rec->{$f}->[$i]}) {  
                                         $out .= '$' . $k .':' . $$rec->{$f}->[$i]->{$k}." ";  
                                 }  
                                 return $out;  
                         } else {  
                                 return $$rec->{$f}->[$i];  
                         }  
                 }  
309          }          }
310            return @out;
311  }  }
312    
313    =head2 prefix
314    
315  =head2 apply_format  Prefix all values with a string
   
 Apply format specified in tag with C<format_name="name"> and  
 C<format_delimiter=";;">.  
316    
317   my $text = $webpac->apply_format($format_name,$format_delimiter,$data);    @v = prefix( 'my_', @v );
   
 Formats can contain C<lookup{...}> if you need them.  
318    
319  =cut  =cut
320    
321  sub apply_format {  sub prefix {
322          my $self = shift;          my $p = shift or die "prefix needs string as first argument";
323            return map { $p . $_ } grep { defined($_) } @_;
324          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);  
325    
326          my $format = $self->_x($self->{'import_xml'}->{'format'}->{$name}->{'content'}) || $log->logdie("can't find format '$name'");  =head2 suffix
327    
328          my @data = split(/\Q$delimiter\E/, $data);  suffix all values with a string
329    
330          my $out = sprintf($format, @data);    @v = suffix( '_my', @v );
         $log->debug("using format $name [$format] on $data to produce: $out");  
331    
332          if ($self->{'lookup_regex'} && $out =~ $self->{'lookup_regex'}) {  =cut
                 return $self->{'lookup'}->lookup($out);  
         } else {  
                 return $out;  
         }  
333    
334    sub suffix {
335            my $s = shift or die "suffix needs string as first argument";
336            return map { $_ . $s } grep { defined($_) } @_;
337  }  }
338    
339  =head2 sort_arr  =head2 surround
340    
341  Sort array ignoring case and html in data  surround all values with a two strings
342    
343   my @sorted = $webpac->sort_arr(@unsorted);    @v = surround( 'prefix_', '_suffix', @v );
344    
345  =cut  =cut
346    
347  sub sort_arr {  sub surround {
348          my $self = shift;          my $p = shift or die "surround need prefix as first argument";
349            my $s = shift or die "surround needs suffix as second argument";
350          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;  
351  }  }
352    
353    =head2 first
354    
355  =head1 INTERNAL METHODS  Return first element
   
 =head2 _sort_by_order  
356    
357  Sort xml tags data structure accoding to C<order=""> attribute.    $v = first( @v );
358    
359  =cut  =cut
360    
361  sub _sort_by_order {  sub first {
362          my $self = shift;          my $r = shift;
363            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;  
364  }  }
365    
366  =head2 _x  =head2 lookup
   
 Convert strings from C<conf/normalize/*.xml> encoding into application  
 specific encoding (optinally specified using C<code_page> to C<new>  
 constructor).  
367    
368   my $text = $n->_x('normalize text string');  Consult lookup hashes for some value
369    
370  This is a stub so that other modules doesn't have to implement it.    @v = lookup( $v );
371      @v = lookup( @v );
372    
373  =cut  =cut
374    
375  sub _x {  sub lookup {
376          my $self = shift;          my $k = shift or return;
377          return shift;          return unless (defined($lookup->{$k}));
378            if (ref($lookup->{$k}) eq 'ARRAY') {
379                    return @{ $lookup->{$k} };
380            } else {
381                    return $lookup->{$k};
382            }
383  }  }
384    
385    =head2 join_with
386    
387  =head1 AUTHOR  Joins walues with some delimiter
388    
389  Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >>    $v = join_with(", ", @v);
   
 =head1 COPYRIGHT & LICENSE  
   
 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.  
390    
391  =cut  =cut
392    
393  1; # End of WebPAC::Normalize  sub join_with {
394            my $d = shift;
395            return join($d, grep { defined($_) && $_ ne '' } @_);
396    }
397    
398    # END
399    1;

Legend:
Removed from v.372  
changed lines
  Added in v.540

  ViewVC Help
Powered by ViewVC 1.1.26