Class: Rexle::Element

Inherits:
Object
  • Object
show all
Includes:
XMLhelper
Defined in:
lib/rexle.rb

Instance Attribute Summary collapse

Instance Method Summary collapse

Methods included from XMLhelper

#doc_pretty_print, #doc_print, #pretty_print, #processing_instructions, #scan_print, #scan_to_a

Constructor Details

#initialize(name = nil, value: nil, attributes: Attributes.new, rexle: nil) ⇒ Element

Returns a new instance of Element.



387
388
389
390
391
392
393
394
395
396
397
398
399
 
# File 'lib/rexle.rb', line 387

def initialize(name=nil, value: nil, attributes: Attributes.new, rexle: nil)

  @rexle = rexle      
  super()

  @name, @attributes = name.to_s, attributes

  raise "Element name must not be blank" unless name
  @child_elements = []
  self.add_text value if value
  #@log = Logger.new('rexle.log','daily')

end

Instance Attribute Details

#child_elements ⇒ Object (readonly)

Returns the value of attribute child_elements.



383
384
385
 
# File 'lib/rexle.rb', line 383

def child_elements
  @child_elements
end

#doc_id ⇒ Object (readonly)

Returns the value of attribute doc_id.



383
384
385
 
# File 'lib/rexle.rb', line 383

def doc_id
  @doc_id
end

#instructions ⇒ Object (readonly)

Returns the value of attribute instructions.



383
384
385
 
# File 'lib/rexle.rb', line 383

def instructions
  @instructions
end

#name ⇒ Object

Returns the value of attribute name.



382
383
384
 
# File 'lib/rexle.rb', line 382

def name
  @name
end

#parent ⇒ Object

Returns the value of attribute parent.



382
383
384
 
# File 'lib/rexle.rb', line 382

def parent
  @parent
end

#value ⇒ Object

Returns the value of attribute value.



382
383
384
 
# File 'lib/rexle.rb', line 382

def value
  @value
end

Instance Method Details

#add(item) ⇒ Object 



844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
 
# File 'lib/rexle.rb', line 844

def add(item)   

  if item.is_a? Rexle::Element then

    if self.doc_id == item.doc_id then

      new_item = item.deep_clone
      add_element new_item
      item.delete
      item = new_item
      new_item
    else
      add_element item
    end
  else
    add_element item
  end

end

#add_attribute(*x) ⇒ Object 



872
873
874
875
876
877
878
879
880
881
882
883
 
# File 'lib/rexle.rb', line 872

def add_attribute(*x)

  procs = {
    Hash: lambda {|x| x[0] || {}},
    String: lambda {|x| Hash[*x]},
    Symbol: lambda {|x| Hash[*x]}
  }

  h = procs[x[0].class.to_s.to_sym].call(x)

  @attributes.merge! h
end

#add_element(item) ⇒ Object 



822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
 
# File 'lib/rexle.rb', line 822

def add_element(item)

  if item.is_a? String then
    @child_elements << String.new(item)

  elsif item.is_a? Rexle::CData then
    @child_elements << item
  elsif item.is_a? Rexle::Comment then
    @child_elements << item                
  elsif item.is_a? Rexle::Element then

    @child_elements << item
    # add a reference from this element (the parent) to the child
    item.parent = self
    item        
      
  elsif item.is_a? Rexle then
    self.add_element(item.root)
  end

end

#add_text(s) ⇒ Object 



885
886
887
888
889
 
# File 'lib/rexle.rb', line 885

def add_text(s)

  self.child_elements << s
  self 
end

#at_css(selector) ⇒ Object 



430
431
432
 
# File 'lib/rexle.rb', line 430

def at_css(selector)
  self.root.element RexleCSS.new(selector).to_xpath
end

#attribute(key) ⇒ Object 



891
892
893
894
895
896
897
898
899
900
 
# File 'lib/rexle.rb', line 891

def attribute(key) 
  
  key = key.to_sym if key.is_a? String
  
  if @attributes[key].is_a? String then
    @attributes[key].gsub('&lt;','<').gsub('&gt;','>') 
  else
    @attributes[key]
  end
end

#attributes ⇒ Object 



902
 
# File 'lib/rexle.rb', line 902

def attributes() @attributes end

#backtrack ⇒ Object 



401
402
403
 
# File 'lib/rexle.rb', line 401

def backtrack()
  BacktrackXPath.new(self)
end

#cdata? ⇒ Boolean

Returns:

  • (Boolean) 


405
406
407
 
# File 'lib/rexle.rb', line 405

def cdata?()
  self.is_a? CData
end

#cdatas ⇒ Object 



904
905
906
 
# File 'lib/rexle.rb', line 904

def cdatas()
  self.children.inject([]){|r,x| x.is_a?(Rexle::CData) ? r << x.to_s : r }
end

#children ⇒ Object 



908
909
910
911
912
913
914
915
916
917
 
# File 'lib/rexle.rb', line 908

def children()

  r =  @child_elements
  
  def r.is_an_empty_string?()
    self.length == 1 and self.first == ''
  end      
  
  return r
end

#children=(a) ⇒ Object 



919
 
# File 'lib/rexle.rb', line 919

def children=(a)   @child_elements = a if a.is_a? Array  end

#clone ⇒ Object 



923
924
925
 
# File 'lib/rexle.rb', line 923

def clone() 
  Element.new(@name, attributes: @attributes) 
end

#contains(raw_args) ⇒ Object 



409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
 
# File 'lib/rexle.rb', line 409

def contains(raw_args)
  #log = Logger.new('rexle.log','daily')
  #log.debug 'inside contains'
  path, raw_val = raw_args.split(',',2)
  val = raw_val.strip[/^["']?.*["']?$/]      
  
  anode = query_xpath(path)

  return [false] if anode.nil? or anode.empty?

  a = scan_contents(anode.first)
  r = [a.grep(/#{val.sub(/^["'](.*)["']$/,'\1')}/).length > 0]

  r.any?
end

#content(options = {}) ⇒ Object 



1049
1050
1051
 
# File 'lib/rexle.rb', line 1049

def content(options={})
  CGI.unescapeHTML(xml(options))
end

#count(path) ⇒ Object 



425
426
427
428
 
# File 'lib/rexle.rb', line 425

def count(path)
  length = query_xpath(path).flatten.compact.length
  length
end

#css(selector) ⇒ Object 



434
435
436
437
438
 
# File 'lib/rexle.rb', line 434

def css(selector)

  selector.split(',')\
              .flat_map{|x| self.root.xpath RexleCSS.new(x).to_xpath}
end

#deep_clone ⇒ Object 



921
 
# File 'lib/rexle.rb', line 921

def deep_clone() Rexle.new(self.xml).root end

#delete(obj = nil) ⇒ Object Also known as: remove 



927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
 
# File 'lib/rexle.rb', line 927

def delete(obj=nil)

  if obj then

    if obj.is_a? String then
      
      self.xpath(obj).each {|e| e.delete}
      
    else

      i = @child_elements.index(obj)
      [@child_elements].each{|x| x.delete_at i} if i          
    end
  else

    self.parent.delete(self) if self.parent
  end
end

#doc_root ⇒ Object 



963
 
# File 'lib/rexle.rb', line 963

def doc_root() @rexle.root                                  end

#each(&blk) ⇒ Object 



964
 
# File 'lib/rexle.rb', line 964

def each(&blk)    self.children.each(&blk)                  end

#each_recursive(&blk) ⇒ Object Also known as: traverse 



965
 
# File 'lib/rexle.rb', line 965

def each_recursive(&blk) recursive_scan(self.children,&blk) end

#element(s) ⇒ Object 



948
949
950
951
 
# File 'lib/rexle.rb', line 948

def element(s)
  r = self.xpath(s)
  r.is_a?(Array) ? r.first : r
end

#elements(s = nil) ⇒ Object 



953
954
955
956
957
958
959
960
961
 
# File 'lib/rexle.rb', line 953

def elements(s=nil)
  procs = {
    NilClass: proc {Elements.new(@child_elements\
                                 .select{|x| x.kind_of? Rexle::Element })},
    String: proc {|x| @child_elements[x]}
  }

  procs[s.class.to_s.to_sym].call(s)      
end

#filter_xpath(raw_path, rlist = [], &blk) ⇒ Object 



503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
 
# File 'lib/rexle.rb', line 503

def filter_xpath(raw_path, rlist=[], &blk)
  #@log.debug 'inside filter_xpath : ' + raw_path.inspect
  path = String.new raw_path

  # is it a function
  fn_match = path.match(/^(\w+)\(["']?([^\)]*)["']?\)(?:\[(.*)\])?$/)
  #@log.debug 'fn_match : ' + fn_match.inspect
  end_fn_match = path.slice!(/\[\w+\(\)\]$/)
  
  if end_fn_match then
    
    m = end_fn_match[1..-4]
    #@log.debug 'its a function'
    [method(m.to_sym).call(xpath path)]
  
  elsif (fn_match and fn_match.captures.first[/^(attribute|@)/]) 

    procs = {

      Array: proc { |x| 
        if block_given? then 
          x.flatten(1) 
        else
          rs = x.flatten
          rs.any?{|x| x == true or x == false} ? rs : rs.uniq(&:object_id) 
        end
      }, 
     String: proc {|x| x},
      Hash: proc {|x| x},
      TrueClass: proc{|x| x},
      FalseClass: proc{|x| x},
      :"Rexle::Element" => proc {|x| [x]}
    }
    bucket = []
    raw_results = path.split('|').map do |xp|

      query_xpath(xp.strip, bucket, &blk)         
    end
    
    results = raw_results

    procs[results.class.to_s.to_sym].call(results) if results              
    
  elsif fn_match.nil?
    
    procs = {

      Array: proc { |x| 
        if block_given? then 
          x.flatten(1) 
        else
          rs = x.flatten
          rs.any?{|x| x == true or x == false} ? rs : rs.uniq(&:object_id) 
        end
      }, 
     String: proc {|x| x},
      Hash: proc {|x| x},
      TrueClass: proc{|x| x},
      FalseClass: proc{|x| x},
      :"Rexle::Element" => proc {|x| [x]}
    }
    bucket = []
    raw_results = path.split('|').map do |xp|

      query_xpath(xp.strip, bucket, &blk)         
    end

    return [true] if !path[/[><]/] and raw_results.flatten.index(true)
    results = raw_results # .flatten.select {|x| x}
    
    procs[results.class.to_s.to_sym].call(results) if results            
    
  else
    
    m, xpath_value, index = fn_match.captures        

    if m == 'text' then
      a = texts()
      return index ? a[index.to_i - 1].to_s : a
    end
    #@log.debug 'before function call'
    raw_results = xpath_value.empty? ? method(m.to_sym).call \
                                : method(m.to_sym).call(xpath_value)

    raw_results

  end

end

#has_elements? ⇒ Boolean

Returns:

  • (Boolean) 


967
 
# File 'lib/rexle.rb', line 967

def has_elements?() !self.elements.empty?                   end

#insert_after(node) ⇒ Object 



968
 
# File 'lib/rexle.rb', line 968

def insert_after(node)   insert(node, 1)                    end

#insert_before(node) ⇒ Object 



969
 
# File 'lib/rexle.rb', line 969

def insert_before(node)  insert(node)                       end

#inspect ⇒ Object 



864
865
866
867
868
869
870
 
# File 'lib/rexle.rb', line 864

def inspect()
  if self.xml.length > 30 then
  "%s ... </>" % self.xml[/<[^>]+>/]
  else
    self.xml
  end  
end

#last(a) ⇒ Object 



970
 
# File 'lib/rexle.rb', line 970

def last(a) a.last                                          end

#map(&blk) ⇒ Object 



971
 
# File 'lib/rexle.rb', line 971

def map(&blk)    self.children.map(&blk)                    end

#max(path) ⇒ Object 



440
441
442
443
 
# File 'lib/rexle.rb', line 440

def max(path) 
  a = query_xpath(path).flatten.select{|x| x.is_a? String}.map(&:to_i)
  a.max 
end

#next_element ⇒ Object Also known as: next_sibling 



461
462
463
464
465
466
467
468
469
 
# File 'lib/rexle.rb', line 461

def next_element()      

  id  = self.object_id
  a = self.parent.child_elements      
  i = a.index {|x| x.object_id == id} + 1
  
  a[i] if  i < a.length
    
end

#notx(bool) ⇒ Object 



473
474
475
476
477
478
479
480
481
 
# File 'lib/rexle.rb', line 473

def notx(bool)
  #log = Logger.new('rexle.log','daily')
  #log.debug 'self ' + self.xml.inspect
  #log.debug 'inside not() ' + bool.inspect
  r = self.xpath(bool).any?
  #log.debug 'r: ' + r.inspect
  #log.debug 'r2: ' + (!r).inspect
  !r
end

#original_clone ⇒ Object 



385
 
# File 'lib/rexle.rb', line 385

alias original_clone clone

#prepend(item) ⇒ Object 



1053
1054
1055
1056
1057
1058
1059
1060
 
# File 'lib/rexle.rb', line 1053

def prepend(item)
  
  @child_elements.unshift item
  
  # add a reference from this element (the parent) to the child
  item.parent = self
  item              
end

#previous_element ⇒ Object Also known as: previous_sibling 



483
484
485
486
487
488
489
490
491
 
# File 'lib/rexle.rb', line 483

def previous_element()      
  
  id  = self.object_id
  a = self.parent.child_elements      
  i = a.index {|x| x.object_id == id} - 1
  
  a[i] if  i >= 0 

end

#query_xpath(raw_xpath_value, rlist = [], &blk) ⇒ Object 



593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
 
# File 'lib/rexle.rb', line 593

def query_xpath(raw_xpath_value, rlist=[], &blk)
  
  #@log.debug 'query_xpath : ' + raw_xpath_value.inspect
  #@log.debug '++ ' + self.xml.inspect

  flag_func = false            

  xpath_value = raw_xpath_value.sub('child::','./')

  if xpath_value[/^[\w\/]+\s*=.*/] then
    #@log.debug 'yah'
    flag_func = true

    xpath_value.sub!(/^\w+\s*=.*/,'.[\0]')
    xpath_value.sub!(/\/([\w]+\s*=.*)/,'[\1]')

  end

  raw_path, raw_condition = xpath_value.sub(/^\.?\/(?!\/)/,'')\
      .match(/([^\[]+)(\[[^\]]+\])?/).captures 

  remaining_path = ($').to_s
  #@log.debug 'remaining_path: ' + remaining_path.inspect

  if remaining_path[/^contains\(/] then
    raw_condition = raw_condition ? raw_condition + '/' + remaining_path \
                                                           : remaining_path
    remaining_path = ''        
  end

  r = raw_path[/^([^\/]+)(?=\/\/)/,1] 
  
  if r then
    a_path = raw_path.split(/(?=\/\/)/,2)
  else
    a_path = raw_path.split('/',2)
  end
  
  condition = raw_condition if a_path.length <= 1 #and not raw_condition[/^\[\w+\(.*\)\]$/]

  if raw_path[0,2] == '//' then
    s = ''
  elsif raw_path == 'text()'        

    a_path.shift
    #return @value
    return self.texts
  else

    attribute = xpath_value[/^(attribute::|@)(.*)/,2] 
  
    return @attributes  if attribute == '*'
    return [@attributes[attribute.to_sym]] if attribute \
              and @attributes and @attributes.has_key?(attribute.to_sym)
    s = a_path.shift
  end      

  # isolate the xpath to return just the path to the current element

  elmnt_path = s[/^([\w:\-\*]+\[[^\]]+\])|[\/]+{,2}[^\/]+/]
  element_part = elmnt_path[/(^@?[^\[]+)?/,1] if elmnt_path

  if element_part then

    unless element_part[/^(@|[@\.\w]+[\s=])/] then
      element_name = element_part[/^[\w:\-\*\.]+/]

    else
      if xpath_value[/^\[/] then
        condition = xpath_value
        element_name = nil
      else
        condition = element_part

        attr_search = format_condition('[' + condition + ']')
        #@log.debug 'attr_search : ' + attr_search.inspect
        return [attribute_search(attr_search, \
                                 self, self.attributes) != nil]
      end

    end

  end

  #element_name ||= '*'
  raw_condition = '' if condition

  attr_search = format_condition(condition) if condition \
                                            and condition.length > 0
  #@log.debug 'attr_search2 : ' + attr_search.inspect
  attr_search2 = xpath_value[/^\[(.*)\]$/,1]

  if attr_search2 then
    #@log.debug 'before attribute_Search'
    r4 = attribute_search(attr_search, self, self.attributes)
    return r4
  end
  
  
  return_elements = []

  if raw_path[0,2] == '//' then

    regex = /\[(\d+)\]$/
    n = xpath_value[regex,1]
    xpath_value.slice!(regex)

    rs = scan_match(self, xpath_value).flatten.compact
    return n ? rs[n.to_i-1] : rs

  else

    if element_name.is_a? String then
      ename, raw_selector = (element_name.split('::',2)).reverse
      
      selector = case raw_selector
        when 'following-sibling' then 1
        when 'preceding-sibling' then -1
      end
      
    else
      ename = element_name
    end        

    if ename != '..' then

      return_elements = @child_elements.map.with_index.select do |x, i|

        next unless x.is_a? Rexle::Element

        (x.name == ename || ename == '.') or  (ename == '*')
      end

      if selector then
        ne = return_elements.inject([]) do |r,x| 
          i = x.last + selector
          if i >= 0 then
            r << i
          else
            r
          end
        end

        return_elements = ne.map {|x| [@child_elements[x], x] if x}
      end
    else
                
      remaining_xpath = raw_path[/\.\.\/(.*)/,1]
      # select the parent element

      r2 =  self.parent.xpath(remaining_xpath)

      return r2
      
    end
  end


  if return_elements.length > 0 then

    if (a_path + [remaining_path]).join.empty? then

      # pass in a block to the filter if it is function contains?
      rlist = return_elements.map.with_index do |x,i| 
        r5 = filter(x, i+1, attr_search, &blk)

        r5
      end.compact

      rlist = rlist[0] if rlist.length == 1

    else

      rlist << return_elements.map.with_index do |x,i| 

        rtn_element = filter(x, i+1, attr_search) do |e| 

          r = e.xpath(a_path.join('/') + raw_condition.to_s \
                + remaining_path, &blk)

          r = e if r.is_a?(Array) and r.first and r.first == true \
                                                          and a_path.empty?

          r
        end

        next if rtn_element.nil? or (rtn_element.is_a? Array \
                                                  and rtn_element.empty?)

        if rtn_element.is_a? Hash then
          rtn_element
        elsif rtn_element.is_a? Array then
          rtn_element
        elsif (rtn_element.is_a? String) || (rtn_element.is_a?(Array) \
                                      and not(rtn_element[0].is_a? String))
          rtn_element
        elsif rtn_element.is_a? Rexle::Element
          rtn_element
        elsif rtn_element  == true
          true
        end
      end

      rlist = rlist.flatten(1) unless rlist.length > 1 \
                                                   and rlist[0].is_a? Array

    end

    rlist.compact! if rlist.is_a? Array

  else

    # strip off the 1st element from the XPath
    new_xpath = xpath_value[/^\/\/[\w:\-]+\/(.*)/,1]

    if new_xpath then
      self.xpath(new_xpath + raw_condition.to_s + remaining_path, \
                                                            rlist,&blk)
    end
  end
 
  rlist = rlist.flatten(1) unless not(rlist.is_a? Array) \
                             or (rlist.length > 1 and rlist[0].is_a? Array)
  rlist = [rlist] if rlist.is_a? Rexle::Element
  rlist = (rlist.length > 0 ? true : false) if flag_func == true

  rlist
end

#root ⇒ Object 



972
 
# File 'lib/rexle.rb', line 972

def root() self                                             end

#text(s = '') ⇒ Object 



974
975
976
977
978
979
980
 
# File 'lib/rexle.rb', line 974

def text(s='')
  
  return self.value if s.empty? 
  
  e = self.element(s)
  e.text if e
end

#texts ⇒ Object 



982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
 
# File 'lib/rexle.rb', line 982

def texts()

 r = @child_elements.select do |x|
    x.is_a? String or x.is_a? Rexle::CData
  end
  
  r.map do |x|
    def x.unescape()
      s = self.to_s.clone
      %w(&lt; < &gt; > &amp; & &pos; ').each_slice(2){|x| s.gsub!(*x)}
      s
    end            
  end
  
  return r
end

#to_a ⇒ Object 



1027
1028
1029
1030
 
# File 'lib/rexle.rb', line 1027

def to_a()
  e = [String.new(self.name), Hash.new(self.attributes), self.value.to_s]
  [*e, *scan_to_a(self.children)]
end

#xml(options = {}) ⇒ Object Also known as: to_s 



1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
 
# File 'lib/rexle.rb', line 1032

def xml(options={})

  h = {
    Hash: lambda {|x|
      o = {pretty: false}.merge(x)
      msg = o[:pretty] == false ? :doc_print : :doc_pretty_print
     
      method(msg).call(self.children)
    },
    String: lambda {|x| 
      r = self.element(x)
      r ? r.xml : ''
    }
  }
  h[options.class.to_s.to_sym].call options
end

#xpath(path, rlist = [], &blk) ⇒ Object 



495
496
497
498
499
500
501
 
# File 'lib/rexle.rb', line 495

def xpath(path, rlist=[], &blk)
  #@log.debug 'inside xpath ' + path.inspect

  r = filter_xpath(path, rlist=[], &blk)
  #@log.debug 'after filter_xpath : ' + r.inspect
  r.is_a?(Array) ? r.compact : r      
end