Class: Amazoned::Parser

Inherits:
Object
  • Object
show all
Defined in:
lib/amazoned/parser.rb

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(response) ⇒ Parser

Returns a new instance of Parser.



5
6
7
8
9
 
# File 'lib/amazoned/parser.rb', line 5

def initialize(response)
  @product_hash = Hash.new
  @response = response
  @html_doc = Nokogiri::HTML(response.body)
end

Instance Attribute Details

#html_doc ⇒ Object (readonly)

Returns the value of attribute html_doc.



3
4
5
 
# File 'lib/amazoned/parser.rb', line 3

def html_doc
  @html_doc
end

#product_hash ⇒ Object

Returns the value of attribute product_hash.



2
3
4
 
# File 'lib/amazoned/parser.rb', line 2

def product_hash
  @product_hash
end

#response ⇒ Object (readonly)

Returns the value of attribute response.



3
4
5
 
# File 'lib/amazoned/parser.rb', line 3

def response
  @response
end

Instance Method Details

#call ⇒ Object 



11
12
13
 
# File 'lib/amazoned/parser.rb', line 11

def call
  parse_response_for_product_details( response )
end

#extract_subcategory_rankings(nokogiri_html) ⇒ Object 



90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
 
# File 'lib/amazoned/parser.rb', line 90

def extract_subcategory_rankings(nokogiri_html)
  # Below is gnarly string manipulation to parse text strings like:
  # "\n    #2\n    in Baby > Baby Care > Health\n    \n    #2\n    in Baby > Baby Care > Pacifiers, Teethers & Teething Relief > Teethers\n    "
  # into:
  # [["2", "Baby > Baby Care > Health"], ["2", "Baby > Baby Care > Pacifiers, Teethers & Teething Relief > Teethers"]]
  nokogiri_html
  .map{|i| i.text}
  .map{|i| i.partition("in")
  .map(&:strip)}
  .map{|i| i - ["in"] }
  .map{|i|
    i.map{|ii|
      ii.gsub("#", "") # remove '#' from '#2'
      .gsub("\u00A0", "") # remove No-Break Space Unicode characters (U+00A0) since Ruby's .strip command won't remove them
    }
  }.each do |i|
    hsh = {}
    hsh[:rank] = i.first.to_i
    hsh[:ladder] = i.last
    product_hash[:best_sellers_rank] << hsh
  end
end

#parse_response_for_product_details(response) ⇒ Object 



15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
 
# File 'lib/amazoned/parser.rb', line 15

def parse_response_for_product_details(response)
  product_hash[:best_sellers_rank] = []

  ########
  # # Parent category Seller Rank Parser
  ########
  parsed_parent_category = html_doc.css('#SalesRank').text.partition("(").first.chop.partition("#").last.partition("in").map(&:strip) - ["in"]
  product_hash[:rank] = parsed_parent_category.first.delete(',').to_i # "903,610" -> 903610
  product_hash[:category] = parsed_parent_category.last

  ########
  # # Subcategory Seller Rank Parser
  ########
  extract_subcategory_rankings( html_doc.css('.zg_hrsr_item') )

  ########
  # # Package Dimension Parser
  ########
  # Package Dimension Parsing Strategy 1:
  product_hash[:package_dimensions] = html_doc.css('.size-weight').children.map{|r| r.text}.reject{|r| !r.match?("inches")}.first

  # Package Dimension Parsing Strategy 2:
  if product_hash[:package_dimensions].blank?

    # Find an index for the string "Package Dimensions" within a string text extraction of the page
    str_index = html_doc.inner_text.index("Package Dimensions")

    unless str_index.nil?

      # Reduce string representing the html page down to a smaller target string including "Package Dimensions" and the weights
      str = html_doc.inner_text[str_index .. str_index + 150]

      # Find within target string an index for where the word "inches" appears, then grab characters around it
      product_hash[:package_dimensions] = str[str.index("inches")- 20.. str.index("inches")+8].strip
    end
  end

  # Package Dimension Parsing Strategy 3:
    response.search('.//*[@class="a-color-secondary a-size-base prodDetSectionEntry"]').map{|n| n.parent}.each do |n|

      # Parse html in each row of Amazon's product details table to get back a string. E.g:  "\n    \n        Best Sellers Rank\n    \n    \n         \n              \n #63 in Toys & Games (See Top 100 in Toys & Games)\n        \n              \n #3 in Toys & Games > Baby & Toddler Toys > Teethers\n        \n              \n        \n    \n    "
      str = n.children.inner_text
      if product_hash[:best_sellers_rank].blank?
        str.match("Best Sellers Rank") do |m|
          # Gnarly string manipulation extracts the array: ["63", "in", "Toys & Games"]
          parsed_parent_category = str.partition("(").first.chop.partition("#").last.partition("in").map(&:strip)

          # From ["63", "in", "Toys & Games"] we only care about first & last parts of this array
          product_hash[:rank] = parsed_parent_category.first.delete(',').to_i
          product_hash[:category] = parsed_parent_category.last

          parsed_categories = str.partition(")").last.split("#").map(&:strip)
          parsed_categories.each do |pc|
            next if pc.blank?
            parsed_category = pc.partition("in").map(&:strip).map{|i| i.gsub("#", "")} - ["in"]

            hsh = {}
            hsh[:rank] = parsed_category.first.delete(',').to_i
            hsh[:ladder] = parsed_category.last
            product_hash[:best_sellers_rank] << hsh
         end
        end
      end

      if product_hash[:product_dimensions].blank?

        # Use pattern matching to extract the product details we care about
        str.match("Product Dimensions") do |m|
          product_hash[:package_dimensions] = str[str.index("inches")- 20.. str.index("inches")+8].strip
        end
      end
    end
  product_hash
end