Class: FormScraperHelper

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

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(url = nil, browser: nil, headless: false, clipb: true, fd: nil, debug: false) ⇒ FormScraperHelper

note: fd corresponds to FakeDataGenerator22 which is optional



16
17
18
19
20
21
22
23
 
# File 'lib/formscraper_helper.rb', line 16

def initialize(url=nil, browser: nil, headless: false, clipb: true,
               fd: nil, debug: false)

  @url, @clipb, @fd, @debug = url, clipb, fd, debug

  @browser = browser ? browser : FerrumWizard.new(url,  headless: headless)

end

Instance Attribute Details

#browser ⇒ Object (readonly)

Returns the value of attribute browser.



12
13
14
 
# File 'lib/formscraper_helper.rb', line 12

def browser
  @browser
end

Instance Method Details

#scrape(body = @browser.body) ⇒ Object 



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
 
# File 'lib/formscraper_helper.rb', line 25

def scrape(body=@browser.body)
  puts 'body: '  + body.inspect if @debug
  doc = Nokorexi.new(body).to_doc

  #a = doc.root.xpath('//input|//select')
  a = doc.root.xpath('//*').select {|x| x.name == 'input' or x.name == 'select'}
  a.reject! do |x|
    x.attributes[:type] == 'hidden' or x.attributes[:style] =~ /display:none/
  end

  @h = a.map do |x|

    key = x.attributes[:name]
    type = x.name

    h = {}
    h[:type] = x.attributes[:type] || type
    h[:xpath] = "//%s[@name=\"%s\"]" % [type, key]
    h[:title] = x.attributes[:title]

    if type == 'select' then
      h[:options] = x.xpath('option').map {|x| x.text.to_s}
    end

    [key, h]

  end.to_h

end

#to_code ⇒ Object 



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
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
 
# File 'lib/formscraper_helper.rb', line 59

def to_code()

s=<<EOF
require 'yaml'
require 'ferrum'
require 'nokorexi'

browser = Ferrum::Browser.new  headless: false
url = '#{@url}'
browser.goto(url)
sleep 2

doc = Nokorexi.new(browser.body).to_doc

# load the YAML document containing the inputs
#filepath = ''
filepath = '/tmp/tmp.yaml'
h = YAML.load(File.read(filepath))
EOF

  @h.each do |key, h|

    puts 'key: ' + key.inspect if @debug

    s += "r = browser.at_xpath('#{h[:xpath]}')\n"

    if h[:type] == 'text' or h[:type] == 'password' then

      var1, s2 = format_var1(h[:title], key)
      s += s2
      s += var1 + " = h['#{var1}']\n"
      s += "r.focus.type #{var1}\n"
      s += "sleep 0.5\n\n"

    elsif h[:type] == 'select'

      var1, s2 = format_var1(h[:title], key)
      s += s2

      s += "# options: #{h[:options].join(', ')}\n"
      s += "#{var1} = h['#{var1}']\n"
      s += 'titles = %w(' + h[:options].join(' ') + ')' + "\n"
      s += 'found = titles.grep /#{' + var1 + '}/i' + "\n"
      s += "n = titles.index(found.first) + 1\n"
      s += "r.focus\n"
      s += "n.times { r.type(:down); sleep 1}\n"
      s += "r.click\n"
      s += "sleep 0.5\n\n"

    elsif h[:type] == 'checkbox'
      s += "r.focus.click\n\n"
    end

  end

  Clipboard.copy s if @clipb
  puts 'generated code copied to clipboard'

  return s

end

#to_h ⇒ Object 



55
56
57
 
# File 'lib/formscraper_helper.rb', line 55

def to_h()
  @h
end

#to_yaml ⇒ Object

creates a YAML document for the inputs



123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
 
# File 'lib/formscraper_helper.rb', line 123

def to_yaml()

  s = '---' + "\n"

  @h.each do |key, h|

    puts 'key: ' + key.inspect if @debug

    if h[:type] == 'text' or h[:type] == 'password' then

      var1, s2 = format_var1(h[:title], key)

      s += s2

      if h[:type] == 'password' then
        @pwd ||= @fd ? @fd.password : 'xxx'
        s += var1 + ": #{@pwd}\n"
      elsif @fd

        found = @fd.lookup var1
        val = found.is_a?(String) ? found : 'xxx'
        s += var1 + ": '#{val}'\n"
      else
        s += var1 + ": xxx\n"
      end

    elsif h[:type] == 'select'

      var1, s2 = format_var1(h[:title], key)

      s += s2
      s += "# options: #{h[:options].join(', ')}\n"
      val = h[:options][1..-1].sample
      s += "#{var1}: '#{val}'\n"

    elsif h[:type] == 'checkbox'

    end

  end

  Clipboard.copy s if @clipb
  puts 'generated YAML copied to clipboard'

  return s

end