Class: Factbook::Sanitizer

Inherits:
Object
  • Object
show all
Includes:
Utils, LogUtils::Logging
Defined in:
lib/factbook/sanitizer.rb

Constant Summary collapse

BEGIN_FACTS_REGEX = 
/<ul\s+
   class="expandcollapse">
/xim
END_FACTS_REGEX =

ignore case; multi-line

/<\/li>\s*
 <\/ul>\s*
 <\/tbody>\s*
 <\/table>
/xim
STYLE_ATTR_REGEX = 
/\s*
  style=('|").+?\1     ## note: use non-greedy match e.g. .+?
/xim
CLASS_ATTR_REGEX =

do NOT allow multi-line - why? why not?

/\s*
  class=('|")(.+?)\1     ## note: use non-greedy match e.g. .+?
/xim
COUNTRY_COMPARISON_REGEX =

todo: add enclosing div too!!!

/
<div>
 <span \s class='category'[^>]*>
   country \s comparison \s to \s the \s world: \s*
 <\/span>
  \s*
 <span \s class='category_data'[^>]*>
  \s*
    <a \s [^>]+>
     .+?
    <\/a>
  \s*
 <\/span>
 <\/div>
/xim
AUDIO_PLAYER_REGEX = 
/
        <div \s class='wrap'>
        <div \s class='audio-player'>
 <audio \s [^>]+>
 <\/audio>
        <\/div>
        <\/div>
/xim

Constants included from Utils

Utils::MONTH_EN_TO_S, Utils::PAGE_INFO_REGEX, Utils::PAGE_LAST_UPDATED_REGEX

Instance Method Summary collapse

Methods included from Utils

#data_to_csv, #encode_utf8, #find_page_info, #find_page_last_updated, #values_to_csv

Instance Method Details

#find_country_profile(html) ⇒ Object 



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
 
# File 'lib/factbook/sanitizer.rb', line 53

def find_country_profile( html )
  ####

  ## remove header (everything before)

  ##   <ul class="expandcollapse">


  pos = html.index( BEGIN_FACTS_REGEX )
  fail "*** no begin facts marker found for page"  if pos.nil?

  puts "  bingo - found BEGIN_FACTS on pos #{pos}"
  html = html[pos..-1]

  pp html[0..100]

  ###

  ## remove footer

  ##  assume everthings after (last list item in unorder list inside a table body)

  ##    </li>

  ##    </ul>

  ##    </tbody></table>


  pos = html.index( END_FACTS_REGEX )
  fail "*** no end facts marker found for page"  if pos.nil?
    
  puts "  bingo - found END_FACTS on pos #{pos}"
  html = html[0...pos] + "</li></ul>\n"        ## note: use ... (not .. to cut-off pos)


  pp html[-200..-1]
  html
end

#sanitize(html_ascii) ⇒ Object 



10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
 
# File 'lib/factbook/sanitizer.rb', line 10

def sanitize( html_ascii )
  ## todo: add option for (html source) encoding - why?? why not??

    
  ## note:

  ##   returns 1) html profile withouth headers, footers, scripts,etc.

  ##           2) page (meta) info e.g. country_name, country_code, last_updated, etc.

  ##           3) errors e.g. list of errors e.g. endcoding errors (invalid byte sequence etc.)


  page_info = PageInfo.new
    
  h = find_page_info( html_ascii )
  page_info.country_code        = h[:country_code]
  page_info.country_name        = h[:country_name]
  page_info.country_affiliation = h[:country_affiliation]
  page_info.region_code         = h[:region_code]
  page_info.region_name         = h[:region_name]

  page_info.last_updated        = find_page_last_updated( html_ascii )


  html_profile_ascii = find_country_profile( html_ascii )    ## cut-off headers, footers, scripts, etc.

  
  ## todo/fix: assume windows 12xx encoding!!!! for factbook - try 

  html, errors = encode_utf8( html_profile_ascii )  ## change encoding to utf-8  (from binary/ascii8bit)


  html = sanitize_profile( html )
  
  [html, page_info, errors]
end

#sanitize_profile(html) ⇒ Object 



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
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
 
# File 'lib/factbook/sanitizer.rb', line 138

def sanitize_profile( html )

  html = html.gsub( STYLE_ATTR_REGEX ) do |m|
          puts "remove style attr:"
          puts "#{m}"
          ''
        end

  html = html.gsub( AUDIO_PLAYER_REGEX ) do |m|
          puts "remove audio player:"
          puts "#{m}"
          ''
        end


  html = html.gsub( COUNTRY_COMPARISON_REGEX ) do |m|
          puts "remove country comparison:"
          puts "#{m}"
          ''
        end

  ## remove/cleanup anchors (a href)

  html = html.gsub( /<a\s+href[^>]*>(.+?)<\/a>/im ) do |_|   ## note: use .+? non-greedy match

    puts " replace anchor (a) href >#{$1}<"
    
    inner_text = $1.dup ## keep a copy

    if inner_text =~ /<img/    ## if includes image remove

      puts "  remove image in anchor"
      ''
    else    ## keep inner text

      inner_text
    end
  end


  ## remove all list e.g. ul/li

  html = html.gsub( /<\/?(li|ul)[^>]*>/im ) do |m|
    puts " remove list >#{m}<"
    ''
  end

  ## clean-up class attrib e.g. remove unknown classes

  html = html.gsub( CLASS_ATTR_REGEX ) do |m|
          puts "cleanup class attr:"
          puts "#{m}"
          
          klasses = $2.split(' ')
          klasses = klasses.select do |klass|
            if ['region', 'category', 'category_data'].include?( klass )
              true
            else
              puts "  remove class #{klass}"
              false
            end
          end

          if klasses.size > 0
            " class='#{klasses.join(' ')}'"   ## note: add leading space!!

          else
            ''   ## remove class attrib completely

          end
        end

   html
end