Module: OutriderTools::Clean

Defined in:
lib/outrider/tools.rb

Class Method Summary collapse

Class Method Details

.file_types(sub = :all) ⇒ Object 



167
168
169
170
171
172
173
174
175
176
177
178
 
# File 'lib/outrider/tools.rb', line 167

def self.file_types sub = :all 
  case sub
  when :all 
    return %w[png jpeg jpg gif svg txt js css zip gz pdf]
  when :images
    return %w[png jpeg jpg gif svg]
  when :pdfs
    return %w[pdf]
  else
    return %w[png jpeg jpg gif svg txt js css zip gz pdf]
  end
end

.process_words_to_array(words = "") ⇒ Object

takes string of words, sorts out duds and returns array



184
185
186
187
188
 
# File 'lib/outrider/tools.rb', line 184

def self.process_words_to_array words = ""
  clean_words = words.split.each do |word|
    word.downcase!
  end
end

.tidy_urls(hrefs, page_uri, domain, files) ⇒ Object 



141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
 
# File 'lib/outrider/tools.rb', line 141

def self.tidy_urls hrefs, page_uri, domain, files 
  # Make these URIs, throwing out problem ones like mailto:
  uris = hrefs.map{ |href| URI.join( page_uri, href ) rescue nil }.compact

  # Pare it down to only those pages that are on the same site
  uris.select!{ |uri| uri.host == domain.host }

  # Throw out links to files (this could be more efficient with regex)
  uris.reject!{ |uri| files.any?{ |ext| uri.path.end_with?(".#{ext}") } }
  
  # Throw out duplicates
  uris.reject!{ |uri| ProjectData.exists?( url: uri.to_s) }

  # Remove #foo fragments so that sub-page links aren't differentiated
  uris.each{ |uri| uri.fragment = nil }

  return uris
  
end

.word_array_to_string(strings) ⇒ Object

takes array of strings and combines them



193
194
195
196
197
198
199
 
# File 'lib/outrider/tools.rb', line 193

def self.word_array_to_string strings
  the_string = ''
  strings.each do |string|
    the_string += string.gsub(/[^a-z0-9\s]/i, '')
  end
  return the_string
end