Class: Staticizer::Crawler

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

Instance Method Summary collapse

Constructor Details

#initialize(initial_page, opts = {}) ⇒ Crawler

Returns a new instance of Crawler.



9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
 
# File 'lib/staticizer/crawler.rb', line 9

def initialize(initial_page, opts = {})
  if initial_page.nil?
    raise ArgumentError, "Initial page required"
  end

  @opts = opts.dup
  @url_queue = []
  @processed_urls = []
  @opts[:output_dir] ||= File.expand_path("crawl/")
  @log = @opts[:logger] || Logger.new(STDOUT)
  @log.level = @opts[:log_level] || Logger::INFO

  if @opts[:aws]
    bucket_name = @opts[:aws].delete(:bucket_name)
    AWS.config(opts[:aws])
    @s3_bucket = AWS::S3.new.buckets[bucket_name]
    @s3_bucket.acl = :public_read
  end

  if @opts[:valid_domains].nil?
    uri = URI.parse(initial_page)
    @opts[:valid_domains] ||= [uri.host]
  end
  add_url(initial_page)
end

Instance Method Details

#add_url(url, info = {}) ⇒ Object 



99
100
101
102
103
104
105
106
107
108
109
110
111
 
# File 'lib/staticizer/crawler.rb', line 99

def add_url(url, info = {})
  if @opts[:filter_url]
    url = @opts[:filter_url].call(url, info)
    return if url.nil?
  else
    regex = "(#{@opts[:valid_domains].join(")|(")})"
    return if url !~ %r{^https?://#{regex}}
  end

  url = url.sub(/#.*$/,'') # strip off any fragments
  return if @url_queue.index {|u| u[0] == url } || @processed_urls.include?(url)
  @url_queue << [url, info]
end

#add_urls(urls, info = {}) ⇒ Object 



79
80
81
 
# File 'lib/staticizer/crawler.rb', line 79

def add_urls(urls, info = {})
  urls.compact.uniq.each {|url| add_url(url, info.dup) }
end

#crawl ⇒ Object 



35
36
37
38
39
40
41
42
43
 
# File 'lib/staticizer/crawler.rb', line 35

def crawl
  @log.info("Starting crawl")
  while(@url_queue.length > 0)
    url, info = @url_queue.shift
    @processed_urls << url
    process_url(url, info)
  end
  @log.info("Finished crawl")
end

#extract_css_urls(css, base_uri) ⇒ Object 



70
71
72
73
74
75
76
77
 
# File 'lib/staticizer/crawler.rb', line 70

def extract_css_urls(css, base_uri)
  css.scan(/url\(([^)]+)\)/).map do |src|
    path = src[0]
    # URLS in css can be wrapped with " or 'ex: url("http:://something/"), strip these
    path = path.strip.gsub(/^['"]/, "").gsub(/['"]$/,"")
    make_absolute(base_uri, path)
  end
end

#extract_hrefs(doc, base_uri) ⇒ Object 



54
55
56
 
# File 'lib/staticizer/crawler.rb', line 54

def extract_hrefs(doc, base_uri)
  doc.xpath("//a/@href").map {|href| make_absolute(base_uri, href) }
end

#extract_images(doc, base_uri) ⇒ Object 



58
59
60
 
# File 'lib/staticizer/crawler.rb', line 58

def extract_images(doc, base_uri)
  doc.xpath("//img/@src").map {|src| make_absolute(base_uri, src) }
end

#extract_links(doc, base_uri) ⇒ Object 



62
63
64
 
# File 'lib/staticizer/crawler.rb', line 62

def extract_links(doc, base_uri)
  doc.xpath("//link/@href").map {|href| make_absolute(base_uri, href) }
end

#extract_scripts(doc, base_uri) ⇒ Object 



66
67
68
 
# File 'lib/staticizer/crawler.rb', line 66

def extract_scripts(doc, base_uri)
  doc.xpath("//script/@src").map {|src| make_absolute(base_uri, src) }
end

#extract_videos(doc, base_uri) ⇒ Object 



45
46
47
48
49
50
51
52
 
# File 'lib/staticizer/crawler.rb', line 45

def extract_videos(doc, base_uri)
  doc.xpath("//video").map do |video|
    sources = video.xpath("//source/@src").map {|src| make_absolute(base_uri, src)}
    poster = video.attributes["poster"].to_s
    make_absolute(base_uri, poster)
    [poster, sources]
  end.flatten.uniq.compact
end

#make_absolute(base_uri, href) ⇒ Object 



83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
 
# File 'lib/staticizer/crawler.rb', line 83

def make_absolute(base_uri, href)
  if href.to_s =~ /^https?/i
    # If the uri is already absolute then don't do anything to it except make spaces to + (otherwise
    # will not retrieve)
    href.to_s.gsub(" ", "+")
  else
    dup_uri = base_uri.dup
    # Remove the query params as otherwise will try use those when making absolute uri
    dup_uri.query = nil
    URI::join(dup_uri.to_s, href).to_s
  end
rescue StandardError => e
  @log.error "Could not make absolute #{dup_uri} - #{href}"
  nil
end

#process_redirect(url, destination_url, opts = {}) ⇒ Object

If we hit a redirect we save the redirect as a meta refresh page TODO: for AWS S3 hosting we could instead create a redirect?



210
211
212
213
 
# File 'lib/staticizer/crawler.rb', line 210

def process_redirect(url, destination_url, opts = {})
  body = "<html><head><META http-equiv='refresh' content='0;URL=\"#{destination_url}\"'></head><body>You are being redirected to <a href='#{destination_url}'>#{destination_url}</a>.</body></html>"
  save_page(body, url, opts)
end

#process_success(response, parsed_uri) ⇒ Object 



186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
 
# File 'lib/staticizer/crawler.rb', line 186

def process_success(response, parsed_uri)
  url = parsed_uri.to_s
  case response['content-type']
  when /css/
    save_page(response, parsed_uri, no_query: true)
    add_urls(extract_css_urls(response.body, parsed_uri), {:type_hint => "css_url"})
  when /html/
    body = response.body
    save_page(body.gsub("https://www.canaan.com", ""), parsed_uri)
    doc = Nokogiri::HTML(body)
    add_urls(extract_videos(doc, parsed_uri), {:type_hint => "video"})
    add_urls(extract_links(doc, parsed_uri), {:type_hint => "link"})
    add_urls(extract_scripts(doc, parsed_uri), {:type_hint => "script"})
    add_urls(extract_images(doc, parsed_uri), {:type_hint => "image"})
    add_urls(extract_hrefs(doc, parsed_uri), {:type_hint => "href"})
    # extract inline style="background-image:url('https://')" type of urls
    add_urls(extract_css_urls(body, parsed_uri), {:type_hint => "css_url"})
  else
    save_page(response, parsed_uri, no_query: true)
  end
end

#process_url(url, info) ⇒ Object

Fetch a URI and save it to disk



216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
 
# File 'lib/staticizer/crawler.rb', line 216

def process_url(url, info)
  @http_connections ||= {}
  parsed_uri = URI(url)

  @log.debug "Fetching #{parsed_uri}"

  # Attempt to use an already open Net::HTTP connection
  key = parsed_uri.host + parsed_uri.port.to_s
  connection = @http_connections[key]
  if connection.nil?
    connection = Net::HTTP.new(parsed_uri.host, parsed_uri.port)
    connection.use_ssl = true if parsed_uri.scheme == "https"
    @http_connections[key] = connection
  end

  request = Net::HTTP::Get.new(parsed_uri.request_uri)
  connection.request(request) do |response|
    case response
    when Net::HTTPSuccess
      process_success(response, parsed_uri)
    when Net::HTTPRedirection
      redirect_url = response['location']
      @log.debug "Processing redirect to #{redirect_url}"
      process_redirect(parsed_uri, redirect_url)
      add_url(redirect_url)
    else
      @log.error "Error #{response.code}:#{response.message} fetching url #{url}"
    end
  end
end

#save_page(response, uri, opts = {}) ⇒ Object 



113
114
115
116
117
118
119
 
# File 'lib/staticizer/crawler.rb', line 113

def save_page(response, uri, opts = {})
  if @opts[:aws]
    save_page_to_aws(response, uri, opts)
  else
    save_page_to_disk(response, uri, opts)
  end
end

#save_page_to_aws(response, uri, opts = {}) ⇒ Object 



170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
 
# File 'lib/staticizer/crawler.rb', line 170

def save_page_to_aws(response, uri, opts = {})
  key = uri.path
  key += "?#{uri.query}" if uri.query
  key = key.gsub(%r{^/},"")
  key = "index.html" if key == ""
  # Upload this file directly to AWS::S3
  opts = {:acl => :public_read}
  opts[:content_type] = response['content-type'] rescue "text/html"
  @log.info "Uploading #{key} to s3 with content type #{opts[:content_type]}"
  if response.respond_to?(:read_body)
    @s3_bucket.objects[key].write(response.read_body, opts)
  else
    @s3_bucket.objects[key].write(response, opts)  
  end      
end

#save_page_to_disk(response, uri, opts = {}) ⇒ Object 



121
122
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
 
# File 'lib/staticizer/crawler.rb', line 121

def save_page_to_disk(response, uri, opts = {})
  path = uri.path
  path += "?#{uri.query}" if uri.query && !opts[:no_query] && !@opts[:no_query]

  path_segments = path.scan(%r{[^/]*/})
  filename = path.include?("/") ? path[path.rindex("/")+1..-1] : path

  current = @opts[:output_dir]
  FileUtils.mkdir_p(current) unless File.exist?(current)

  # Create all the directories necessary for this file
  path_segments.each do |segment|
    current = File.join(current, "#{segment}").sub(%r{/$},'')
    if File.file?(current)
      # If we are trying to create a directory and there already is a file
      # with the same name add a .d to the file since we can't create
      # a directory and file with the same name in the file system
      dirfile = current + ".d"
      FileUtils.mv(current, dirfile)
      FileUtils.mkdir(current)
      FileUtils.cp(dirfile, File.join(current, "/index.html"))
    elsif !File.exists?(current)
      FileUtils.mkdir(current)
    end
  end

  body = response.respond_to?(:read_body) ? response.read_body : response
  body = @opts[:process_body].call(body, uri, opts) if @opts[:process_body]
  outfile = File.join(current, "/#{filename}")

  if filename == ""
    indexfile = File.join(outfile, "/index.html")
    return if opts[:no_overwrite] && File.exists?(indexfile)
    @log.info "Saving #{indexfile}"
    File.open(indexfile, "wb") {|f| f << body }
  elsif File.directory?(outfile)
    dirfile = outfile + ".d"
    outfile = File.join(outfile, "/index.html")
    return if opts[:no_overwrite] && File.exists?(outfile)
    @log.info "Saving #{dirfile}"
    File.open(dirfile, "wb") {|f| f << body }
    FileUtils.cp(dirfile, outfile)
  else
    return if opts[:no_overwrite] && File.exists?(outfile)
    @log.info "Saving #{outfile}"
    File.open(outfile, "wb") {|f| f << body }
  end
end