Class: SiteDiff::Crawler

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

Overview

SiteDiff Crawler.

Defined Under Namespace

Classes: Info

Constant Summary collapse

DEFAULT_DEPTH = 
3

Instance Method Summary collapse

Constructor Details

#initialize(hydra, base, interval, include_regex, exclude_regex, depth = DEFAULT_DEPTH, curl_opts = UriWrapper::DEFAULT_CURL_OPTS, debug: true, &block) ⇒ Crawler

Create a crawler with a base URL



18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
 
# File 'lib/sitediff/crawler.rb', line 18

def initialize(hydra, base,
               interval,
               include_regex,
               exclude_regex,
               depth = DEFAULT_DEPTH,
               curl_opts = UriWrapper::DEFAULT_CURL_OPTS,
               debug: true,
               &block)
  @hydra = hydra
  @base_uri = Addressable::URI.parse(base)
  @base = base
  @interval = interval
  @include_regex = include_regex
  @exclude_regex = exclude_regex
  @found = Set.new
  @callback = block
  @curl_opts = curl_opts
  @debug = debug

  add_uri('', depth, referrer: '/')
end

Instance Method Details

#add_uri(rel, depth, referrer = '') ⇒ Object

Handle a newly found relative URI



41
42
43
44
45
46
47
48
49
50
 
# File 'lib/sitediff/crawler.rb', line 41

def add_uri(rel, depth, referrer = '')
  return if @found.include? rel

  @found << rel

  wrapper = UriWrapper.new(@base + rel, @curl_opts, debug: @debug, referrer:)
  wrapper.queue(@hydra) do |res|
    fetched_uri(rel, depth, res)
  end
end

#fetched_uri(rel, depth, res) ⇒ Object

Handle the fetch of a URI



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
89
90
91
92
93
94
95
 
# File 'lib/sitediff/crawler.rb', line 53

def fetched_uri(rel, depth, res)
  if res.error
    SiteDiff.log(res.error, :error)
    return
  elsif !res.content
    SiteDiff.log('Response is missing content. Treating as an error.', :error)
    return
  end

  base = Addressable::URI.parse(@base + rel)
  doc = Nokogiri::HTML(res.content)

  # Call the callback
  info = Info.new(
    relative: rel,
    uri: base,
    read_result: res,
    document: doc
  )
  # Insert delay to limit fetching rate
  if @interval != 0
    SiteDiff.log("Waiting #{@interval} milliseconds.", :info)
    sleep(@interval / 1000.0)
  end
  @callback[info]

  return unless depth >= 1

  # Find links
  links = find_links(doc)
  uris = links.map { |l| resolve_link(base, l) }.compact
  uris = filter_links(uris)

  # Make them relative
  rels = uris.map { |u| relativize_link(u) }

  # Queue them in turn
  rels.each do |r|
    next if @found.include? r

    add_uri(r, depth - 1, rel)
  end
end

#filter_links(uris) ⇒ Object

Filter out links we don’t want. Links passed in are absolute URIs.



127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
 
# File 'lib/sitediff/crawler.rb', line 127

def filter_links(uris)
  uris.find_all do |u|
    is_sub_uri = (u.host == @base_uri.host) &&
                 u.path.start_with?(@base_uri.path)
    next unless is_sub_uri

    # puts "Trying regex #{u.path}"
    is_included = @include_regex.nil? ? false : @include_regex.match(u.path)
    is_excluded = @exclude_regex.nil? ? false : @exclude_regex.match(u.path)
    if is_excluded && !is_included
      SiteDiff.log "Ignoring excluded URL #{u.path}", :info
    end
    is_included || !is_excluded
  end
end

#find_links(doc) ⇒ Object

Return a list of string links found on a page.



122
123
124
 
# File 'lib/sitediff/crawler.rb', line 122

def find_links(doc)
  doc.xpath('//a[@href]').map { |e| e['href'] }
end

#relativize_link(uri) ⇒ Object

Make a link relative to @base_uri



107
108
109
110
111
112
113
114
115
116
117
118
119
 
# File 'lib/sitediff/crawler.rb', line 107

def relativize_link(uri)
  # fullPath = uri.path
  # if uri.query
  #   fullPath += "?" + uri.query
  # end
  #
  # if uri.fragment
  #   fullPath += "#" + uri.fragment
  # end
  # fullPath.gsub(@base_uri.path, "")
  #
  uri.path.slice(@base_uri.path.length, uri.path.length)
end

#resolve_link(base, rel) ⇒ Object

Resolve a potentially-relative link. Return nil on error.



98
99
100
101
102
103
104
 
# File 'lib/sitediff/crawler.rb', line 98

def resolve_link(base, rel)
  rel = rel.strip
  base + rel
rescue Addressable::URI::InvalidURIError
  SiteDiff.log "skipped invalid URL: '#{rel}' (at #{base})", :warning
  nil
end