Class: HTTPFetcher

Inherits:
Object
  • Object
show all
Defined in:
lib/feed2imap/httpfetcher.rb

Overview

Class used to retrieve the feed over HTTP

Instance Method Summary collapse

Instance Method Details

#fetch(url, lastcheck) ⇒ Object 



118
119
120
121
 
# File 'lib/feed2imap/httpfetcher.rb', line 118

def fetch(url, lastcheck)
  uri = URI::parse(url)
  return fetcher(uri, uri, lastcheck, MAXREDIR)
end

#fetcher(baseuri, uri, lastcheck, recursion) ⇒ Object 



44
45
46
47
48
49
50
51
52
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
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
 
# File 'lib/feed2imap/httpfetcher.rb', line 44

def fetcher(baseuri, uri, lastcheck, recursion)
  proxy_host = nil
  proxy_port = nil
  proxy_user = nil
  proxy_pass = nil
  if ENV['http_proxy']
      proxy_uri = URI.parse(ENV['http_proxy'])
      proxy_host = proxy_uri.host
      proxy_port = proxy_uri.port
      proxy_user, proxy_pass = proxy_uri.userinfo.split(/:/) if proxy_uri.userinfo
  end

  http = Net::HTTP::Proxy(proxy_host,
                          proxy_port,
                          proxy_user,
                          proxy_pass ).new(uri.host, uri.port)
  http.read_timeout = @timeout
  http.open_timeout = @timeout
  if uri.scheme == 'https'
    http.use_ssl = true 
    http.verify_mode = OpenSSL::SSL::VERIFY_NONE
  end
  if defined?(Feed2Imap)
    useragent = "Feed2Imap v#{Feed2Imap.version} http://home.gna.org/feed2imap/"
  else
    useragent = 'Feed2Imap http://home.gna.org/feed2imap/'
  end

  headers = {
    'User-Agent' => useragent,
    'Accept-Encoding' => 'gzip',
  }
  if lastcheck != Time::at(0)
    headers.merge!('If-Modified-Since' => lastcheck.httpdate)
  end
  req = Net::HTTP::Get::new(uri.request_uri, headers)
  if uri.userinfo
    , pw = uri.userinfo.split(':')
    req.basic_auth(, pw)
  # workaround. eg. wikini redirects and loses auth info.
  elsif uri.host == baseuri.host and baseuri.userinfo
    , pw = baseuri.userinfo.split(':')
    req.basic_auth(, pw)
  end
  begin
    response = http.request(req)
  rescue Timeout::Error
    raise "Timeout while fetching #{baseuri.to_s}"
  end
  case response
  when Net::HTTPSuccess
    case response['Content-Encoding']
    when 'gzip'
      return Zlib::GzipReader.new(StringIO.new(response.body)).read
    else
      return response.body
    end
  when Net::HTTPRedirection
    # if not modified
    if Net::HTTPNotModified === response
      puts "HTTPNotModified on #{uri}" if HTTPDEBUG
      return nil 
    end
    if recursion > 0
      redir = URI::join(uri.to_s, response['location'])
      return fetcher(baseuri, redir, lastcheck, recursion - 1)
    else
      raise "Too many redirections while fetching #{baseuri.to_s}"
    end
  else
    raise "#{response.code}: #{response.message} while fetching #{baseuri.to_s}"
  end
end

#timeout=(value) ⇒ Object 



40
41
42
 
# File 'lib/feed2imap/httpfetcher.rb', line 40

def timeout=(value)
  @timeout = value
end