class Mechanize::HTTP::Agent

An HTTP (and local disk access) user agent. This class is an implementation detail and is subject to change at any time.

Attributes

max_file_buffer[RW]

Responses larger than this will be written to a Tempfile instead of stored in memory. Setting this to nil disables creation of Tempfiles.

Public Class Methods

new() click to toggle source

Creates a new Mechanize HTTP user agent. The user agent is an implementation detail of mechanize and its API may change at any time.

# File lib/mechanize/http/agent.rb, line 119
def initialize
  @conditional_requests     = true
  @context                  = nil
  @content_encoding_hooks   = []
  @cookie_jar               = Mechanize::CookieJar.new
  @follow_meta_refresh      = false
  @follow_meta_refresh_self = false
  @gzip_enabled             = true
  @history                  = Mechanize::History.new
  @keep_alive               = true
  @max_file_buffer          = 100_000 # 5MB for response bodies
  @open_timeout             = nil
  @post_connect_hooks       = []
  @pre_connect_hooks        = []
  @read_timeout             = nil
  @redirect_ok              = true
  @redirection_limit        = 20
  @request_headers          = {}
  @robots                   = false
  @user_agent               = nil
  @webrobots                = nil

  # HTTP Authentication
  @authenticate_parser  = Mechanize::HTTP::WWWAuthenticateParser.new
  @authenticate_methods = Hash.new do |methods, uri|
    methods[uri] = Hash.new do |realms, auth_scheme|
      realms[auth_scheme] = []
    end
  end
  @digest_auth          = Net::HTTP::DigestAuth.new
  @digest_challenges    = {}
  @password             = nil # HTTP auth password
  @user                 = nil # HTTP auth user
  @domain               = nil # NTLM HTTP domain

  # SSL
  @pass = nil

  @scheme_handlers = Hash.new { |h, scheme|
    h[scheme] = lambda { |link, page|
      raise Mechanize::UnsupportedSchemeError, scheme
    }
  }

  @scheme_handlers['http']      = lambda { |link, page| link }
  @scheme_handlers['https']     = @scheme_handlers['http']
  @scheme_handlers['relative']  = @scheme_handlers['http']
  @scheme_handlers['file']      = @scheme_handlers['http']

  @http = Net::HTTP::Persistent.new 'mechanize'
  @http.idle_timeout = 5
  @http.keep_alive   = 300
end

Public Instance Methods

fetch(uri, method = :get, headers = {}) click to toggle source

Retrieves uri and parses it into a page or other object according to PluggableParser. If the URI is an HTTP or HTTPS scheme URI the given HTTP method is used to retrieve it, along with the HTTP headers, request params and HTTP referer.

redirects tracks the number of redirects experienced when retrieving the page. If it is over the #redirection_limit an error will be raised.

# File lib/mechanize/http/agent.rb, line 181
def fetch uri, method = :get, headers = {}, params = [],
          referer = current_page, redirects = 0
  referer_uri = referer ? referer.uri : nil

  uri = resolve uri, referer

  uri, params = resolve_parameters uri, method, params

  request = http_request uri, method, params

  connection = connection_for uri

  request_auth             request, uri

  disable_keep_alive       request
  enable_gzip              request

  request_language_charset request
  request_cookies          request, uri
  request_host             request, uri
  request_referer          request, uri, referer_uri
  request_user_agent       request
  request_add_headers      request, headers

  pre_connect              request

  # Consult robots.txt
  if robots && uri.is_a?(URI::HTTP)
    robots_allowed?(uri) or raise Mechanize::RobotsDisallowedError.new(uri)
  end

  # Add If-Modified-Since if page is in history
  page = visited_page(uri)

  if (page = visited_page(uri)) and page.response['Last-Modified']
    request['If-Modified-Since'] = page.response['Last-Modified']
  end if(@conditional_requests)

  # Specify timeouts if given
  connection.open_timeout = @open_timeout if @open_timeout
  connection.read_timeout = @read_timeout if @read_timeout

  request_log request

  response_body_io = nil

  # Send the request
  response = connection.request(uri, request) { |res|
    response_log res

    response_body_io = response_read res, request, uri

    res
  }

  hook_content_encoding response, uri, response_body_io

  response_body_io = response_content_encoding response, response_body_io if
    request.response_body_permitted?

  post_connect uri, response, response_body_io

  page = response_parse response, response_body_io, uri

  response_cookies response, uri, page

  meta = response_follow_meta_refresh response, uri, page, redirects
  return meta if meta

  case response
  when Net::HTTPSuccess
    if robots && page.is_a?(Mechanize::Page)
      page.parser.noindex? and raise Mechanize::RobotsDisallowedError.new(uri)
    end

    page
  when Mechanize::FileResponse
    page
  when Net::HTTPNotModified
    log.debug("Got cached page") if log
    visited_page(uri) || page
  when Net::HTTPRedirection
    response_redirect response, method, page, redirects, referer
  when Net::HTTPUnauthorized
    response_authenticate(response, page, uri, request, headers, params,
                          referer)
  else
    raise Mechanize::ResponseCodeError.new(page), "Unhandled response"
  end
end
proxy_uri() click to toggle source

URI for a proxy connection

# File lib/mechanize/http/agent.rb, line 274
def proxy_uri
  @http.proxy_uri
end
retry_change_requests() click to toggle source

Retry non-idempotent requests?

# File lib/mechanize/http/agent.rb, line 279
def retry_change_requests
  @http.retry_change_requests
end
retry_change_requests=(retri) click to toggle source

Retry non-idempotent requests

# File lib/mechanize/http/agent.rb, line 285
def retry_change_requests= retri
  @http.retry_change_requests = retri
end

HTTP Authentication

↑ top

Attributes

domain[RW]
password[RW]
user[RW]

Headers

↑ top

Attributes

conditional_requests[RW]

Disables If-Modified-Since conditional requests (enabled by default)

gzip_enabled[RW]

Is gzip compression of requests enabled?

request_headers[RW]

A hash of request headers to be used for every request

user_agent[R]

The User-Agent header to send

Public Instance Methods

user_agent=(user_agent) click to toggle source
# File lib/mechanize/http/agent.rb, line 291
def user_agent= user_agent
  @webrobots = nil if user_agent != @user_agent
  @user_agent = user_agent
end

History

↑ top

Attributes

history[RW]

history of requests made

Public Instance Methods

back() click to toggle source

Equivalent to the browser back button. Returns the most recent page visited.

# File lib/mechanize/http/agent.rb, line 300
def back
  @history.pop
end
current_page() click to toggle source

Returns the latest page loaded by the agent

# File lib/mechanize/http/agent.rb, line 307
def current_page
  @history.last
end
max_history() click to toggle source
# File lib/mechanize/http/agent.rb, line 311
def max_history
  @history.max_size
end
max_history=(length) click to toggle source
# File lib/mechanize/http/agent.rb, line 315
def max_history=(length)
  @history.max_size = length
end
visited_page(url) click to toggle source

Returns a visited page for the url passed in, otherwise nil

# File lib/mechanize/http/agent.rb, line 320
def visited_page url
  @history.visited_page resolve url
end

Hooks

↑ top

Attributes

content_encoding_hooks[R]

A list of hooks to call to handle the content-encoding of a request.

post_connect_hooks[R]

A list of hooks to call after retrieving a response. Hooks are called with the agent and the response returned.

pre_connect_hooks[R]

A list of hooks to call before making a request. Hooks are called with the agent and the request to be performed.

Public Instance Methods

hook_content_encoding(response, uri, response_body_io) click to toggle source
# File lib/mechanize/http/agent.rb, line 326
def hook_content_encoding response, uri, response_body_io
  @content_encoding_hooks.each do |hook|
    hook.call self, uri, response, response_body_io
  end
end
post_connect(uri, response, body_io) { |agent, uri, response, body| ... } click to toggle source

Invokes hooks added to #post_connect_hooks after a response is returned and the response body is handled.

Yields the context, the uri for the request, the response and the response body.

# File lib/mechanize/http/agent.rb, line 339
def post_connect uri, response, body_io # :yields: agent, uri, response, body
  @post_connect_hooks.each do |hook|
    begin
      hook.call self, uri, response, body_io.read
    ensure
      body_io.rewind
    end
  end
end
pre_connect(request) { |agent, request| ... } click to toggle source

Invokes hooks added to #pre_connect_hooks before a request is made. Yields the agent and the request that will be performed to each hook.

# File lib/mechanize/http/agent.rb, line 353
def pre_connect request # :yields: agent, request
  @pre_connect_hooks.each do |hook|
    hook.call self, request
  end
end

Redirection

↑ top

Attributes

follow_meta_refresh[RW]

Follow HTML meta refresh and HTTP Refresh. If set to :anywhere meta refresh tags outside of the head element will be followed.

follow_meta_refresh_self[RW]

Follow an HTML meta refresh that has no “url=” in the content attribute.

Defaults to false to prevent infinite refresh loops.

redirect_ok[RW]

Controls how this agent deals with redirects. The following values are allowed:

:all, true

All 3xx redirects are followed (default)

:permanent

Only 301 Moved Permanantly redirects are followed

false

No redirects are followed

redirection_limit[RW]

Maximum number of redirects to follow

Request

↑ top

Public Instance Methods

connection_for(uri) click to toggle source
# File lib/mechanize/http/agent.rb, line 361
def connection_for uri
  case uri.scheme.downcase
  when 'http', 'https' then
    return @http
  when 'file' then
    return Mechanize::FileConnection.new
  end
end
content_encoding_gunzip(body_io) click to toggle source

Decodes a gzip-encoded body_io. If it cannot be decoded, inflate is tried followed by raising an error.

# File lib/mechanize/http/agent.rb, line 374
def content_encoding_gunzip body_io
  log.debug('gzip response') if log

  zio = Zlib::GzipReader.new body_io
  out_io = auto_io 'mechanize-gunzip', 16384, zio
  zio.finish

  return out_io
rescue Zlib::Error
  log.error('unable to gunzip response, trying raw inflate') if log

  body_io.rewind
  body_io.read 10

  begin
    return inflate body_io, -Zlib::MAX_WBITS
  rescue Zlib::Error => e
    log.error("unable to gunzip response: #{e}") if log
    raise
  end
ensure
  zio.close if zio and not zio.closed?
  body_io.close unless body_io.closed?
end
content_encoding_inflate(body_io) click to toggle source

Decodes a deflate-encoded body_io. If it cannot be decoded, raw inflate is tried followed by raising an error.

# File lib/mechanize/http/agent.rb, line 403
def content_encoding_inflate body_io
  log.debug('deflate body') if log

  return inflate body_io
rescue Zlib::Error
  log.error('unable to inflate response, trying raw deflate') if log

  body_io.rewind

  begin
    return inflate body_io, -Zlib::MAX_WBITS
  rescue Zlib::Error => e
    log.error("unable to inflate response: #{e}") if log
    raise
  end
ensure
  body_io.close
end
disable_keep_alive(request) click to toggle source
# File lib/mechanize/http/agent.rb, line 422
def disable_keep_alive request
  request['connection'] = 'close' unless @keep_alive
end
enable_gzip(request) click to toggle source
# File lib/mechanize/http/agent.rb, line 426
def enable_gzip request
  request['accept-encoding'] = if @gzip_enabled
                                 'gzip,deflate,identity'
                               else
                                 'identity'
                               end
end
http_request(uri, method, params = nil) click to toggle source
# File lib/mechanize/http/agent.rb, line 434
def http_request uri, method, params = nil
  case uri.scheme.downcase
  when 'http', 'https' then
    klass = Net::HTTP.const_get(method.to_s.capitalize)

    request ||= klass.new(uri.request_uri)
    request.body = params.first if params

    request
  when 'file' then
    Mechanize::FileRequest.new uri
  end
end
request_add_headers(request, headers = {}) click to toggle source
# File lib/mechanize/http/agent.rb, line 448
def request_add_headers request, headers = {}
  @request_headers.each do |k,v|
    request[k] = v
  end

  headers.each do |field, value|
    case field
    when :etag              then request["ETag"] = value
    when :if_modified_since then request["If-Modified-Since"] = value
    when Symbol then
      raise ArgumentError, "unknown header symbol #{field}"
    else
      request[field] = value
    end
  end
end
request_auth(request, uri) click to toggle source
# File lib/mechanize/http/agent.rb, line 465
def request_auth request, uri
  base_uri = uri + '/'
  schemes = @authenticate_methods[base_uri]

  if realm = schemes[:digest].find { |r| r.uri == base_uri } then
    request_auth_digest request, uri, realm, base_uri, false
  elsif realm = schemes[:iis_digest].find { |r| r.uri == base_uri } then
    request_auth_digest request, uri, realm, base_uri, true
  elsif schemes[:basic].find { |r| r.uri == base_uri } then
    request.basic_auth @user, @password
  end
end
request_auth_digest(request, uri, realm, base_uri, iis) click to toggle source
# File lib/mechanize/http/agent.rb, line 478
def request_auth_digest request, uri, realm, base_uri, iis
  challenge = @digest_challenges[realm]

  uri.user = @user
  uri.password = @password

  auth = @digest_auth.auth_header uri, challenge.to_s, request.method, iis
  request['Authorization'] = auth
end
request_cookies(request, uri) click to toggle source
# File lib/mechanize/http/agent.rb, line 488
def request_cookies request, uri
  return if @cookie_jar.empty? uri

  cookies = @cookie_jar.cookies uri

  return if cookies.empty?

  request.add_field 'Cookie', cookies.join('; ')
end
request_host(request, uri) click to toggle source
# File lib/mechanize/http/agent.rb, line 498
def request_host request, uri
  port = [80, 443].include?(uri.port.to_i) ? nil : uri.port
  host = uri.host

  request['Host'] = [host, port].compact.join ':'
end
request_language_charset(request) click to toggle source
# File lib/mechanize/http/agent.rb, line 505
def request_language_charset request
  request['accept-charset']  = 'ISO-8859-1,utf-8;q=0.7,*;q=0.7'
  request['accept-language'] = 'en-us,en;q=0.5'
end
request_log(request) click to toggle source

Log specified headers for the request

# File lib/mechanize/http/agent.rb, line 511
def request_log request
  return unless log

  log.info("#{request.class}: #{request.path}")

  request.each_header do |k, v|
    log.debug("request-header: #{k} => #{v}")
  end
end
request_referer(request, uri, referer) click to toggle source

Sets a Referer header. Fragment part is removed as demanded by RFC 2616 14.36, and user information part is removed just like major browsers do.

# File lib/mechanize/http/agent.rb, line 524
def request_referer request, uri, referer
  return unless referer
  return if 'https' == referer.scheme.downcase and
            'https' != uri.scheme.downcase
  if referer.fragment || referer.user || referer.password
    referer = referer.dup
    referer.fragment = referer.user = referer.password = nil
  end
  request['Referer'] = referer
end
request_user_agent(request) click to toggle source
# File lib/mechanize/http/agent.rb, line 535
def request_user_agent request
  request['User-Agent'] = @user_agent if @user_agent
end
resolve(uri, referer = current_page) click to toggle source
# File lib/mechanize/http/agent.rb, line 539
def resolve(uri, referer = current_page)
  referer_uri = referer && referer.uri
  if uri.is_a?(URI)
    uri = uri.dup
  elsif uri.nil?
    if referer_uri
      return referer_uri
    end
    raise ArgumentError, "absolute URL needed (not nil)"
  else
    url = uri.to_s.strip
    if url.empty?
      if referer_uri
        return referer_uri.dup.tap { |u| u.fragment = nil }
      end
      raise ArgumentError, "absolute URL needed (not #{uri.inspect})"
    end

    url.gsub!(%r[^#{0.chr}-#{126.chr}]/) { |match|
      if RUBY_VERSION >= "1.9.0"
        Mechanize::Util.uri_escape(match)
      else
        sprintf('%%%X', match.unpack($KCODE == 'UTF8' ? 'U' : 'C')[0])
      end
    }

    escaped_url = Mechanize::Util.html_unescape(
      url.split(%r((?:%[0-9A-Fa-f]{2})+|#)/).each_slice(2).map { |x, y|
        "#{WEBrick::HTTPUtils.escape(x)}#{y}"
      }.join('')
    )

    begin
      uri = URI.parse(escaped_url)
    rescue
      uri = URI.parse(WEBrick::HTTPUtils.escape(escaped_url))
    end
  end

  scheme = uri.relative? ? 'relative' : uri.scheme.downcase
  uri = @scheme_handlers[scheme].call(uri, referer)

  if referer_uri
    if uri.path.length == 0 && uri.relative?
      uri.path = referer_uri.path
    end
  end

  uri.path = '/' if uri.path.length == 0

  if uri.relative?
    raise ArgumentError, "absolute URL needed (not #{uri})" unless
      referer_uri

    if referer.respond_to?(:bases) && referer.parser &&
        (lbase = referer.bases.last) && lbase.uri && lbase.uri.absolute?
      base = lbase
    else
      base = nil
    end

    uri = referer_uri + (base ? base.uri : referer_uri) + uri
    # Strip initial "/.." bits from the path
    uri.path.sub!(%r^(\/\.\.)+(?=\/)/, '')
  end

  unless ['http', 'https', 'file'].include?(uri.scheme.downcase)
    raise ArgumentError, "unsupported scheme: #{uri.scheme}"
  end

  uri
end
resolve_parameters(uri, method, parameters) click to toggle source
# File lib/mechanize/http/agent.rb, line 612
def resolve_parameters uri, method, parameters
  case method
  when :head, :get, :delete, :trace then
    if parameters and parameters.length > 0
      uri.query ||= ''
      uri.query << '&' if uri.query.length > 0
      uri.query << Mechanize::Util.build_query_string(parameters)
    end

    return uri, nil
  end

  return uri, parameters
end

Response

↑ top

Public Instance Methods

get_meta_refresh(response, uri, page) click to toggle source
# File lib/mechanize/http/agent.rb, line 629
def get_meta_refresh response, uri, page
  return nil unless @follow_meta_refresh

  if page.respond_to?(:meta_refresh) and
     (redirect = page.meta_refresh.first) then
    [redirect.delay, redirect.href] unless
      not @follow_meta_refresh_self and redirect.link_self
  elsif refresh = response['refresh']
    delay, href, link_self = Mechanize::Page::MetaRefresh.parse refresh, uri
    raise Mechanize::Error, 'Invalid refresh http header' unless delay
    [delay.to_f, href] unless
      not @follow_meta_refresh_self and link_self
  end
end
response_authenticate(response, page, uri, request, headers, params, referer) click to toggle source
# File lib/mechanize/http/agent.rb, line 644
def response_authenticate(response, page, uri, request, headers, params,
                          referer)
  raise Mechanize::UnauthorizedError, page unless @user || @password

  www_authenticate = response['www-authenticate']

  raise Mechanize::UnauthorizedError, page unless www_authenticate

  challenges = @authenticate_parser.parse www_authenticate

  if challenge = challenges.find { |c| c.scheme =~ %r^Digest$/ } then
    realm = challenge.realm uri

    auth_scheme = if response['server'] =~ %rMicrosoft-IIS/ then
                    :iis_digest
                  else
                    :digest
                  end

    existing_realms = @authenticate_methods[realm.uri][auth_scheme]

    raise Mechanize::UnauthorizedError, page if
      existing_realms.include? realm

    existing_realms << realm
    @digest_challenges[realm] = challenge
  elsif challenge = challenges.find { |c| c.scheme == 'NTLM' } then
    existing_realms = @authenticate_methods[uri + '/'][:ntlm]

    raise Mechanize::UnauthorizedError, page if
      existing_realms.include?(realm) and not challenge.params

    existing_realms << realm

    if challenge.params then
      type_2 = Net::NTLM::Message.decode64 challenge.params

      type_3 = type_2.response({ :user => @user, :password => @password, :domain => @domain },
                               { :ntlmv2 => true }).encode64

      headers['Authorization'] = "NTLM #{type_3}"
    else
      type_1 = Net::NTLM::Message::Type1.new.encode64
      headers['Authorization'] = "NTLM #{type_1}"
    end
  elsif challenge = challenges.find { |c| c.scheme == 'Basic' } then
    realm = challenge.realm uri

    existing_realms = @authenticate_methods[realm.uri][:basic]

    raise Mechanize::UnauthorizedError, page if
      existing_realms.include? realm

    existing_realms << realm
  else
    raise Mechanize::UnauthorizedError, page
  end

  fetch uri, request.method.downcase.to_sym, headers, params, referer
end
response_content_encoding(response, body_io) click to toggle source
# File lib/mechanize/http/agent.rb, line 705
def response_content_encoding response, body_io
  length = response.content_length ||
    case body_io
    when Tempfile, IO then
      body_io.stat.size
    else
      body_io.length
    end

  return body_io if length.zero?

  out_io = case response['Content-Encoding']
           when nil, 'none', '7bit' then
             body_io
           when 'deflate' then
             content_encoding_inflate body_io
           when 'gzip', 'x-gzip' then
             content_encoding_gunzip body_io
           else
             raise Mechanize::Error,
               "unsupported content-encoding: #{response['Content-Encoding']}"
           end

  out_io.flush
  out_io.rewind

  out_io
rescue Zlib::Error => e
  message = "error handling content-encoding #{response['Content-Encoding']}:"
  message << " #{e.message} (#{e.class})"
  raise Mechanize::Error, message
ensure
  begin
    if Tempfile === body_io and
       (StringIO === out_io or out_io.path != body_io.path) then
      body_io.close! 
    end
  rescue IOError
    # HACK ruby 1.8 raises IOError when closing the stream
  end
end
response_cookies(response, uri, page) click to toggle source
# File lib/mechanize/http/agent.rb, line 747
def response_cookies response, uri, page
  if Mechanize::Page === page and page.body =~ %rSet-Cookie/
    page.search('//head/meta[@http-equiv="Set-Cookie"]').each do |meta|
      save_cookies(uri, meta['content'])
    end
  end

  header_cookies = response.get_fields 'Set-Cookie'

  return unless header_cookies

  header_cookies.each do |set_cookie|
    save_cookies(uri, set_cookie)
  end
end
response_follow_meta_refresh(response, uri, page, redirects) click to toggle source
# File lib/mechanize/http/agent.rb, line 774
def response_follow_meta_refresh response, uri, page, redirects
  delay, new_url = get_meta_refresh(response, uri, page)
  return nil unless delay
  new_url = new_url ? resolve(new_url, page) : uri

  raise Mechanize::RedirectLimitReachedError.new(page, redirects) if
    redirects + 1 > @redirection_limit

  sleep delay
  @history.push(page, page.uri)
  fetch new_url, :get, {}, [],
        Mechanize::Page.new, redirects
end
response_log(response) click to toggle source
# File lib/mechanize/http/agent.rb, line 788
def response_log response
  return unless log

  log.info("status: #{response.class} #{response.http_version} "               "#{response.code} #{response.message}")

  response.each_header do |k, v|
    log.debug("response-header: #{k} => #{v}")
  end
end
response_parse(response, body_io, uri) click to toggle source
# File lib/mechanize/http/agent.rb, line 799
def response_parse response, body_io, uri
  @context.parse uri, response, body_io
end
response_read(response, request, uri) click to toggle source
# File lib/mechanize/http/agent.rb, line 803
def response_read response, request, uri
  content_length = response.content_length

  if use_tempfile? content_length then
    body_io = Tempfile.new 'mechanize-raw'
    body_io.unlink
    body_io.binmode if defined? body_io.binmode
  else
    body_io = StringIO.new
  end

  body_io.set_encoding Encoding::BINARY if body_io.respond_to? :set_encoding
  total = 0

  begin
    response.read_body { |part|
      total += part.length

      if StringIO === body_io and use_tempfile? total then
        new_io = Tempfile.new 'mechanize-raw'
        new_io.unlink
        new_io.binmode

        new_io.write body_io.string

        body_io = new_io
      end

      body_io.write(part)
      log.debug("Read #{part.length} bytes (#{total} total)") if log
    }
  rescue Net::HTTP::Persistent::Error => e
    body_io.rewind
    raise Mechanize::ResponseReadError.new(e, response, body_io, uri,
                                           @context)
  end

  body_io.flush
  body_io.rewind

  raise Mechanize::ResponseCodeError, response if
    Net::HTTPUnknownResponse === response

  content_length = response.content_length

  unless Net::HTTP::Head === request or Net::HTTPRedirection === response then
    raise EOFError, "Content-Length (#{content_length}) does not match "                        "response body length (#{body_io.length})" if
      content_length and content_length != body_io.length
  end

  body_io
end
response_redirect(response, method, page, redirects, referer = current_page) click to toggle source
# File lib/mechanize/http/agent.rb, line 857
def response_redirect response, method, page, redirects, referer = current_page
  case @redirect_ok
  when true, :all
    # shortcut
  when false, nil
    return page
  when :permanent
    return page unless Net::HTTPMovedPermanently === response
  end

  log.info("follow redirect to: #{response['Location']}") if log

  raise Mechanize::RedirectLimitReachedError.new(page, redirects) if
    redirects + 1 > @redirection_limit

  redirect_method = method == :head ? :head : :get

  @history.push(page, page.uri)
  new_uri = resolve response['Location'].to_s, page

  fetch new_uri, redirect_method, {}, [], referer, redirects + 1
end
save_cookies(uri, set_cookie) click to toggle source
# File lib/mechanize/http/agent.rb, line 763
def save_cookies(uri, set_cookie)
  log = log()  # reduce method calls
  Mechanize::Cookie.parse(uri, set_cookie, log) { |c|
    if @cookie_jar.add(uri, c)
      log.debug("saved cookie: #{c}") if log
    else
      log.debug("rejected cookie: #{c}") if log
    end
  }
end

Robots

↑ top

Attributes

robots[R]

When true, this agent will consult the site’s robots.txt for each access.

Public Instance Methods

robots=(value) click to toggle source
# File lib/mechanize/http/agent.rb, line 889
def robots= value
  require 'webrobots' if value
  @webrobots = nil if value != @robots
  @robots = value
end
robots_allowed?(uri) click to toggle source

Tests if this agent is allowed to access url, consulting the site’s robots.txt.

# File lib/mechanize/http/agent.rb, line 899
def robots_allowed? uri
  return true if uri.request_uri == '/robots.txt'

  webrobots.allowed? uri
end
robots_disallowed?(url) click to toggle source

Opposite of robots_allowed?

# File lib/mechanize/http/agent.rb, line 907
def robots_disallowed? url
  !robots_allowed? url
end
robots_error(url) click to toggle source

Returns an error object if there is an error in fetching or parsing robots.txt of the site url.

# File lib/mechanize/http/agent.rb, line 913
def robots_error(url)
  webrobots.error(url)
end
robots_error!(url) click to toggle source

Raises the error if there is an error in fetching or parsing robots.txt of the site url.

# File lib/mechanize/http/agent.rb, line 919
def robots_error!(url)
  webrobots.error!(url)
end
robots_reset(url) click to toggle source

Removes robots.txt cache for the site url.

# File lib/mechanize/http/agent.rb, line 924
def robots_reset(url)
  webrobots.reset(url)
end
webrobots() click to toggle source
# File lib/mechanize/http/agent.rb, line 928
def webrobots
  @webrobots ||= WebRobots.new(@user_agent, :http_get => method(:get_robots))
end

SSL

↑ top

Attributes

pass[RW]

OpenSSL key password

Public Instance Methods

ca_file() click to toggle source

Path to an OpenSSL CA certificate file

# File lib/mechanize/http/agent.rb, line 935
def ca_file
  @http.ca_file
end
ca_file=(ca_file) click to toggle source

Sets the path to an OpenSSL CA certificate file

# File lib/mechanize/http/agent.rb, line 940
def ca_file= ca_file
  @http.ca_file = ca_file
end
cert_store() click to toggle source

The SSL certificate store used for validating connections

# File lib/mechanize/http/agent.rb, line 945
def cert_store
  @http.cert_store
end
cert_store=(cert_store) click to toggle source

Sets the SSL certificate store used for validating connections

# File lib/mechanize/http/agent.rb, line 950
def cert_store= cert_store
  @http.cert_store = cert_store
end
certificate() click to toggle source

The client X509 certificate

# File lib/mechanize/http/agent.rb, line 955
def certificate
  @http.certificate
end
certificate=(certificate) click to toggle source

Sets the client certificate to given X509 certificate. If a path is given the certificate will be loaded and set.

# File lib/mechanize/http/agent.rb, line 961
def certificate= certificate
  certificate = if OpenSSL::X509::Certificate === certificate then
                  certificate
                else
                  OpenSSL::X509::Certificate.new File.read certificate
                end

  @http.certificate = certificate
end
private_key() click to toggle source

An OpenSSL private key or the path to a private key

# File lib/mechanize/http/agent.rb, line 972
def private_key
  @http.private_key
end
private_key=(private_key) click to toggle source

Sets the client’s private key

# File lib/mechanize/http/agent.rb, line 977
def private_key= private_key
  private_key = if OpenSSL::PKey::PKey === private_key then
                  private_key
                else
                  OpenSSL::PKey::RSA.new File.read(private_key), @pass
                end

  @http.private_key = private_key
end
ssl_version() click to toggle source

SSL version to use

# File lib/mechanize/http/agent.rb, line 988
def ssl_version
  @http.ssl_version
end
ssl_version=(ssl_version) click to toggle source

Sets the SSL version to use

# File lib/mechanize/http/agent.rb, line 993
def ssl_version= ssl_version
  @http.ssl_version = ssl_version
end
verify_callback() click to toggle source

A callback for additional certificate verification. See OpenSSL::SSL::SSLContext#verify_callback

The callback can be used for debugging or to ignore errors by always returning true. Specifying nil uses the default method that was valid when the SSLContext was created

# File lib/mechanize/http/agent.rb, line 1003
def verify_callback
  @http.verify_callback
end
verify_callback=(verify_callback) click to toggle source

Sets the certificate verify callback

# File lib/mechanize/http/agent.rb, line 1008
def verify_callback= verify_callback
  @http.verify_callback = verify_callback
end
verify_mode() click to toggle source

How to verify SSL connections. Defaults to VERIFY_PEER

# File lib/mechanize/http/agent.rb, line 1013
def verify_mode
  @http.verify_mode
end
verify_mode=(verify_mode) click to toggle source

Sets the mode for verifying SSL connections

# File lib/mechanize/http/agent.rb, line 1018
def verify_mode= verify_mode
  @http.verify_mode = verify_mode
end

Timeouts

↑ top

Attributes

keep_alive[RW]

Set to false to disable HTTP/1.1 keep-alive requests

open_timeout[RW]

Length of time to wait until a connection is opened in seconds

read_timeout[RW]

Length of time to attempt to read data from the server

Public Instance Methods

idle_timeout() click to toggle source

Reset connections that have not been used in this many seconds

# File lib/mechanize/http/agent.rb, line 1025
def idle_timeout
  @http.idle_timeout
end
idle_timeout=(timeout) click to toggle source

Sets the connection idle timeout for persistent connections

# File lib/mechanize/http/agent.rb, line 1030
def idle_timeout= timeout
  @http.idle_timeout = timeout
end

Utility

↑ top

Attributes

context[RW]

The context parses responses into pages

scheme_handlers[RW]

Handlers for various URI schemes

Public Instance Methods

auto_io(name, read_size, input_io) { |chunk| ... } click to toggle source

Creates a new output IO by reading input_io in read_size chunks. If the output is over the #max_file_buffer size a Tempfile with name is created.

If a block is provided, each chunk of input_io is yielded for further processing.

# File lib/mechanize/http/agent.rb, line 1044
def auto_io name, read_size, input_io
  out_io = StringIO.new

  out_io.set_encoding Encoding::BINARY if out_io.respond_to? :set_encoding

  until input_io.eof? do
    if StringIO === out_io and use_tempfile? out_io.size then
      new_io = Tempfile.new name
      new_io.unlink
      new_io.binmode

      new_io.write out_io.string
      out_io = new_io
    end

    chunk = input_io.read read_size
    chunk = yield chunk if block_given?

    out_io.write chunk
  end

  out_io.rewind

  out_io
end
inflate(compressed, window_bits = nil) click to toggle source
# File lib/mechanize/http/agent.rb, line 1070
def inflate compressed, window_bits = nil
  inflate = Zlib::Inflate.new window_bits

  out_io = auto_io 'mechanize-inflate', 1024, compressed do |chunk|
    inflate.inflate chunk
  end

  out_io.write inflate.finish

  out_io
ensure
  inflate.close
end
log() click to toggle source
# File lib/mechanize/http/agent.rb, line 1084
def log
  @context.log
end
set_proxy(addr, port, user = nil, pass = nil) click to toggle source

Sets the proxy address, port, user, and password addr should be a host, with no “http://”, port may be a port number, service name or port number string.

# File lib/mechanize/http/agent.rb, line 1093
def set_proxy addr, port, user = nil, pass = nil
  unless addr and port then
    @http.proxy = nil

    return
  end

  unless Integer === port then
    begin
      port = Socket.getservbyname port
    rescue SocketError
      begin
        port = Integer port
      rescue ArgumentError
        raise ArgumentError, "invalid value for port: #{port.inspect}"
      end
    end
  end

  proxy_uri = URI "http://#{addr}"
  proxy_uri.port = port
  proxy_uri.user     = user if user
  proxy_uri.password = pass if pass

  @http.proxy = proxy_uri
end
use_tempfile?(size) click to toggle source
# File lib/mechanize/http/agent.rb, line 1120
def use_tempfile? size
  return false unless @max_file_buffer
  return false unless size

  size >= @max_file_buffer
end