diamond-mechanize 2.2 → 2.3

data/lib/mechanize/page/image.rb

@@ -0,0 +1,30 @@
+##
+# An image element on an HTML page
+
+class Mechanize::Page::Image
+  attr_reader :node
+  attr_reader :page
+
+  def initialize(node, page)
+    @node = node
+    @page = page
+  end
+
+  def src
+    @node['src']
+  end
+
+  def url
+    case src
+    when %r{^https?://}
+      src
+    else 
+      if page.bases[0]
+        (page.bases[0].href + src).to_s
+      else
+        (page.uri + src).to_s
+      end
+    end
+  end
+end
+

data/lib/mechanize/page/label.rb

@@ -0,0 +1,20 @@
+##
+# A form label on an HTML page
+
+class Mechanize::Page::Label
+  attr_reader :node
+  attr_reader :text
+  attr_reader :page
+  alias :to_s :text
+
+  def initialize(node, page)
+    @node = node
+    @text = node.inner_text
+    @page = page
+  end
+
+  def for
+    (id = @node['for']) && page.search("##{id}") || nil
+  end
+end
+

data/lib/mechanize/page/link.rb

@@ -0,0 +1,98 @@
+##
+# This class encapsulates links.  It contains the text and the URI for
+# 'a' tags parsed out of an HTML page.  If the link contains an image,
+# the alt text will be used for that image.
+#
+# For example, the text for the following links with both be 'Hello World':
+#
+#   <a href="http://example">Hello World</a>
+#   <a href="http://example"><img src="test.jpg" alt="Hello World"></a>
+
+class Mechanize::Page::Link
+  attr_reader :node
+  attr_reader :href
+  attr_reader :attributes
+  attr_reader :page
+  alias :referer :page
+
+  def initialize(node, mech, page)
+    @node       = node
+    @attributes = node
+    @href       = node['href']
+    @mech       = mech
+    @page       = page
+    @text       = nil
+    @uri        = nil
+  end
+
+  # Click on this link
+  def click
+    @mech.click self
+  end
+
+  # This method is a shorthand to get link's DOM id.
+  # Common usage:
+  #   page.link_with(:dom_id => "links_exact_id")
+  def dom_id
+    node['id']
+  end
+
+  # This method is a shorthand to get a link's DOM class
+  # Common usage:
+  #   page.link_with(:dom_class => "links_exact_class")
+  def dom_class
+    node['class']
+  end
+
+  def pretty_print(q) # :nodoc:
+    q.object_group(self) {
+      q.breakable; q.pp text
+      q.breakable; q.pp href
+    }
+  end
+
+  alias inspect pretty_inspect # :nodoc:
+
+  # A list of words in the rel attribute, all lower-cased.
+  def rel
+    @rel ||= (val = attributes['rel']) ? val.downcase.split(' ') : []
+  end
+
+  # Test if the rel attribute includes +kind+.
+  def rel? kind
+    rel.include? kind
+  end
+
+  # The text content of this link
+  def text
+    return @text if @text
+
+    @text = @node.inner_text
+
+    # If there is no text, try to find an image and use it's alt text
+    if (@text.nil? or @text.empty?) and imgs = @node.search('img') then
+      @text = imgs.map do |e|
+        e['alt']
+      end.join
+    end
+
+    @text
+  end
+
+  alias :to_s :text
+
+  # A URI for the #href for this link.  The link is first parsed as a raw
+  # link.  If that fails parsing an escaped link is attepmted.
+
+  def uri
+    @uri ||= if @href then
+               begin
+                 URI.parse @href
+               rescue URI::InvalidURIError
+                 URI.parse WEBrick::HTTPUtils.escape @href
+               end
+             end
+  end
+
+end
+

data/lib/mechanize/page/meta_refresh.rb

@@ -0,0 +1,68 @@
+##
+# This class encapsulates a meta element with a refresh http-equiv.  Mechanize
+# treats meta refresh elements just like 'a' tags.  MetaRefresh objects will
+# contain links, but most likely will have no text.
+
+class Mechanize::Page::MetaRefresh < Mechanize::Page::Link
+
+  ##
+  # Time to wait before next refresh
+
+  attr_reader :delay
+
+  ##
+  # This MetaRefresh links did not contain a url= in the content attribute and
+  # links to itself.
+
+  attr_reader :link_self
+
+  ##
+  # Matches the content attribute of a meta refresh element.  After the match:
+  #
+  #   $1:: delay
+  #   $3:: url
+
+  CONTENT_REGEXP = /^\s*(\d+\.?\d*)(;|;\s*url=\s*['"]?(\S*?)['"]?)?\s*$/i
+
+  ##
+  # Parses the delay and url from the content attribute of a meta refresh
+  # element.  Parse requires the uri of the current page to infer a url when
+  # no url is specified.
+  #
+  # Returns an array of [delay, url]. (both in string)
+  #
+  # Returns nil if the delay and url cannot be parsed.
+
+  def self.parse content, base_uri
+    return unless content =~ CONTENT_REGEXP
+
+    link_self = $3.nil? || $3.empty?
+    delay, refresh_uri = $1, $3
+
+    dest = base_uri
+    dest += refresh_uri if refresh_uri
+
+    return delay, dest, link_self
+  end
+
+  def self.from_node node, page, uri
+    http_equiv = node['http-equiv']
+    return unless http_equiv and http_equiv.downcase == 'refresh'
+
+    delay, uri, link_self = parse node['content'], uri
+
+    return unless delay
+
+    new node, page, delay, uri.to_s, link_self
+  end
+
+  def initialize node, page, delay, href, link_self = false
+    super node, page.mech, page
+
+    @delay     = delay =~ /\./ ? delay.to_f : delay.to_i
+    @href      = href
+    @link_self = link_self
+  end
+
+end
+

data/lib/mechanize/parser.rb

@@ -0,0 +1,173 @@
+##
+# The parser module provides standard methods for accessing the headers and
+# content of a response that are shared across pluggable parsers.
+
+module Mechanize::Parser
+
+  extend Forwardable
+
+  special_filenames = Regexp.union %w[
+    AUX
+    COM1
+    COM2
+    COM3
+    COM4
+    COM5
+    COM6
+    COM7
+    COM8
+    COM9
+    CON
+    LPT1
+    LPT2
+    LPT3
+    LPT4
+    LPT5
+    LPT6
+    LPT7
+    LPT8
+    LPT9
+    NUL
+    PRN
+  ]
+
+  ##
+  # Special filenames that must be escaped
+
+  SPECIAL_FILENAMES = /\A#{special_filenames}/i
+
+  ##
+  # The URI this file was retrieved from
+
+  attr_accessor :uri
+
+  ##
+  # The Mechanize::Headers for this file
+
+  attr_accessor :response
+
+  alias header response
+
+  ##
+  # The HTTP response code
+
+  attr_accessor :code
+
+  ##
+  # :method: [](header)
+  #
+  # Access HTTP +header+ by name
+
+  def_delegator :header, :[], :[]
+
+  ##
+  # :method: []=(header, value)
+  #
+  # Set HTTP +header+ to +value+
+
+  def_delegator :header, :[]=, :[]=
+
+  ##
+  # :method: key?(header)
+  #
+  # Is the named +header+ present?
+
+  def_delegator :header, :key?, :key?
+
+  ##
+  # :method: each
+  #
+  # Enumerate HTTP headers
+
+  def_delegator :header, :each, :each
+
+  ##
+  # :method: each
+  #
+  # Enumerate HTTP headers in capitalized (canonical) form
+
+  def_delegator :header, :canonical_each, :canonical_each
+
+  ##
+  # Extracts the filename from a Content-Disposition header in the #response
+  # or from the URI.  If +full_path+ is true the filename will include the
+  # host name and path to the resource, otherwise a filename in the current
+  # directory is given.
+
+  def extract_filename full_path = @full_path
+    handled = false
+
+    if @uri then
+      uri = @uri
+      uri += 'index.html' if uri.path.end_with? '/'
+
+      path     = uri.path.split(/\//)
+      filename = path.pop || 'index.html'
+    else
+      path     = []
+      filename = 'index.html'
+    end
+
+    # Set the filename
+    if disposition = @response['content-disposition'] then
+      content_disposition =
+        Mechanize::HTTP::ContentDispositionParser.parse disposition
+
+      if content_disposition then
+        filename = content_disposition.filename
+        filename = filename.split(/[\\\/]/).last
+        handled = true
+      end
+    end
+
+    if not handled and @uri then
+      filename << '.html' unless filename =~ /\./
+      filename << "?#{@uri.query}" if @uri.query
+    end
+
+    if SPECIAL_FILENAMES =~ filename then
+      filename = "_#{filename}"
+    end
+
+    filename = filename.tr "\x00-\x20<>:\"/\\|?*", '_'
+
+    @filename = if full_path then
+                  File.join @uri.host, path, filename
+                else
+                  filename
+                end
+  end
+
+  ##
+  # Creates a Mechanize::Header from the Net::HTTPResponse +response+.
+  #
+  # This allows the Net::HTTPResponse to be garbage collected sooner.
+
+  def fill_header response
+    @response = Mechanize::Headers.new
+
+    response.each { |k,v|
+      @response[k] = v
+    } if response
+
+    @response
+  end
+
+  ##
+  # Finds a free filename based on +filename+, but is not race-free
+
+  def find_free_name filename
+    filename = @filename unless filename
+
+    number = 1
+
+    while File.exist? filename do
+      filename = "#{@filename}.#{number}"
+      number += 1
+    end
+
+    filename
+  end
+
+end
+

data/lib/mechanize/pluggable_parsers.rb

@@ -0,0 +1,144 @@
+require 'mechanize/file'
+require 'mechanize/file_saver'
+require 'mechanize/page'
+
+##
+# This class is used to register and maintain pluggable parsers for Mechanize
+# to use.
+#
+# Mechanize allows different parsers for different content types.  Mechanize
+# uses PluggableParser to determine which parser to use for any content type.
+# To use your own pluggable parser or to change the default pluggable parsers,
+# register them with this class.
+#
+# The default parser for unregistered content types is Mechanize::File.
+#
+# The module Mechanize::Parser provides basic functionality for any content
+# type, so you may use it in custom parsers you write.  For small files you
+# wish to perform in-memory operations on, you should subclass
+# Mechanize::File.  For large files you should subclass Mechanize::Download as
+# the content is only loaded into memory in small chunks.
+#
+# == Example
+#
+# To create your own parser, just create a class that takes four parameters in
+# the constructor.  Here is an example of registering a pluggable parser that
+# handles CSV files:
+#
+#   require 'csv'
+#
+#   class CSVParser < Mechanize::File
+#     attr_reader :csv
+#
+#     def initialize uri = nil, response = nil, body = nil, code = nil
+#       super uri, response, body, code
+#       @csv = CSV.parse body
+#     end
+#   end
+#
+#   agent = Mechanize.new
+#   agent.pluggable_parser.csv = CSVParser
+#   agent.get('http://example.com/test.csv')  # => CSVParser
+#
+# Now any response with a content type of 'text/csv' will initialize a
+# CSVParser and return that object to the caller.
+#
+# To register a pluggable parser for a content type that pluggable parser does
+# not know about, use the hash syntax:
+#
+#   agent.pluggable_parser['text/something'] = SomeClass
+#
+# To set the default parser, use #default:
+#
+#   agent.pluggable_parser.default = Mechanize::Download
+#
+# Now all unknown content types will be saved to disk and not loaded into
+# memory.
+
+class Mechanize::PluggableParser
+
+  CONTENT_TYPES = {
+    :html  => 'text/html',
+    :wap   => 'application/vnd.wap.xhtml+xml',
+    :xhtml => 'application/xhtml+xml',
+    :pdf   => 'application/pdf',
+    :csv   => 'text/csv',
+    :xml   => 'text/xml',
+  }
+
+  attr_accessor :default
+
+  def initialize
+    @parsers = {
+      CONTENT_TYPES[:html]  => Mechanize::Page,
+      CONTENT_TYPES[:xhtml] => Mechanize::Page,
+      CONTENT_TYPES[:wap]   => Mechanize::Page,
+    }
+
+    @default = Mechanize::File
+  end
+
+  ##
+  # Returns the parser registered for the given +content_type+
+
+  def parser(content_type)
+    content_type.nil? ? default : @parsers[content_type] || default
+  end
+
+  def register_parser(content_type, klass) # :nodoc:
+    @parsers[content_type] = klass
+  end
+
+  ##
+  # Registers +klass+ as the parser for text/html and application/xhtml+xml
+  # content
+
+  def html=(klass)
+    register_parser(CONTENT_TYPES[:html], klass)
+    register_parser(CONTENT_TYPES[:xhtml], klass)
+  end
+
+  ##
+  # Registers +klass+ as the parser for application/xhtml+xml content
+
+  def xhtml=(klass)
+    register_parser(CONTENT_TYPES[:xhtml], klass)
+  end
+
+  ##
+  # Registers +klass+ as the parser for application/pdf content
+
+  def pdf=(klass)
+    register_parser(CONTENT_TYPES[:pdf], klass)
+  end
+
+  ##
+  # Registers +klass+ as the parser for text/csv content
+
+  def csv=(klass)
+    register_parser(CONTENT_TYPES[:csv], klass)
+  end
+
+  ##
+  # Registers +klass+ as the parser for text/xml content
+
+  def xml=(klass)
+    register_parser(CONTENT_TYPES[:xml], klass)
+  end
+
+  ##
+  # Retrieves the parser for +content_type+ content
+
+  def [](content_type)
+    @parsers[content_type]
+  end
+
+  ##
+  # Sets the parser for +content_type+ content to +klass+
+
+  def []=(content_type, klass)
+    @parsers[content_type] = klass
+  end
+
+end
+