Reputation: 123
I'm trying to write a class "web" in Ruby 2.0.0 that inherits from GEXF::Graph, but I am unable to get the Graph methods like Web.define_node_attribute to work. I'm a new ruby programmer, so I expect I'm doing something goofy. Thanks.
webrun.rb
require 'rubygems'
require 'gexf'
require 'anemone'
require 'mechanize'
require_relative 'web'
web = Web.new
web.define_node_attribute(:url)
web.define_node_attribute(:links,
:type => GEXF::Attribute::BOOLEAN,
:default => true)
web.rb
require 'rubygems'
require 'gexf'
require 'anemone'
require 'mechanize'
class Web < GEXF::Graph
attr_accessor :root
attr_accessor :pages
def initialize
@pages = Array.new
end
def pages
@pages
end
def add page
@pages << page
end
def parse uri, protocol = 'http:', domain = 'localhost', file = 'index.html'
u = uri.split('/')
if n = /^(https?:)/.match(u[0])
protocol = n[0]
u.shift()
end
if u[0] == ''
u.shift()
end
if n = /([\w\.]+\.(org|com|net))/.match(u[0])
domain = n[0]
u.shift()
end
if n = /(.*\.(html?|gif))/.match(u[-1])
file = n[0]
u.pop()
end
cnt = 0
while u[cnt] == '..' do
cnt = cnt + 1
u.shift()
end
while cnt > 0 do
cnt = cnt - 1
u.shift()
end
directory = '/'+u.join('/')
puts "protocol: " + protocol + " domain: " + domain + \
" directory: " + directory + " file: " + file
protocol + "//" + domain + directory + (directory[-1] == '/' ? '/' : '') + file
end
def crawl
Anemone.crawl(@root) do |anemone|
anemone.on_every_page do |sitepage|
add sitepage
end
end
end
def save file
f = File.open(file, mode = "w")
f.write(to_xml)
f.close()
end
end
Upvotes: 1
Views: 62
Reputation: 1375
The issue is that you are monkey-patching the GEXF::Graph
initialize method without calling super on it. What you did was essentially 'write-over' the initialize method that needed to be called. To fix this, change your initialize method to call the super method first:
def initialize
super
@pages = Array.new
end
Upvotes: 1