require 'nokogiri'

require 'open-uri'

require './post.rb'

class Scraper

IDDIGITS = 7

LEVELWIDTH = 40

def initialize(url)

@url = url # url of the hackernews post

# html_file is not necessary

html_file = open(@url)

@doc = Nokogiri::HTML(html_file)

end

def scrape

get_post

get_comments

print_stats

end

private

def get_post

# should these initializing steps be moved to post.rb?

# pass the doc to Post.new?

title = @doc.css('title').text

# url of the article link

url = @doc.css('tr.athing>td.title>a')[0]['href']

# take the text with the class 'score' and split at the space

# take the first part, then convert to integer

points = @doc.css('.score').text.split(" ")[0].to_i

# extracting the last 7 characters of the url only works with a web address

item_id = @url[-IDDIGITS..-1]

@post = Post.new(title, url, points, item_id)

end

def get_comments

# first 'tr.athing' is the title block

comments = @doc.css('tr.athing')

comments.shift

comments.each {|comment|

user_name = comment.css('.comhead > a:first-child').text

time = comment.css('.comhead > a:nth-child(2)').text

# content is bracketed by \n

content = comment.css('.comment').text

level = comment.css('img')[0]['width'].to_i/LEVELWIDTH

comment = Comment.new(user_name, time, content, level)

@post.add_comment(comment)

}

end

def print_stats

puts @post.info

end

end

scraper = Scraper.new(ARGV[0])

scraper.scrape