攻略サイトをねこぞぎ奪う

こんなん作った。かなり手抜きだけど

require 'net/http'
require 'socket'
require 'URL.rb'

exit if ARGV.length != 1

def mkdir(dir)
    Dir.mkdir(dir) unless File.exist?(dir)
end

def parse_html(url)
    puts url.to_s
    
    tokens = []

    begin
        Net::HTTP.start(url.domain, 80) { |http|
            str = ""
    
            response = http.get(url.site)
            response.body.split(/([<>])/m).each { |t|
                case t
                when /</
                    tokens.push(str) if str != ""
                    str = t
            
                when />/
                    tokens.push(str + t)
                    str = ""

                else
                    str += t
                end
            }
    
            tokens.push(str) if str != ""
    
            tokens
        }
    rescue
        puts "[" + url.to_s + "]取得失敗..." 
    end
end

def put_html(url, tokens)
    anchor = []
    
    open(url.filename, "w") { |f|
        tokens.each { |token|
            case token
            when /<link/i
                elements = parse_tag(token)
                
                if elements["rel"] == "stylesheet"
                    resource_url = url.make_url(elements["href"])
                    load_resource(resource_url, "css")
                    
                    elements["href"] = "./css/" + resource_url.filename
                    f.print "<link" + make_tag(elements) + ">"
                else
                    f.print token
                end
            
            when /<img/i
                elements = parse_tag(token)
                
                resource_url = url.make_url(elements["src"])
                
                # 同一domainのときだけ、リソースを取得する
                if resource_url.domain == url.domain
                    load_resource(resource_url, "img")
                
                    elements["src"] = "./img/" + resource_url.filename
                    f.print "<img" + make_tag(elements) + ">"
                end
                
            when /<a/i
                elements = parse_tag(token)
                
                if url.level < 1 && elements.include?("href")
                    anchor_url = url.make_url(elements["href"])
                    
                    if anchor_url.domain == url.domain
                        anchor.push(anchor_url)
                        elements["href"] = anchor_url.filename
                        f.print "<a" + make_tag(elements) + ">"
                    else
                        f.print token
                    end
                else
                    f.print token
                end
                
            else
                f.print token
            end
        }
    }
    
    anchor
end

def parse_tag(str)
    elements = {}
    
    str.scan(/\w+\s*=['"].+?['"]/).each { |token|
        elements[token.split("=")[0].strip.downcase] = token.split("=")[1].gsub(/['"]/, "").strip
    }

    elements
end

def make_tag(elements)
    str = ""
    elements.each { |key, value|
        str += " #{key}=\"#{value}\""
    }
    str
end

def load_resource(url, type)
    open((type.nil?? "" : "./#{type}/") + url.filename, "wb") { |f|
        Net::HTTP.start(url.domain, 80) { 
            |http| f.print http.get(url.site).body
        }
    }
end





Net::HTTP.version_1_2   # おまじない

urls = [URL.new(ARGV[0])]

mkdir("img")
mkdir("css")

urls.each { |url|
    tokens = parse_html(url)
    urls.push(put_html(url, tokens))
    urls.flatten!; urls.compact!
}

URLクラスはこんなんです

class URL
    def initialize(url, level = 0)
        @level = level
        
        url = url[("http://".length)..-1] if url.length >= "http://".length && url =~ /^http:\/\//
        
        @str = url.split(/\//, -1)
        @str.push("") if @str.length == 1
        
        @str[-1] = @str[-1].split(/\?/)[0] if @str[-1].include?("?")
    end
    
    def domain
        @str[0]
    end
    
    def site
        "/" + @str[1..-1].join("/")
    end
    
    def filename
        if @str[-1] == ""
            "index.html"
        else
            @str[-1]
        end
    end
    
    def level
        @level
    end
    
    def make_url(url)
        case url
        # 完全URL
        when /^http:\/\//
            URL.new(url, self.level + 1)
            
        # 絶対URL
        when /^\// 
            URL.new(self.domain + url, self.level + 1)
        # 相対URL
        else 
            URL.new(@str[0...-1].join("/") + "/" + url, self.level + 1)
        end
    end
    
    def to_s
        @str.join("/")
    end
end