postmodern / spidr

A versatile Ruby web spidering library that can spider a site, multiple domains, certain links or infinitely. Spidr is designed to be fast and easy to use.
MIT License
806 stars 109 forks source link
crawler ruby scraper spider spider-links web web-crawler web-scraper web-scraping web-spider

Spidr

CI

Description

Spidr is a versatile Ruby web spidering library that can spider a site, multiple domains, certain links or infinitely. Spidr is designed to be fast and easy to use.

Features

Examples

Start spidering from a URL:

Spidr.start_at('http://tenderlovemaking.com/') do |agent|
  # ...
end

Spider a host:

Spidr.host('solnic.eu') do |agent|
  # ...
end

Spider a domain (and any sub-domains):

Spidr.domain('ruby-lang.org') do |agent|
  # ...
end

Spider a site:

Spidr.site('http://www.rubyflow.com/') do |agent|
  # ...
end

Spider multiple hosts:

Spidr.start_at('http://company.com/', hosts: ['company.com', /host[\d]+\.company\.com/]) do |agent|
  # ...
end

Do not spider certain links:

Spidr.site('http://company.com/', ignore_links: [%{^/blog/}]) do |agent|
  # ...
end

Do not spider links on certain ports:

Spidr.site('http://company.com/', ignore_ports: [8000, 8010, 8080]) do |agent|
  # ...
end

Do not spider links blacklisted in robots.txt:

Spidr.site('http://company.com/', robots: true) do |agent|
  # ...
end

Print out visited URLs:

Spidr.site('http://www.rubyinside.com/') do |spider|
  spider.every_url { |url| puts url }
end

Build a URL map of a site:

url_map = Hash.new { |hash,key| hash[key] = [] }

Spidr.site('http://intranet.com/') do |spider|
  spider.every_link do |origin,dest|
    url_map[dest] << origin
  end
end

Print out the URLs that could not be requested:

Spidr.site('http://company.com/') do |spider|
  spider.every_failed_url { |url| puts url }
end

Finds all pages which have broken links:

url_map = Hash.new { |hash,key| hash[key] = [] }

spider = Spidr.site('http://intranet.com/') do |spider|
  spider.every_link do |origin,dest|
    url_map[dest] << origin
  end
end

spider.failures.each do |url|
  puts "Broken link #{url} found in:"

  url_map[url].each { |page| puts "  #{page}" }
end

Search HTML and XML pages:

Spidr.site('http://company.com/') do |spider|
  spider.every_page do |page|
    puts ">>> #{page.url}"

    page.search('//meta').each do |meta|
      name = (meta.attributes['name'] || meta.attributes['http-equiv'])
      value = meta.attributes['content']

      puts "  #{name} = #{value}"
    end
  end
end

Print out the titles from every page:

Spidr.site('https://www.ruby-lang.org/') do |spider|
  spider.every_html_page do |page|
    puts page.title
  end
end

Print out every HTTP redirect:

Spidr.host('company.com') do |spider|
  spider.every_redirect_page do |page|
    puts "#{page.url} -> #{page.headers['Location']}"
  end
end

Find what kinds of web servers a host is using, by accessing the headers:

servers = Set[]

Spidr.host('company.com') do |spider|
  spider.all_headers do |headers|
    servers << headers['server']
  end
end

Pause the spider on a forbidden page:

Spidr.host('company.com') do |spider|
  spider.every_forbidden_page do |page|
    spider.pause!
  end
end

Skip the processing of a page:

Spidr.host('company.com') do |spider|
  spider.every_missing_page do |page|
    spider.skip_page!
  end
end

Skip the processing of links:

Spidr.host('company.com') do |spider|
  spider.every_url do |url|
    if url.path.split('/').find { |dir| dir.to_i > 1000 }
      spider.skip_link!
    end
  end
end

Requirements

Install

$ gem install spidr

License

See {file:LICENSE.txt} for license information.