Spidr

• Homepage
• Source
• Issues
• Mailing List

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

• Follows:
  • a tags.
  • iframe tags.
  • frame tags.
  • Cookie protected links.
  • HTTP 300, 301, 302, 303 and 307 Redirects.
  • Meta-Refresh Redirects.
  • HTTP Basic Auth protected links.
• Black-list or white-list URLs based upon:
  • URL scheme.
  • Host name
  • Port number
  • Full link
  • URL extension
  • Optional /robots.txt support.
• Provides callbacks for:
  • Every visited Page.
  • Every visited URL.
  • Every visited URL that matches a specified pattern.
  • Every origin and destination URI of a link.
  • Every URL that failed to be visited.
• Provides action methods to:
  • Pause spidering.
  • Skip processing of pages.
  • Skip processing of links.
• Restore the spidering queue and history from a previous session.
• Custom User-Agent strings.
• Custom proxy settings.
• HTTPS support.

Examples

Start spidering from a URL:

Spidr.start_at('https://www.ruby-lang.org/en/') do |agent|
  # ...
end

Spider a host:

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

Spider a domain (and any sub-domains):

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

Spider a site:

Spidr.site('https://www.ruby-lang.org/') do |agent|
  # ...
end

Spider multiple hosts:

Spidr.start_at('https://www.ruby-lang.org/en/', hosts: ['ruby-lang.org', /.*\.ruby-lang\.org\z/]) do |agent|
  # ...
end

Do not spider certain links:

Spidr.site('https://www.ruby-lang.org/', ignore_links: [%r{\A/blog/}]) do |agent|
  # ...
end

Do not spider links on certain ports:

Spidr.site('https://www.ruby-lang.org/', ignore_ports: [8000, 8010, 8080]) do |agent|
  # ...
end

Do not spider links blacklisted in robots.txt:

Spidr.site('https://www.ruby-lang.org/', robots: true) do |agent|
  # ...
end

Print out visited URLs:

Spidr.site('https://www.ruby-lang.org/') 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('https://www.ruby-lang.org/') 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('https://www.ruby-lang.org/') 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('https://www.ruby-lang.org/') 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('https://www.ruby-lang.org/') 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('www.ruby-lang.org') 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('www.ruby-lang.org') do |spider|
  spider.all_headers do |headers|
    servers << headers['server']
  end
end

Pause the spider on a forbidden page:

Spidr.host('www.ruby-lang.org') do |spider|
  spider.every_forbidden_page do |page|
    spider.pause!
  end
end

Skip the processing of a page:

Spidr.host('www.ruby-lang.org') do |spider|
  spider.every_missing_page do |page|
    spider.skip_page!
  end
end

Skip the processing of links:

Spidr.host('www.ruby-lang.org') do |spider|
  spider.every_url do |url|
    if url.path.split('/').find { |dir| dir.to_i > 1000 }
      spider.skip_link!
    end
  end
end

Requirements

• ruby >= 2.0.0
• nokogiri ~> 1.3

Install

$ gem install spidr

License

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