huyha85/opengraph_parser

Simple Ruby Parser library for parsing OpenGraph protocol (http://ogp.me)

OpengraphParser

OpengraphParser is a simple Ruby library for parsing Open Graph protocol information from a website. Learn more about the protocol at: http://ogp.me

Installation

  gem install opengraph_parser

or add to Gemfile

  gem "opengraph_parser"

Usage

Parsing an URL

og = OpenGraph.new("http://ogp.me")
og.title # => "Open Graph protocol"
og.type # => "website"
og.url # => "http://ogp.me/"
og.description # => "The Open Graph protocol enables any web page to become a rich object in a social graph."
og.images # => ["http://ogp.me/logo.png"]

You can also get other Open Graph metadata as:

og.metadata # => {"og:image:type"=>"image/png", "og:image:width"=>"300", "og:image:height"=>"300"}

Parsing a HTML document

og = OpenGraph.new(html_string)

Custom header fields

In some cases you may need to change fields in HTTP request header for an URL

og = OpenGraph.new("http://opg.me", { :headers => {'User-Agent' => 'Custom User Agent'} })

Fallback

If you try to parse Open Graph information for a website that doesn’t have any Open Graph metadata, the library will try to find other information in the website as the following rules:

• <title> for title
• <meta name="description"> for description
• <link rel="image_src"> or all <img> tags for images

You can disable this fallback lookup by passing false to init method:

og = OpenGraph.new("http://ogp.me", false)

Contributing to opengraph_parser

• Check out the latest master to make sure the feature hasn't been implemented or the bug hasn't been fixed yet.
• Check out the issue tracker to make sure someone already hasn't requested it and/or contributed it.
• Fork the project.
• Start a feature/bugfix branch.
• Commit and push until you are happy with your contribution.
• Make sure to add tests for it. This is important so I don't break it in a future version unintentionally.
• Please try not to mess with the Rakefile, version, or history. If you want to have your own version, or is otherwise necessary, that is fine, but please isolate to its own commit so I can cherry-pick around it.

Copyright

Copyright (c) 2013 Huy Ha. See LICENSE.txt for further details.

opengraph_parser's Issues

Is this gem unmaintained?

Hi there!

I have a question to @huyha85. Do you continue to maintain this gem? I want to push some changes to this repository. Will you accept PR and push the new version to Rubygems?

Reads streams forever

Pointing opengraph_parser to a big file or stream will make it read that until terminates, causing high memory usage and eventually a DoS if user input is passed to it.

Obvious solution would adding a Timeout.timeout, but since that's inherently thread unsafe, please don't do it.

A viable solution I see would be making a head request and looking at the Content-Type header. If it's not text/html don't continue. Other heuristics could be applied there, for example checking Content-Length.

2d828be completely broke http

Introduced in 2d828be, lib/redirect_follower.rb#28

Unfortunately, get_response doesn't work that way: ruby-doc.org. Passing a second argument (headers hash, empty or not) causes it to use the first one as hostname, which doesn't work.

Minimal example: try RedirectFollower.new('http://google.com').resolve with and without https.

Request to ssl encrypted pages are not working.

I tried your gem and it workes quite well. But it would be nice if you could also support ssl encrypted urls, because they are wide spreaded.

I would suggest the following change in line 15:

require 'net/https'

uri = URI.parse(URI.escape(url))
if uri.scheme == 'https'
  https = Net::HTTP.new(uri.host, 443)
  https.use_ssl = true
  https.verify_mode = OpenSSL::SSL::VERIFY_PEER
  self.response = https.request_get(uri.request_uri)
else
  self.response = Net::HTTP.get_response(uri)
end

Would be great seeing this integrated, currently can't find time to build a pull request.

How can I get metatags by property value?

Hi, not an issue, just a question :)

require 'opengraph_parser'
fb = OpenGraph.new 'https://www.facebook.com/smashmag/')
puts fb.metadata[property="al:ios:url"]# => nothing

require 'opengraph_parser'

fb = OpenGraph.new('https://www.facebook.com/smashmag/')

puts fb.metadata# => {:title=>[{:_value=>"Smashing Magazine"}], :description=>[{:_value=>"Smashing Magazine. 291 378 Me gusta. Smashing Magazine delivers useful and innovative information to Web designers and developers."}], :image=>[{:_value=>"https://scontent-mad1-1.xx.fbcdn.net/v/t1.0-1/p200x200/17352146_10155179903597490_6515514170610847005_n.png?oh=a76c1ec8e8c526aa557545968c5a03ea&oe=598C27D2"}], :url=>[{:_value=>"https://www.facebook.com/smashmag/"}]}

The info I am trying to get is <meta property="al:ios:url" content="fb://page/?id=45576747489" />
Is that possible?

Raise error on 404

Currently it does nothing.

Could also respond to body?.

Handle URLs with anchors

Anchor tags are a valid part of an URI but shouldn't be included in the request.

url = "http://www.cnet.com/news/pi-top-the-3d-printable-raspberry-pi-laptop-anyone-can-build/#ftag=CAD590a51e"

require 'uri'
require 'open-uri'
require 'nokogiri'

p Nokogiri::HTML(open URI.parse(url), &:read).css('title').text #=> "Pi-Top: The 3D-printable Raspberry Pi laptop anyone can build - CNET"

require 'opengraph_parser'
p OpenGraph.new(url).title #=> "Page Not Found (404) - CNET"

include spec directory in gem file itself

most gems include the tests. It would help us package the gem directly in debian. Now we take the github snapshots from tags. Can you please include the spec directory in the gem itself?