link_thumbnailer 2.5.2 → 2.6.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 49bbee58589eb5ba7a8e91f3747116510331160e
-  data.tar.gz: 8dde81b01a30960aa69febd8fe42f3bcb824b6a9
+  metadata.gz: 70bf3449f58e41e93e1fca01e64835b3fe150d91
+  data.tar.gz: 8db948511fb227d94e6afbf6721222539edd13a3
 SHA512:
-  metadata.gz: 2031071a1077e7ad8915bfa11183e75f23a6e4e7d6a7700f661e599fa164fb2424dfa6fabd36576ca184ed97028199175d9087978514751528a68a93bc8a4c0e
-  data.tar.gz: 4806f8bb2bdd828739e1359bcf0bf784be459f69c7d13f1ec14efbff4738cf5d770447d6f0b181f8119c548266a0c98a756936ff19e651949e5e33f6d17add01
+  metadata.gz: 63ab6ab4ed55873cb612257507089af4fbd8e6ed3e941e46cfb92057fa59f31a92ed3f69a9b880e189ff4296e967964dcf40f46ffc36bb4fd650517de74079e4
+  data.tar.gz: 28ebc45714d8695ba0b2d912858aeba54bf4fb5326318b65855d3bbc7f060c4cb004196c89224b4ce6f407dec4e9adfbb4b69d4af1413f43cb6742afad290dac

data/CHANGELOG.md

@@ -1,3 +1,7 @@
+# 2.6.0
+
+- Introduce new `raise_on_invalid_format` option (false by default) to raise `LinkThumbnailer::FormatNotSupported` if http `Content-Type` is invalid. Fixes #61 and #64.
+
 # 2.5.2
 
 - Fix OpenURI::HTTPError exception raised when video_info gem is not able to parse video metadata. Fixes #60.

data/lib/link_thumbnailer/configuration.rb

@@ -25,7 +25,7 @@ module LinkThumbnailer
     attr_accessor :redirect_limit, :blacklist_urls, :user_agent,
                   :verify_ssl, :http_open_timeout, :http_read_timeout, :attributes,
                   :graders, :description_min_length, :positive_regex, :negative_regex,
-                  :image_limit, :image_stats
+                  :image_limit, :image_stats, :raise_on_invalid_format
 
     alias_method :http_timeout, :http_open_timeout
     alias_method :http_timeout=, :http_open_timeout=
@@ -58,6 +58,7 @@ module LinkThumbnailer
       @negative_regex = /combx|comment|com-|contact|foot|footer|footnote|masthead|media|meta|outbrain|promo|related|scroll|shoutbox|sidebar|sponsor|shopping|tags|tool|widget|modal/i
       @image_limit    = 5
       @image_stats    = true
+      @raise_on_invalid_format = false
     end
 
 	end

data/lib/link_thumbnailer/exceptions.rb

@@ -2,5 +2,6 @@ module LinkThumbnailer
   class Exceptions      < StandardError; end
   class RedirectLimit   < Exceptions; end
   class BadUriFormat    < Exceptions; end
+  class FormatNotSupported < Exceptions; end
   class ScraperInvalid  < Exceptions; end
 end

data/lib/link_thumbnailer/processor.rb

@@ -1,3 +1,4 @@
+require 'pry'
 require 'delegate'
 require 'uri'
 require 'net/http/persistent'
@@ -53,8 +54,11 @@ module LinkThumbnailer
       headers           = {}
       headers['Cookie'] = response['Set-Cookie'] if response['Set-Cookie'].present?
 
+      raise ::LinkThumbnailer::FormatNotSupported.new(response['Content-Type']) unless valid_response_format?(response)
+
       case response
-      when ::Net::HTTPSuccess then response.body
+      when ::Net::HTTPSuccess
+        response.body
       when ::Net::HTTPRedirection
         call(
           resolve_relative_url(response['location']),
@@ -102,6 +106,17 @@ module LinkThumbnailer
       url.is_a?(::URI::HTTP)
     end
 
+    def valid_response_format?(response)
+      return true unless config.raise_on_invalid_format
+      return true if response['Content-Type'] =~ /text\/html/
+      return true if response['Content-Type'] =~ /application\/html/
+      return true if response['Content-Type'] =~ /application\/xhtml\+xml/
+      return true if response['Content-Type'] =~ /application\/xml/
+      return true if response['Content-Type'] =~ /text\/xml/
+      return true if response['Content-Type'] =~ /text\/plain/
+      false
+    end
+
     def url=(url)
       @url = ::URI.parse(url.to_s)
     end

data/lib/link_thumbnailer/version.rb

@@ -1,3 +1,3 @@
 module LinkThumbnailer
-  VERSION = '2.5.2'
+  VERSION = '2.6.0'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: link_thumbnailer
 version: !ruby/object:Gem::Version
-  version: 2.5.2
+  version: 2.6.0
 platform: ruby
 authors:
 - Pierre-Louis Gottfrois
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-06-04 00:00:00.000000000 Z
+date: 2015-06-21 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activesupport