spix_parser 1.7.3 → 1.7.5

data/lib/spix_parser/parser.rb

@@ -32,10 +32,10 @@ module Spix
     extend self
 
     def format_links(options)
-      text     = Sanitizer.html_decode options[:text]
-      site_url = Sanitizer.html_decode options[:site_url]
+      text     = options[:text]
+      site_url = options[:site_url]
 
-      parse_links(text)
+      parse_links(text, site_url)
       parse_images(text, site_url)
 
       text
@@ -45,66 +45,75 @@ module Spix
 
     def join_attributes(attrs)
       attrs.map do |attr, value|
-        %Q[#{attr}="#{value.to_s.gsub(/"/, """)}"] unless value.blank?
-      end.compact.join(" ")
+        %Q[#{attr}="#{value.to_s}"] unless value.blank?
+      end.compact.join(" ").gsub(/"/, """)
     end
 
-    def parse_attrs(str)
-      attrs = {}
-      return attrs unless str || str.respond_to?(:scan)
+    def parse_attrs(str, options = {})
+      attrs_to_add = options.delete(:adding) 
+      allowed = options.delete(:allowed_attrs)
 
-      match_by_spaces = str !~ /'|"/
-      if match_by_spaces
-        # Make sure to match the last html attribute.
+      if match_by_spaces = str =~ /'|"|"|"/
+        value_regexp = %r{\s*#{$&}(.*?)#{$&}}
+        encloser = $&.gsub(/'/, '"')
+      else
+        encloser = "\"" 
         str += " "
         value_regexp = /\s*(.*?)\s/
-      else
-        value_regexp = /\s*["'](.*?)["']/
       end
       attribute_regexp = /\b([a-zA-Z0-9:]+)\s*/
 
-      str.scan(/#{attribute_regexp}=#{value_regexp}/im) do
-        attrs[$1.to_s.downcase] = $2
-      end
+      finded_attrs = {} 
+      result = str.gsub(/#{attribute_regexp}=#{value_regexp}/im) { 
+        attribute, value = $1.downcase, $2
+
+        next if value.blank? or not allowed.include?(attribute.to_sym)
 
-      attrs
+        parsed_value = yield(attribute, value) || value 
+
+        finded_attrs[attribute.to_sym] = parsed_value
+
+        %{#{attribute}=#{enclose parsed_value, encloser} }
+
+      } + " " + (
+        attrs_to_add.to_a - finded_attrs.to_a
+      ).map {|k,v| "#{k}=#{enclose v, encloser}" }.join(" ")
+      return [result.split.join(" "), finded_attrs.keys]
     end
 
-    def parse_links(text)
-      text.gsub!(/(<a\s+([^>]+)>)/uim) do |match|
-        attrs = parse_attrs($2.to_s)
+    def enclose(str, encloser)
+      "#{encloser}#{str}#{encloser}"
+    end
 
-        # just parse these attributes
+    def parse_links(text, site_url)
+      allowed = [:href, :title, :target, :rel]
+      text.gsub!(/<a\s+([^>]+)>/uim) do |match|
         attrs = {
-          :href   => attrs["href"],
-          :title  => attrs["title"],
           :target => "_blank",
           :rel    => "external nofollow"
         }
 
-        "<a #{join_attributes(attrs)}>"
+        parsed_attribute, attrs = parse_attrs($1.to_s, :adding => attrs, :allowed_attrs => allowed ) do |attr, value|
+          parse_relative_source(value, site_url) if attr == "href"
+        end
+
+        %{<a #{parsed_attribute}>}
       end
     end
 
     def parse_images(text, site_url)
-      text.gsub!(/(<img(.*?)\/?>)/uim) do |match|
-        attrs = parse_attrs($2.to_s)
-
-        # just parse these attributes
-        attrs = {
-          :src    => parse_relative_image_source(attrs["src"], site_url),
-          :alt    => attrs["alt"],          
-          :title  => attrs["title"],
-          :style  => attrs["style"],
-          :width  => attrs["width"],          
-          :height => attrs["height"]
-        }
-
-        "<img #{join_attributes(attrs)} />" if attrs[:src].present?
+      allowed = [:src, :alt, :title, :style, :width, :height]
+      text.gsub!(/<img(.*?)\/?>/uim) do |match|
+        parsed_attribute, attrs = parse_attrs($1.to_s, :allowed_attrs => allowed) do |attr, value|
+          parse_relative_source(value, site_url) if attr == "src"
+        end
+        %{<img #{parsed_attribute} />} if attrs.include?(:src)
       end
     end
 
-    def parse_relative_image_source(src, site_url)
+    def parse_relative_source(src, site_url)
+      src = Sanitizer.html_decode(src) if src
+      site_url = Sanitizer.html_decode(site_url) if site_url
       if src.present? && site_url
         begin
           src = URI.parse(src)

data/lib/spix_parser/version.rb

@@ -4,7 +4,7 @@ module Spix
     module Version
       MAJOR = 1
       MINOR = 7
-      TINY  = 3
+      TINY  = 5
 
       def self.current_version
         "#{MAJOR}.#{MINOR}.#{TINY}"

data/lib/spix_parser/wrappers/entry.rb

@@ -47,8 +47,8 @@ module Spix
 
       def uid
         uid = self.url || ""
-        uid += self.encoded_raw_content_for :title
-        uid += self.encoded_raw_content[0..25]
+        uid += encoded_raw_content_for :title
+        uid += encoded_raw_content[0..25]
         uid.to_sha1
       end
       memoize(:uid)

data/spec/spix_parser/parser_spec.rb

@@ -30,7 +30,7 @@ describe Spix::Parser do
 
     it "should parse correctly images with absolute with another domain" do
       feed = Spix::Parser.parse(load_fixture('feed_with_absolute_images_from_another_domain.atom'), :mode => :local)
-      feed.feed_items.first.content[/<img.*src=["'](.*?)["'].*\/>/, 1].should == "http://oglobo.globo.com/fotos/2011/07/06/06_MHB_ballmer.jpg"   
+      feed.feed_items.first.content[/<img.*src=&#034;(.*?)&#034;.*\/>/, 1].should == "http://oglobo.globo.com/fotos/2011/07/06/06_MHB_ballmer.jpg"   
     end
   end
   

data/spec/spix_parser/utils_spec.rb

@@ -4,6 +4,21 @@ require 'spec_helper'
 describe Spix::Utils do
   describe ".format_links" do
     context "html containing links" do
+
+      it "parses link tags with html escaped quote (") and absolute sources" do
+        input_html = %q[<div><a href=&quot;/foo/bar.html&quot; title=&quot;FooBar!&quot;>FooBar!</a></div>]
+
+        Spix::Utils.format_links(:text => input_html, :site_url => "http://busk.com/").should ==
+          %q[<div><a href=&quot;http://busk.com/foo/bar.html&quot; title=&quot;FooBar!&quot; target=&quot;_blank&quot; rel=&quot;external nofollow&quot;>FooBar!</a></div>]
+      end
+
+      it "parses link tags with absolute sources" do
+          input_html = %q[<div><a href="/foo/bar.html" title="FooBar!" target="_blank" rel="external nofollow">FooBar!</a></div>]
+
+        Spix::Utils.format_links(:text => input_html, :site_url => "http://busk.com/").should ==
+           %q[<div><a href="http://busk.com/foo/bar.html" title="FooBar!" target="_blank" rel="external nofollow">FooBar!</a></div>]
+      end
+
       it "parsers links in the given html string adding rel and target" do
         input_html = %q[<div><a href="foo/bar.html" title="FooBar!">FooBar!</a></div>]
 
@@ -21,23 +36,22 @@ describe Spix::Utils do
       it "parses links with simple quotes" do
         input_html = %q[<div><a href='foo/bar.html' title='FooBar!'>FooBar!</a></div>]
 
-        Spix::Utils.format_links(:text => input_html).should ==
-          %q[<div><a href="foo/bar.html" title="FooBar!" target="_blank" rel="external nofollow">FooBar!</a></div>]
+        Spix::Utils.format_links(:text => input_html, :site_url => "http://busk.com/").should ==
+          %q[<div><a href="http://busk.com/foo/bar.html" title="FooBar!" target="_blank" rel="external nofollow">FooBar!</a></div>]
       end
 
-      # TODO: should we strip these extra &quot; ?
       it "parses links with html escaped quote (&quot;)" do
         input_html = %q[<div><a href=&quot;foo/bar.html&quot; title=&quot;FooBar!&quot;>FooBar!</a></div>]
 
-        Spix::Utils.format_links(:text => input_html).should ==
-          %q[<div><a href="&quot;foo/bar.html&quot;" title="&quot;FooBar!&quot;" target="_blank" rel="external nofollow">FooBar!</a></div>]
+        Spix::Utils.format_links(:text => input_html, :site_url => "http://busk.com/").should ==
+          %q[<div><a href=&quot;http://busk.com/foo/bar.html&quot; title=&quot;FooBar!&quot; target=&quot;_blank&quot; rel=&quot;external nofollow&quot;>FooBar!</a></div>]
       end
 
       it "parses links with html attributes without quotes, based on spaces" do
         input_html = %q[<div><a href=foo/bar.html title=FooBar!>FooBar!</a></div>]
 
-        Spix::Utils.format_links(:text => input_html).should ==
-          %q[<div><a href="foo/bar.html" title="FooBar!" target="_blank" rel="external nofollow">FooBar!</a></div>]
+        Spix::Utils.format_links(:text => input_html, :site_url => "http://busk.com/").should ==
+          %q[<div><a href="http://busk.com/foo/bar.html" title="FooBar!" target="_blank" rel="external nofollow">FooBar!</a></div>]
       end
 
       it "parses links with html attributes having spaces before or after the equal sign" do
@@ -67,7 +81,7 @@ describe Spix::Utils do
         input_html = %q[<div><img src="images/bar.jpg" title="FooBar!" alt="FooBar!" width="100" height="200" /></div>]
 
         Spix::Utils.format_links(:text => input_html).should ==
-          %q[<div><img src="images/bar.jpg" alt="FooBar!" title="FooBar!" width="100" height="200" /></div>]
+          %q[<div><img src="images/bar.jpg" title="FooBar!" alt="FooBar!" width="100" height="200" /></div>]
       end
 
       it "parses image tags removing other invalid html attributes" do
@@ -138,23 +152,37 @@ describe Spix::Utils do
       it "parses image tags with simple quotes" do
         input_html = %q[<div><img src='images/bar.jpg' title='FooBar!' /></div>]
 
-        Spix::Utils.format_links(:text => input_html).should ==
-          %q[<div><img src="images/bar.jpg" title="FooBar!" /></div>]
+        Spix::Utils.format_links(:text => input_html, :site_url => "http://busk.com/").should ==
+          %q[<div><img src="http://busk.com/images/bar.jpg" title="FooBar!" /></div>]
       end
 
       # TODO: should we strip these extra &quot; ?
       it "parses image tags with html escaped quote (&quot;)" do
         input_html = %q[<div><img src=&quot;images/bar.jpg&quot; title=&quot;FooBar!&quot; /></div>]
 
-        Spix::Utils.format_links(:text => input_html).should ==
-          %q[<div><img src="&quot;images/bar.jpg&quot;" title="&quot;FooBar!&quot;" /></div>]
+        Spix::Utils.format_links(:text => input_html, :site_url => "http://busk.com/").should ==
+          %q[<div><img src=&quot;http://busk.com/images/bar.jpg&quot; title=&quot;FooBar!&quot; /></div>]
+      end
+
+      it "parses image tags with html escaped quote (&quot;) and absolute sources" do
+        input_html = %q[<div><img src=&quot;http://busk.com/images/bar.jpg&quot; title=&quot;FooBar!&quot; /></div>]
+
+        Spix::Utils.format_links(:text => input_html, :site_url => "http://busk.com/").should ==
+          %q[<div><img src=&quot;http://busk.com/images/bar.jpg&quot; title=&quot;FooBar!&quot; /></div>]
+      end
+
+      it "parses image tags with absolute sources" do
+        input_html = %q[<div><img src="/images/bar.jpg" title="FooBar!" /></div>]
+
+        Spix::Utils.format_links(:text => input_html, :site_url => "http://busk.com/").should ==
+          %q[<div><img src="http://busk.com/images/bar.jpg" title="FooBar!" /></div>]
       end
 
       it "parses image tags with html attributes without quotes, based on spaces" do
         input_html = %q[<div><img src=images/bar.jpg title=FooBar! /></div>]
 
-        Spix::Utils.format_links(:text => input_html).should ==
-          %q[<div><img src="images/bar.jpg" title="FooBar!" /></div>]
+        Spix::Utils.format_links(:text => input_html, :site_url => "http://busk.com/").should ==
+          %q[<div><img src="http://busk.com/images/bar.jpg" title="FooBar!" /></div>]
       end
 
       it "parses image tags with html attributes having spaces before or after the equal sign" do

metadata

@@ -2,7 +2,7 @@
 name: spix_parser
 version: !ruby/object:Gem::Version 
   prerelease: 
-  version: 1.7.3
+  version: 1.7.5
 platform: ruby
 authors: 
 - Marcio Lopes de Faria