hpricot_scrub 0.3.0 → 0.3.1

data/CHANGELOG.txt

@@ -1,3 +1,11 @@
+2008-06-03 Mina Naguib <mina.hpricotscrub@naguib.ca>
+	Release 0.3.1
+  	- Allow an element rule to be a Proc (which is expected to return one of
+	  the other non-proc rules (false/:strip/true/attr_rules).  This allows the
+	  building of highly custom filtering rules (for example filter out <B>
+	  tags unless their parent is a <P>...)
+	- Slight cleanup of internal documentation
+
 2008-01-11 Mina Naguib <mina.hpricotscrub@naguib.ca>
 	Release 0.3.0
 	Large overhaul of the module's logic to mimic most of perl's HTML::Scrubber

data/lib/hpricot_scrub/hpricot_scrub.rb

@@ -38,15 +38,15 @@ module Hpricot
       #
       # Merge+delete legacy config keys
       #
-      # :remove_tags
+      # :remove_tags -> :elem_rules (false)
       (config.delete(:remove_tags) || []).each do |tag|
         config[:elem_rules][tag] = false unless config[:elem_rules].has_key?(tag)
       end
-      # :allow_tags
+      # :allow_tags -> :elem_rules (true)
       (config.delete(:allow_tags) || []).each do |tag|
         config[:elem_rules][tag] = true unless config[:elem_rules].has_key?(tag)
       end
-      # :allow_attributes
+      # :allow_attributes -> :default_attribute_rule (procs)
       (config.delete(:allow_attributes) || []).each do |attribute|
         #
         # Add it to the default attribute rule
@@ -174,13 +174,14 @@ module Hpricot
     # Scrubs the element according to the given config
     # The relevant config key is :elem_rules.  It is expected to be a Hash having String HTML tag names as keys, and a rule as values
     # The rule value dictates what happens to the element.  The following logic is used:
-    #   If the rules is false, the element is removed
+    #   If the rule is false/nil, the element is removed along with all it's children
     #   If the rule is :strip, the element is stripped (the element itself is deleted and its children are promoted upwards to where it was)
+    #   If the rule is a proc, the proc is called (and given the element itself) - the proc's expected to return a valid rule that matches this documentation
     #   Otherwise the element is kept
     #
     #  If the element name (HTML tag) was not found in :elem_rules, the default rule in config key :default_elem_rule is used
     #
-    # After the above is done, scrub_attributes is called if the element was kept.  The rule is passed to it as it's assumed to be the attribute rules (see Hpricot::Scrub.keep_attribute?) to apply to the attributes, UNLESS the rule was explicitly "true", in which case the config key :default_attribute_rule is passed.
+    # After the above is done, if the element was kept, it's time to clean up its attributes so scrub_attributes is called.  The rule is passed to it as it's assumed to be the attribute rules (see Hpricot::Scrub.keep_attribute?) to apply to the attributes, UNLESS the rule was explicitly "true", in which case the config key :default_attribute_rule is passed.
     #
     # This is recursive and will do all the above to all the children of the element as well.
     #
@@ -194,6 +195,10 @@ module Hpricot
 
       rule = config[:elem_rules].has_key?(name) ? config[:elem_rules][name] : config[:default_elem_rule]
 
+      while rule.is_a?(Proc)
+        rule = rule.call(self)
+      end
+
       if !rule
         remove
       elsif rule == :strip

data/lib/hpricot_scrub/version.rb

@@ -2,7 +2,7 @@ module HpricotScrub #:nodoc:
   module VERSION #:nodoc:
     MAJOR = 0
     MINOR = 3
-    TINY  = 0
+    TINY  = 1
 
     STRING = [MAJOR, MINOR, TINY].join('.')
   end

data/test/hpricot_scrub_test.rb

@@ -15,7 +15,9 @@ class HpricotScrubTest < Test::Unit::TestCase
         "a" =>  {
           "href"  =>  %r|^https?://|i
         },
-        "b" => true,
+        "b" => Proc.new do |bold_element|
+          bold_element.parent.name == "p" ? true : :strip
+        end,
         "body"  =>  {
           "lang" => %w(en es fr)
         },
@@ -67,7 +69,6 @@ class HpricotScrubTest < Test::Unit::TestCase
   def test_elem_rule_keep
      @scrubbed_docs.each_with_index do |doc, i|
        assert_equal @docs[i].search("//a").length, doc.search("//a").length
-       assert_equal @docs[i].search("//b").length, doc.search("//b").length
        assert_equal @docs[i].search("//img").length,  doc.search("//img").length
      end
   end
@@ -86,6 +87,12 @@ class HpricotScrubTest < Test::Unit::TestCase
      end
   end
 
+  def test_elem_rule_proc
+     @scrubbed_docs.each_with_index do |doc, i|
+       assert_equal @docs[i].search("//p/b").length, doc.search("//b").length
+     end
+  end
+
   def test_attr_default_rule_removes
      @scrubbed_docs.each do |doc|
        assert_equal 0, doc.search("*[@mce_src]").length

metadata

@@ -3,8 +3,8 @@ rubygems_version: 0.9.4
 specification_version: 1
 name: hpricot_scrub
 version: !ruby/object:Gem::Version 
-  version: 0.3.0
-date: 2008-04-22 00:00:00 -05:00
+  version: 0.3.1
+date: 2008-06-04 00:00:00 -05:00
 summary: Scrub HTML with Hpricot
 require_paths: 
 - lib
@@ -64,7 +64,16 @@ dependencies:
   version_requirement: 
   version_requirements: !ruby/object:Gem::Version::Requirement 
     requirements: 
-    - - ">="
+    - - ">"
       - !ruby/object:Gem::Version 
-        version: "0.5"
+        version: 0.0.0
+    version: 
+- !ruby/object:Gem::Dependency 
+  name: ">= 0.5"
+  version_requirement: 
+  version_requirements: !ruby/object:Gem::Version::Requirement 
+    requirements: 
+    - - ">"
+      - !ruby/object:Gem::Version 
+        version: 0.0.0
     version: