Allow the crawler to skip links with query parameters.

This is done by calling 'skip_query_string' on the yielded anemone object.

lib/anemone/core.rb

@@ -100,6 +100,14 @@ def skip_links_like(*patterns)
       @skip_link_patterns.concat [patterns].flatten.compact
       self
     end
+
+    #
+    # Setting this skips all links with a query string (?param=value part).
+    #
+    def skip_query_string
+      @skip_query_string = true
+      self
+    end
 
     #
     # Add a block to be executed on every Page as they are encountered
@@ -249,7 +257,15 @@ def visit_link?(link, from_page = nil)
         too_deep = false
       end
 
-      !@pages.has_page?(link) && !skip_link?(link) && allowed && !too_deep
+      !@pages.has_page?(link) && !skip_link?(link) && !skip_query_string?(link) && allowed && !too_deep
+    end
+
+    #
+    # Returns +true+ if *link* should not be visited because
+    # it has a query string and +skip_query_string+ is set.
+    #
+    def skip_query_string?(link)
+      @skip_query_string && link.query
     end
 
     #

spec/core_spec.rb

@@ -71,6 +71,19 @@ module Anemone
         core.should have(2).pages
         core.pages.keys.should_not include(pages[2].url)
       end
+
+      it "should be able to skip links with query strings" do
+        pages = []
+        pages << FakePage.new('0', :links => ['1?foo=1', '2'])
+        pages << FakePage.new('1?foo=1')
+        pages << FakePage.new('2')
+
+        core = Anemone.crawl(pages[0].url, @opts) do |a|
+          a.skip_query_string
+        end
+
+        core.should have(2).pages
+      end
 
       it "should be able to skip links based on a RegEx" do
         pages = []