Merge pull request #1568 from nitoyon/url-escape

lib/jekyll/page.rb

@@ -135,7 +135,7 @@ module Jekyll
     #
     # Returns the destination file path String.
     def destination(dest)
-      path = Jekyll.sanitized_path(dest, url)
+      path = Jekyll.sanitized_path(dest, URL.unescape_path(url))
       path = File.join(path, "index.html") if url =~ /\/$/
       path
     end

lib/jekyll/post.rb

@@ -208,10 +208,10 @@ module Jekyll
         :year        => date.strftime("%Y"),
         :month       => date.strftime("%m"),
         :day         => date.strftime("%d"),
-        :title       => CGI.escape(slug),
+        :title       => slug,
         :i_day       => date.strftime("%d").to_i.to_s,
         :i_month     => date.strftime("%m").to_i.to_s,
-        :categories  => (categories || []).map { |c| URI.escape(c.to_s) }.join('/'),
+        :categories  => (categories || []).map { |c| c.to_s }.join('/'),
         :short_month => date.strftime("%b"),
         :y_day       => date.strftime("%j"),
         :output_ext  => output_ext
@@ -260,7 +260,7 @@ module Jekyll
     # Returns destination file path String.
     def destination(dest)
       # The url needs to be unescaped in order to preserve the correct filename
-      path = Jekyll.sanitized_path(dest, CGI.unescape(url))
+      path = Jekyll.sanitized_path(dest, URL.unescape_path(url))
       path = File.join(path, "index.html") if path[/\.html$/].nil?
       path
     end

lib/jekyll/url.rb

@@ -1,3 +1,5 @@
+require 'uri'
+
 # Public: Methods that generate a URL for a resource such as a Post or a Page.
 #
 # Examples
@@ -44,7 +46,7 @@ module Jekyll
     # Returns the _unsanitizied_ String URL
     def generate_url
       @placeholders.inject(@template) do |result, token|
-        result.gsub(/:#{token.first}/, token.last)
+        result.gsub(/:#{token.first}/, self.class.escape_path(token.last))
       end
     end
 
@@ -65,5 +67,43 @@ module Jekyll
 
       url
     end
+
+    # Escapes a path to be a valid URL path segment
+    #
+    # path - The path to be escaped.
+    #
+    # Examples:
+    #
+    #   URL.escape_path("/a b")
+    #   # => "/a%20b"
+    #
+    # Returns the escaped path.
+    def self.escape_path(path)
+      # Because URI.escape doesn't escape '?', '[' and ']' by defaut,
+      # specify unsafe string (except unreserved, sub-delims, ":", "@" and "/").
+      #
+      # URI path segment is defined in RFC 3986 as follows:
+      #   segment       = *pchar
+      #   pchar         = unreserved / pct-encoded / sub-delims / ":" / "@"
+      #   unreserved    = ALPHA / DIGIT / "-" / "." / "_" / "~"
+      #   pct-encoded   = "%" HEXDIG HEXDIG
+      #   sub-delims    = "!" / "$" / "&" / "'" / "(" / ")"
+      #                 / "*" / "+" / "," / ";" / "="
+      URI.escape(path, /[^a-zA-Z\d\-._~!$&\'()*+,;=:@\/]/)
+    end
+
+    # Unescapes a URL path segment
+    #
+    # path - The path to be unescaped.
+    #
+    # Examples:
+    #
+    #   URL.unescape_path("/a%20b")
+    #   # => "/a b"
+    #
+    # Returns the unescaped path.
+    def self.unescape_path(path)
+      URI.unescape(path)
+    end
   end
 end

test/source/+/%# +.md

@@ -0,0 +1,6 @@
+---
+layout: default
+title : Page name with non-alphabetic character 
+---
+Line 1
+{{ page.title }}

test/source/_posts/2014-03-22-escape-+ %20[].markdown

@@ -0,0 +1,6 @@
+---
+layout: default
+title: Plus space percent
+---
+
+Signs are nice

test/test_filters.rb

@@ -125,7 +125,7 @@ class TestFilters < Test::Unit::TestCase
           case g["name"]
           when "default"
             assert g["items"].is_a?(Array), "The list of grouped items for 'default' is not an Array."
-            assert_equal 4, g["items"].size
+            assert_equal 5, g["items"].size
           when "nil"
             assert g["items"].is_a?(Array), "The list of grouped items for 'nil' is not an Array."
             assert_equal 2, g["items"].size

test/test_page.rb

@@ -30,6 +30,11 @@ class TestPage < Test::Unit::TestCase
         assert_equal false, @page.published?
       end
 
+      should "create url with non-alphabetic characters" do
+        @page = setup_page('+', '%# +.md')
+        assert_equal "/+/%25%23%20+.html", @page.url
+      end
+
       context "in a directory hierarchy" do
         should "create url based on filename" do
           @page = setup_page('/contacts', 'bar.html')
@@ -174,6 +179,15 @@ class TestPage < Test::Unit::TestCase
         assert File.exists?(File.join(dest_dir, '+', 'plus+in+url'))
       end
 
+      should "write even when permalink has '%# +'" do
+        page = setup_page('+', '%# +.md')
+        do_render(page)
+        page.write(dest_dir)
+
+        assert File.directory?(dest_dir)
+        assert File.exists?(File.join(dest_dir, '+', '%# +.html'))
+      end
+
       should "write properly without html extension" do
         page = setup_page('contacts.html')
         page.site.permalink_style = :pretty

test/test_post.rb

@@ -86,13 +86,20 @@ class TestPost < Test::Unit::TestCase
         end
       end
 
-      should "CGI escape urls" do
+      should "escape urls" do
         @post.categories = []
         @post.process("2009-03-12-hash-#1.markdown")
         assert_equal "/2009/03/12/hash-%231.html", @post.url
         assert_equal "/2009/03/12/hash-#1", @post.id
       end
 
+      should "escape urls with non-alphabetic characters" do
+        @post.categories = []
+        @post.process("2014-03-22-escape-+ %20[].markdown")
+        assert_equal "/2014/03/22/escape-+%20%2520%5B%5D.html", @post.url
+        assert_equal "/2014/03/22/escape-+ %20[]", @post.id
+      end
+
       should "respect permalink in yaml front matter" do
         file = "2008-12-03-permalinked-post.textile"
         @post.process(file)
@@ -533,6 +540,26 @@ class TestPost < Test::Unit::TestCase
           assert File.exists?(File.join(dest_dir, '2008', '10', '18', 'foo-bar.html'))
         end
 
+        should "write properly when url has hash" do
+          post = setup_post("2009-03-12-hash-#1.markdown")
+          do_render(post)
+          post.write(dest_dir)
+
+          assert File.directory?(dest_dir)
+          assert File.exists?(File.join(dest_dir, '2009', '03', '12',
+                                        'hash-#1.html'))
+        end
+
+        should "write properly when url has space" do
+          post = setup_post("2014-03-22-escape-+ %20[].markdown")
+          do_render(post)
+          post.write(dest_dir)
+
+          assert File.directory?(dest_dir)
+          assert File.exists?(File.join(dest_dir, '2014', '03', '22',
+                                        'escape-+ %20[].html'))
+        end
+
         should "write properly without html extension" do
           post = setup_post("2008-10-18-foo-bar.textile")
           post.site.permalink_style = ":title"

test/test_site.rb

@@ -159,6 +159,7 @@ class TestSite < Test::Unit::TestCase
       @site.process
       # files in symlinked directories may appear twice
       sorted_pages = %w(
+        %#\ +.md
         .htaccess
         about.html
         bar.html