Fix removal of meaningful white-space

Fixes #4

lib/html_sanitize_ex/scrubber.ex

@@ -9,9 +9,15 @@ defmodule HtmlSanitizeEx.Scrubber do
 
   def scrub(html, scrubber_module) do
     html
+    |> before_scrub
     |> scrubber_module.before_scrub
     |> HtmlSanitizeEx.Parser.parse
     |> HtmlSanitizeEx.Traverser.traverse(scrubber_module)
     |> HtmlSanitizeEx.Parser.to_html
   end
+
+  defp before_scrub(html) do
+    html
+    |> String.replace(~r/(>)(\ +)(<)/, "\\1&#32;\\3")
+  end
 end

lib/html_sanitize_ex/scrubber/html5.ex

@@ -33,6 +33,8 @@ defmodule HtmlSanitizeEx.Scrubber.HTML5 do
   Meta.allow_tag_with_these_attributes "a", ["accesskey", "class", "contenteditable", "contextmenu", "dir", "draggable", "dropzone", "hidden", "id", "inert", "itemid", "itemprop", "itemref", "itemscope", "itemtype", "lang", "role", "spellcheck", "tabindex", "title", "translate",
                                               "target", "ping", "rel", "media", "hreflang", "type"]
 
+  Meta.allow_tag_with_these_attributes "b", ["accesskey", "class", "contenteditable", "contextmenu", "dir", "draggable", "dropzone", "hidden", "id", "inert", "itemid", "itemprop", "itemref", "itemscope", "itemtype", "lang", "role", "spellcheck", "tabindex", "title", "translate"]
+
   Meta.allow_tag_with_these_attributes "h1", ["accesskey", "class", "contenteditable", "contextmenu", "dir", "draggable", "dropzone", "hidden", "id", "inert", "itemid", "itemprop", "itemref", "itemscope", "itemtype", "lang", "role", "spellcheck", "tabindex", "title", "translate"]
   Meta.allow_tag_with_these_attributes "h2", ["accesskey", "class", "contenteditable", "contextmenu", "dir", "draggable", "dropzone", "hidden", "id", "inert", "itemid", "itemprop", "itemref", "itemscope", "itemtype", "lang", "role", "spellcheck", "tabindex", "title", "translate"]
   Meta.allow_tag_with_these_attributes "h3", ["accesskey", "class", "contenteditable", "contextmenu", "dir", "draggable", "dropzone", "hidden", "id", "inert", "itemid", "itemprop", "itemref", "itemscope", "itemtype", "lang", "role", "spellcheck", "tabindex", "title", "translate"]

mix.lock

@@ -1,3 +1,3 @@
-%{"inch_ex": {:hex, :inch_ex, "0.4.0-dev"},
-  "mochiweb": {:hex, :mochiweb, "2.12.2"},
-  "poison": {:hex, :poison, "1.4.0"}}
+%{"inch_ex": {:hex, :inch_ex, "0.5.1", "c1c18966c935944cbb2d273796b36e44fab3c54fd59f906ff026a686205b4e14", [:mix], [{:poison, "~> 1.5 or ~> 2.0", [hex: :poison, optional: false]}]},
+  "mochiweb": {:hex, :mochiweb, "2.12.2", "80804ad342afa3d7f3524040d4eed66ce74b17a555de454ac85b07c479928e46", [:make, :rebar], []},
+  "poison": {:hex, :poison, "2.1.0", "f583218ced822675e484648fa26c933d621373f01c6c76bd00005d7bd4b82e27", [:mix], []}}

test/basic_html_test.exs

@@ -99,7 +99,7 @@ defmodule HtmlSanitizeExScrubberBasicHTMLTest do
   @tag href_scrubbing: true
   test "test_strip_links_with_line_feed_and_uppercase_tag" do
     input = "<a href='almost'>on my mind</a> <A href='almost'>all day long</A>"
-    assert "<a href=\"almost\">on my mind</a><a href=\"almost\">all day long</a>" == basic_html_sanitize(input)
+    assert "<a href=\"almost\">on my mind</a> <a href=\"almost\">all day long</a>" == basic_html_sanitize(input)
   end
 
   @tag href_scrubbing: true

test/html5_test.exs

@@ -5,6 +5,11 @@ defmodule HtmlSanitizeExScrubberHTML5Test do
     HtmlSanitizeEx.html5(text)
   end
 
+  test "strips nothing" do
+    input = "This <b>is</b> <b>an</b> <i>example</i> of <u>space</u> eating."
+    assert input == full_html_sanitize(input)
+  end
+
   test "leaves the allowed tags alone" do
     input = "<h1 class=\"heading\">hello world!</h1>"
     expected = "<h1 class=\"heading\">hello world!</h1>"

test/no_scrub_test.exs

@@ -0,0 +1,13 @@
+defmodule HtmlSanitizeExScrubberNoScrubTest do
+  use ExUnit.Case
+
+  defp no_scrub_sanitize(text) do
+    HtmlSanitizeEx.noscrub(text)
+  end
+
+  test "strips nothing" do
+    input = "This <b>is</b> <b>an</b> <i>example</i> of <u>space</u> eating."
+    expected = "This <b>is</b> <b>an</b> <i>example</i> of <u>space</u> eating."
+    assert expected == no_scrub_sanitize(input)
+  end
+end