Add function to convert node tree back to HTML

CHANGELOG.md

@@ -12,6 +12,7 @@ This project adheres to [Semantic Versioning](http://semver.org/).
 - Add the child combinator to `Floki.find/2`.
 - Add the adjacent sibling combinator to `Floki.find/2`.
 - Add the general adjacent sibling combinator to `Floki.find/2`.
+- Add `Floki.raw_html/2`
 
 ## [0.4.1] - 2015-09-18
 

README.md

@@ -43,6 +43,10 @@ Floki.find(html, "#content")
 Floki.find(html, "p.headline")
 # => [{"p", [{"class", "headline"}], ["Floki"]}]
 
+Floki.find(html, "p.headline") 
+|> Floki.raw_html
+# => <p class="headline">Floki</p>
+
 
 Floki.find(html, "a")
 # => [{"a", [{"href", "http://github.com/philss/floki"}], ["Github page"]},
@@ -129,6 +133,14 @@ Floki.find(html, ".example")
 # => [{"div", [{"class", "example"}], []}]
 ```
 
+To convert your node tree back to raw HTML (spaces are ignored):
+
+```elixir
+Floki.find(html, ".example") 
+|> Flok.raw_html
+# =>  <div class="example"></div>
+```
+
 To fetch some attribute from elements, try:
 
 ```elixir

lib/floki.ex

@@ -77,6 +77,48 @@ defmodule Floki do
     Parser.parse(html)
   end
 
+  @self_closing_tags ["area", "base", "br", "col", "command", "embed", "hr", "img", "input", "keygen", "link", "mete", "param", "source", "track", "wbr"]
+
+  @doc """
+  Converts node tree to raw HTML (spaces are ignored).
+
+  ## Examples
+
+      iex> Floki.parse(~s(<div class="wrapper">my content</div>)) |> Floki.raw_html
+      ~s(<div class="wrapper">my content</div>)
+
+  """
+
+  def raw_html(tuple) when is_tuple(tuple), do: raw_html([tuple])
+  def raw_html([], html), do: html
+  def raw_html([value|tail], html) when is_bitstring(value), do: value
+  def raw_html([node|tail], html \\ "") do 
+    {elem, attrs, value} = node
+    raw_html(tail, html <> tag_for(elem, attrs |> tag_attrs, value))
+  end
+
+  defp tag_attrs(attr_list) do
+    attr_list
+    |> Enum.reduce("", fn(c,t) -> ~s(#{t} #{elem(c,0)}="#{elem(c,1)}") end)
+    |> String.strip
+  end
+
+  defp tag_for(elem, attrs, value) do
+    if Enum.member?(@self_closing_tags, elem) do
+      if attrs != "" do
+        "<#{elem} #{attrs}/>"
+      else
+        "<#{elem}/>"
+      end
+    else
+      if attrs != "" do
+        "<#{elem} #{attrs}>#{raw_html(value)}</#{elem}>"
+      else
+        "<#{elem}>#{raw_html(value)}</#{elem}>"
+      end
+    end
+  end
+
   @doc """
   Find elements inside a HTML tree or string.
 

test/floki_test.exs

@@ -108,6 +108,19 @@ defmodule FlokiTest do
       }
   end
 
+  @basic_html """
+    <div id="content">
+      <p>
+        <a href="uol.com.br" class="bar">
+          <span>UOL</span>
+          <img src="foo.png"/>
+        </a>
+      </p>
+      <strong>ok</strong>
+      <br/>
+    </div>
+  """
+
   # Floki.parse/1
 
   test "parse html_without_html_tag" do
@@ -119,6 +132,23 @@ defmodule FlokiTest do
     ]
   end
 
+  # Floki.raw_html/2
+
+  test "raw_html" do
+    raw_html = Floki.parse(@basic_html) |> Floki.raw_html
+    assert raw_html == String.split(@basic_html, "\n") |> Enum.map(&(String.strip(&1))) |> Enum.join("")
+  end
+
+  test "raw_html (html with data attributes)" do
+    raw_html = Floki.parse(@html_with_data_attributes) |> Floki.raw_html
+    assert raw_html == String.split(raw_html, "\n") |> Enum.map(&(String.strip(&1))) |> Enum.join("")
+  end
+
+  test "raw_html (after find)" do
+    raw_html = Floki.parse(@basic_html) |> Floki.find("a") |> Floki.raw_html
+    assert raw_html == ~s(<a href="uol.com.br" class="bar"><span>UOL</span><img src="foo.png"/></a>)
+  end
+
   # Floki.find/2 - Classes
 
   test "find elements with a given class" do