diff --git a/microdata.py b/microdata.py
index 19581a1..6132bba 100755
--- a/microdata.py
+++ b/microdata.py
@@ -153,6 +153,11 @@ def __repr__(self):
     'time':     'datetime',
 }
 
+block_elements = [
+    "div",
+    "p",
+    "h1", "h2", "h3", "h4", "h5", "h6"
+]
 
 def _find_items(e):
     items = []
@@ -232,8 +237,15 @@ def _text(e):
         chunks.append(e.data)
     elif hasattr(e, 'tagName') and e.tagName == 'script':
         return ''
+    elif hasattr(e, 'tagName') and e.tagName == 'br':
+        chunks.append("\n")
+
     for child in e.childNodes:
         chunks.append(_text(child))
+
+    if hasattr(e, 'tagName') and e.tagName in block_elements:
+        chunks.append("\n")
+    
     return ''.join(chunks)
 
 
diff --git a/test-data/example-dirty.html b/test-data/example-dirty.html
new file mode 100644
index 0000000..7bf66d7
--- /dev/null
+++ b/test-data/example-dirty.html
@@ -0,0 +1,40 @@
+<!doctype html>
+<html>
+    <!-- shamelessly stolen from http://schema.org/Person -->
+    <head>
+        <title>person example</title>
+    </head>
+
+    <body>
+    <div itemid="http://www.xyz.edu/~jane" itemscope itemtype="http://schema.org/Person">
+      <span itemprop="name">Jane Doe</span>
+      <img src="janedoe.jpg" itemprop="image" />
+
+      <span itemprop="jobTitle">Professor</span>
+      <div itemprop="description">A Professor that likes<br>Linebreaks</div>
+      <div itemprop="address" itemscope itemtype="http://schema.org/PostalAddress">
+        <span itemprop="streetAddress">
+            <p>20341 Whitworth Institute</p><p>405 N. Whitworth</p>
+          <!-- Comment Node -->
+          <script>
+            // Unrelated text
+          </script>
+        </span>
+        <span itemprop="addressLocality">Seattle</span>,
+        <span itemprop="addressRegion">WA</span>
+        <span itemprop="postalCode">98052</span>
+      </div>
+      <span itemprop="telephone">(425) 123-4567</span>
+      <a href="mailto:jane-doe@xyz.edu" itemprop="email">jane-doe@xyz.edu</a>
+
+      Jane's home page:
+      <a href="http://www.janedoe.com" itemprop="url">janedoe.com</a>
+
+      Graduate students:
+      <a href="http://www.xyz.edu/students/alicejones.html" itemprop="colleagues">
+        Alice Jones</a>
+      <a href="http://www.xyz.edu/students/bobsmith.html" itemprop="colleagues">
+        Bob Smith</a>
+    </div>
+    </body>
+</html>
diff --git a/test.py b/test.py
index bf1a6c6..703419b 100644
--- a/test.py
+++ b/test.py
@@ -108,6 +108,23 @@ def test_skip_level(self):
         self.assertEqual(len(items), 1)
         self.assertEqual(items[0].name, "Jane Doe")
 
+    def test_newlines(self):
+        items = get_items(open("test-data/example-dirty.html"))
+        # this html should have just one main item
+        self.assertTrue(len(items), 1)
+
+        item = items[0]
+
+        # description contains a br tag so it should have a newline
+        self.assertEqual(item.description.strip(), "A Professor that likes\nLinebreaks")
+
+        self.assertEqual(item.address.itemtype, [URI("http://schema.org/PostalAddress")])
+        # street adress should contain newlines because p is a block element
+        self.assertEqual(item.address.streetAddress.strip(), "20341 Whitworth Institute\n405 N. Whitworth")
+        self.assertEqual(item.address.addressLocality, "Seattle")
+
+        item = items[0]
+
     def test_parse_multiple_props(self):
         items = get_items(open("test-data/multiple-props.html"))
 
