Namespaced attributes are equal if they correspond to the same string.

4 files changed, 56 insertions, 2 deletions

diff --git a/bs4/element.py b/bs4/element.py
index fdb90e0..7e5810a 100644
--- a/bs4/element.py
+++ b/bs4/element.py
@@ -24,11 +24,22 @@ def _alias(attr):
 
 class NamespacedAttribute(object):
 
-    def __init__(self, namespace_abbreviation, name, namespace):
+    def __init__(self, namespace_abbreviation, name, namespace=None):
         self.namespace_abbreviation = namespace_abbreviation
         self.name = name
         self.namespace = namespace
 
+    def __eq__(self, other):
+        if isinstance(other, NamespacedAttribute):
+            return (
+                self.namespace_abbreviation == other.namespace_abbreviation
+                and self.name == other.name
+                and self.namespace == other.namespace)
+        elif isinstance(other, basestring):
+            return str(self) == other
+        else:
+            return super(NamespacedAttribute, self).__eq__(other)
+
     def __str__(self):
         name = self.name
         if self.namespace_abbreviation:
diff --git a/bs4/testing.py b/bs4/testing.py
index dc20812..b2ca180 100644
--- a/bs4/testing.py
+++ b/bs4/testing.py
@@ -358,6 +358,20 @@ class HTMLTreeBuilderSmokeTest(object):
         # For the rest of the story, see TestSubstitutions in
         # test_tree.py.
 
+class XMLTreeBuilderSmokeTest(object):
+
+    def test_namespaces_are_preserved(self):
+        markup = '<root xmlns:a="http://www.example.com/" xmlns:b="http://example.net/"><a:foo>This tag is in the a namespace</a:foo><b:foo>This tag is in the b namespace</b:foo></root>'
+        soup = self.soup(markup)
+        root = soup.root
+        import pdb; pdb.set_trace()
+        self.assertEquals("http://www.example.com/", root['xmlns:a'])
+        self.assertEquals("http://www.example.net/", root['xmlns:b'])
+
+
+        pass
+
+
 
 def skipIf(condition, reason):
    def nothing(test, *args, **kwargs):
diff --git a/bs4/tests/test_lxml.py b/bs4/tests/test_lxml.py
index 92b7389..27ec570 100644
--- a/bs4/tests/test_lxml.py
+++ b/bs4/tests/test_lxml.py
@@ -14,6 +14,7 @@ from bs4.testing import skipIf
 from bs4.tests import test_htmlparser
 from bs4.testing import (
     HTMLTreeBuilderSmokeTest,
+    XMLTreeBuilderSmokeTest,
     SoupTest,
     skipIf,
 )
@@ -35,3 +36,14 @@ class LXMLTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest):
             "<p>foo&#x10000000000000;bar</p>", "<p>foobar</p>")
         self.assertSoupEquals(
             "<p>foo&#1000000000;bar</p>", "<p>foobar</p>")
+
+@skipIf(
+    not LXML_PRESENT,
+    "lxml seems not to be present, not testing its XML tree builder.")
+class LXMLXMLTreeBuilderSmokeTest(SoupTest, XMLTreeBuilderSmokeTest):
+    """See ``HTMLTreeBuilderSmokeTest``."""
+
+    @property
+    def default_builder(self):
+        return LXMLTreeBuilderForXML()
+
diff --git a/bs4/tests/test_soup.py b/bs4/tests/test_soup.py
index 2b7c003..e9aaa78 100644
--- a/bs4/tests/test_soup.py
+++ b/bs4/tests/test_soup.py
@@ -3,7 +3,10 @@
 
 import unittest
 from bs4 import BeautifulSoup
-from bs4.element import SoupStrainer
+from bs4.element import (
+    SoupStrainer,
+    NamespacedAttribute,
+    )
 from bs4.dammit import EntitySubstitution, UnicodeDammit
 from bs4.testing import SoupTest
 import warnings
@@ -233,3 +236,17 @@ class TestUnicodeDammit(unittest.TestCase):
             msg = w[0].message
             self.assertTrue(isinstance(msg, UnicodeWarning))
             self.assertTrue("Some characters could not be decoded" in str(msg))
+
+
+class TestNamedspacedAttribute(SoupTest):
+
+    def test_attribute_is_equivalent_to_colon_separated_string(self):
+        a = NamespacedAttribute("a", "b")
+        self.assertEqual("a:b", a)
+
+    def test_attributes_are_equivalent_if_all_members_identical(self):
+        a = NamespacedAttribute("a", "b", "c")
+        b = NamespacedAttribute("a", "b", "c")
+        self.assertEqual(a, b)
+        b.namespace = "d"
+        self.assertNotEqual(a, b)