fix a unicode_literals regression in ihatexml.py

ambv · ambv · commit 7399617f011e · 2013-05-16T13:37:01.000+02:00
diff --git a/html5lib/ihatexml.py b/html5lib/ihatexml.py
@@ -260,7 +260,7 @@ def fromXmlName(self, name):
         return name
 
     def escapeChar(self, char):
-        replacement = "U" + hex(ord(char))[2:].upper().rjust(5, "0")
+        replacement = "U%05X" % (ord(char),)
         self.replaceCache[char] = replacement
         return replacement
 
diff --git a/html5lib/tests/test_parser2.py b/html5lib/tests/test_parser2.py
@@ -16,6 +16,10 @@ class MoreParserTests(unittest.TestCase):
 
     def setUp(self):
         self.dom_tree = treebuilders.getTreeBuilder("dom")
+        try:
+            self.lxml_tree = treebuilders.getTreeBuilder("lxml")
+        except ImportError:
+            self.lxml_tree = None
 
     def test_assertDoctypeCloneable(self):
         parser = html5parser.HTMLParser(tree=self.dom_tree)
@@ -27,6 +31,12 @@ def test_line_counter(self):
         parser = html5parser.HTMLParser(tree=self.dom_tree)
         parser.parse("<pre>\nx\n&gt;\n</pre>")
 
+    def test_ihatexml(self):
+        if not self.lxml_tree:
+            return
+        parser = html5parser.HTMLParser(tree=self.lxml_tree)
+        parser.parse(b'<p xml:lang="pl">Witam wszystkich')
+
     def test_namespace_html_elements_0_dom(self):
         parser = html5parser.HTMLParser(tree=self.dom_tree, namespaceHTMLElements=True)
         doc = parser.parse("<html></html>")
