lxml fixes

jgraham · jgraham · commit 4bf7e885c951 · 2008-01-28T22:17:27.000Z
--HG--
extra : convert_revision : svn%3Aacbfec75-9323-0410-a652-858a13e371e0/trunk%401093
diff --git a/src/html5lib/treebuilders/etree_lxml.py b/src/html5lib/treebuilders/etree_lxml.py
@@ -170,13 +170,17 @@ def insertRoot(self, name):
         if self.doctype:
             docStr += "<!DOCTYPE %s"%self.doctype.name
             if self.doctype.publicId is not None:
-                docStr += "PUBLIC %s"%self.doctype.publicId
+                docStr += ' PUBLIC "%s"'%self.doctype.publicId
             if self.doctype.systemId:
-                docStr += "SYSTEM %s"%self.doctype.systemId
+                docStr += '  "%s"'%self.doctype.systemId
             docStr += ">"
         docStr += "<html></html>"
         
-        root = etree.fromstring(docStr)
+        try:
+            root = etree.fromstring(docStr)
+        except etree.XMLSyntaxError:
+            print docStr
+            raise
         
         #Create the root document and add the ElementTree to it
         self.document = self.documentClass()
diff --git a/tests/test_parser.py b/tests/test_parser.py
@@ -1,11 +1,13 @@
 import os
 import sys
 import traceback
-
 import StringIO
 import unittest
-from support import html5lib_test_files, TestData, convert, convertExpected
+import warnings
+
+warnings.simplefilter("error")
 
+from support import html5lib_test_files, TestData, convert, convertExpected
 from html5lib import html5parser, treebuilders, constants
 
 treeTypes = {"simpletree":treebuilders.getTreeBuilder("simpletree"),
@@ -34,8 +36,11 @@
         pass
     
 try:
-    import lxml.etree as lxml
-    treeTypes['lxml'] = treebuilders.getTreeBuilder("etree", lxml, fullTree=True)
+    try:
+        import lxml.html as lxml
+    except ImportError:
+        import lxml.etree as lxml
+    treeTypes['lxml'] = treebuilders.getTreeBuilder("lxml", lxml, fullTree=True)
 except ImportError:
     pass
 
@@ -70,7 +75,11 @@ def runParserTest(self, innerHTML, input, expected, errors, treeClass):
             if innerHTML:
                 document = p.parseFragment(StringIO.StringIO(input), innerHTML)
             else:
-                document = p.parse(StringIO.StringIO(input))
+                try:
+                    document = p.parse(StringIO.StringIO(input))
+                except constants.DataLossWarning:
+                    sys.stderr.write("Test input causes known dataloss, skipping")
+                    return 
         except:
             errorMsg = "\n".join(["\n\nInput:", input, "\nExpected:", expected,
                                   "\nTraceback:", traceback.format_exc()])
