[LNT] Python 3 support: adapt XML parsing with extra entities

Thomas Preud'homme · Thomas Preud'homme · commit 593510ca3828 · 2019-12-05T12:30:50.000Z
Summary: Test server/ui/V4Pages.py parses XML with non standard entities using the ElementTree's XML parser. However the method it uses to declare the non standard entities are no longer available in Python 3's ElementTree. This commit instead modifies the DOCTYPE element in the XML directly before passing it to ElementTree's XML parser. Reviewers: cmatthews, hubert.reinterpretcast, kristof.beyls, leandron, PrzemekWirkus Reviewed By: cmatthews Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D68986
diff --git a/tests/server/ui/V4Pages.py b/tests/server/ui/V4Pages.py
@@ -109,10 +109,12 @@ def dump_html(html_string):
 
 def get_xml_tree(html_string):
     try:
-        parser = ET.XMLParser()
-        parser.parser.UseForeignDTD(True)
-        parser.entity.update((x, unichr(i)) for x, i in name2codepoint.items())
-        tree = ET.fromstring(html_string, parser=parser)
+        entities_defs = []
+        for x, i in name2codepoint.items():
+            entities_defs.append('  <!ENTITY {x} "&#{i};">'.format(**locals()))
+        docstring = "<!DOCTYPE html [\n{}\n]>".format('\n'.join(entities_defs))
+        html_string = html_string.replace("<!DOCTYPE html>", docstring, 1)
+        tree = ET.fromstring(html_string)
     except:  # noqa FIXME: figure out what we expect this to throw.
         dump_html(html_string)
         raise
