[Python-checkins] CVS: python/dist/src/Lib/test test_sgmllib.py,1.1,1.2

[Fred L. Drake]

Update of /cvsroot/python/python/dist/src/Lib/test
In directory usw-pr-cvs1:/tmp/cvs-serv15870/test

Modified Files:
	test_sgmllib.py 
Log Message:
Added several new tests to check the behavior with respect to doctype
declarations and weird markup that we used to accept & ignore that recent
versions raised an exception for; the original behavior has been restored
and augmented (the user can decide what to do if they care; the default is
to ignore it as done in early versions).


Index: test_sgmllib.py
===================================================================
RCS file: /cvsroot/python/python/dist/src/Lib/test/test_sgmllib.py,v
retrieving revision 1.1
retrieving revision 1.2
diff -C2 -d -r1.1 -r1.2
*** test_sgmllib.py	2001/07/16 18:52:40	1.1
--- test_sgmllib.py	2001/09/24 20:22:09	1.2
***************
*** 55,58 ****
--- 55,61 ----
          self.append(("pi", data))
  
+     def unknown_decl(self, decl):
+         self.append(("unknown decl", decl))
+ 
  
  class CDATAEventCollector(EventCollector):
***************
*** 66,75 ****
      collector = EventCollector
  
!     def check_events(self, source, expected_events):
          parser = self.collector()
!         for s in source:
!             parser.feed(s)
!         parser.close()
!         events = parser.get_events()
          if events != expected_events:
              self.fail("received events did not match expected events\n"
--- 69,90 ----
      collector = EventCollector
  
!     def get_events(self, source):
          parser = self.collector()
!         try:
!             for s in source:
!                 parser.feed(s)
!             parser.close()
!         except:
!             #self.events = parser.events
!             raise
!         return parser.get_events()
! 
!     def check_events(self, source, expected_events):
!         try:
!             events = self.get_events(source)
!         except:
!             import sys
!             #print >>sys.stderr, pprint.pformat(self.events)
!             raise
          if events != expected_events:
              self.fail("received events did not match expected events\n"
***************
*** 88,91 ****
--- 103,131 ----
                        % (source, pprint.pformat(parser.get_events())))
  
+     def test_doctype_decl_internal(self):
+         inside = """\
+ DOCTYPE html PUBLIC '-//W3C//DTD HTML 4.01//EN'
+              SYSTEM 'http://www.w3.org/TR/html401/strict.dtd' [
+   <!ELEMENT html - O EMPTY>
+   <!ATTLIST html
+       version CDATA #IMPLIED
+       profile CDATA 'DublinCore'>
+   <!NOTATION datatype SYSTEM 'http://xml.python.org/notations/python-module'>
+   <!ENTITY myEntity 'internal parsed entity'>
+   <!ENTITY anEntity SYSTEM 'http://xml.python.org/entities/something.xml'>
+   <!ENTITY % paramEntity 'name|name|name'>
+   %paramEntity;
+   <!-- comment -->
+ ]"""
+         self.check_events(["<!%s>" % inside], [
+             ("decl", inside),
+             ])
+ 
+     def test_doctype_decl_external(self):
+         inside = "DOCTYPE html PUBLIC '-//W3C//DTD HTML 4.01//EN'"
+         self.check_events("<!%s>" % inside, [
+             ("decl", inside),
+             ])
+ 
      def test_underscore_in_attrname(self):
          # SF bug #436621
***************
*** 133,136 ****
--- 173,186 ----
              ])
  
+     def test_bare_ampersands(self):
+         self.check_events("this text & contains & ampersands &", [
+             ("data", "this text & contains & ampersands &"),
+             ])
+ 
+     def test_bare_pointy_brackets(self):
+         self.check_events("this < text > contains < bare>pointy< brackets", [
+             ("data", "this < text > contains < bare>pointy< brackets"),
+             ])
+ 
      def test_attr_syntax(self):
          output = [
***************
*** 157,160 ****
--- 207,218 ----
              ])
  
+     def test_illegal_declarations(self):
+         s = 'abc<!spacer type="block" height="25">def'
+         self.check_events(s, [
+             ("data", "abc"),
+             ("unknown decl", 'spacer type="block" height="25"'),
+             ("data", "def"),
+             ])
+ 
      def test_weird_starttags(self):
          self.check_events("<a<a>", [
***************
*** 197,200 ****
--- 255,266 ----
              ])
  
+     def test_illegal_declarations(self):
+         s = 'abc<!spacer type="block" height="25">def'
+         self.check_events(s, [
+             ("data", "abc"),
+             ("unknown decl", 'spacer type="block" height="25"'),
+             ("data", "def"),
+             ])
+ 
      # XXX These tests have been disabled by prefixing their names with
      # an underscore.  The first two exercise outstanding bugs in the
***************
*** 241,243 ****
  
  
! test_support.run_unittest(SGMLParserTestCase)
--- 307,314 ----
  
  
! def test_main():
!     test_support.run_unittest(SGMLParserTestCase)
! 
! 
! if __name__ == "__main__":
!     test_main()