Unit tests for RFC4518 stringprep

tests/test_stringprep.py

@@ -5,12 +5,37 @@ from ldapserver.rfc4518_stringprep import prepare, MatchingType, SubstringType
 
 class TestStringprep(unittest.TestCase):
 	def test_map(self):
+		# [...] COMBINING GRAPHEME JOINER (U+034F) [...] code points are also mapped to nothing.
+		self.assertEqual(prepare(' foo\u034Fbar ', MatchingType.EXACT_STRING, SubstringType.NONE), ' foobar ')
+		# [...] LINE FEED (LF) (U+000A) [...] are mapped to SPACE (U+0020).
 		self.assertEqual(prepare(' foo\n\rbar ', MatchingType.EXACT_STRING, SubstringType.NONE), ' foo  bar ')
-		self.assertEqual(prepare('foo\u200Bbar', MatchingType.EXACT_STRING, SubstringType.NONE), ' foobar ')
+		# For case ignore, numeric, and stored prefix string matching rules,
+		# characters are case folded per B.2 of [RFC3454].
+		self.assertEqual(prepare(' FooBar ', MatchingType.CASE_IGNORE_STRING, SubstringType.NONE), ' foobar ')
+		# Not that a valid numeric string can contain any characters affected by case-folding?!
+		self.assertEqual(prepare('FooBar', MatchingType.NUMERIC_STRING, SubstringType.NONE), 'foobar')
 
-	# TODO: systematic test cases
+	def test_normalize(self):
+		self.assertEqual(prepare(' \u00C5 ', MatchingType.EXACT_STRING, SubstringType.NONE), ' \u00C5 ')
+		self.assertEqual(prepare(' \u212B ', MatchingType.EXACT_STRING, SubstringType.NONE), ' \u00C5 ')
+		self.assertEqual(prepare(' \u0041\u030A ', MatchingType.EXACT_STRING, SubstringType.NONE), ' \u00C5 ')
+
+	def test_check_prohibited(self):
+		with self.assertRaises(ValueError):
+			prepare(' foo \uFFFD bar ', MatchingType.EXACT_STRING, SubstringType.NONE)
+
+	def test_insignificant_characters(self):
+		self.assertEqual(prepare('foo bar', MatchingType.EXACT_STRING, SubstringType.NONE), ' foo  bar ')
+		# Test special case of SPACE followed by combining mark
+		self.assertEqual(prepare('foo \u030A bar', MatchingType.EXACT_STRING, SubstringType.NONE), ' foo \u030A  bar ')
+		self.assertEqual(prepare(' \u030A foobar', MatchingType.EXACT_STRING, SubstringType.NONE), '  \u030A  foobar ')
+		self.assertEqual(prepare('foobar \u030A', MatchingType.EXACT_STRING, SubstringType.NONE), ' foobar \u030A ')
+		# Not that a numeric string or a telephone number can contain any combining
+		# marks, but the RFC says that SPACES followed by combining marks are
+		# special, so ...?!
+		self.assertEqual(prepare('foo \u030A bar', MatchingType.NUMERIC_STRING), 'foo \u030Abar')
+		self.assertEqual(prepare('foo \u030A bar', MatchingType.TELEPHONE_NUMBER), 'foo \u030Abar')
 
-	def test_examples(self):
 		# Examples from RFC4518 for "Insignificant Character Handling"
 		self.assertEqual(prepare('foo bar  ', MatchingType.EXACT_STRING, SubstringType.NONE), ' foo  bar ')
 		self.assertEqual(prepare('foo bar  ', MatchingType.EXACT_STRING, SubstringType.INITIAL), ' foo  bar ')