htsworkflow/util/test/test_rdfhelp.py

   1 import os
   2 import types
   3 from unittest2 import TestCase
   4
   5 from datetime import datetime
   6
   7 from htsworkflow.util.rdfhelp import \
   8      add_default_schemas, \
   9      blankOrUri, \
  10      dcNS, \
  11      dump_model, \
  12      fromTypedNode, \
  13      get_model, \
  14      guess_parser, \
  15      guess_parser_by_extension, \
  16      load_string_into_model, \
  17      owlNS, \
  18      rdfNS, \
  19      rdfsNS, \
  20      remove_schemas, \
  21      toTypedNode, \
  22      stripNamespace, \
  23      simplify_uri, \
  24      sanitize_literal, \
  25      xsdNS
  26
  27 try:
  28     import RDF
  29
  30     class TestRDFHelp(TestCase):
  31         def test_from_none(self):
  32           self.failUnlessEqual(fromTypedNode(None), None)
  33
  34         def test_typed_node_boolean(self):
  35             node = toTypedNode(True)
  36             self.failUnlessEqual(node.literal_value['string'], u'1')
  37             self.failUnlessEqual(str(node.literal_value['datatype']),
  38                                  'http://www.w3.org/2001/XMLSchema#boolean')
  39
  40         def test_bad_boolean(self):
  41             node = RDF.Node(literal='bad', datatype=xsdNS['boolean'].uri)
  42             self.failUnlessRaises(ValueError, fromTypedNode, node)
  43
  44         def test_typed_node_string(self):
  45             node = toTypedNode('hello')
  46             self.failUnlessEqual(node.literal_value['string'], u'hello')
  47             self.failUnless(node.literal_value['datatype'] is None)
  48
  49         def test_typed_real_like(self):
  50             num = 3.14
  51             node = toTypedNode(num)
  52             self.failUnlessEqual(fromTypedNode(node), num)
  53
  54         def test_typed_integer(self):
  55             num = 3
  56             node = toTypedNode(num)
  57             self.failUnlessEqual(fromTypedNode(node), num)
  58             self.failUnlessEqual(type(fromTypedNode(node)), type(num))
  59
  60         def test_typed_node_string(self):
  61             s = "Argh matey"
  62             node = toTypedNode(s)
  63             self.failUnlessEqual(fromTypedNode(node), s)
  64             self.failUnlessEqual(type(fromTypedNode(node)), types.UnicodeType)
  65
  66         def test_blank_or_uri_blank(self):
  67             node = blankOrUri()
  68             self.failUnlessEqual(node.is_blank(), True)
  69
  70         def test_blank_or_uri_url(self):
  71             s = 'http://google.com'
  72             node = blankOrUri(s)
  73             self.failUnlessEqual(node.is_resource(), True)
  74             self.failUnlessEqual(str(node.uri), s)
  75
  76         def test_blank_or_uri_node(self):
  77             s = RDF.Node(RDF.Uri('http://google.com'))
  78             node = blankOrUri(s)
  79             self.failUnlessEqual(node.is_resource(), True)
  80             self.failUnlessEqual(node, s)
  81
  82         def test_unicode_node_roundtrip(self):
  83             literal = u'\u5927'
  84             roundtrip = fromTypedNode(toTypedNode(literal))
  85             self.failUnlessEqual(roundtrip, literal)
  86             self.failUnlessEqual(type(roundtrip), types.UnicodeType)
  87
  88         def test_datetime_no_microsecond(self):
  89             dateTimeType = xsdNS['dateTime'].uri
  90             short_isostamp = '2011-12-20T11:44:25'
  91             short_node = RDF.Node(literal=short_isostamp,
  92                                  datatype=dateTimeType)
  93             short_datetime = datetime(2011,12,20,11,44,25)
  94
  95             self.assertEqual(fromTypedNode(short_node), short_datetime)
  96             self.assertEqual(toTypedNode(short_datetime), short_node)
  97             self.assertEqual(fromTypedNode(toTypedNode(short_datetime)),
  98                              short_datetime)
  99
 100         def test_datetime_with_microsecond(self):
 101             dateTimeType = xsdNS['dateTime'].uri
 102             long_isostamp = '2011-12-20T11:44:25.081776'
 103             long_node = RDF.Node(literal=long_isostamp,
 104                                  datatype=dateTimeType)
 105             long_datetime = datetime(2011,12,20,11,44,25,81776)
 106
 107             self.assertEqual(fromTypedNode(long_node), long_datetime)
 108             self.assertEqual(toTypedNode(long_datetime), long_node)
 109             self.assertEqual(fromTypedNode(toTypedNode(long_datetime)),
 110                              long_datetime)
 111
 112         def test_strip_namespace_uri(self):
 113             nsOrg = RDF.NS('example.org/example#')
 114             nsCom = RDF.NS('example.com/example#')
 115
 116             term = 'foo'
 117             node = nsOrg[term]
 118             self.failUnlessEqual(stripNamespace(nsOrg, node), term)
 119             self.failUnlessEqual(stripNamespace(nsCom, node), None)
 120             self.failUnlessEqual(stripNamespace(nsOrg, node.uri), term)
 121
 122         def test_strip_namespace_exceptions(self):
 123             nsOrg = RDF.NS('example.org/example#')
 124             nsCom = RDF.NS('example.com/example#')
 125
 126             node = toTypedNode('bad')
 127             self.failUnlessRaises(ValueError, stripNamespace, nsOrg, node)
 128             self.failUnlessRaises(ValueError, stripNamespace, nsOrg, nsOrg)
 129
 130         def test_simplify_uri(self):
 131             DATA = [('http://asdf.org/foo/bar', 'bar'),
 132                     ('http://asdf.org/foo/bar#bleem', 'bleem'),
 133                     ('http://asdf.org/foo/bar/', 'bar'),
 134                     ('http://asdf.org/foo/bar?was=foo', 'was=foo')]
 135
 136             for uri, expected in DATA:
 137                 self.assertEqual(simplify_uri(uri), expected)
 138
 139             for uri, expected in DATA:
 140                 n = RDF.Uri(uri)
 141                 self.assertEqual(simplify_uri(n), expected)
 142
 143             for uri, expected in DATA:
 144                 n = RDF.Node(RDF.Uri(uri))
 145                 self.assertEqual(simplify_uri(n), expected)
 146
 147             # decoding literals is questionable
 148             n = toTypedNode('http://foo/bar')
 149             self.assertRaises(ValueError, simplify_uri, n)
 150
 151         def test_owl_import(self):
 152             path, name = os.path.split(__file__)
 153             #loc = 'file://'+os.path.abspath(path)+'/'
 154             loc = os.path.abspath(path)+'/'
 155             model = get_model()
 156             fragment = '''
 157 @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
 158 @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
 159 @prefix owl: <http://www.w3.org/2002/07/owl#> .
 160
 161 _:a owl:imports "{loc}extra.turtle" .
 162 '''.format(loc=loc)
 163             load_string_into_model(model, 'turtle', fragment, loc)
 164             tc = RDF.Node(RDF.Uri('http://jumpgate.caltech.edu/wiki/TestCase'))
 165             query = RDF.Statement(tc, rdfsNS['label'], None)
 166             result = list(model.find_statements(query))
 167             self.failUnlessEqual(len(result), 1)
 168             self.failUnlessEqual(str(result[0].object), 'TestCase')
 169
 170         def test_sanitize_literal_text(self):
 171             self.failUnlessRaises(ValueError, sanitize_literal, "hi")
 172             hello_text = "hello"
 173             hello_none = RDF.Node(hello_text)
 174             self.failUnlessEqual(str(sanitize_literal(hello_none)),
 175                                  hello_text)
 176             hello_str = RDF.Node(literal=hello_text,
 177                                  datatype=xsdNS['string'].uri)
 178             hello_clean = sanitize_literal(hello_str)
 179             self.failUnlessEqual(hello_clean.literal_value['string'],
 180                                  hello_text)
 181
 182         def test_sanitize_literal_empty_string(self):
 183             value = ""
 184             value_node = RDF.Node(value)
 185             self.assertEqual(str(sanitize_literal(value_node)), value)
 186
 187         def test_sanitize_literal_html(self):
 188             hello = "hello <a onload='javascript:alert(\"foo\");' href='http://google.com'>google.com</a>, whats up?"
 189             hello_clean = 'hello <a href="http://google.com">google.com</a>, whats up?'
 190             hello_node = RDF.Node(literal=hello,
 191                                   datatype=xsdNS['string'].uri)
 192             hello_sanitized = sanitize_literal(hello_node)
 193             self.failUnlessEqual(hello_sanitized.literal_value['string'],
 194                                  hello_clean)
 195
 196             hostile = "hi <b>there</b><script type='text/javascript>alert('boo');</script><a href='javascript:alert('poke')>evil</a> scammer"
 197             hostile_node = RDF.Node(hostile)
 198             hostile_sanitized = sanitize_literal(hostile_node)
 199             # so it drops the stuff after the javascript link.
 200             # I suppose it could be worse
 201             hostile_result = """hi <b>there</b>"""
 202             self.failUnlessEqual(str(hostile_sanitized), hostile_result)
 203
 204         def test_guess_parser_from_file(self):
 205             DATA = [
 206                 ('/a/b/c.rdf', 'rdfxml'),
 207                 ('/a/b/c.xml', 'rdfxml'),
 208                 ('/a/b/c.html', 'rdfa'),
 209                 ('/a/b/c.turtle', 'turtle'),
 210                 ('http://foo.bar/bleem.turtle', 'turtle')]
 211             for path, parser in DATA:
 212                 self.assertEqual(guess_parser_by_extension(path), parser)
 213                 self.assertEqual(guess_parser(None, path), parser)
 214
 215             DATA = [
 216                 ('application/rdf+xml', 'http://a.org/b/c', 'rdfxml'),
 217                 ('application/x-turtle', 'http://a.org/b/c', 'turtle'),
 218                 ('text/html', 'http://a.org/b/c', 'rdfa'),
 219                 ('text/html', 'http://a.org/b/c.html', 'rdfa'),
 220                 ('text/plain', 'http://a.org/b/c.turtle', 'turtle'),
 221                 ('text/plain', 'http://a.org/b/c', 'guess')
 222             ]
 223             for contenttype, url, parser in DATA:
 224                 self.assertEqual(guess_parser(contenttype, url), parser)
 225
 226     class TestRDFSchemas(TestCase):
 227         def test_rdf_schema(self):
 228             """Does it basically work?
 229             """
 230             model = get_model()
 231             self.assertEqual(model.size(), 0)
 232             add_default_schemas(model)
 233             self.assertTrue(model.size() > 0)
 234             remove_schemas(model)
 235             self.assertEqual(model.size(), 0)
 236
 237         def test_included_schemas(self):
 238             model = get_model()
 239             add_default_schemas(model)
 240
 241             # rdf test
 242             s = RDF.Statement(rdfNS[''], dcNS['title'], None)
 243             title = model.get_target(rdfNS[''], dcNS['title'])
 244             self.assertTrue(title is not None)
 245
 246             s = RDF.Statement(rdfNS['Property'], rdfNS['type'], rdfsNS['Class'])
 247             self.assertTrue(model.contains_statement(s))
 248
 249             # rdfs test
 250             s = RDF.Statement(rdfsNS['Class'], rdfNS['type'], rdfsNS['Class'])
 251             self.assertTrue(model.contains_statement(s))
 252
 253             s = RDF.Statement(owlNS['inverseOf'], rdfNS['type'],
 254                               rdfNS['Property'])
 255             self.assertTrue(model.contains_statement(s))
 256
 257
 258 except ImportError, e:
 259     print "Unable to test rdfhelp"
 260
 261 def suite():
 262     from unittest2 import TestSuite, defaultTestLoader
 263     suite = TestSuite()
 264     suite.addTests(defaultTestLoader.loadTestsFromTestCase(TestRDFHelp))
 265     suite.addTests(defaultTestLoader.loadTestsFromTestCase(TestRDFSchemas))
 266     return suite
 267
 268
 269 if __name__ == "__main__":
 270     from unittest2 import main
 271     main(defaultTest="suite")