htsworkflow/util/test/test_rdfhelp.py

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