htsworkflow/frontend/samples/test_samples.py

   1 import datetime
   2
   3 try:
   4     import json
   5 except ImportError, e:
   6     import simplejson as json
   7
   8 from django.test import TestCase
   9 from django.test.utils import setup_test_environment, \
  10      teardown_test_environment
  11 from django.db import connection
  12 from django.conf import settings
  13
  14 from htsworkflow.frontend.samples.models import \
  15         Affiliation, \
  16         ExperimentType, \
  17         Species, \
  18         Library
  19
  20 from htsworkflow.frontend.samples.views import \
  21      library_dict, \
  22      library_json
  23
  24 from htsworkflow.frontend.auth import apidata
  25 from htsworkflow.util.conversion import unicode_or_none
  26 from htsworkflow.util.ethelp import validate_xhtml
  27
  28 class LibraryTestCase(TestCase):
  29     fixtures = ['initial_data.json',
  30                 'woldlab.json',
  31                 'test_samples.json']
  32
  33     def setUp(self):
  34         create_db(self)
  35
  36     def testOrganism(self):
  37         self.assertEquals(self.library_10001.organism(), 'human')
  38
  39     def testAffiliations(self):
  40         self.library_10001.affiliations.add(self.affiliation_alice)
  41         self.library_10002.affiliations.add(
  42                 self.affiliation_alice,
  43                 self.affiliation_bob
  44         )
  45         self.failUnless(len(self.library_10001.affiliations.all()), 1)
  46         self.failUnless(self.library_10001.affiliation(), 'Alice')
  47
  48         self.failUnless(len(self.library_10002.affiliations.all()), 2)
  49         self.failUnless(self.library_10001.affiliation(), 'Alice, Bob')
  50
  51
  52 class SampleWebTestCase(TestCase):
  53     """
  54     Test returning data from our database in rest like ways.
  55     (like returning json objects)
  56     """
  57     fixtures = ['initial_data.json',
  58                 'woldlab.json',
  59                 'test_samples.json']
  60
  61     def test_library_info(self):
  62         for lib in Library.objects.all():
  63             lib_dict = library_dict(lib.id)
  64             url = '/samples/library/%s/json' % (lib.id,)
  65             lib_response = self.client.get(url, apidata)
  66             self.failUnlessEqual(lib_response.status_code, 200)
  67             lib_json = json.loads(lib_response.content)
  68
  69             for d in [lib_dict, lib_json]:
  70                 # amplified_from_sample is a link to the library table,
  71                 # I want to use the "id" for the data lookups not
  72                 # the embedded primary key.
  73                 # It gets slightly confusing on how to implement sending the right id
  74                 # since amplified_from_sample can be null
  75                 #self.failUnlessEqual(d['amplified_from_sample'], lib.amplified_from_sample)
  76                 self.failUnlessEqual(d['antibody_id'], lib.antibody_id)
  77                 self.failUnlessEqual(d['cell_line_id'], lib.cell_line_id)
  78                 self.failUnlessEqual(d['cell_line'], unicode_or_none(lib.cell_line))
  79                 self.failUnlessEqual(d['experiment_type'], lib.experiment_type.name)
  80                 self.failUnlessEqual(d['experiment_type_id'], lib.experiment_type_id)
  81                 self.failUnlessEqual(d['gel_cut_size'], lib.gel_cut_size)
  82                 self.failUnlessEqual(d['hidden'], lib.hidden)
  83                 self.failUnlessEqual(d['id'], lib.id)
  84                 self.failUnlessEqual(d['insert_size'], lib.insert_size)
  85                 self.failUnlessEqual(d['library_name'], lib.library_name)
  86                 self.failUnlessEqual(d['library_species'], lib.library_species.scientific_name)
  87                 self.failUnlessEqual(d['library_species_id'], lib.library_species_id)
  88                 self.failUnlessEqual(d['library_type_id'], lib.library_type_id)
  89                 if lib.library_type_id is not None:
  90                     self.failUnlessEqual(d['library_type'], lib.library_type.name)
  91                 else:
  92                     self.failUnlessEqual(d['library_type'], None)
  93                     self.failUnlessEqual(d['made_for'], lib.made_for)
  94                     self.failUnlessEqual(d['made_by'], lib.made_by)
  95                     self.failUnlessEqual(d['notes'], lib.notes)
  96                     self.failUnlessEqual(d['replicate'], lib.replicate)
  97                     self.failUnlessEqual(d['stopping_point'], lib.stopping_point)
  98                     self.failUnlessEqual(d['successful_pM'], lib.successful_pM)
  99                     self.failUnlessEqual(d['undiluted_concentration'],
 100                                          unicode(lib.undiluted_concentration))
 101                 # some specific tests
 102                 if lib.id == '10981':
 103                     # test a case where there is no known status
 104                     lane_set = {u'status': u'Unknown',
 105                                 u'paired_end': True,
 106                                 u'read_length': 75,
 107                                 u'lane_number': 1,
 108                                 u'lane_id': 1193,
 109                                 u'flowcell': u'303TUAAXX',
 110                                 u'status_code': None}
 111                     self.failUnlessEqual(len(d['lane_set']), 1)
 112                     self.failUnlessEqual(d['lane_set'][0], lane_set)
 113                 elif lib.id == '11016':
 114                     # test a case where there is a status
 115                     lane_set = {u'status': 'Good',
 116                                 u'paired_end': True,
 117                                 u'read_length': 75,
 118                                 u'lane_number': 5,
 119                                 u'lane_id': 1197,
 120                                 u'flowcell': u'303TUAAXX',
 121                                 u'status_code': 2}
 122                     self.failUnlessEqual(len(d['lane_set']), 1)
 123                     self.failUnlessEqual(d['lane_set'][0], lane_set)
 124
 125
 126     def test_invalid_library_json(self):
 127         """
 128         Make sure we get a 404 if we request an invalid library id
 129         """
 130         response = self.client.get('/samples/library/nottheone/json', apidata)
 131         self.failUnlessEqual(response.status_code, 404)
 132
 133
 134     def test_invalid_library(self):
 135         response = self.client.get('/library/nottheone/')
 136         self.failUnlessEqual(response.status_code, 404)
 137
 138
 139     def test_library_no_key(self):
 140         """
 141         Make sure we get a 302 if we're not logged in
 142         """
 143         response = self.client.get('/samples/library/10981/json')
 144         self.failUnlessEqual(response.status_code, 403)
 145         response = self.client.get('/samples/library/10981/json', apidata)
 146         self.failUnlessEqual(response.status_code, 200)
 147
 148     def test_library_rdf(self):
 149         import RDF
 150         from htsworkflow.util.rdfhelp import get_model, \
 151              dump_model, \
 152              fromTypedNode, \
 153              load_string_into_model, \
 154              rdfNS, \
 155              libraryOntology
 156         model = get_model()
 157
 158         response = self.client.get('/library/10981/')
 159         self.assertEqual(response.status_code, 200)
 160         content = response.content
 161         load_string_into_model(model, 'rdfa', content)
 162
 163         body = """prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
 164         prefix libns: <http://jumpgate.caltech.edu/wiki/LibraryOntology#>
 165
 166         select ?library ?name ?library_id ?gel_cut ?made_by
 167         where {
 168            ?library a libns:library ;
 169                     libns:name ?name ;
 170                     libns:library_id ?library_id ;
 171                     libns:gel_cut ?gel_cut ;
 172                     libns:made_by ?made_by
 173         }"""
 174         query = RDF.SPARQLQuery(body)
 175         for r in query.execute(model):
 176             self.assertEqual(fromTypedNode(r['library_id']), u'10981')
 177             self.assertEqual(fromTypedNode(r['name']),
 178                              u'Paired End Multiplexed Sp-BAC')
 179             self.assertEqual(fromTypedNode(r['gel_cut']), 400)
 180             self.assertEqual(fromTypedNode(r['made_by']), u'Igor')
 181
 182         state = validate_xhtml(content)
 183         if state is not None:
 184             self.assertTrue(state)
 185
 186         # validate a library page.
 187         from htsworkflow.util.rdfhelp import add_default_schemas
 188         from htsworkflow.util.rdfinfer import Infer
 189         add_default_schemas(model)
 190         inference = Infer(model)
 191         errmsgs = list(inference.run_validation())
 192         self.assertEqual(len(errmsgs), 0)
 193
 194     def test_library_index_rdfa(self):
 195         from htsworkflow.util.rdfhelp import \
 196              add_default_schemas, get_model, load_string_into_model, \
 197              dump_model
 198         from htsworkflow.util.rdfinfer import Infer
 199
 200         model = get_model()
 201         add_default_schemas(model)
 202         inference = Infer(model)
 203
 204         response = self.client.get('/library/')
 205         self.assertEqual(response.status_code, 200)
 206         load_string_into_model(model, 'rdfa', response.content)
 207
 208         errmsgs = list(inference.run_validation())
 209         self.assertEqual(len(errmsgs), 0)
 210
 211         body =  """prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
 212         prefix libns: <http://jumpgate.caltech.edu/wiki/LibraryOntology#>
 213
 214         select ?library ?library_id ?name ?species ?species_name
 215         where {
 216            ?library a libns:Library .
 217            OPTIONAL { ?library libns:library_id ?library_id . }
 218            OPTIONAL { ?library libns:species ?species .
 219                       ?species libns:species_name ?species_name . }
 220            OPTIONAL { ?library libns:name ?name . }
 221         }"""
 222         bindings = set(['library', 'library_id', 'name', 'species', 'species_name'])
 223         query = RDF.SPARQLQuery(body)
 224         count = 0
 225         for r in query.execute(model):
 226             count += 1
 227             for name, value in r.items():
 228                 self.assertTrue(name in bindings)
 229                 self.assertTrue(value is not None)
 230
 231         self.assertEqual(count, len(Library.objects.filter(hidden=False)))
 232
 233         state = validate_xhtml(response.content)
 234         if state is not None: self.assertTrue(state)
 235
 236
 237 # The django test runner flushes the database between test suites not cases,
 238 # so to be more compatible with running via nose we flush the database tables
 239 # of interest before creating our sample data.
 240 def create_db(obj):
 241     obj.species_human = Species.objects.get(pk=8)
 242     obj.experiment_rna_seq = ExperimentType.objects.get(pk=4)
 243     obj.affiliation_alice = Affiliation.objects.get(pk=1)
 244     obj.affiliation_bob = Affiliation.objects.get(pk=2)
 245
 246     Library.objects.all().delete()
 247     obj.library_10001 = Library(
 248         id = "10001",
 249         library_name = 'C2C12 named poorly',
 250         library_species = obj.species_human,
 251         experiment_type = obj.experiment_rna_seq,
 252         creation_date = datetime.datetime.now(),
 253         made_for = 'scientist unit 2007',
 254         made_by = 'microfludics system 7321',
 255         stopping_point = '2A',
 256         undiluted_concentration = '5.01',
 257         hidden = False,
 258     )
 259     obj.library_10001.save()
 260     obj.library_10002 = Library(
 261         id = "10002",
 262         library_name = 'Worm named poorly',
 263         library_species = obj.species_human,
 264         experiment_type = obj.experiment_rna_seq,
 265         creation_date = datetime.datetime.now(),
 266         made_for = 'scientist unit 2007',
 267         made_by = 'microfludics system 7321',
 268         stopping_point = '2A',
 269         undiluted_concentration = '5.01',
 270         hidden = False,
 271     )
 272     obj.library_10002.save()
 273
 274 try:
 275     import RDF
 276     HAVE_RDF = True
 277
 278     rdfNS = RDF.NS("http://www.w3.org/1999/02/22-rdf-syntax-ns#")
 279     xsdNS = RDF.NS("http://www.w3.org/2001/XMLSchema#")
 280     libNS = RDF.NS("http://jumpgate.caltech.edu/wiki/LibraryOntology#")
 281 except ImportError,e:
 282     HAVE_RDF = False
 283
 284
 285 class TestRDFaLibrary(TestCase):
 286     fixtures = ['initial_data.json',
 287                 'woldlab.json',
 288                 'test_samples.json']
 289
 290     def test_parse_rdfa(self):
 291         model = get_rdf_memory_model()
 292         parser = RDF.Parser(name='rdfa')
 293         url = '/library/10981/'
 294         lib_response = self.client.get(url)
 295         self.failIfEqual(len(lib_response.content), 0)
 296
 297         parser.parse_string_into_model(model,
 298                                        lib_response.content,
 299                                        'http://localhost'+url)
 300         # http://jumpgate.caltech.edu/wiki/LibraryOntology#affiliation>
 301         self.check_literal_object(model, ['Bob'], p=libNS['affiliation'])
 302         self.check_literal_object(model, ['Multiplexed'], p=libNS['experiment_type'])
 303         self.check_literal_object(model, ['400'], p=libNS['gel_cut'])
 304         self.check_literal_object(model, ['Igor'], p=libNS['made_by'])
 305         self.check_literal_object(model, ['Paired End Multiplexed Sp-BAC'], p=libNS['name'])
 306         self.check_literal_object(model, ['Drosophila melanogaster'], p=libNS['species_name'])
 307
 308         self.check_uri_object(model,
 309                               [u'http://localhost/lane/1193'],
 310                               p=libNS['has_lane'])
 311
 312         fc_uri = RDF.Uri('http://localhost/flowcell/303TUAAXX/')
 313         self.check_literal_object(model,
 314                                   [u"303TUAAXX"],
 315                                   s=fc_uri, p=libNS['flowcell_id'])
 316
 317     def check_literal_object(self, model, values, s=None, p=None, o=None):
 318         statements = list(model.find_statements(
 319             RDF.Statement(s,p,o)))
 320         self.failUnlessEqual(len(statements), len(values),
 321                         "Couln't find %s %s %s" % (s,p,o))
 322         for s in statements:
 323             self.failUnless(s.object.literal_value['string'] in values)
 324
 325
 326     def check_uri_object(self, model, values, s=None, p=None, o=None):
 327         statements = list(model.find_statements(
 328             RDF.Statement(s,p,o)))
 329         self.failUnlessEqual(len(statements), len(values),
 330                         "Couln't find %s %s %s" % (s,p,o))
 331         for s in statements:
 332             self.failUnless(unicode(s.object.uri) in values)
 333
 334
 335
 336 def get_rdf_memory_model():
 337     storage = RDF.MemoryStorage()
 338     model = RDF.Model(storage)
 339     return model
 340
 341 def suite():
 342     from unittest import TestSuite, defaultTestLoader
 343     suite = TestSuite()
 344     suite.addTests(defaultTestLoader.loadTestsFromTestCase(LibraryTestCase))
 345     suite.addTests(defaultTestLoader.loadTestsFromTestCase(SampleWebTestCase))
 346     suite.addTests(defaultTestLoader.loadTestsFromTestCase(TestRDFaLibrary))
 347     return suite
 348
 349 if __name__ == "__main__":
 350     from unittest import main
 351     main(defaultTest="suite")