htsworkflow/frontend/samples/test_samples.py

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