def test_removal_of_bios(self):
# one of the biographies of person1 has been removed
# 1. the biography should not be there anymore
# 2. the bioport_id should still exist
# 3. if we have no more biographies for this person, the person should disappear
repo = self.repo
url1 = os.path.abspath(os.path.join(THIS_DIR, 'data/knaw/list.xml'))
url2 = os.path.abspath(os.path.join(THIS_DIR, 'data/knaw_changed/list.xml'))
src = Source(id=u'test', url=url1, description='knaw test dinges')
repo.add_source(src)
repo.download_biographies(src)
# we should have 5 persons from this source
self.assertEqual(len(repo.get_persons(source_id=src.id)), 5)
# one of them has id test/002
person2 = repo.get_biography(local_id='test/002').get_person()
bioport_id3 = person2.bioport_id
# now we change the biographies that are available in some of the sources
src.url = url2
repo.download_biographies(src)
# now the biography of person2 does not exist anymore
self.assertEqual(len(list(repo.get_biographies(local_id='test/002'))), 0)
# and neither does person2 himself
assert person2.bioport_id not in [p.bioport_id for p in repo.get_persons()], [(p.bioport_id, p.get_biographies()) for p in repo.get_persons()]
# we still, however, should have the bioport_id in our repository
self.assertEqual(repo.db.get_bioport_id(biography_id='test/002'), person2.bioport_id)
# now we re-download the old data
src.url = url1
repo.download_biographies(src)
# and we have the 5 old persons again
self.assertEqual(len(repo.get_persons(source_id=src.id)), 5)
# and 002 is back
self.assertEqual(len(list(repo.get_biographies(local_id='test/002'))), 1)
# it is very important that our newly re-donloaed test/002 has the same bioport id as before
self.assertEqual(repo.get_biography(local_id='test/002').get_person().bioport_id, bioport_id3)
# now, we do the same exercise of removing and re-adding, but this time we identify test/002 with test/005
person5 = repo.get_biography(local_id='test/005').get_person()
new_person = repo.identify(person5, person2)
# we have now one person less
self.assertEqual(len(repo.get_persons(source_id=src.id)), 4)
self.assertEqual(len(new_person.get_biographies()), 2)
src.url = url2
repo.download_biographies(src)
self.assertEqual(len(list(repo.get_biographies(source_id=src.id))), 5)
self.assertEqual(len(repo.get_persons(source_id=src.id)), 5, [p.get_biographies() for p in repo.get_persons(source_id=src.id)])
new_person = repo.get_person(bioport_id=new_person.bioport_id)
self.assertEqual(len(new_person.get_biographies()), 1)
def test_download_changed_bios(self):
repo = self.repo
url = os.path.abspath(os.path.join(THIS_DIR, 'data/knaw/list.xml'))
# add some sources
SOURCE_ID = u'test1'
BASE = 10
src = Source(id=SOURCE_ID, url=url, description='knaw test dinges')
repo.add_source(src)
repo.download_biographies(src)
self.assertEqual(len(repo.get_persons(source_id=SOURCE_ID)), 5)
self.assertEqual(len(repo.get_persons()), BASE + 5)
old_persons = [repo.get_person(bioport_id) for bioport_id in repo.get_bioport_ids()]
self.assertFalse([x for x in old_persons if x.status == '1'])
some_person = old_persons[0]
for person in old_persons:
# TODO: check why sometimes some_person.status is a string
self.assertEqual(person.status, STATUS_NEW)
person.record.status = STATUS_DIFFICULT
person.save()
# if we download the information at our source, nothing should have changed
repo.download_illustrations(src)
self.assertEqual(len(repo.get_persons(source_id=SOURCE_ID)), 5)
self.assertEqual(len(repo.get_persons()), BASE + 5)
self.assertEqual(set([p.bioport_id for p in old_persons]), set([p.bioport_id for p in repo.get_persons()]))
# now we donwload the biographies a second time, and again nothing should ahve changed
repo.download_biographies(src)
self.assertEqual(len(repo.get_persons(source_id=SOURCE_ID)), 5)
self.assertEqual(len(repo.get_persons()), BASE + 5)
self.assertEqual(set([p.bioport_id for p in old_persons]), set([p.bioport_id for p in repo.get_persons()]))
# also the status should remain the same
self.assertEqual(STATUS_DIFFICULT, repo.get_person(bioport_id=some_person.bioport_id).status)
# now we change the biographies that are available in some of the sources, and download again
url2 = os.path.abspath(os.path.join(THIS_DIR, 'data/knaw_changed/list.xml'))
src.url = url2
repo.download_biographies(src)
# the new url has one biograrphy less, and one new one.
# we remove the one that has disappeared, so the number of persons should now be the same as it was previously
# 1 has remained exactly the same
# 2 has disappeared
# 3 has changed location - it is found in 006.xml
# 4 has changed name
# 5 has changed location, but should have the same data
# 7 is a new entry
self.assertEqual(len(repo.get_persons(source_id=SOURCE_ID)), 5,)
# all of the old bioport_ids should still be available
for p in old_persons:
assert p.bioport_id in repo.get_bioport_ids()
# create a dictionary with the persons from our test-source, accessabilble from their ID
persons = {}
for person in repo.get_persons(source_id=SOURCE_ID):
knaw_bio = None
for bio in person.get_biographies():
if bio.get_source().id == SOURCE_ID:
knaw_bio = bio
person.knaw_bio = knaw_bio
bio_id = knaw_bio.get_id().split('/')[1]
for i in range(1, 10):
if bio_id == '00%s' % i:
persons[i] = person
# person 2 has disappeared
self.assertTrue(2 not in persons, persons)
# 3 has changed location - it is found in 006.xml.
assert persons[3].knaw_bio.source_url.endswith('006.xml'), persons[3].knaw_bio.source_url
# even in a different location, its status remains the same
self.assertEqual(persons[3].status, STATUS_DIFFICULT)
# person 4 has a changed name
assert 'changed' in persons[4].name(), persons[4].name()
# even if his name has changes, its status remains the same
self.assertEqual(persons[4].status, STATUS_DIFFICULT)
# person 5 has changed location
assert persons[5].knaw_bio.source_url.endswith('005a.xml'), persons[5].knaw_bio.source_url
self.assertEqual(persons[5].status, STATUS_DIFFICULT)
# person 7 is a new entry
self.assertEqual(persons[7].status, STATUS_NEW)
