def test_search_by_tag_or_tag(wa_video):
catcha = wa_video
common_tag_value = 'testing_tag_even'
common_tag = make_wa_tag(common_tag_value)
for i in [1, 2, 3, 4, 5]:
c = deepcopy(catcha)
c['id'] = '{}{}'.format(catcha['id'], i)
tag = make_wa_tag(tagname='testing_tag{}'.format(i))
c['body']['items'].append(tag)
if i % 2 == 0:
c['body']['items'].append(common_tag)
x = CRUD.create_anno(c)
payload = make_jwt_payload(user=catcha['creator']['id'])
request = make_json_request(
method='get',
# to send a list of userids
query_string='tag={}&tag={}'.format(common_tag_value, 'testing_tag3'))
request.catchjwt = payload
response = search_api(request)
resp = json.loads(response.content.decode('utf-8'))
assert response.status_code == 200
assert resp['total'] == 3
assert len(resp['rows']) == 3
for a in resp['rows']:
if not Catcha.has_tag(a, 'testing_tag3'):
assert Catcha.has_tag(a, common_tag_value) is True
def test_search_by_target_source_ok(wa_audio, wa_image):
catcha1 = wa_audio
catcha2 = wa_image
tsource = catcha1['target']['items'][0]['source']
ttype = catcha1['target']['items'][0]['type']
for i in [1, 2, 3, 4]:
for catcha in [catcha1, catcha2]:
c = deepcopy(catcha)
c['id'] = '{}{}'.format(catcha['id'], i)
x = CRUD.create_anno(c)
payload = make_jwt_payload()
request = make_json_request(
method='get', query_string='target_source={}'.format(tsource))
request.catchjwt = payload
response = search_api(request)
resp = json.loads(response.content.decode('utf-8'))
assert response.status_code == 200
assert resp['total'] == 4
for a in resp['rows']:
assert Catcha.has_target_source(a, tsource, ttype)
assert Catcha.has_target_source(a, tsource)
def handle(self, *args, **kwargs):
filepath = kwargs['filepath']
with open(filepath, 'r') as f:
annojs_list = json.load(f)
catcha_list = []
failed = []
for ajs in annojs_list:
# workaround for input with really old catch-annojs
if 'media' in ajs:
if 'ranges' in ajs and len(ajs['ranges']) > 0:
if 'start' not in ajs['ranges'][0]:
ajs['ranges'][0]['start'] = ''
ajs['ranges'][0]['end'] = ''
else:
ajs['error_msg'] = 'missing _media_ property'
failed.append(ajs)
continue # input format not acceptable
try:
catcha = Catcha.normalize(ajs)
except Exception as e:
ajs['error_msg'] = 'normalize error: {}'.format(str(e))
failed.append(ajs)
else:
catcha_list.append(catcha)
resp = {
'total_success': len(catcha_list),
'total_failed': len(failed),
'catcha_list': catcha_list,
'annojs_failed_list': failed,
}
print(json.dumps(resp, indent=4))
def test_create_compat_annojs(js_text):
js = js_text
c = Consumer._default_manager.create()
payload = make_jwt_payload(apikey=c.consumer, user=js['user']['id'])
token = make_encoded_token(c.secret_key, payload)
client = Client()
compat_create_url = reverse('compat_create')
response = client.post(compat_create_url,
data=json.dumps(js),
HTTP_X_ANNOTATOR_AUTH_TOKEN=token,
content_type='application/json')
assert response.status_code == 200
resp = json.loads(response.content.decode('utf-8'))
assert resp['id'] != js['id'] # should not preserve `id` in input annojs
assert resp['user']['id'] == payload['userId']
assert set(resp['tags']) == set(js['tags'])
assert resp['contextId'] == js['contextId']
x = Anno._default_manager.get(pk=resp['id'])
assert x.creator_id == payload['userId']
catcha = AnnoJS.convert_to_catcha(resp)
assert Catcha.are_similar(catcha, x.serialized)
def test_body_sanitize():
body_unsafe_text = [
' < script same_attr=blah other_attr="pooh"></scritp>',
'<script>',
'something <\tscript\t somethingelse="{}">'.format('blah'),
]
catcha = make_wa_object()
for b_text in body_unsafe_text:
catcha['body']['items'][0]['value'] = b_text
with pytest.raises(InvalidInputWebAnnotationError) as e:
safe = Catcha.safe_body_text_value(catcha)
catcha['body']['items'][0]['value'] = \
'body of annotation that is safe and has no script tags.'
safe = Catcha.safe_body_text_value(catcha)
assert safe
def convert_to_catcha(annojs_list):
catcha_list = []
error_list = []
for annojs in annojs_list:
try:
catcha = Catcha.normalize(annojs)
except Exception as e:
click.echo('CONVERSION ERROR: {} -- '.format(e))
error_list.append(annojs)
raise e
else:
catcha_list.append(catcha)
return (catcha_list, error_list)
def test_search_replies_catcha_ok(wa_list):
anno_list = []
for wa in wa_list:
x = CRUD.create_anno(wa)
anno_list.append(x)
c = Consumer._default_manager.create()
payload = make_jwt_payload(apikey=c.consumer)
token = make_encoded_token(c.secret_key, payload)
client = Client()
# create some replies
reply_to = anno_list[0]
wa_replies = []
for i in range(1, 5):
wa = make_wa_object(age_in_hours=1,
media=ANNO,
reply_to=reply_to.anno_id,
user=payload['userId'])
wa_replies.append(wa)
response = client.post('{}'.format(
reverse('crud_api', kwargs={'anno_id': wa['id']})),
data=json.dumps(wa),
HTTP_AUTHORIZATION='Token {}'.format(token),
content_type='application/json')
assert response.status_code == 200
# search for the replies
search_url = ('{}?context_id={}&collection_id={}&media=Annotation&'
'limit=-1&target_source_id={}').format(
reverse('create_or_search'),
reply_to.raw['platform']['context_id'],
reply_to.raw['platform']['collection_id'],
reply_to.anno_id)
response = client.get(search_url,
HTTP_AUTHORIZATION='token {}'.format(token))
assert response.status_code == 200
resp = response.json()
assert resp['total'] == 4
for catcha in resp['rows']:
assert catcha['creator']['id'] == payload['userId']
# find target with media type Annotation
target_item = Catcha.fetch_target_item_by_media(catcha, ANNO)
assert target_item is not None
assert target_item['source'] == reply_to.anno_id
def test_search_replies_js_with_uri(js_list):
anno_list = []
for js in js_list:
wa = Catcha.normalize(js)
x = CRUD.create_anno(wa)
anno_list.append(x)
c = Consumer._default_manager.create()
payload = make_jwt_payload(apikey=c.consumer)
token = make_encoded_token(c.secret_key, payload)
client = Client()
# create some replies
reply_to = anno_list[0]
js_replies = []
compat_create_url = reverse('compat_create')
for i in range(1, 5):
js = make_annotatorjs_object(age_in_hours=1,
media=ANNO,
reply_to=reply_to.anno_id,
user=payload['userId'])
js_replies.append(js)
response = client.post(compat_create_url,
data=json.dumps(js),
HTTP_X_ANNOTATOR_AUTH_TOKEN=token,
content_type='application/json')
assert response.status_code == 200
# search for the replies
search_url = ('{}?context_id={}&collection_id={}&media=comment&'
'limit=-1&uri=someFakeId&parentid={}').format(
reverse('compat_search'),
reply_to.raw['platform']['context_id'],
reply_to.raw['platform']['collection_id'],
reply_to.anno_id)
response = client.get(search_url, HTTP_X_ANNOTATOR_AUTH_TOKEN=token)
assert response.status_code == 200
resp = response.json()
assert resp['total'] == 4
for annojs in resp['rows']:
assert annojs['media'] == 'comment'
assert annojs['parent'] == reply_to.anno_id
assert annojs['user']['id'] == payload['userId']
def test_long_annotatorjs():
here = os.path.abspath(os.path.dirname(__file__))
# filename = os.path.join(here, 'annojs_3K_sorted.json')
# filename = os.path.join(here, 'annojs_HxAT101.json')
filename = os.path.join(here, 'annojs_sample_1.json')
sample = readfile_into_jsonobj(filename)
created_list = []
failed_to_create = []
client = Client()
c = Consumer._default_manager.create()
no_media = []
for js in sample:
if 'media' not in js:
no_media.append(js['id'])
failed_to_create.append(js['id'])
continue
# prep and remove insipient props for compare easiness
# js['id'] = str(js['id'])
js['uri'] = str(js['uri'])
del(js['archived'])
del(js['deleted'])
if 'citation' in js:
del(js['citation'])
if 'quote' in js and not js['quote']:
del(js['quote'])
if 'parent' not in js:
js['parent'] = '0'
if 'contextId' not in js:
js['contextId'] = 'unknown'
if 'collectionId' not in js:
js['collectionId'] = 'unknown'
payload = make_jwt_payload(
apikey=c.consumer, user=js['user']['id'])
token = make_encoded_token(c.secret_key, payload)
url = reverse('compat_create')
response = client.post(
url, data=json.dumps(js),
HTTP_X_ANNOTATOR_AUTH_TOKEN=token,
content_type='application/json')
'''
if response.status_code != 200:
print('failed to create js({}): {}\n{}'.format(
js['id'], response.content,
json.dumps(js, sort_keys=True, indent=4)))
failed_to_create.append(js['id'])
assert response.status_code == 200
else:
resp = json.loads(response.content)
'''
assert response.status_code == 200
resp = json.loads(response.content)
how_many = len(no_media)
if how_many > 0:
print('******* found ({}) annotations with no_media: {}'.format(
len(no_media), no_media))
# able to insert all annotatorjs! now comparing
counter = 0
pulled_from_db = 0
for js in sample:
if js['id'] in failed_to_create:
print('skipping not created anno({})'.format(js['id']))
continue # skip if could not create
created_anno = Anno._default_manager.get(pk=js['id'])
pulled_from_db += 1
created_js = AnnoJS.convert_from_anno(created_anno)
if AnnoJS.are_similar(js, created_js):
catcha = AnnoJS.convert_to_catcha(js)
assert Catcha.are_similar(catcha, created_anno.serialized)
else:
# sometimes the tags are repeated or not present!
js['tags'] = list(set(js['tags'])) if 'tags' in js else []
created_js['tags'] = created_js['tags'] if 'tags' in created_js else []
if AnnoJS.are_similar(js, created_js):
print('---------- AnnoJS similar({}), after tags sorted:'.format(js['id']))
else:
counter += 1
'''
print('---------- AnnoJS not similar({}):'.format(js['id']))
print('---------- (->) {}'.format(json.dumps(js,
sort_keys=True,
indent=4)))
print('---------- (<-) {}'.format(json.dumps(created_js,
sort_keys=True,
indent=4)))
'''
assert counter == 0
def test_catcha_normalize_ok(wa_image):
catch = wa_image
catcha = Catcha.normalize(catch)
assert json.dumps(catch, sort_keys=True,
indent=4) == json.dumps(catcha, sort_keys=True, indent=4)