def test_creation_without_api_key(client: Client) -> None:
"""Test whether creation without sending the API key whilst logged in is forbidden."""
client, headers, _ = setup_user_client(client,
is_staff=False,
is_grafeas_staff=False)
result = client.get(reverse("volunteer-list"), USER_CREATION_DATA)
assert result.status_code == status.HTTP_403_FORBIDDEN
def test_create_no_coc(self, client: Client) -> None:
"""Test that no transcription can be created without accepting the CoC."""
client, headers, user = setup_user_client(client)
user.accepted_coc = False
user.save()
submission = create_submission()
data = {
"submission_id": submission.id,
"username": user.username,
"original_id": "ABC",
"source": submission.source.name,
"url": "https://example.com",
"text": "test content",
}
result = client.post(
reverse("transcription-list"),
json.dumps(data),
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_403_FORBIDDEN
assert Transcription.objects.count() == 0
def test_with_blacklisted_user(self, client: Client) -> None:
"""Test whether a creation with a blacklisted user is rejected."""
client, headers, user = setup_user_client(client)
user.blacklisted = True
user.save()
submission = create_submission()
data = {
"submission_id": submission.id,
"username": user.username,
"original_id": "ABC",
"source": submission.source.name,
"url": "https://example.com",
"text": "test content",
}
result = client.post(
reverse("transcription-list"),
json.dumps(data),
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_423_LOCKED
assert Transcription.objects.count() == 0
def test_time_frames(
self,
client: Client,
time_frame: str,
dates: List[datetime],
results: List[Dict[str, Union[str, int]]],
) -> None:
"""Verify that the time_frame parameter properly changes the response."""
client, headers, user = setup_user_client(client, id=123456)
for date in dates:
create_transcription(
create_submission(completed_by=user, complete_time=date),
user,
create_time=make_aware(date),
)
result = client.get(
reverse("submission-rate") +
f"?time_frame={time_frame}&completed_by=123456",
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_200_OK
response = result.json()
assert response["results"] == results
def test_report_not_removed(self, client: Client) -> None:
"""Verify that reporting sends a message to Slack."""
mock = MagicMock()
slack_client.chat_postMessage = mock
client, headers, user = setup_user_client(client)
submission = create_submission(id=3)
assert not submission.removed_from_queue
assert not submission.report_reason
assert not submission.report_slack_channel_id
assert not submission.report_slack_message_ts
data = {"reason": "Violation of ALL the rules"}
result = client.patch(reverse("submission-report",
args=[submission.id]),
json.dumps(data),
content_type="application/json",
**headers)
submission.refresh_from_db()
assert result.status_code == status.HTTP_201_CREATED
assert not submission.removed_from_queue
assert submission.report_reason == "Violation of ALL the rules"
def test_ocr_on_create_with_cannot_ocr_flag(
self, client: Client, settings: SettingsWrapper) -> None:
"""Verify the OCR process exits early if the cannot_ocr flag is already set."""
settings.ENABLE_OCR = True
settings.IMAGE_DOMAINS = ["example.com"]
assert Transcription.objects.count() == 0
client, headers, _ = setup_user_client(client)
source = get_default_test_source()
data = {
"original_id": "spaaaaace",
"source": source.pk,
"content_url": "http://example.com/a.jpg",
"cannot_ocr": "True",
}
with patch("api.models.process_image", return_value={"text":
"AAA"}) as mock:
# mock it anyway just in case this fails -- we don't want to actually
# call OCR
result = client.post(
reverse("submission-list"),
data,
content_type="application/json",
**headers,
)
mock.assert_not_called()
assert result.status_code == status.HTTP_201_CREATED
assert Transcription.objects.count() == 0
def test_submission_create_with_full_args(self, client: Client) -> None:
"""Test whether creation with all arguments is successful."""
client, headers, _ = setup_user_client(client)
source = get_default_test_source()
data = {
"original_id": "spaaaaace",
"source": source.pk,
"url": "http://example.com",
"tor_url": "http://example.com/tor",
"content_url": "http://a.com",
"title": "This is a Submission",
"nsfw": False,
}
result = client.post(
reverse("submission-list"),
data,
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_201_CREATED
submission = Submission.objects.get(id=result.json()["id"])
assert submission.original_id == data["original_id"]
assert submission.source == source
assert submission.url == data["url"]
assert submission.tor_url == data["tor_url"]
assert submission.content_url == data["content_url"]
assert submission.nsfw == data["nsfw"]
assert submission.title == data["title"]
def test_post_errors(self, client: Client, error: int,
expected_redirect: str) -> None:
"""Verify that errors from the API translate correctly to responses."""
client, _, user = setup_user_client(client)
add_social_auth_to_user(user)
submission = create_submission(
original_id=int(random.random() * 1000),
content_url="http://imgur.com",
title="a",
claimed_by=user,
)
class Response:
...
response = Response()
response.status_code = error
with patch("api.views.submission.SubmissionViewSet.done",
lambda a, b, c: response):
response = client.post(
reverse("transcribe_submission",
kwargs={"submission_id": submission.id}),
data={"transcription": "AAA"},
)
assert reverse(expected_redirect) in response.url
def test_claim_already_claimed_other_user(self, client: Client) -> None:
"""Test a claim on a Submission already claimed by another user.
This should be prevented and return an error.
"""
BlossomUser.objects.all().delete()
client, headers, user = setup_user_client(client,
id=1,
username="******")
other_user = create_user(id=2, username="******")
submission = create_submission(claimed_by=other_user)
data = {"username": user.username}
result = client.patch(
reverse("submission-claim", args=[submission.id]),
json.dumps(data),
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_409_CONFLICT
claimed_by = result.json()
assert claimed_by["id"] == 2
assert claimed_by["username"] == "user_2"
submission.refresh_from_db()
assert submission.claimed_by == other_user
def test_summary_blacklisted_user(self, client: Client) -> None:
"""Test that a blacklisted user is reported as having 0 gamma."""
client, headers, user = setup_user_client(client)
user.blacklisted = True
user.save()
for _ in range(3):
create_submission(completed_by=user)
assert Submission.objects.filter(completed_by=user).count() == 3
result = client.get(
reverse("volunteer-summary") + f"?username={user.username}",
**headers)
assert result.status_code == status.HTTP_200_OK
assert result.json().get("username") == user.username
assert result.json()["gamma"] == 0
user.blacklisted = False
user.save()
result = client.get(
reverse("volunteer-summary") + f"?username={user.username}",
**headers)
assert result.json()["gamma"] == 3
def test_check_reddit_for_missing_information(
self, client: Client, settings: SettingsWrapper) -> None:
"""Verify that if information is missing we will check Reddit for it."""
client, _, user = setup_user_client(client)
add_social_auth_to_user(user)
settings.ENABLE_REDDIT = True
class RedditSubmission:
class Response:
over_18 = True
title = "AAA"
def submission(self, **kwargs: Any) -> Response:
"""Return a mocked response from Reddit."""
return self.Response()
with patch("app.middleware.configure_reddit",
lambda a: RedditSubmission()):
submission = create_submission(
original_id=int(random.random() * 1000),
content_url="http://imgur.com",
)
client.get(reverse("choose_transcription"))
submission.refresh_from_db()
assert submission.title == "AAA"
assert submission.nsfw is True
def test_is_returning_transcriber(
self,
client: Client,
recent_gamma: int,
total_gamma: int,
expected: bool,
) -> None:
"""Test whether returning transcribers are determined correctly."""
# Mock both the total gamma
with patch(
"authentication.models.BlossomUser.gamma",
new_callable=PropertyMock,
return_value=total_gamma,
):
# Mock the Slack client to catch the sent messages by the function under test.
client, headers, user = setup_user_client(client)
now = datetime.datetime.now(tz=pytz.UTC)
# Create the recent transcriptions
for i in range(0, recent_gamma):
submission = create_submission(
id=i + 100,
claimed_by=user,
completed_by=user,
complete_time=now,
)
create_transcription(submission,
user,
id=i + 100,
create_time=now)
assert _is_returning_transcriber(user) == expected
def test_list_with_null_filters(
self, client: Client, filter_str: str, result_count: int
) -> None:
"""Verify that attributes can be filtered by null."""
client, headers, user = setup_user_client(client, id=123)
today = timezone.now()
create_submission(
id=1,
claimed_by=user,
completed_by=user,
claim_time=today,
complete_time=today,
title="Test Submission",
url="https://example.org",
tor_url="https://example.org",
content_url="https://example.org",
redis_id="abc",
)
create_submission(id=2)
result = client.get(
reverse("submission-list") + f"?{filter_str}",
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_200_OK
assert len(result.json()["results"]) == result_count
def test_creation_with_normal_user(client: Client) -> None:
"""Test whether creation is not allowed to a user which is not a staff member."""
client, headers, _ = setup_user_client(client,
is_grafeas_staff=False,
is_staff=False)
result = client.post(reverse("volunteer-list"), USER_CREATION_DATA)
assert result.status_code == status.HTTP_403_FORBIDDEN
def test_random_none_available(self, client: Client) -> None:
"""Test that no transcription is returned when there is none available."""
client, headers, user = setup_user_client(client)
result = client.get(reverse("transcription-review-random"), **headers)
assert not result.content
assert result.status_code == status.HTTP_200_OK
def test_list_with_contains_filters(self, client: Client, filter_str: str,
result_count: int) -> None:
"""Test whether the transcription text can be searched."""
client, headers, user = setup_user_client(client, id=123)
submission = create_submission(id=1)
create_transcription(
submission,
user,
id=2,
text="This is a very interesting text and such.",
)
create_transcription(
submission,
user,
id=3,
text="A text is a form of literature.",
)
create_transcription(
submission,
user,
id=4,
text="Bla bla bla bla.",
)
result = client.get(
reverse("transcription-list") + f"?{filter_str}",
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_200_OK
assert len(result.json()["results"]) == result_count
def test_list_with_null_filters(self, client: Client, filter_str: str,
result_count: int) -> None:
"""Verify that filtering for null works correctly."""
client, headers, user = setup_user_client(client, id=123)
submission = create_submission(id=1)
create_transcription(
submission,
user,
id=2,
original_id="abc",
url="https://example.org",
text="Test Transcription",
)
create_transcription(submission,
user,
id=3,
original_id=None,
url=None,
text=None)
result = client.get(
reverse("transcription-list") + f"?{filter_str}",
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_200_OK
assert len(result.json()["results"]) == result_count
def test_list_with_time_filters(self, client: Client) -> None:
"""Verify that the transcriptions can be filtered by time."""
client, headers, user = setup_user_client(client)
dates = [
datetime(2021, 1, 1),
datetime(2021, 2, 1),
datetime(2021, 2, 3),
datetime(2021, 5, 10),
]
for date in dates:
create_transcription(create_submission(),
user,
create_time=make_aware(date))
result = client.get(
reverse("transcription-list") +
"?create_time__gte=2021-02-01T00:00:00Z" +
"&create_time__lte=2021-04-01T00:00:00Z",
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_200_OK
assert len(result.json()["results"]) == 2
def test_failed_ocr_on_create(self, client: Client,
settings: SettingsWrapper) -> None:
"""Verify that a new submission completes the OCR process."""
settings.ENABLE_OCR = True
settings.IMAGE_DOMAINS = ["example.com"]
assert Transcription.objects.count() == 0
client, headers, _ = setup_user_client(client)
source = get_default_test_source()
data = {
"original_id": "spaaaaace",
"source": source.pk,
"content_url": "http://example.com/a.jpg",
}
with patch("api.models.process_image", return_value=None) as mock:
result = client.post(
reverse("submission-list"),
data,
content_type="application/json",
**headers,
)
mock.assert_called_once()
assert result.status_code == status.HTTP_201_CREATED
assert Transcription.objects.count() == 0
assert result.json().get("cannot_ocr") is True
def test_create(self, client: Client) -> None:
"""Test whether the creation functions correctly when invoked correctly."""
client, headers, user = setup_user_client(client)
submission = create_submission()
data = {
"submission_id": submission.id,
"username": user.username,
"original_id": "ABC",
"source": submission.source.name,
"url": "https://example.com",
"text": "test content",
}
result = client.post(
reverse("transcription-list"),
json.dumps(data),
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_201_CREATED
transcription = Transcription.objects.get(id=result.json()["id"])
assert transcription.submission == submission
assert transcription.source == submission.source
assert transcription.author == user
assert transcription.original_id == data["original_id"]
assert transcription.url == data["url"]
assert transcription.text == data["text"]
def test_rate_count_aggregation(
self,
client: Client,
data: List[Dict[str, Union[str, int]]],
url_additions: str,
different_result: List[Dict],
) -> None:
"""Test if the number of transcriptions per day is aggregated correctly."""
client, headers, user = setup_user_client(client, id=123456)
for obj in data:
date = make_aware(
datetime.strptime(obj.get("date"), "%Y-%m-%dT%H:%M:%SZ"))
for _ in range(obj.get("count")):
create_transcription(
create_submission(completed_by=user, complete_time=date),
user,
create_time=date,
)
if not url_additions:
url_additions = "?completed_by=123456"
else:
url_additions += "&completed_by=123456"
result = client.get(
reverse("submission-rate") + url_additions,
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_200_OK
rates = result.json()["results"]
if different_result:
assert rates == different_result
else:
assert rates == data
def test_create_with_tz_aware_timestamp(self, client: Client) -> None:
"""Test whether the creation functions correctly when invoked correctly."""
# TODO: Remove me when we remove the ability to create with create_time variable
client, headers, user = setup_user_client(client)
submission = create_submission()
timestamp = "2021-11-28T13:00:05.985314+00:00"
data = {
"submission_id": submission.id,
"username": user.username,
"original_id": "ABC",
"create_time": timestamp,
"source": submission.source.name,
"url": "https://example.com",
"text": "test content",
}
result = client.post(
reverse("transcription-list"),
json.dumps(data),
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_201_CREATED
transcription = Transcription.objects.get(id=result.json()["id"])
assert transcription.create_time.isoformat() == timestamp
def test_filtered_leaderboard(self, client: Client,) -> None:
"""Test if the submissions for the rate is calculated on are filtered."""
BlossomUser.objects.all().delete()
date_joined = datetime(2021, 11, 3, tzinfo=pytz.UTC)
client, headers, user = setup_user_client(
client, id=1, username="******", date_joined=date_joined, is_volunteer=True
)
# Submissions before filter
for _ in range(4):
create_submission(
completed_by=user, complete_time=make_aware(datetime(2021, 11, 3))
)
# Submissions after filter
for _ in range(7):
create_submission(
completed_by=user, complete_time=make_aware(datetime(2021, 11, 5))
)
results = client.get(
reverse("submission-leaderboard")
+ "?user_id=1&complete_time__gte=2021-11-04T00:00:00Z",
content_type="application/json",
**headers,
)
assert results.status_code == status.HTTP_200_OK
results = results.json()
assert results["user"] == {
"id": 1,
"username": "******",
"gamma": 7,
"rank": 1,
"date_joined": "2021-11-03T00:00:00Z",
}
def test_search_with_ordering_filter(
self,
client: Client,
ordering: str,
create_times: List[datetime],
expected_times: List[datetime],
) -> None:
"""Verify that listing items with specified orderings works correctly."""
client, headers, user = setup_user_client(client)
for time in create_times:
create_transcription(create_time=make_aware(time),
submission=create_submission(),
user=user)
result = client.get(
reverse("transcription-list") + f"?ordering={ordering}",
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_200_OK
result_times = [
datetime.strptime(obj.get("create_time"), "%Y-%m-%dT%H:%M:%SZ")
for obj in result.json()["results"]
]
assert result_times == expected_times
def test_list_with_filters(self, client: Client) -> None:
"""Verify that listing all submissions works correctly."""
Source.objects.all().delete() # clear out system ones for test
client, headers, _ = setup_user_client(client)
Source.objects.get_or_create(name="AAA")
Source.objects.get_or_create(name="BBB")
Source.objects.get_or_create(name="CCC")
result = client.get(reverse("source-list"),
content_type="application/json",
**headers)
assert result.status_code == status.HTTP_200_OK
assert len(result.json()["results"]) == 3
result = client.get(
reverse("source-list") + "?name=AAA",
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_200_OK
assert len(result.json()["results"]) == 1
assert result.json()["results"][0]["name"] == "AAA"
def test_in_progress_custom_time(self, client: Client) -> None:
"""Verify that passing a different time changes the returned submissions."""
client, headers, user = setup_user_client(client)
reddit, _ = Source.objects.get_or_create(name="reddit")
submission = create_submission(
claimed_by=user,
claim_time=timezone.now() - timezone.timedelta(hours=2),
source=reddit,
)
# will default to four hours, should return nothing
result = client.get(
reverse("submission-in-progress") + "?source=reddit",
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_200_OK
assert len(result.json()) == 0
# we'll set a custom hour and now we should get the submission
result = client.get(
reverse("submission-in-progress") + "?source=reddit&hours=1",
content_type="application/json",
**headers,
)
assert result.status_code == status.HTTP_200_OK
assert len(result.json()) == 1
assert result.json()[0]["id"] == submission.id
def test_find_in_progress(client: Client, url: str, expected: bool) -> None:
"""Verify that an in-progress posts can be found by its URLs."""
client, headers, user = setup_user_client(client,
id=123,
username="******")
submission = create_submission(
claimed_by=user,
url="https://reddit.com/r/antiwork/comments/q1tlcf/work_is_work/",
tor_url=
"https://reddit.com/r/TranscribersOfReddit/comments/q1tnhc/antiwork_image_work_is_work/",
content_url="https://i.redd.it/upwchc4bqhr71.jpg",
title="Work is work",
)
result = client.get(
reverse("find") + f"?url={url}",
content_type="application/json",
**headers,
)
if expected:
assert result.status_code == status.HTTP_200_OK
actual = result.json()
assert actual["submission"]["id"] == submission.id
assert actual["author"]["id"] == user.id
else:
assert result.status_code == status.HTTP_404_NOT_FOUND
def test_transcribot_limit_param(self, client: Client) -> None:
"""Verify that adding the `limit` QSP modifies the results."""
client, headers, _ = setup_user_client(client)
user_model = get_user_model()
transcribot = user_model.objects.get(username="******")
submission1 = create_submission(source="reddit", original_id="A")
submission2 = create_submission(source="reddit", original_id="B")
submission3 = create_submission(source="reddit", original_id="C")
create_transcription(submission1, transcribot, original_id=None)
create_transcription(submission2, transcribot, original_id=None)
create_transcription(submission3, transcribot, original_id=None)
result = client.get(
reverse("submission-get-transcribot-queue") +
"?source=reddit&limit=none",
content_type="application/json",
**headers,
).json()
assert len(result["data"]) == 3
result = client.get(
reverse("submission-get-transcribot-queue") +
"?source=reddit&limit=1",
content_type="application/json",
**headers,
).json()
assert len(result["data"]) == 1
assert result["data"][0]["id"] == submission1.id
def test_get_transcribot_queue(self, client: Client) -> None:
"""Test that the OCR queue endpoint returns the correct data."""
client, headers, _ = setup_user_client(client)
user_model = get_user_model()
transcribot = user_model.objects.get(username="******")
result = client.get(
reverse("submission-get-transcribot-queue") + "?source=reddit",
content_type="application/json",
**headers,
).json()
assert len(result["data"]) == 0 # there are no posts to work on
submission = create_submission(source="reddit")
result = client.get(
reverse("submission-get-transcribot-queue") + "?source=reddit",
content_type="application/json",
**headers,
).json()
# now there's a submission without a transcribot transcription
assert len(result["data"]) == 0
create_transcription(submission, transcribot, original_id=None)
result = client.get(
reverse("submission-get-transcribot-queue") + "?source=reddit",
content_type="application/json",
**headers,
).json()
# now the submission has a transcribot entry
assert len(result["data"]) == 1
def test_yeet_with_no_user(self, client: Client) -> None:
"""Verify that yeeting without a count only deletes one."""
client, headers, user = setup_user_client(client)
response = client.post(reverse("submission-yeet"),
content_type="application/json",
**headers)
assert response.status_code == 400
