async def test_get_users_by_id(writer: TwitterDataWriter, reader: TwitterDataReader):
conversation_id = "users by id test"
obama_message = generate_messages(
1, random_datestring(), "", conversation_id, OBAMA
)[0]
dog_rates_message = generate_messages(
1, random_datestring(), "", conversation_id, DOG_RATES
)[0]
amazing_phil_message = generate_messages(
1, random_datestring(), "", conversation_id, AMAZINGPHIL
)[0]
for message in (obama_message, dog_rates_message, amazing_phil_message):
writer.add_message(message, True)
await writer.finalize()
assert reader.get_users_by_id((OBAMA, AMAZINGPHIL)) == [
ArchivedUserSummary(
str(OBAMA),
"",
str(OBAMA),
DBRead.DEFAULT_DISPLAY_NAME,
DBRead.USER_AVATAR_DEFAULT_URL,
False,
False,
),
ArchivedUserSummary(
str(AMAZINGPHIL),
"",
str(AMAZINGPHIL),
DBRead.DEFAULT_DISPLAY_NAME,
DBRead.USER_AVATAR_DEFAULT_URL,
False,
False,
),
]
assert reader.get_users_by_id((DOG_RATES,), sidecar=False) == [
ArchivedUser(
str(DOG_RATES),
"",
"dog_rates",
"We Rate Dogs",
f"{DBRead.AVATAR_API_URL}{DOG_RATES}.jpg",
True,
False,
1,
"sample bio",
"",
dog_rates_message["createdAt"],
dog_rates_message["createdAt"],
)
]
async def test_conversations_by_user(
writer: TwitterDataWriter, reader: TwitterDataReader
):
"""create a conversation or two with random users and then like 30 with a
particular user. make sure they come back in the right order. this test also
doesn't bother testing pagination"""
dummy_convo = generate_messages(
2,
random_2000s_datestring(),
random_2010s_datestring(),
"won't be returned",
OBAMA,
AMAZINGPHIL,
)
messages_with_dog_rates = []
for how_many in range(30, 0, -1):
messages_with_dog_rates += generate_messages(
how_many,
random_2000s_datestring(),
random_2010s_datestring(),
get_random_text(),
DOG_RATES,
)
descending_ids = []
prev_id = ""
for message in messages_with_dog_rates:
if message["conversationId"] != prev_id:
descending_ids.append(message["conversationId"])
prev_id = message["conversationId"]
shuffle(messages_with_dog_rates)
for message in dummy_convo:
writer.add_message(message)
for message in messages_with_dog_rates:
writer.add_message(message, True)
await writer.finalize()
assert [
c.id for c in reader.get_conversations_by_user(DOG_RATES, 1)
] == descending_ids[: DBRead.CONVERSATIONS_PER_PAGE]
async def test_onesided_individual_conversation(writer: TwitterDataWriter):
conversation_id = "one-sided conversation"
talking_user = DOG_RATES
start_time = "2010-06-06T01:00:00.543Z"
end_time = "2010-09-06T01:00:00.345Z"
messages = generate_messages(
5, start_time, end_time, conversation_id, talking_user, MAIN_USER_ID
)
for message in messages:
writer.add_message(message)
await writer.finalize()
assert writer.execute(
"select * from conversations where id=?;", (conversation_id,)
).fetchone() == (
conversation_id,
"individual",
None,
5,
0,
int(talking_user),
start_time,
end_time,
0,
None,
2,
0,
)
assert check_dog_rates(writer) == 5
assert writer.execute(
"select * from participants where conversation=? and participant=?;",
(conversation_id, int(talking_user)),
).fetchone() == (
int(talking_user),
conversation_id,
5,
None,
None,
None,
)
assert writer.execute(
"select * from participants where conversation=? and participant=?;",
(conversation_id, MAIN_USER_ID),
).fetchone() == (
MAIN_USER_ID,
conversation_id,
0,
None,
None,
None,
)
async def test_othersided_individual_conversation(writer: TwitterDataWriter):
start_time = "2015-10-10T07:07:07.777Z"
end_time = "2020-10-10T07:07:07.777Z"
conversation_id = "you-talking"
messages = generate_messages(
8, start_time, end_time, conversation_id, MAIN_USER_ID, DOG_RATES
)
for message in messages:
writer.add_message(message)
await writer.finalize()
assert writer.execute(
"select * from conversations where id=?;", (conversation_id,)
).fetchone() == (
conversation_id,
"individual",
None,
8,
8,
DOG_RATES,
start_time,
end_time,
1,
None,
2,
0,
)
assert check_dog_rates(writer) == 0
assert writer.execute(
"select * from participants where conversation=? and participant=?;",
(conversation_id, DOG_RATES),
).fetchone() == (
DOG_RATES,
conversation_id,
0,
None,
None,
None,
)
assert writer.execute(
"select * from participants where conversation=? and participant=?;",
(conversation_id, MAIN_USER_ID),
).fetchone() == (
MAIN_USER_ID,
conversation_id,
8,
None,
None,
None,
)
async def test_set_user_nicknotes(
writer: TwitterDataWriter, reader: TwitterDataReader
):
writer.add_message(
generate_messages(1, random_2010s_datestring(), "", "whatevs", OBAMA)[0],
True,
)
await writer.finalize()
reader.set_user_nickname(OBAMA, "obama")
assert reader.get_users_by_id((OBAMA,))[0].nickname == "obama"
reader.set_user_notes(OBAMA, "was president")
assert reader.get_users_by_id((OBAMA,), False)[0].notes == "was president"
async def test_conversation_notes(
writer: TwitterDataWriter, reader: TwitterDataReader
):
writer.add_message(
generate_messages(
1, random_2000s_datestring(), "", "a-conversation", MAIN_USER_ID
)[0],
True,
)
await writer.finalize()
reader.set_conversation_notes("a-conversation", "this is a conversation")
assert (
reader.get_conversation_by_id("a-conversation").notes
== "this is a conversation"
)
async def test_text_search(writer: TwitterDataWriter, reader: TwitterDataReader):
texts = ("i am home", "home i am", "being at home", "home is here")
messages = generate_messages(
20,
random_2000s_datestring(),
random_2010s_datestring(),
"searchyconvo",
MAIN_USER_ID,
AMAZINGPHIL,
)
for i in range(len(texts)):
messages[i]["text"] = texts[i]
for message in messages:
writer.add_message(message)
await writer.finalize()
results = reader.traverse_messages(after="beginning", search="home")["results"]
assert len(results) == len(texts)
async def test_conversations_by_time(
writer: TwitterDataWriter, reader: TwitterDataReader
):
page_length = DBRead.CONVERSATIONS_PER_PAGE
group_ids = [get_random_text() for _ in range(page_length + 2)]
group_messages = sum(
[
generate_messages(
2,
random_2000s_datestring(),
random_2010s_datestring(),
group_id,
10101,
)
for group_id in group_ids
],
[],
)
individual_ids = [get_random_text() for _ in range(page_length + 2)]
individual_messages = sum(
[
generate_messages(
2,
random_2000s_datestring(),
random_2010s_datestring(),
individual_id,
MAIN_USER_ID,
next(unique_id),
)
for individual_id in individual_ids
],
[],
)
messages = individual_messages + group_messages
# since every conversation has two messages, every other message starting with
# the first one is the first in a conversation, and every other message starting
# with the second one is the last; the firsts will end up in ascending_ids and
# the lasts in descending_ids
ascending_ids = [
m["conversationId"]
for m in sorted(messages[::2], key=lambda x: x["createdAt"])
]
descending_ids = [
m["conversationId"]
for m in sorted(messages[1::2], key=lambda x: x["createdAt"], reverse=True)
]
# the same conversation ids should be in both lists; just sorted according to
# different times
assert set(ascending_ids) == set(descending_ids)
for message in individual_messages:
writer.add_message(message)
for message in group_messages:
writer.add_message(message, True)
await writer.finalize()
# check !group and !individual results:
assert reader.get_conversations_by_time(1, group=False, individual=False) == []
# check group + individual conversation results:
asc_conversations_p1 = reader.get_conversations_by_time(1)
assert ascending_ids[0:page_length] == [x.id for x in asc_conversations_p1]
asc_conversations_p2 = reader.get_conversations_by_time(2)
assert ascending_ids[page_length : page_length * 2] == [
x.id for x in asc_conversations_p2
]
dsc_conversations = reader.get_conversations_by_time(1, asc=False)
assert [x.id for x in dsc_conversations] == descending_ids[0:page_length]
# check individual conversation results:
iac = reader.get_conversations_by_time(1, group=False)
assert [i.id for i in iac] == [x for x in ascending_ids if x in individual_ids][
0:page_length
]
# check group conversation results:
gdc = reader.get_conversations_by_time(2, asc=False, individual=False)
assert [i.id for i in gdc] == [x for x in descending_ids if x in group_ids][
page_length : page_length * 2
]
async def test_normal_individual_conversation(connected_writer: TwitterDataWriter):
users = (MAIN_USER_ID, DOG_RATES)
user0_span = ("2020-01-01T10:00:00.100Z", "2020-01-10T10:00:00.100Z")
user1_span = ("2020-01-02T10:00:00.100Z", "2020-01-09T10:00:00.100Z")
conversation_id = "simple_conversation"
side1 = generate_messages(
10,
user0_span[0],
user0_span[1],
conversation_id,
users[0],
users[1],
)
side2 = generate_messages(
10,
user1_span[0],
user1_span[1],
conversation_id,
users[1],
users[0],
)
for message in sorted(side1 + side2, key=lambda x: x["createdAt"]):
connected_writer.add_message(message)
await connected_writer.finalize()
assert connected_writer.execute(
"select * from conversations where id=?;", (conversation_id,)
).fetchone() == (
conversation_id,
"individual",
None,
20,
10,
int(users[1]),
user0_span[0],
user0_span[1],
1,
None,
2,
0,
)
assert check_dog_rates(connected_writer) == 10
assert connected_writer.execute(
"select * from participants where conversation=? and participant=?;",
(conversation_id, int(users[0])),
).fetchone() == (
int(users[0]),
conversation_id,
10,
None,
None,
None,
)
assert connected_writer.execute(
"select * from participants where conversation=? and participant=?;",
(conversation_id, int(users[1])),
).fetchone() == (
int(users[1]),
conversation_id,
10,
None,
None,
None,
)
