def test_mailman_chain():
url = "http://mail.scipy.org/pipermail/ipython-dev/"
arx = archive.Archive(url)
arx.save("test.csv")
arx2 = archive.load("test.csv")
arx3 = archive.Archive("ipython-dev")
assert arx.data.shape == arx.data.shape
os.remove("test.csv")
def test_mailman_chain():
name = "bigbang-dev-test.txt"
#archive loaded from mbox
arx = archive.Archive(name, archive_dir="tests/data", mbox=True)
arx.save("test.csv")
#archive loaded from stored csv
arx2 = archive.load("test.csv")
print arx.data.dtypes
print arx.data.shape
assert arx.data.shape == arx2.data.shape, \
"Original and restored archives are different shapes"
assert (arx2.data.index == arx.data.index).all(), \
"Original and restored archives have nonidentical indices"
assert [t.get_num_messages() for t in arx.get_threads()] == [3,1,2], \
"Thread message count in mbox archive is off"
assert [t.get_num_messages() for t in arx2.get_threads()] == [3,1,2], \
"Thread message count in restored archive is off"
# smoke test entity resolution
arx2.resolve_entities()
os.remove("test.csv")
def test_empty_list_compute_activity_issue_246():
test_df_csv_path = os.path.join(CONFIG.test_data_path,
'empty-archive-df.csv')
df = pd.read_csv(test_df_csv_path)
with assert_raises(mailman.MissingDataException):
empty_archive = archive.Archive(df)
activity = empty_archive.get_activity()
def test__empty_list_compute_activity_issue_246(self):
test_df_csv_path = os.path.join(
CONFIG.test_data_path, "empty-archive-df.csv"
)
df = pd.read_csv(test_df_csv_path)
with self.assertRaises(archive.MissingDataException):
empty_archive = archive.Archive(df)
empty_archive.get_activity()
def test_email_entity_resolution(self):
name = "2001-November.txt"
arx = archive.Archive(name, archive_dir="tests/data", mbox=True)
e = process.resolve_sender_entities(arx.get_activity(resolved=False))
utils.repartition_dataframe(arx.get_activity(), e)
self.assertTrue(True, msg="email entity resolution crashed")
def test_clean_message():
name = "2001-November.txt"
arx = archive.Archive(name, archive_dir="tests/data", mbox=True)
body = arx.data['Body']['<*****@*****.**>']
assert "But seemingly it is even stranger than this." in body, \
"Selected wrong message"
assert "Is it a problem of lapack3.0 of of" in body, \
"Quoted text is not in uncleaned message"
assert "Is it a problem of lapack3.0 of of" not in utils.clean_message(body), \
"Quoted text is in cleaned message"
def test_mailman_chain():
url = "http://mail.scipy.org/pipermail/ipython-dev/"
mailman.collect_from_url(url)
mailman.unzip_archive(url)
arx = archive.Archive(url)
arx.save("test.csv")
arx2 = archive.load("test.csv")
assert arx.data.shape == arx.data.shape
os.remove("test.csv")
def test__open_list_archives(self):
data = archive.open_list_archives(
archive_name="3GPP_TSG_SA_WG4_EVS.mbox",
archive_dir=CONFIG.test_data_path + "3GPP_mbox/",
mbox=True,
)
assert len(data.columns.values) == 6
assert len(data.index.values) == 50
data = archive.open_list_archives(
archive_name="3GPP_mbox",
archive_dir=CONFIG.test_data_path,
mbox=False,
)
assert len(data.columns.values) == 6
assert len(data.index.values) == 108
## Testing add_affilation
rel_email_affil = pd.DataFrame.from_records(
[
{
"email": "*****@*****.**",
"affiliation": "TestOrg",
"min_date": dateutil.parser.parse("2019-04-18"),
"max_date": dateutil.parser.parse("2021-04-18"),
}
]
)
arx = archive.Archive(data)
arx.add_affiliation(rel_email_affil)
assert (
arx.data.loc["[email protected]"][
"affiliation"
]
== "TestOrg"
)
assert (
arx.data.loc["*****@*****.**"]["affiliation"]
is None
)
def test_mailman_chain(self):
name = "bigbang-dev-test.txt"
# archive loaded from mbox
arx = archive.Archive(name,
archive_dir=CONFIG.test_data_path,
mbox=True)
arx.save("test.csv")
# archive loaded from stored csv
arx2 = archive.load("test.csv")
print(arx.data.dtypes)
print(arx.data.shape)
self.assertTrue(
arx.data.shape == arx2.data.shape,
msg="Original and restored archives are different shapes",
)
self.assertTrue(
(arx2.data.index == arx.data.index).all(),
msg="Original and restored archives have nonidentical indices",
)
self.assertTrue(
[t.get_num_messages() for t in arx.get_threads()] == [3, 1, 2],
msg="Thread message count in mbox archive is off",
)
self.assertTrue(
[t.get_num_messages() for t in arx2.get_threads()] == [3, 1, 2],
msg="Thread message count in restored archive is off",
)
# smoke test entity resolution
arx2.resolve_entities()
os.remove("test.csv")
def test_clean_message(self):
name = "2001-November.txt"
arx = archive.Archive(name,
archive_dir=CONFIG.test_data_path,
mbox=True)
body = arx.data["Body"]["<*****@*****.**>"]
self.assertTrue(
"But seemingly it is even stranger than this." in body,
msg="Selected wrong message",
)
self.assertTrue(
"Is it a problem of lapack3.0 of of" in body,
msg="Quoted text is not in uncleaned message",
)
self.assertTrue(
"Is it a problem of lapack3.0 of of"
not in utils.clean_message(body),
msg="Quoted text is in cleaned message",
)
