def test_read_events_num(self):
df = DataFrame(self.data, columns=self.short_channels)
s3_file_path = f"s3://{self.bucket_name}/test.fcs"
df.to_fcs(s3_file_path)
num = read_events_num(s3_file_path)
assert num == len(self.data)
def test_data(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, self.name)
df = DataFrame(self.data, columns=self.channels)
df.to_fcs(filename)
fcs = Fcs.from_file(filename)
assert np.array_equal(fcs.values, self.data)
def test_rename_channels(self):
cols = pd.MultiIndex.from_tuples(
list(zip(self.short_channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
s3_file_path = f"s3://{self.bucket_name}/test.fcs"
df.to_fcs(s3_file_path)
rename_channels(s3_file_path, {
"a": "a_1",
"d": "d_1"
},
"short",
allow_rewrite=True)
dfr = DataFrame.from_fcs(s3_file_path, channel_type="multi")
np.testing.assert_array_equal(
["a_1", "b", "c", "d_1"],
dfr.columns.get_level_values("short").values,
)
np.testing.assert_array_equal(
list("ABCD"),
dfr.columns.get_level_values("long").values,
)
ln = ""
for i in range(30):
ln = ln + str(i)
rename_channels(s3_file_path, {"C": ln}, "long", allow_rewrite=True)
assert ["A", "B", ln, "D"] == read_channels(s3_file_path, "long")
def test_rename_channels_alt(self):
with TmpDir() as dir_:
df = DataFrame(
self.data,
columns=pd.MultiIndex.from_tuples(
zip(self.short_channels, self.long_channels),
names=["short", "long"],
),
)
filename = os.path.join(dir_, self.name)
df.to_fcs(filename)
ln = ""
for i in range(30):
ln = ln + str(i)
with pytest.raises(ValueError) as err:
rename_channels(filename, {"a_di": ln}, "short")
assert (
str(err.value) ==
"New channel names are too long causing overlap with Data Segment."
)
rename_channels(filename, {"a_di": ln},
"short",
allow_rewrite=True)
df2 = DataFrame.from_fcs(filename)
assert_almost_equal(df.values, df2.values)
def test_count(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, self.name)
df = DataFrame(self.data, columns=self.channels)
df.to_fcs(filename)
fcs = Fcs.from_file(filename)
assert fcs.count == len(self.data)
def test_read_fcs(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, self.name)
df = DataFrame(self.data, columns=self.short_channels)
df.to_fcs(filename)
df2 = read_fcs(filename)
assert_frame_equal(df, df2)
def test_rename_channels_alt(self):
cols = pd.MultiIndex.from_tuples(
list(zip(self.short_channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
s3_file_path = f"s3://{self.bucket_name}/test.fcs"
df.to_fcs(s3_file_path)
with pytest.raises(ValueError) as e:
rename_channels(s3_file_path,
dict(zip(self.short_channels, list("abcd"))),
"short")
assert str(
e.value) == "S3 url is only supported when allow_rewrite is set."
ln = ""
for i in range(30):
ln = ln + str(i)
with pytest.raises(ValueError) as e:
rename_channels(s3_file_path, {"c": ln}, "short")
assert (
str(e.value) ==
"New channel names are too long causing overlap with Data Segment."
)
def test_read_text_segment(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, self.name)
cols = pd.MultiIndex.from_tuples(
list(zip(self.channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
df.to_fcs(filename)
seg = Fcs.read_text_segment(filename)
assert seg.pnn == self.channels
assert seg.pns == self.long_channels
def test_read_channels(self):
cols = pd.MultiIndex.from_tuples(
list(zip(self.short_channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
s3_file_path = f"s3://{self.bucket_name}/test.fcs"
df.to_fcs(s3_file_path)
assert read_channels(s3_file_path, "short") == self.short_channels
assert read_channels(s3_file_path, "long") == self.long_channels
np.testing.assert_array_equal(read_channels(s3_file_path, "multi"),
cols)
def test_to_fcs(self):
with TmpDir() as dir_:
cols = pd.MultiIndex.from_tuples(
list(zip(self.short_channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
df.to_fcs(f"s3://{self.bucket_name}/test.fcs")
filename = os.path.join(dir_, "test.fcs")
self.s3.download_file(self.bucket_name, "test.fcs", filename)
dfr = DataFrame.from_fcs(filename, channel_type="multi")
assert_frame_equal(df, dfr)
def test_read_events_num(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, "test.fcs")
cols = pd.MultiIndex.from_tuples(
list(zip(self.short_channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
df.to_fcs(filename)
with open(filename, "rb") as fp:
num = read_events_num(fp)
assert num == len(self.data)
def test_to_fcs(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, "test.fcs")
cols = pd.MultiIndex.from_tuples(
list(zip(self.short_channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
with open(filename, "wb") as fp:
df.to_fcs(fp)
dfr = DataFrame.from_fcs(filename, channel_type="multi")
assert_frame_equal(df, dfr)
def test_read_channels(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, "test.fcs")
cols = pd.MultiIndex.from_tuples(
list(zip(self.short_channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
df.to_fcs(filename)
assert read_channels(filename, "short") == self.short_channels
assert read_channels(filename, "long") == self.long_channels
np.testing.assert_array_equal(read_channels(filename, "multi"),
cols)
def test_write_fcs(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, self.name)
df = DataFrame(self.data, columns=self.short_channels)
write_fcs(df, filename, long_names=self.long_channels)
df = DataFrame.from_fcs(filename, "multi")
np.testing.assert_array_equal(df.values, self.data)
np.testing.assert_array_equal(
df.columns,
pd.MultiIndex.from_tuples(
zip(self.short_channels, self.long_channels),
names=["short", "long"],
),
)
def test_read_events_num(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, self.name)
df = DataFrame(self.data, columns=self.short_channels)
write_fcs(df, filename, long_names=self.long_channels)
num = read_events_num(filename)
assert num == 2
def test_read_text_segment(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, "test.fcs")
cols = pd.MultiIndex.from_tuples(
list(zip(self.channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
df.to_fcs(filename)
self.s3.upload_file(filename, self.bucket_name, "test.fcs")
parse_func = create_open_func(S3ReadBuffer, bucket=self.bucket_name)
with parse_func("test.fcs") as fp:
seg = Fcs.read_text_segment(fp)
assert seg.pnn == self.channels
assert seg.pns == self.long_channels
def test_from_fcs(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, "test.fcs")
cols = pd.MultiIndex.from_tuples(
list(zip(self.channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
df.to_fcs(filename)
self.s3.upload_file(filename, self.bucket_name, "test.fcs")
parse_func = create_open_func(S3ReadBuffer, bucket=self.bucket_name)
with parse_func("test.fcs") as fp:
fcs = Fcs.from_file(fp)
assert fcs.short_channels == self.channels
assert fcs.long_channels == self.long_channels
assert fcs.count == len(self.data)
assert np.array_equal(fcs.values, self.data)
def test_rename_channels(self):
with TmpDir() as dir_:
filename = os.path.join(dir_, "test.fcs")
cols = pd.MultiIndex.from_tuples(
list(zip(self.short_channels, self.long_channels)),
names=["short", "long"],
)
df = DataFrame(self.data, columns=cols)
df.to_fcs(filename)
with open(filename, "rb+") as fp:
rename_channels(fp, dict(zip(self.short_channels,
list("wxyz"))), "short")
with open(filename, "rb+") as fp:
rename_channels(fp, dict(zip(self.long_channels,
list("WXYZ"))), "long")
assert read_channels(filename, "short") == list("wxyz")
assert read_channels(filename, "long") == list("WXYZ")
def test_rename_channels(self):
with TmpDir() as dir_:
df = DataFrame(
self.data,
columns=pd.MultiIndex.from_tuples(
zip(self.short_channels, self.long_channels),
names=["short", "long"],
),
)
filename = os.path.join(dir_, self.name)
df.to_fcs(filename)
rename_channels(filename,
dict(zip(self.short_channels, list("abcd"))),
"short")
short_channels = read_channels(filename, "short")
assert short_channels == list("abcd")
rename_channels(filename,
dict(zip(self.long_channels, list("ABCD"))),
"long")
long_channels = read_channels(filename, "long")
assert long_channels == list("ABCD")
data = [[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]]
columns = pd.MultiIndex.from_tuples(list(zip("abc", "ABC")),
names=["short", "long"])
df = DataFrame(data, columns=columns)
filename = os.path.join(dir_, self.name)
df.to_fcs(filename)
rename_channels(filename, {
"a": "a_1",
"b": "b_1",
"c": "c_1"
},
channel_type="short")
short_channels = read_channels(filename, "short")
long_channels = read_channels(filename, "long")
assert short_channels == ["a_1", "b_1", "c_1"]
assert long_channels == list("ABC")