diff --git a/test/common.py b/test/common.py
new file mode 100644
index 0000000000000000000000000000000000000000..d4bd05aea77f167216908b40690ae8fdaf963cbf
--- /dev/null
+++ b/test/common.py
@@ -0,0 +1,14 @@
+#! /usr/bin/env python
+# -*- coding: utf-8 -*-
+
+import numpy as np
+import pandas as pd
+
+
+def initData(cols=2, start_date="2017-01-01", end_date="2017-12-31", freq="1h"):
+ dates = pd.date_range(start="2017-01-01", end="2017-12-31", freq="1h")
+ data = {}
+ dummy = np.arange(len(dates))
+ for col in range(1, cols+1):
+ data[f"var{col}"] = dummy*(col)
+ return pd.DataFrame(data, index=dates)
diff --git a/test/test_core.py b/test/test_core.py
index 171143f71a368a72d17ddf8100cbc1f693067f97..397132eefa4cae1228beffa9e689f140b3978ea1 100644
--- a/test/test_core.py
+++ b/test/test_core.py
@@ -5,102 +5,136 @@ import pytest
import numpy as np
import pandas as pd
-from core import runner, flagNext
-from funcs.functions import flagGeneric, Params
+from core import runner, flagNext, prepareMeta
from config import Fields
-from flagger import SimpleFlagger, DmpFlagger
+from flagger import SimpleFlagger, DmpFlagger, PositionalFlagger
from .testfuncs import initData
-def initMeta(data):
- dates = data.index
- variables = data.columns
- randg = np.random.randint
- start_dates = [dates[randg(0, (len(dates)//2)-1)] for _ in variables]
- end_dates = [dates[randg(len(dates)//2, len(dates) - 1 )] for _ in variables]
- tests = ["generic, {func: abs(this) + 1 > 0}"] * len(variables)
- return pd.DataFrame({Fields.VARNAME: data.columns,
- Fields.STARTDATE: start_dates,
- Fields.ENDDATE: end_dates,
- Fields.FLAGS: tests})
+TESTFLAGGERS = [
+ SimpleFlagger(), DmpFlagger(), # PositionalFlagger()
+]
-def test_temporalPartitioning():
+@pytest.mark.parametrize("flagger", TESTFLAGGERS)
+def test_temporalPartitioning(flagger):
+ """
+ Check if the time span in meta is respected
+ """
+ data = initData(3)
+ var1, var2, var3, *_ = data.columns
+ split_date = data.index[len(data.index)//2]
+ tests = ["range, {min: -2, max: -1}",
+ "generic, {func: this <= sum(this)}",
+ "generic, {func: this <= sum(this)}"]
+
+ meta = prepareMeta(
+ pd.DataFrame(
+ {Fields.VARNAME: [var1, var2, var3],
+ Fields.STARTDATE: [None, None, split_date],
+ Fields.ENDDATE: [None, split_date, None],
+ Fields.FLAGS: tests}),
+ data)
- data = initData()
- meta = initMeta(data)
- flagger = SimpleFlagger()
pdata, pflags = runner(meta, flagger, data)
fields = [Fields.VARNAME, Fields.STARTDATE, Fields.ENDDATE]
for _, row in meta.iterrows():
vname, start_date, end_date = row[fields]
- fchunk = pflags[vname].dropna()
+ fchunk = pflags.loc[flagger.isFlagged(pflags[vname]), vname]
assert fchunk.index.min() == start_date, "different start dates"
assert fchunk.index.max() == end_date, "different end dates"
-def test_flagNextFill():
- flagger = SimpleFlagger()
- data = initData().iloc[:, 1]
- flags = flagger.emptyFlags(data)
-
- idx = [0, 1, 2]
- flags.iloc[idx] = flagger.setFlag(flags.iloc[idx])
-
- n = 4
- fflags = flagNext(flagger, flags.copy(), 4)
- result_idx = np.unique(np.where(pd.notnull(fflags))[0])
- expected_idx = np.arange(min(idx), max(idx) + n + 1)
- assert (result_idx == expected_idx).all()
-
+@pytest.mark.parametrize("flagger", TESTFLAGGERS)
+def test_missingConfig(flagger):
+ """
+ Test if variables available in the dataset but not the config
+ are handled correctly, i.e. are ignored
+ """
+ data = initData(2)
+ var1, var2, *_ = data.columns
+ meta = prepareMeta(
+ pd.DataFrame(
+ {Fields.VARNAME: [var1],
+ Fields.FLAGS: ["range, {min: -9999, max: 9999}"]}),
+ data)
-def test_flagNextOverwrite():
- flagger = SimpleFlagger()
- data = initData().iloc[:, 0]
- flags = flagger.emptyFlags(data)
+ pdata, pflags = runner(meta, flagger, data)
+ assert var1 in pdata and var2 not in pflags
+
+
+@pytest.mark.parametrize("flagger", TESTFLAGGERS)
+def test_missingVariable(flagger):
+ """
+ Test if variables available in the config but not dataset
+ are handled correctly, i.e. are ignored
+ """
+ data = initData(1)
+ var, *_ = data.columns
+ meta = prepareMeta(
+ pd.DataFrame(
+ {Fields.VARNAME: [var, "empty"],
+ Fields.FLAGS: ["range, {min: -9999, max: 9999}",
+ "range, {min: -9999, max: 9999}"]}),
+ data)
- flags.iloc[0::3] = flagger.setFlag(flags.iloc[0::3], 1)
- flags.iloc[2::3] = flagger.setFlag(flags.iloc[2::3], 2)
+ pdata, pflags = runner(meta, flagger, data)
+ assert (data.columns == [var]).all()
+
+
+@pytest.mark.parametrize("flagger", TESTFLAGGERS)
+def test_assignVariable(flagger):
+ """
+ Test the assign keyword, a variable present in the configuration, but not
+ dataset will be added to output flags
+ """
+ data = initData(1)
+ var1, *_ = data.columns
+ var2 = "empty"
+ meta = prepareMeta(
+ pd.DataFrame(
+ {Fields.VARNAME: [var1, var2],
+ Fields.FLAGS: ["range, {min: -9999, max: 9999}",
+ f"generic, {{func: isflagged({var2}), assign: True}}"]}),
+ data)
- fflags = flagNext(flagger, flags.copy().iloc[:], 4)
- assert ((fflags.values[pd.isnull(flags)] == 1).all(axis=None))
+ pdata, pflags = runner(meta, flagger, data)
+ if isinstance(pflags.columns, pd.MultiIndex):
+ cols = (pflags
+ .columns.get_level_values(0)
+ .drop_duplicates())
+ assert (cols == [var1, var2]).all()
+ else:
+ assert (pflags.columns == [var1, var2]).all()
-def test_flagNextMulticolumn():
- flagger = DmpFlagger()
- data = initData().iloc[:, 0]
- flags = flagger.emptyFlags(data)
- var, *_ = flags.columns.get_level_values(0)
- flags.loc[data.index[0::3], var] = flagger.setFlag(
- flags.loc[data.index[0::3], var], "DOUBTFUL")
+@pytest.mark.parametrize("flagger", TESTFLAGGERS)
+def test_flagNext(flagger):
+ """
+ Test if the flagNext functionality works as expected
+ """
+ data = initData().iloc[:, 1]
+ flags = flagger.initFlags(data)
- flags.loc[data.index[2::3], var] = flagger.setFlag(
- flags.loc[data.index[2::3], var], "BAD")
+ idx = [0, 1, 2]
+ flags.iloc[idx] = flagger.setFlag(flags.iloc[idx])
+ n = 4
fflags = flagNext(flagger, flags.copy(), 4)
- assert ((fflags.values[pd.isnull(flags)] == 1).all(axis=None))
-
-
-def test_flagGenericFailure():
- flagger = SimpleFlagger()
- data = initData()
- flags = flagger.emptyFlags(data)
- var1, var2, *_ = data.columns
-
- # expression does not return a result of identical shape
- with pytest.raises(TypeError):
- flagGeneric(data, flags, var2, flagger, **{Params.FUNC: f"sum({var1})"})
-
- # expression does not return a boolean result
- with pytest.raises(TypeError):
- flagGeneric(data, flags, var2, flagger, **{Params.FUNC: f"{var1}"})
+ result_idx = np.unique(np.where(flagger.isFlagged(fflags))[0])
+ expected_idx = np.arange(min(idx), max(idx) + n + 1)
+ assert (result_idx == expected_idx).all()
if __name__ == "__main__":
- test_temporalPartitioning()
- test_flagNextFill()
- test_flagNextOverwrite()
- test_flagNextMulticolumn()
- test_flagGenericFailure()
+
+ # NOTE: PositionalFlagger is currently broken, going to fix it when needed
+ # for flagger in [SimpleFlagger, PositionalFlagger, DmpFlagger]:
+ for flagger in [SimpleFlagger(), DmpFlagger()]:
+ test_temporalPartitioning(flagger)
+ test_flagNext(flagger)
+ test_missingConfig(flagger)
+ test_missingVariable(flagger)
+ test_assignVariable(flagger)
diff --git a/test/test_evaluator.py b/test/test_evaluator.py
index 80d3608fa088aab24c2bcb0a318401043525bdb7..da8ac555045dbaeecfa5c06d4b7df79317d074dd 100644
--- a/test/test_evaluator.py
+++ b/test/test_evaluator.py
@@ -12,7 +12,7 @@ from dsl import evalExpression
def test_evaluationBool():
data = initData()
flagger = SimpleFlagger()
- flags = flagger.emptyFlags(data, 0)
+ flags = flagger.initFlags(data, 0)
var1, var2, *_ = data.columns
tests = [
@@ -37,7 +37,7 @@ def test_evaluationBool():
def test_missingIdentifier():
data = initData()
flagger = SimpleFlagger()
- flags = flagger.emptyFlags(data)
+ flags = flagger.initFlags(data)
tests = ["func(var2) < 5", "var3 != NODATA"]
for test in tests:
with pytest.raises(NameError):
@@ -47,7 +47,7 @@ def test_missingIdentifier():
def test_flagPropagation():
data = initData()
flagger = SimpleFlagger()
- flags = flagger.emptyFlags(data, 0)
+ flags = flagger.initFlags(data, 0)
flags.iloc[::5] = flagger.setFlag(flags.iloc[::5])
var1, var2, *_ = data.columns
diff --git a/test/test_generic.py b/test/test_generic.py
index dce7285231c2b782f2c76bc827344106a2a5a040..fd95ca93789e6a39dd3d88a123df02bf0efedfe8 100644
--- a/test/test_generic.py
+++ b/test/test_generic.py
@@ -3,11 +3,13 @@
import numpy as np
import pandas as pd
+import pytest
from .testfuncs import initData
from dsl import evalExpression
from flagger import SimpleFlagger
+from funcs.functions import flagGeneric, Params
def test_ismissing():
@@ -19,7 +21,7 @@ def test_ismissing():
data.iloc[(len(data)//2)+1:, 0] = nodata
flagger = SimpleFlagger()
- flags = flagger.emptyFlags(data)
+ flags = flagger.initFlags(data)
var1, var2, *_ = data.columns
@@ -37,7 +39,7 @@ def test_isflagged():
flagger = SimpleFlagger()
data = initData()
- flags = flagger.emptyFlags(data, 0)
+ flags = flagger.initFlags(data, 0)
var1, var2, *_ = data.columns
flags.iloc[::2, 0] = flagger.setFlag(flags.iloc[::2, 0])
@@ -51,11 +53,11 @@ def test_isflagged():
assert (flagged == idx).all
-def test_isflaggedNonstandard():
+def test_isflaggedArgument():
flagger = SimpleFlagger()
data = initData()
- flags = flagger.emptyFlags(data, 0)
+ flags = flagger.initFlags(data, 0)
var1, var2, *_ = data.columns
flags.iloc[::2, 0] = flagger.setFlag(flags.iloc[::2, 0], -9)
@@ -69,7 +71,25 @@ def test_isflaggedNonstandard():
assert (flagged == idx).all
+def test_flagFailure():
+ flagger = SimpleFlagger()
+ data = initData()
+ flags = flagger.initFlags(data)
+ var1, var2, *_ = data.columns
+
+ # expression does not return a result of identical shape
+ with pytest.raises(TypeError):
+ flagGeneric(data, flags, var2, flagger,
+ **{Params.FUNC: f"sum({var1})"})
+
+ # need a test for missing variables
+ with pytest.raises(NameError):
+ flagGeneric(data, flags, var2, flagger,
+ **{Params.FUNC: f"sum({var1 + 'x'})"})
+
+
if __name__ == "__main__":
test_ismissing()
test_isflagged()
- test_isflaggedNonstandard()
+ test_isflaggedArgument()
+ test_flagFailure()
diff --git a/test/testfuncs.py b/test/testfuncs.py
deleted file mode 100644
index 3e60913bfe538c35345dec4cb78da2404cc52cd1..0000000000000000000000000000000000000000
--- a/test/testfuncs.py
+++ /dev/null
@@ -1,14 +0,0 @@
-#! /usr/bin/env python
-# -*- coding: utf-8 -*-
-
-import numpy as np
-import pandas as pd
-
-
-def initData(start_date="2017-01-01", end_date="2017-12-31", freq="1h"):
- dates = pd.date_range(start="2017-01-01", end="2017-12-31", freq="1h")
- data = pd.DataFrame(
- data={"var1": np.arange(len(dates)),
- "var2": np.arange(len(dates), len(dates)*2)},
- index=dates)
- return data