Try to fix more tests

Author: znicholls

docs/source/notebooks/ops.py

@@ -217,7 +217,7 @@
 # [Pint](https://pint.readthedocs.io) and some other work, this operation
 # is also unit aware.
 #
-# There are two methods of integration available, `cumtrapz` and `cumsum`.
+# There are two methods of integration available, `cumulative_trapezoid` and `cumsum`.
 # The method that should be used depends on the data you are integrating,
 # specifically whether the data are piecewise linear or piecewise constant.
 #
@@ -226,18 +226,18 @@
 #
 # Other output such as effective radiative forcing, concentrations or decadal
 # timeseries of emissions, represent a  point estimate or an average over a period.
-# These timeseries are therefore piecewise linear and should use the `cumtrapz`
+# These timeseries are therefore piecewise linear and should use the `cumulative_trapezoid`
 # method.
 
 # %%
 with warnings.catch_warnings():
     # Ignore warning about nans in the historical timeseries
     warnings.filterwarnings("ignore", module="scmdata.ops")
 
-    # Radiative Forcings are piecewise-linear so `cumtrapz` should be used
+    # Radiative Forcings are piecewise-linear so `cumulative_trapezoid` should be used
     erf_integral = (
         db_forcing.filter(variable="Effective Radiative Forcing")
-        .cumtrapz()
+        .cumulative_trapezoid()
         .convert_unit("TJ / m^2")
     )
 
@@ -416,13 +416,7 @@
 )
 erf_total_for_shift_rel_to_ref_period_shifted[
     "label"
-] = "rel. to {} - {} (median of {} - {} mean adjusted to {})".format(
-    ref_period[0],
-    ref_period[-1],
-    evaluation_period[0],
-    evaluation_period[-1],
-    target,
-)
+] = f"rel. to {ref_period[0]} - {ref_period[-1]} (median of {evaluation_period[0]} - {evaluation_period[-1]} mean adjusted to {target})"  # noqa: E501
 
 # %%
 pdf = run_append(

pyproject.toml

@@ -267,6 +267,7 @@ authorized_licenses = [
     "mit",
     "mit license",
     "Mozilla Public License 2.0 (MPL 2.0)",
+    "psf-2.0",
     "python software foundation",
     "python software foundation license",
     "zpl 2.1",

src/scmdata/ops.py

@@ -5,6 +5,7 @@
 `Pint's Pandas interface <https://pint.readthedocs.io/en/0.13/pint-pandas.html>`_
 to handle unit conversions automatically
 """
+
 import warnings
 
 import numpy as np
@@ -113,10 +114,9 @@ def _perform_op(base, other, op, use_pint_units=True):
                 elif op == "subtract":
                     out.append(base[col] - other[col])
 
-            except KeyError:
-                raise KeyError(  # noqa: TRY200
-                    f"No equivalent in `other` for {list(zip(col_names, col))}"
-                )
+            except KeyError as exc:
+                msg = f"No equivalent in `other` for {list(zip(col_names, col))}"
+                raise KeyError(msg) from exc
 
         elif op == "multiply":
             out.append(base[col] * other[col])
@@ -227,17 +227,17 @@ def subtract(self, other, op_cols, **kwargs):
     >>> fos_minus_afolu.head()
     time                                                                  2010-01-01  2020-01-01
     model     region   scenario  unit       variable
-    idealised World|NH idealised gigatC / a Emissions|CO2|Fossil - AFOLU      -0.001       3.995
-              World|SH idealised gigatC / a Emissions|CO2|Fossil - AFOLU       1.997       5.993
+    idealised World|NH idealised gigatC / yr Emissions|CO2|Fossil - AFOLU      -0.001       3.995
+              World|SH idealised gigatC / yr Emissions|CO2|Fossil - AFOLU       1.997       5.993
 
     >>> nh = start.filter(region="World|NH")
     >>> sh = start.filter(region="World|SH")
     >>> nh_minus_sh = nh.subtract(sh, op_cols={"region": "World|NH - SH"})
     >>> nh_minus_sh.head()
     time                                                               2010-01-01  2020-01-01
     model     region        scenario  unit       variable
-    idealised World|NH - SH idealised gigatC / a Emissions|CO2|Fossil        -2.0        -2.0
-                                      megatC / a Emissions|CO2|AFOLU         -2.0        -2.0
+    idealised World|NH - SH idealised gigatC / yr Emissions|CO2|Fossil        -2.0        -2.0
+                                      megatC / yr Emissions|CO2|AFOLU         -2.0        -2.0
     """
     out = _perform_op(
         prep_for_op(self, op_cols, self.meta.columns, **kwargs),
@@ -336,17 +336,17 @@ def add(self, other, op_cols, **kwargs):
     >>> fos_plus_afolu.head()
     time                                                                  2010-01-01  2020-01-01
     model     region   scenario  unit       variable
-    idealised World|NH idealised gigatC / a Emissions|CO2|Fossil + AFOLU       0.001       4.005
-              World|SH idealised gigatC / a Emissions|CO2|Fossil + AFOLU       2.003       6.007
+    idealised World|NH idealised gigatC / yr Emissions|CO2|Fossil + AFOLU       0.001       4.005
+              World|SH idealised gigatC / yr Emissions|CO2|Fossil + AFOLU       2.003       6.007
 
     >>> nh = start.filter(region="World|NH")
     >>> sh = start.filter(region="World|SH")
     >>> nh_plus_sh = nh.add(sh, op_cols={"region": "World|NH + SH"})
     >>> nh_plus_sh.head()
     time                                                               2010-01-01  2020-01-01
     model     region        scenario  unit       variable
-    idealised World|NH + SH idealised gigatC / a Emissions|CO2|Fossil         2.0        10.0
-                                      megatC / a Emissions|CO2|AFOLU          4.0        12.0
+    idealised World|NH + SH idealised gigatC / yr Emissions|CO2|Fossil         2.0        10.0
+                                      megatC / yr Emissions|CO2|AFOLU          4.0        12.0
     """
     out = _perform_op(
         prep_for_op(self, op_cols, self.meta.columns, **kwargs),
@@ -544,14 +544,14 @@ def divide(self, other, op_cols, **kwargs):
               World|SH idealised MtC / yr Emissions|CO2|AFOLU         3.0         7.0
 
     >>> fos_divide_afolu = fos.divide(
-    ...     afolu, op_cols={"variable": "Emissions|CO2|Fossil / AFOLU"}
+    ...     afolu, op_cols={"variable": "Emissions|CO2|Fossil / yrFOLU"}
     ... )
     >>> # The rows align and the units are handled automatically
     >>> fos_divide_afolu.convert_unit("dimensionless").head()
     time                                                                     2010-01-01  2020-01-01
     model     region   scenario  unit          variable
-    idealised World|NH idealised dimensionless Emissions|CO2|Fossil / AFOLU    0.000000  800.000000
-              World|SH idealised dimensionless Emissions|CO2|Fossil / AFOLU  666.666667  857.142857
+    idealised World|NH idealised dimensionless Emissions|CO2|Fossil / yrFOLU    0.000000  800.000000
+              World|SH idealised dimensionless Emissions|CO2|Fossil / yrFOLU  666.666667  857.142857
 
     >>> nh = start.filter(region="World|NH")
     >>> sh = start.filter(region="World|SH")
@@ -597,7 +597,7 @@ def cumsum(self, out_var=None, check_annual=True):
     arbitrary day/month of the year has not been implemented, if that would be
     useful raise an issue on GitHub.
 
-    If the timeseries are piecewise-linear, :meth:`cumtrapz` should be used instead.
+    If the timeseries are piecewise-linear, :meth:`cumulative_trapezoid` should be used instead.
 
     Parameters
     ----------
@@ -618,7 +618,7 @@ def cumsum(self, out_var=None, check_annual=True):
 
     See Also
     --------
-    :func:`cumtrapz`
+    :func:`cumulative_trapezoid`
 
     Raises
     ------
@@ -677,7 +677,7 @@ def cumsum(self, out_var=None, check_annual=True):
     return out
 
 
-def cumtrapz(self, out_var=None):
+def cumulative_trapezoid(self, out_var=None):
     """
     Integrate with respect to time using the trapezoid rule
 
@@ -743,7 +743,9 @@ def cumtrapz(self, out_var=None):
     # some thinking about unit handling
     _initial = 0.0
     out = pd.DataFrame(
-        scipy.integrate.cumtrapz(y=ts, x=times_in_s, axis=1, initial=_initial)
+        scipy.integrate.cumulative_trapezoid(
+            y=ts, x=times_in_s, axis=1, initial=_initial
+        )
     )
     out.index = ts.index
     out.columns = ts.columns
@@ -790,7 +792,7 @@ def integrate(self, out_var=None):
     See Also
     --------
     :meth:`cumsum`
-    :meth:`cumtrapz`
+    :meth:`cumulative_trapezoid`
 
     Raises
     ------
@@ -805,14 +807,14 @@ def integrate(self, out_var=None):
         data will also contain nans.
     DeprecationWarning
         This function has been deprecated in preference to :meth:`cumsum` and
-        :meth:`cumtrapz`.
+        :meth:`cumulative_trapezoid`.
     """
     warnings.warn(
-        "integrate has been deprecated in preference of cumsum and cumtrapz",
+        "integrate has been deprecated in preference of cumsum and cumulative_trapezoid",
         DeprecationWarning,
     )
 
-    return cumtrapz(self, out_var)
+    return cumulative_trapezoid(self, out_var)
 
 
 def delta_per_delta_time(self, out_var=None):
@@ -1144,7 +1146,7 @@ def inject_ops_methods(cls):
         ("divide", divide),
         ("integrate", integrate),
         ("cumsum", cumsum),
-        ("cumtrapz", cumtrapz),
+        ("cumulative_trapezoid", cumulative_trapezoid),
         ("delta_per_delta_time", delta_per_delta_time),
         ("linear_regression", linear_regression),
         ("linear_regression_gradient", linear_regression_gradient),

tests/unit/test_ops.py

@@ -99,7 +99,7 @@ def test_no_scipy(scm_run):
     with pytest.raises(
         ImportError, match="scipy is not installed. Run 'pip install scipy'"
     ):
-        scm_run.cumtrapz()
+        scm_run.cumulative_trapezoid()
 
 
 @OPS_MARK
@@ -113,10 +113,10 @@ def test_single_timeseries(op, base_single_scmrun, other_single_scmrun):
     exp_ts["variable"] = "Emissions|CO2|AFOLU"
 
     if op in ["add", "subtract"]:
-        exp_ts["unit"] = "gigatC / a"
+        exp_ts["unit"] = "gigatC / yr"
 
     elif op == "multiply":
-        exp_ts["unit"] = "gigatC ** 2 / a ** 2"
+        exp_ts["unit"] = "gigatC ** 2 / yr ** 2"
 
     elif op == "divide":
         exp_ts["unit"] = "dimensionless"
@@ -170,7 +170,7 @@ def test_different_unit_error():
 
     error_msg = re.escape(
         "Cannot convert from 'kelvin' ([temperature]) to "
-        "'gigatC' ([carbon] * [mass])"
+        "'gigatC' ([mass] * [carbon])"
     )
     with pytest.raises(DimensionalityError, match=error_msg):
         base.add(other, op_cols={"variable": "Warming plus Cumulative emissions CO2"})
@@ -262,10 +262,10 @@ def test_scalar_ops_pint(op):
     exp = ScmRun(exp_ts)
 
     if op in ["add", "subtract"]:
-        exp["unit"] = "gigatC / a"
+        exp["unit"] = "gigatC / yr"
 
     elif op == "multiply":
-        exp["unit"] = "gigatC * megatC / a ** 2"
+        exp["unit"] = "gigatC * megatC / yr ** 2"
 
     elif op == "divide":
         exp["unit"] = "gigatC / megatC"
@@ -315,7 +315,7 @@ def test_scalar_multiply_pint_by_run():
     exp_ts = perform_pint_op(start, scalar, "multiply_inverse")
     exp = ScmRun(exp_ts)
 
-    exp["unit"] = "megatC * gigatC / a**2"
+    exp["unit"] = "megatC * gigatC / yr**2"
 
     res = scalar * start
 
@@ -330,8 +330,8 @@ def test_scalar_ops_pint_wrong_unit(op):
     )
 
     error_msg = re.escape(
-        "Cannot convert from 'gigatC / a' ([carbon] * [mass] / [time]) "
-        "to 'CH4 * megametric_ton / a' ([mass] * [methane] / [time])"
+        "Cannot convert from 'gigatC / yr' ([mass] * [carbon] / [time]) "
+        "to 'megametric_ton * CH4 / yr' ([mass] * [methane] / [time])"
     )
     with pytest.raises(DimensionalityError, match=error_msg):
         if op == "add":
@@ -356,10 +356,10 @@ def test_vector_ops_pint(op):
     exp = ScmRun(exp_ts)
 
     if op in ["add", "subtract"]:
-        exp["unit"] = "gigatC / a"
+        exp["unit"] = "gigatC / yr"
 
     elif op == "multiply":
-        exp["unit"] = "gigatC * megatC / a ** 2"
+        exp["unit"] = "gigatC * megatC / yr ** 2"
 
     elif op == "divide":
         exp["unit"] = "gigatC / megatC"
@@ -391,8 +391,8 @@ def test_vector_ops_pint_wrong_unit(op, start_unit):
     )
 
     error_msg = re.escape(
-        "Cannot convert from 'gigatC / a' ([carbon] * [mass] / [time]) "
-        "to 'CH4 * megametric_ton / a' ([mass] * [methane] / [time])"
+        "Cannot convert from 'gigatC / yr' ([mass] * [carbon] / [time]) "
+        "to 'megametric_ton * CH4 / yr' ([mass] * [methane] / [time])"
     )
     with pytest.raises(DimensionalityError, match=error_msg):
         if op == "add":
@@ -549,11 +549,11 @@ def test_wrong_length_ops(op):
 # We can add initial back if use case arises. At the moment I can't see an easy
 # way to make the units behave.
 # @pytest.mark.parametrize("initial", (None, 0, 1, -1.345))
-def test_cumtrapz(out_var):
+def test_cumulative_trapezoid(out_var):
     dat = [1, 2, 3]
     start = get_single_ts(data=dat, index=[1, 2, 3], unit="GtC / yr")
 
-    res = start.cumtrapz(out_var=out_var)
+    res = start.cumulative_trapezoid(out_var=out_var)
 
     if out_var is None:
         exp_var = ("Cumulative " + start["variable"]).values
@@ -570,10 +570,10 @@ def test_cumtrapz(out_var):
     )
 
 
-def test_cumtrapz_time_handling_big_jumps():
+def test_cumulative_trapezoid_time_handling_big_jumps():
     start = get_single_ts(data=[1, 2, 3], index=[10, 20, 50], unit="GtC / yr")
 
-    res = start.cumtrapz()
+    res = start.cumulative_trapezoid()
 
     npt.assert_allclose(
         res.values.squeeze(),
@@ -582,12 +582,12 @@ def test_cumtrapz_time_handling_big_jumps():
     )
 
 
-def test_cumtrapz_time_handling_all_over_jumps():
+def test_cumulative_trapezoid_time_handling_all_over_jumps():
     start = get_single_ts(
         data=[1, 2, 3, 3, 1.8], index=[10, 10.1, 11, 20, 50], unit="GtC / yr"
     )
 
-    res = start.cumtrapz()
+    res = start.cumulative_trapezoid()
 
     first = 0
     second = first + 1.5 * 0.1
@@ -603,7 +603,7 @@ def test_cumtrapz_time_handling_all_over_jumps():
     "method,exp",
     [
         ("cumsum", [1, 3, 6, np.nan, np.nan, np.nan, np.nan, np.nan]),
-        ("cumtrapz", [0, 1.5, 4, np.nan, np.nan, np.nan, np.nan, np.nan]),
+        ("cumulative_trapezoid", [0, 1.5, 4, np.nan, np.nan, np.nan, np.nan, np.nan]),
     ],
 )
 def test_integration_nan_handling(method, exp):
@@ -633,7 +633,7 @@ def test_integration_nan_handling(method, exp):
 @pytest.mark.xfail(
     _check_pandas_less_110(), reason="pandas<=1.1.0 does not have rtol argument"
 )
-def test_cumtrapz_multiple_ts():
+def test_cumulative_trapezoid_multiple_ts():
     variables = ["Emissions|CO2", "Heat Uptake", "Temperature"]
     start = get_multiple_ts(
         data=np.array([[1, 2, 3], [-1, -2, -3], [0, 5, 10]]).T,
@@ -642,7 +642,7 @@ def test_cumtrapz_multiple_ts():
         unit=["Mt CO2 / yr", "W / m^2", "K"],
     )
 
-    res = start.cumtrapz()
+    res = start.cumulative_trapezoid()
 
     exp = get_single_ts(
         data=np.array([[0, 7.5, 45], [0, -7.5, -45], [0, 12.5, 125]]).T,
@@ -730,7 +730,9 @@ def test_integrate_deprecated():
     dat = [1, 2, 3]
     start = get_single_ts(data=dat, index=[2020, 2021, 2024], unit="GtC / yr")
 
-    match = "integrate has been deprecated in preference of cumsum and cumtrapz"
+    match = (
+        "integrate has been deprecated in preference of cumsum and cumulative_trapezoid"
+    )
     with pytest.warns(DeprecationWarning, match=match):
         start.integrate()
 

tests/unit/test_pyam_compat.py

@@ -19,11 +19,7 @@ def test_to_int_value_error(test_iam_df):
         )
 
     else:
-        error_msg = re.escape(
-            f'time data "{bad_val}" '
-            "doesn't match format "
-            '"%Y/%m/%d", at position 4. You might want to try:'
-        )
+        error_msg = re.escape(f"Invalid time domain: {bad_val}")
 
     with pytest.raises(ValueError, match=error_msg):
         LongDatetimeIamDataFrame(idf)