Configure specific warnings (#401)

ssolson · akeeste · web-flow · commit fb005ce3f61e · 2025-08-19T14:36:06.000-05:00
This PR removes the ignore all future warnings added when some packages
were suggesting updating to numpy 2.0 but mhkit needed other packages to
update before that could happen.

---------

Co-authored-by: akeeste &lt;akeeste@sandia.gov&gt;
diff --git a/examples/ADCP_Delft3D_TRTS_example.ipynb b/examples/ADCP_Delft3D_TRTS_example.ipynb
diff --git a/mhkit/__init__.py b/mhkit/__init__.py
@@ -6,8 +6,10 @@
 
 _rmc()
 
-# Ignore future warnings
-_warn.simplefilter(action="ignore", category=FutureWarning)
+# Use targeted warning configuration
+from mhkit.warnings import configure_warnings
+
+configure_warnings()
 
 __version__ = "v0.9.0"
 
diff --git a/mhkit/acoustics/sel.py b/mhkit/acoustics/sel.py
@@ -131,7 +131,7 @@ def sound_exposure_level(
     # from weighted mean square values
     band = spsd.sel(freq=slice(fmin, fmax))
     w = w.sel(freq=slice(fmin, fmax))
-    exposure = np.trapz(band * w, band["freq"])
+    exposure = np.trapezoid(band * w, band["freq"])
 
     # Sound exposure level (L_{E,p}) = (L_{p,rms} + 10log10(t))
     sel = 10 * np.log10(exposure / reference) + 10 * np.log10(
diff --git a/mhkit/acoustics/spl.py b/mhkit/acoustics/spl.py
@@ -108,7 +108,7 @@ def sound_pressure_level(
     # Mean square sound pressure in a specified frequency band from mean square values
     band = spsd.sel(freq=slice(fmin, fmax))
     freqs = band["freq"]
-    pressure_squared = np.trapz(band, freqs)
+    pressure_squared = np.trapezoid(band, freqs)
 
     # Mean square sound pressure level
     mspl = 10 * np.log10(pressure_squared / reference)
@@ -197,7 +197,7 @@ def _band_sound_pressure_level(
         else:
             spsd_slc = spsd.sel(freq=slice(*band_range))
 
-        pressure_squared.loc[{"freq_bins": key}] = np.trapz(spsd_slc, x)
+        pressure_squared.loc[{"freq_bins": key}] = np.trapezoid(spsd_slc, x)
 
     # Mean square sound pressure level in dB rel 1 uPa
     mspl = 10 * np.log10(pressure_squared / reference)
diff --git a/mhkit/dolfyn/adp/discharge.py b/mhkit/dolfyn/adp/discharge.py
@@ -229,7 +229,7 @@ def _calc_discharge(vel, x, depth, surface_zoff=None):
             sign = -1
         else:
             sign = 1
-        return sign * np.trapz(np.trapz(vel, depth, axis=0), x)
+        return sign * np.trapezoid(np.trapezoid(vel, depth, axis=0), x)
 
     # Extrapolate to bed
     vel = ds["vel"].copy()
@@ -265,7 +265,7 @@ def _calc_discharge(vel, x, depth, surface_zoff=None):
             (0.5 * rho * speed**3).mean().item()
         )  # kg/m^3 * m^3/s^3 = kg/s^3 = W/m^2
         hydraulic_depth = abs(
-            np.trapz((water_depth - surface_offset)[_xinds], xy[0][_xinds])
+            np.trapezoid((water_depth - surface_offset)[_xinds], xy[0][_xinds])
         ) / (
             xy[0][_xinds].max() - xy[0][_xinds].min()
         )  # area / surface-width
diff --git a/mhkit/dolfyn/adv/turbulence.py b/mhkit/dolfyn/adv/turbulence.py
@@ -559,7 +559,7 @@ def _integral_TE01(self, I_tke, theta):
         x = np.arange(-20, 20, 1e-2)  # I think this is a long enough range.
         out = np.empty_like(I_tke.flatten())
         for i, (b, t) in enumerate(zip(I_tke.flatten(), theta.flatten())):
-            out[i] = np.trapz(
+            out[i] = np.trapezoid(
                 cbrt(x**2 - 2 / b * np.cos(t) * x + b ** (-2)) * np.exp(-0.5 * x**2),
                 x,
             )
@@ -684,7 +684,7 @@ def integral_length_scales(self, a_cov, U_mag, fs=None):
         T_int = np.zeros(acov.shape[:2])
         for i in range(3):
             for t in range(a_cov["time"].size):
-                T_int[i, t] = np.trapz(acov[i, t][: zero_crossing[i, t]], dx=1 / fs)
+                T_int[i, t] = np.trapezoid(acov[i, t][: zero_crossing[i, t]], dx=1 / fs)
 
         L_int = U_mag.values * T_int
 
diff --git a/mhkit/dolfyn/time.py b/mhkit/dolfyn/time.py
@@ -123,18 +123,17 @@ def epoch2date(ep_time, offset_hr=0, to_str=False):
     elif not isinstance(ep_time, (np.ndarray, list)):
         ep_time = [ep_time]
 
-    ######### IMPORTANT #########
-    # Note the use of `utcfromtimestamp` here, rather than `fromtimestamp`
-    # This is CRITICAL! See the difference between those functions here:
-    #    https://docs.python.org/3/library/datetime.html#datetime.datetime.fromtimestamp
-    # Long story short: `fromtimestamp` used system-specific timezone
-    # info to calculate the datetime object, but returns a
-    # timezone-agnostic object.
     if offset_hr != 0:
         delta = timedelta(hours=offset_hr)
-        time = [datetime.utcfromtimestamp(t) + delta for t in ep_time]
+        time = [
+            datetime.fromtimestamp(t, timezone.utc).replace(tzinfo=None) + delta
+            for t in ep_time
+        ]
     else:
-        time = [datetime.utcfromtimestamp(t) for t in ep_time]
+        time = [
+            datetime.fromtimestamp(t, timezone.utc).replace(tzinfo=None)
+            for t in ep_time
+        ]
 
     if to_str:
         time = date2str(time)
diff --git a/mhkit/river/io/d3d.py b/mhkit/river/io/d3d.py
@@ -570,7 +570,7 @@ def variable_interpolation(
 
             if len(idx[0]):
                 for i in idx[0]:
-                    transformed_data[var][i] = interp.griddata(
+                    transformed_data.loc[i, var] = interp.griddata(
                         data_raw[var][["x", "y", "waterdepth"]],
                         data_raw[var][var],
                         [points["x"][i], points["y"][i], points["waterdepth"][i]],
@@ -815,7 +815,7 @@ def turbulent_intensity(
 
         if len(idx[0]):
             for i in idx[0]:
-                turbulent_data[var][i] = interp.griddata(
+                turbulent_data.loc[i, var] = interp.griddata(
                     turbulent_data_raw[var][["x", "y", "waterdepth"]],
                     turbulent_data_raw[var][var],
                     [points["x"][i], points["y"][i], points["waterdepth"][i]],
@@ -837,7 +837,7 @@ def turbulent_intensity(
     zero_ind = neg_index[0][zero_bool]
     non_zero_ind = neg_index[0][~zero_bool]
     turbulent_data.loc[zero_ind, "turkin1"] = np.zeros(len(zero_ind))
-    turbulent_data.loc[non_zero_ind, "turkin1"] = [np.nan] * len(non_zero_ind)
+    turbulent_data.loc[non_zero_ind, "turkin1"] = np.nan
 
     turbulent_data["turbulent_intensity"] = (
         np.sqrt(2 / 3 * turbulent_data["turkin1"]) / u_mag * 100
diff --git a/mhkit/river/resource.py b/mhkit/river/resource.py
@@ -311,7 +311,7 @@ def energy_produced(
     # Sample range for pdf
     x = np.linspace(edges.min(), edges.max(), 1000)
     # Calculate the expected value of power
-    expected_power = np.trapz(x * hist_dist.pdf(x), x=x)
+    expected_power = np.trapezoid(x * hist_dist.pdf(x), x=x)
     # Note: Built-in Expected Value method often throws warning
     # EV = hist_dist.expect(lb=edges.min(), ub=edges.max())
     # Calculate energy
diff --git a/mhkit/tests/wave/io/test_cdip.py b/mhkit/tests/wave/io/test_cdip.py
@@ -77,8 +77,8 @@ def test_dates_to_timestamp(self):
             self.test_nc, start_date=start_date, end_date=end_date
         )
 
-        start_dt = datetime.utcfromtimestamp(start_stamp).replace(tzinfo=pytz.UTC)
-        end_dt = datetime.utcfromtimestamp(end_stamp).replace(tzinfo=pytz.UTC)
+        start_dt = datetime.fromtimestamp(start_stamp, pytz.UTC)
+        end_dt = datetime.fromtimestamp(end_stamp, pytz.UTC)
 
         self.assertEqual(start_dt, start_date)
         self.assertEqual(end_dt, end_date)
diff --git a/mhkit/tidal/io/noaa.py b/mhkit/tidal/io/noaa.py
@@ -501,7 +501,12 @@ def _xml_to_dataframe(response: requests.Response) -> tuple[pd.DataFrame, dict]:
         [pd.DataFrame(obs.attrib, index=[0]) for obs in data], ignore_index=True
     )
 
-    df["t"] = pd.to_datetime(df.t)
+    try:
+        df["t"] = pd.to_datetime(pd.to_numeric(df.t), unit="ms")
+    except ValueError:
+        # Don't convert df.t to numeric if its a datetime formatted string
+        df["t"] = pd.to_datetime(df.t)
+
     df = df.set_index("t")
     df.drop_duplicates(inplace=True)
 
@@ -547,7 +552,7 @@ def read_noaa_json(filename: str, to_pandas: bool = True) -> tuple[pd.DataFrame,
         # Remainder is DataFrame
         data = pd.DataFrame.from_dict(json_data)
         # Convert from epoch to date time
-        data.index = pd.to_datetime(data.index, unit="ms")
+        data.index = pd.to_datetime(pd.to_numeric(data.index), unit="ms")
 
     except ValueError:  # using cache.py format
         if "metadata" in json_data:
diff --git a/mhkit/warnings.py b/mhkit/warnings.py
@@ -0,0 +1,26 @@
+import warnings
+
+# Only suppress specific, reviewed warnings here.
+# Example: Suppress a known FutureWarning from a specific dependency
+# warnings.filterwarnings(
+#     "ignore",
+#     category=FutureWarning,
+#     module=r"^some_dependency\.module$",
+#     message=r"This is a known harmless future warning."
+# )
+
+# Add more targeted filters as needed, after review.
+
+
+def configure_warnings():
+    """
+    Call this function at package import to apply MHKiT's targeted warning filters.
+    """
+    # Example: Uncomment and edit below to suppress a specific warning
+    # warnings.filterwarnings(
+    #     "ignore",
+    #     category=FutureWarning,
+    #     module=r"^some_dependency\.module$",
+    #     message=r"This is a known harmless future warning."
+    # )
+    pass
diff --git a/mhkit/wave/graphics.py b/mhkit/wave/graphics.py
@@ -839,7 +839,10 @@ def plot_boxplot(Hs, buoy_title=None):
     bp2 = plt.subplot(gs[1, :])
     meanprops = dict(linewidth=2.5, marker="|", markersize=25)
     bp2_example = bp2.boxplot(
-        bp_sample2, vert=False, flierprops=flierprops, medianprops=medianprops
+        bp_sample2,
+        orientation="horizontal",
+        flierprops=flierprops,
+        medianprops=medianprops,
     )
     sample_mean = 2.3
     bp2.scatter(sample_mean, 1, marker="|", color="g", linewidths=1.0, s=200)
diff --git a/mhkit/wave/io/ndbc.py b/mhkit/wave/io/ndbc.py
@@ -1,7 +1,7 @@
 import os
 from collections import OrderedDict as _OrderedDict
 from collections import defaultdict as _defaultdict
-from io import BytesIO
+from io import BytesIO, StringIO
 import re
 import requests
 import zlib
@@ -19,6 +19,9 @@
     convert_nested_dict_and_pandas,
 )
 
+# Set pandas option to opt-in to future behavior
+pd.set_option("future.no_silent_downcasting", True)
+
 
 def read_file(file_name, missing_values=["MM", 9999, 999, 99], to_pandas=True):
     """
@@ -102,21 +105,25 @@ def read_file(file_name, missing_values=["MM", 9999, 999, 99], to_pandas=True):
             header=None,
             names=header,
             comment="#",
-            parse_dates=[parse_vals],
         )
     # If first line is not commented, then the first row can be used as header
     else:
-        data = pd.read_csv(
-            file_name, sep="\\s+", header=0, comment="#", parse_dates=[parse_vals]
-        )
+        data = pd.read_csv(file_name, sep="\\s+", header=0, comment="#")
 
     # Convert index to datetime
     date_column = "_".join(parse_vals)
+    data[date_column] = (
+        data[parse_vals].apply(lambda val: val.astype("string")).agg(" ".join, axis=1)
+    )
+
     data["Time"] = pd.to_datetime(data[date_column], format=date_format)
     data.index = data["Time"].values
+
     # Remove date columns
     del data[date_column]
     del data["Time"]
+    for val in parse_vals:
+        del data[val]
 
     # If there was a row of units, convert to dictionary
     if units_exist:
@@ -126,7 +133,11 @@ def read_file(file_name, missing_values=["MM", 9999, 999, 99], to_pandas=True):
 
     # Convert columns to numeric data if possible, otherwise leave as string
     for column in data:
-        data[column] = pd.to_numeric(data[column], errors="ignore")
+        try:
+            data[column] = pd.to_numeric(data[column])
+        except (ValueError, TypeError):
+            # Keep as string if conversion fails
+            pass
 
     # Convert column names to float if possible (handles frequency headers)
     # if there is non-numeric name, just leave all as strings.
@@ -136,7 +147,8 @@ def read_file(file_name, missing_values=["MM", 9999, 999, 99], to_pandas=True):
         data.columns = data.columns
 
     # Replace indicated missing values with nan
-    data.replace(missing_values, np.nan, inplace=True)
+    data = data.replace(missing_values, np.nan)
+    data = data.infer_objects(copy=False)
 
     if not to_pandas:
         data = convert_to_dataset(data)
@@ -234,7 +246,7 @@ def available_data(
             msg = f"request.get({ndbc_data}) failed by returning code of {response.status_code}"
             raise Exception(msg)
 
-        filenames = pd.read_html(response.text)[0].Name.dropna()
+        filenames = pd.read_html(StringIO(response.text))[0].Name.dropna()
         buoys = _parse_filenames(parameter, filenames)
 
         available_data = buoys.copy(deep=True)

Original file line number	Diff line number	Diff line change
`@@ -77,8 +77,8 @@ def test_dates_to_timestamp(self):`
`77`	`77`	`self.test_nc, start_date=start_date, end_date=end_date`
`78`	`78`	`)`
`79`	`79`
`80`		`- start_dt = datetime.utcfromtimestamp(start_stamp).replace(tzinfo=pytz.UTC)`
`81`		`- end_dt = datetime.utcfromtimestamp(end_stamp).replace(tzinfo=pytz.UTC)`
	`80`	`+ start_dt = datetime.fromtimestamp(start_stamp, pytz.UTC)`
	`81`	`+ end_dt = datetime.fromtimestamp(end_stamp, pytz.UTC)`
`82`	`82`
`83`	`83`	`self.assertEqual(start_dt, start_date)`
`84`	`84`	`self.assertEqual(end_dt, end_date)`