ENH: Warn on unused arguments to resample (pandas-dev#62101)

Author: jbrockmendel

pandas/core/resample.py

@@ -2185,6 +2185,22 @@ def __init__(
         else:
             freq = to_offset(freq)
 
+        if not isinstance(freq, Tick):
+            if offset is not None:
+                warnings.warn(
+                    "The 'offset' keyword does not take effect when resampling "
+                    "with a 'freq' that is not Tick-like (h, m, s, ms, us, ns)",
+                    RuntimeWarning,
+                    stacklevel=find_stack_level(),
+                )
+            if origin != "start_day":
+                warnings.warn(
+                    "The 'origin' keyword does not take effect when resampling "
+                    "with a 'freq' that is not Tick-like (h, m, s, ms, us, ns)",
+                    RuntimeWarning,
+                    stacklevel=find_stack_level(),
+                )
+
         end_types = {"ME", "YE", "QE", "BME", "BYE", "BQE", "W"}
         rule = freq.rule_code
         if rule in end_types or ("-" in rule and rule[: rule.find("-")] in end_types):

pandas/tests/resample/test_datetime_index.py

@@ -885,14 +885,14 @@ def test_resample_origin_epoch_with_tz_day_vs_24h(unit):
     random_values = np.random.default_rng(2).standard_normal(len(rng))
     ts_1 = Series(random_values, index=rng)
 
-    result_1 = ts_1.resample("D", origin="epoch").mean()
+    result_1 = ts_1.resample("D").mean()
     result_2 = ts_1.resample("24h", origin="epoch").mean()
     tm.assert_series_equal(result_1, result_2, check_freq=False)
     # GH#41943 check_freq=False bc Day and Hour(24) no longer compare as equal
 
     # check that we have the same behavior with epoch even if we are not timezone aware
     ts_no_tz = ts_1.tz_localize(None)
-    result_3 = ts_no_tz.resample("D", origin="epoch").mean()
+    result_3 = ts_no_tz.resample("D").mean()
     result_4 = ts_no_tz.resample("24h", origin="epoch").mean()
     tm.assert_series_equal(result_1, result_3.tz_localize(rng.tz), check_freq=False)
     tm.assert_series_equal(result_1, result_4.tz_localize(rng.tz), check_freq=False)
@@ -901,7 +901,7 @@ def test_resample_origin_epoch_with_tz_day_vs_24h(unit):
     start, end = "2000-10-01 23:30:00+0200", "2000-12-02 00:30:00+0200"
     rng = date_range(start, end, freq="7min").as_unit(unit)
     ts_2 = Series(random_values, index=rng)
-    result_5 = ts_2.resample("D", origin="epoch").mean()
+    result_5 = ts_2.resample("D").mean()
     result_6 = ts_2.resample("24h", origin="epoch").mean()
     tm.assert_series_equal(result_1.tz_localize(None), result_5.tz_localize(None))
     tm.assert_series_equal(result_1.tz_localize(None), result_6.tz_localize(None))
@@ -910,6 +910,7 @@ def test_resample_origin_epoch_with_tz_day_vs_24h(unit):
 def test_resample_origin_with_day_freq_on_dst(unit):
     # GH 31809
     tz = "America/Chicago"
+    msg = "The '(origin|offset)' keyword does not take effect"
 
     def _create_series(values, timestamps, freq="D"):
         return Series(
@@ -927,7 +928,9 @@ def _create_series(values, timestamps, freq="D"):
 
     expected = _create_series([24.0, 25.0], ["2013-11-02", "2013-11-03"])
     for origin in ["epoch", "start", "start_day", start, None]:
-        result = ts.resample("D", origin=origin).sum()
+        warn = RuntimeWarning if origin != "start_day" else None
+        with tm.assert_produces_warning(warn, match=msg):
+            result = ts.resample("D", origin=origin).sum()
         tm.assert_series_equal(result, expected)
 
     # test complex behavior of origin/offset in a DST context
@@ -939,7 +942,8 @@ def _create_series(values, timestamps, freq="D"):
     # GH#61985 changed this to behave like "B" rather than "24h"
     expected_ts = ["2013-11-03 00:00-05:00"]
     expected = _create_series([25.0], expected_ts)
-    result = ts.resample("D", origin="start", offset="-2h").sum()
+    with tm.assert_produces_warning(RuntimeWarning, match=msg):
+        result = ts.resample("D", origin="start", offset="-2h").sum()
     tm.assert_series_equal(result, expected)
 
     expected_ts = ["2013-11-02 22:00-05:00", "2013-11-03 21:00-06:00"]
@@ -950,17 +954,20 @@ def _create_series(values, timestamps, freq="D"):
     # GH#61985 changed this to behave like "B" rather than "24h"
     expected_ts = ["2013-11-03 00:00-05:00"]
     expected = _create_series([25.0], expected_ts)
-    result = ts.resample("D", origin="start", offset="2h").sum()
+    with tm.assert_produces_warning(RuntimeWarning, match=msg):
+        result = ts.resample("D", origin="start", offset="2h").sum()
     tm.assert_series_equal(result, expected)
 
     expected_ts = ["2013-11-03 00:00-05:00"]
     expected = _create_series([25.0], expected_ts)
-    result = ts.resample("D", origin="start", offset="-1h").sum()
+    with tm.assert_produces_warning(RuntimeWarning, match=msg):
+        result = ts.resample("D", origin="start", offset="-1h").sum()
     tm.assert_series_equal(result, expected)
 
     expected_ts = ["2013-11-03 00:00-05:00"]
     expected = _create_series([25.0], expected_ts)
-    result = ts.resample("D", origin="start", offset="1h").sum()
+    with tm.assert_produces_warning(RuntimeWarning, match=msg):
+        result = ts.resample("D", origin="start", offset="1h").sum()
     tm.assert_series_equal(result, expected)
 
 
@@ -2022,9 +2029,8 @@ def test_resample_empty_series_with_tz():
     df = DataFrame({"ts": [], "values": []}).astype(
         {"ts": "datetime64[ns, Atlantic/Faroe]"}
     )
-    result = df.resample("2MS", on="ts", closed="left", label="left", origin="start")[
-        "values"
-    ].sum()
+    rs = df.resample("2MS", on="ts", closed="left", label="left")
+    result = rs["values"].sum()
 
     expected_idx = DatetimeIndex(
         [], freq="2MS", name="ts", dtype="datetime64[ns, Atlantic/Faroe]"

pandas/tests/resample/test_period_index.py

@@ -921,10 +921,10 @@ def test_resample_with_offset_month(self):
         ser = Series(np.arange(len(pi)), index=pi)
         msg = "Resampling with a PeriodIndex is deprecated"
         with tm.assert_produces_warning(FutureWarning, match=msg):
-            rs = ser.resample("M", offset="3h")
+            rs = ser.resample("M")
         result = rs.mean()
         result = result.to_timestamp("M")
-        expected = ser.to_timestamp().resample("ME", offset="3h").mean()
+        expected = ser.to_timestamp().resample("ME").mean()
         # TODO: is non-tick the relevant characteristic? (GH 33815)
         expected.index = expected.index._with_freq(None)
         tm.assert_series_equal(result, expected)