Added improvements in to_datetime Error reporting message - Outofbounds error message (pandas-dev#47849)

* Revert "addressing reviews CL#1"

This reverts commit 69dc2d4.

* improving OutOfBoundsDatetime exception messages

* improving OutOfBoundsDatetime exception messages

* improving OutOfBoundsDatetime exception messages

* improving OutOfBoundsDatetime exception messages

* improving OutOfBoundsDatetime exception messages

* testcase update

* improving OutOfBoundsDatetime exception messages

* review comments improving OutOfBoundsDatetime exception messages

* testcase update

* addressed testcase review comments

* Revert "testcase update"

This reverts commit 6da730d.

* precommit code format changes

* precommit code format changes

Author: dannyi96

pandas/_libs/tslib.pyx

@@ -652,7 +652,8 @@ cpdef array_to_datetime(
                     else:
                         raise TypeError(f"{type(val)} is not convertible to datetime")
 
-            except OutOfBoundsDatetime:
+            except OutOfBoundsDatetime as ex:
+                ex.args = (str(ex) + f" present at position {i}", )
                 if is_coerce:
                     iresult[i] = NPY_NAT
                     continue

pandas/tests/base/test_constructors.py

@@ -157,7 +157,7 @@ def test_constructor_datetime_outofbound(self, a, klass):
 
         # Explicit dtype specified
         # Forced conversion fails for all -> all cases raise error
-        msg = "Out of bounds"
+        msg = "Out of bounds|Out of bounds .* present at position 0"
         with pytest.raises(pd.errors.OutOfBoundsDatetime, match=msg):
             klass(a, dtype="datetime64[ns]")
 

pandas/tests/indexes/datetimes/test_constructors.py

@@ -521,7 +521,7 @@ def test_construction_outofbounds(self):
         # coerces to object
         tm.assert_index_equal(Index(dates), exp)
 
-        msg = "Out of bounds nanosecond timestamp"
+        msg = "Out of bounds .* present at position 0"
         with pytest.raises(OutOfBoundsDatetime, match=msg):
             # can't create DatetimeIndex
             DatetimeIndex(dates)

pandas/tests/indexes/datetimes/test_scalar_compat.py

@@ -38,7 +38,7 @@ def test_dti_date(self):
     @pytest.mark.parametrize("data", [["1400-01-01"], [datetime(1400, 1, 1)]])
     def test_dti_date_out_of_range(self, data):
         # GH#1475
-        msg = "Out of bounds nanosecond timestamp: 1400-01-01 00:00:00"
+        msg = "Out of bounds .* present at position 0"
         with pytest.raises(OutOfBoundsDatetime, match=msg):
             DatetimeIndex(data)
 

pandas/tests/tools/test_to_datetime.py

@@ -688,9 +688,10 @@ def test_to_datetime_dt64s(self, cache, dt):
         "dt", [np.datetime64("1000-01-01"), np.datetime64("5000-01-02")]
     )
     def test_to_datetime_dt64s_out_of_bounds(self, cache, dt):
-        msg = f"Out of bounds nanosecond timestamp: {dt}"
+        msg = "Out of bounds .* present at position 0"
         with pytest.raises(OutOfBoundsDatetime, match=msg):
             to_datetime(dt, errors="raise")
+        msg = f"Out of bounds nanosecond timestamp: {dt}"
         with pytest.raises(OutOfBoundsDatetime, match=msg):
             Timestamp(dt)
         assert to_datetime(dt, errors="coerce", cache=cache) is NaT
@@ -973,13 +974,13 @@ def test_datetime_outofbounds_scalar(self, value, format, infer):
         assert res is NaT
 
         if format is not None:
-            msg = "is a bad directive in format|Out of bounds nanosecond timestamp"
+            msg = "is a bad directive in format|Out of bounds .* present at position 0"
             with pytest.raises(ValueError, match=msg):
                 to_datetime(
                     value, errors="raise", format=format, infer_datetime_format=infer
                 )
         else:
-            msg = "Out of bounds nanosecond timestamp"
+            msg = "Out of bounds .* present at position 0"
             with pytest.raises(OutOfBoundsDatetime, match=msg):
                 to_datetime(
                     value, errors="raise", format=format, infer_datetime_format=infer
@@ -1700,7 +1701,7 @@ def test_to_datetime_barely_out_of_bounds(self):
         # in an in-bounds datetime
         arr = np.array(["2262-04-11 23:47:16.854775808"], dtype=object)
 
-        msg = "Out of bounds nanosecond timestamp"
+        msg = "Out of bounds .* present at position 0"
         with pytest.raises(OutOfBoundsDatetime, match=msg):
             to_datetime(arr)
 
@@ -2593,7 +2594,10 @@ def test_invalid_origins_tzinfo(self):
     @pytest.mark.parametrize("format", [None, "%Y-%m-%d %H:%M:%S"])
     def test_to_datetime_out_of_bounds_with_format_arg(self, format):
         # see gh-23830
-        msg = "Out of bounds nanosecond timestamp"
+        msg = (
+            "Out of bounds nanosecond timestamp: 2417-10-27 00:00:00 "
+            "present at position 0"
+        )
         with pytest.raises(OutOfBoundsDatetime, match=msg):
             to_datetime("2417-10-27 00:00:00", format=format)
 

pandas/tests/tslibs/test_array_to_datetime.py

@@ -125,7 +125,7 @@ def test_coerce_outside_ns_bounds(invalid_date, errors):
     kwargs = {"values": arr, "errors": errors}
 
     if errors == "raise":
-        msg = "Out of bounds nanosecond timestamp"
+        msg = "Out of bounds .* present at position 0"
 
         with pytest.raises(ValueError, match=msg):
             tslib.array_to_datetime(**kwargs)
@@ -170,7 +170,9 @@ def test_to_datetime_barely_out_of_bounds():
     # Close enough to bounds that dropping nanos
     # would result in an in-bounds datetime.
     arr = np.array(["2262-04-11 23:47:16.854775808"], dtype=object)
-    msg = "Out of bounds nanosecond timestamp: 2262-04-11 23:47:16"
+    msg = (
+        "Out of bounds nanosecond timestamp: 2262-04-11 23:47:16 present at position 0"
+    )
 
     with pytest.raises(tslib.OutOfBoundsDatetime, match=msg):
         tslib.array_to_datetime(arr)