Added improvements in to_datetime Error reporting message (#47995)

* #16757: improvement of to_datetime errors

* The exceptions from to_datetime(errors='raise') could include information about the exception

* testcase updates

* testcase updates

* testcase updates

* testcase updates

Author: dannyi96

pandas/_libs/tslib.pyx

@@ -595,7 +595,7 @@ cpdef array_to_datetime(
                                 continue
                             elif is_raise:
                                 raise ValueError(
-                                    f"time data {val} doesn't match format specified"
+                                    f"time data \"{val}\" at position {i} doesn't match format specified"
                                 )
                             return values, tz_out
 
@@ -611,7 +611,7 @@ cpdef array_to_datetime(
                             if is_coerce:
                                 iresult[i] = NPY_NAT
                                 continue
-                            raise TypeError("invalid string coercion to datetime")
+                            raise TypeError(f"invalid string coercion to datetime for \"{val}\" at position {i}")
 
                         if tz is not None:
                             seen_datetime_offset = True

pandas/_libs/tslibs/parsing.pyx

@@ -286,7 +286,7 @@ def parse_datetime_string(
         datetime dt
 
     if not _does_string_look_like_datetime(date_string):
-        raise ValueError('Given date string not likely a datetime.')
+        raise ValueError(f'Given date string {date_string} not likely a datetime')
 
     if does_string_look_like_time(date_string):
         # use current datetime as default, not pass _DEFAULT_DATETIME
@@ -320,7 +320,7 @@ def parse_datetime_string(
     except TypeError:
         # following may be raised from dateutil
         # TypeError: 'NoneType' object is not iterable
-        raise ValueError('Given date string not likely a datetime.')
+        raise ValueError(f'Given date string {date_string} not likely a datetime')
 
     return dt
 
@@ -396,7 +396,7 @@ cdef parse_datetime_string_with_reso(
         int out_tzoffset
 
     if not _does_string_look_like_datetime(date_string):
-        raise ValueError('Given date string not likely a datetime.')
+        raise ValueError(f'Given date string {date_string} not likely a datetime')
 
     parsed, reso = _parse_delimited_date(date_string, dayfirst)
     if parsed is not None:

pandas/tests/scalar/period/test_period.py

@@ -301,7 +301,7 @@ def test_invalid_arguments(self):
         with pytest.raises(ValueError, match=msg):
             Period(month=1)
 
-        msg = "Given date string not likely a datetime"
+        msg = "Given date string -2000 not likely a datetime"
         with pytest.raises(ValueError, match=msg):
             Period("-2000", "A")
         msg = "day is out of range for month"

pandas/tests/tools/test_to_datetime.py

@@ -951,7 +951,7 @@ def test_datetime_invalid_scalar(self, value, format, infer):
         msg = (
             "is a bad directive in format|"
             "second must be in 0..59|"
-            "Given date string not likely a datetime"
+            f"Given date string {value} not likely a datetime"
         )
         with pytest.raises(ValueError, match=msg):
             to_datetime(
@@ -1003,7 +1003,7 @@ def test_datetime_invalid_index(self, values, format, infer):
 
         msg = (
             "is a bad directive in format|"
-            "Given date string not likely a datetime|"
+            f"Given date string {values[0]} not likely a datetime|"
             "second must be in 0..59"
         )
         with pytest.raises(ValueError, match=msg):
@@ -2220,7 +2220,7 @@ def test_day_not_in_month_raise(self, cache):
 
     @pytest.mark.parametrize("arg", ["2015-02-29", "2015-02-32", "2015-04-31"])
     def test_day_not_in_month_raise_value(self, cache, arg):
-        msg = f"time data {arg} doesn't match format specified"
+        msg = f'time data "{arg}" at position 0 doesn\'t match format specified'
         with pytest.raises(ValueError, match=msg):
             to_datetime(arg, errors="raise", format="%Y-%m-%d", cache=cache)