add missing return

Author: MarcoGorelli

doc/source/user_guide/timedeltas.rst

@@ -424,6 +424,7 @@ Similarly to other of the datetime-like indices, ``DatetimeIndex`` and ``PeriodI
 Selections work similarly, with coercion on string-likes and slices:
 
 .. ipython:: python
+   :okwarning:
 
    s["1 day":"2 day"]
    s["1 day 01:00:00"]
@@ -432,6 +433,7 @@ Selections work similarly, with coercion on string-likes and slices:
 Furthermore you can use partial string selection and the range will be inferred:
 
 .. ipython:: python
+   :okwarning:
 
    s["1 day":"1 day 5 hours"]
 

doc/source/whatsnew/v1.5.0.rst

@@ -770,6 +770,7 @@ Other Deprecations
 - Deprecated the argument ``na_sentinel`` in :func:`factorize`, :meth:`Index.factorize`, and :meth:`.ExtensionArray.factorize`; pass ``use_na_sentinel=True`` instead to use the sentinel ``-1`` for NaN values and ``use_na_sentinel=False`` instead of ``na_sentinel=None`` to encode NaN values (:issue:`46910`)
 - Deprecated :meth:`DataFrameGroupBy.transform` not aligning the result when the UDF returned DataFrame (:issue:`45648`)
 - Clarified warning from :func:`to_datetime` when delimited dates can't be parsed in accordance to specified ``dayfirst`` argument (:issue:`46210`)
+- Emit warning from :func:`to_datetime` when delimited dates can't be parsed in accordance to specified ``dayfirst`` argument even for dates where leading zero is omitted (e.g. ``31/1/2001``) (:issue:`47880`)
 - Deprecated :class:`Series` and :class:`Resampler` reducers (e.g. ``min``, ``max``, ``sum``, ``mean``) raising a ``NotImplementedError`` when the dtype is non-numric and ``numeric_only=True`` is provided; this will raise a ``TypeError`` in a future version (:issue:`47500`)
 - Deprecated :meth:`Series.rank` returning an empty result when the dtype is non-numeric and ``numeric_only=True`` is provided; this will raise a ``TypeError`` in a future version (:issue:`47500`)
 - Deprecated argument ``errors`` for :meth:`Series.mask`, :meth:`Series.where`, :meth:`DataFrame.mask`, and :meth:`DataFrame.where` as ``errors`` had no effect on this methods (:issue:`47728`)

pandas/_libs/tslibs/parsing.pyx

@@ -99,8 +99,14 @@ cdef:
     int MAX_DAYS_IN_MONTH = 31, MAX_MONTH = 12
 
 
-cdef inline bint _is_not_delimiter(const char ch):
-    return strchr(delimiters, ch) == NULL
+cdef inline bint _is_delimiter(const char ch):
+    return strchr(delimiters, ch) != NULL
+
+
+cdef inline int _parse_1digit(const char* s):
+    cdef int result = 0
+    result += getdigit_ascii(s[0], -10) * 1
+    return result
 
 
 cdef inline int _parse_2digit(const char* s):
@@ -151,18 +157,37 @@ cdef inline object _parse_delimited_date(str date_string, bint dayfirst):
         bint can_swap = 0
 
     buf = get_c_string_buf_and_size(date_string, &length)
-    if length == 10:
+    if length == 10 and _is_delimiter(buf[2]) and _is_delimiter(buf[5]):
         # parsing MM?DD?YYYY and DD?MM?YYYY dates
-        if _is_not_delimiter(buf[2]) or _is_not_delimiter(buf[5]):
-            return None, None
         month = _parse_2digit(buf)
         day = _parse_2digit(buf + 3)
         year = _parse_4digit(buf + 6)
         reso = 'day'
         can_swap = 1
-    elif length == 7:
+    elif length == 9 and _is_delimiter(buf[1]) and _is_delimiter(buf[4]):
+        # parsing M?DD?YYYY and D?MM?YYYY dates
+        month = _parse_1digit(buf)
+        day = _parse_2digit(buf + 2)
+        year = _parse_4digit(buf + 5)
+        reso = 'day'
+        can_swap = 1
+    elif length == 9 and _is_delimiter(buf[2]) and _is_delimiter(buf[4]):
+        # parsing MM?D?YYYY and DD?M?YYYY dates
+        month = _parse_2digit(buf)
+        day = _parse_1digit(buf + 3)
+        year = _parse_4digit(buf + 5)
+        reso = 'day'
+        can_swap = 1
+    elif length == 8 and _is_delimiter(buf[1]) and _is_delimiter(buf[3]):
+        # parsing M?D?YYYY and D?M?YYYY dates
+        month = _parse_1digit(buf)
+        day = _parse_1digit(buf + 2)
+        year = _parse_4digit(buf + 4)
+        reso = 'day'
+        can_swap = 1
+    elif length == 7 and _is_delimiter(buf[2]):
         # parsing MM?YYYY dates
-        if buf[2] == b'.' or _is_not_delimiter(buf[2]):
+        if buf[2] == b'.':
             # we cannot reliably tell whether e.g. 10.2010 is a float
             # or a date, thus we refuse to parse it here
             return None, None
@@ -238,6 +263,18 @@ cdef inline bint does_string_look_like_time(str parse_string):
 
     return 0 <= hour <= 23 and 0 <= minute <= 59
 
+from pandas.util._exceptions import find_stack_level
+
+
+def du_parse_with_warning(*args, **kwargs):
+    parsed = du_parse(*args, **kwargs)
+    warnings.warn(
+        "Parsing datetime strings without a format specified, "
+        "please specify a format to avoid unexpected results",
+        stacklevel=find_stack_level(),
+    )
+    return parsed
+
 
 def parse_datetime_string(
     # NB: This will break with np.str_ (GH#32264) even though
@@ -265,8 +302,12 @@ def parse_datetime_string(
 
     if does_string_look_like_time(date_string):
         # use current datetime as default, not pass _DEFAULT_DATETIME
-        dt = du_parse(date_string, dayfirst=dayfirst,
-                      yearfirst=yearfirst, **kwargs)
+        dt = du_parse_with_warning(
+            date_string,
+            dayfirst=dayfirst,
+            yearfirst=yearfirst,
+            **kwargs,
+        )
         return dt
 
     dt, _ = _parse_delimited_date(date_string, dayfirst)
@@ -282,8 +323,13 @@ def parse_datetime_string(
         pass
 
     try:
-        dt = du_parse(date_string, default=_DEFAULT_DATETIME,
-                      dayfirst=dayfirst, yearfirst=yearfirst, **kwargs)
+        dt = du_parse_with_warning(
+            date_string,
+            default=_DEFAULT_DATETIME,
+            dayfirst=dayfirst,
+            yearfirst=yearfirst,
+            **kwargs,
+        )
     except TypeError:
         # following may be raised from dateutil
         # TypeError: 'NoneType' object is not iterable
@@ -681,7 +727,11 @@ def try_parse_dates(
             date = datetime.now()
             default = datetime(date.year, date.month, 1)
 
-        parse_date = lambda x: du_parse(x, dayfirst=dayfirst, default=default)
+        parse_date = lambda x: du_parse_with_warning(
+            x,
+            dayfirst=dayfirst,
+            default=default,
+        )
 
         # EAFP here
         try:
@@ -728,13 +778,17 @@ def try_parse_date_and_time(
             date = datetime.now()
             default = datetime(date.year, date.month, 1)
 
-        parse_date = lambda x: du_parse(x, dayfirst=dayfirst, default=default)
+        parse_date = lambda x: du_parse_with_warning(
+            x,
+            dayfirst=dayfirst,
+            default=default,
+        )
 
     else:
         parse_date = date_parser
 
     if time_parser is None:
-        parse_time = lambda x: du_parse(x)
+        parse_time = lambda x: du_parse_with_warning(x)
 
     else:
         parse_time = time_parser
@@ -955,7 +1009,7 @@ def guess_datetime_format(dt_str, bint dayfirst=False):
         datetime_attrs_to_format.insert(0, day_attribute_and_format)
 
     try:
-        parsed_datetime = du_parse(dt_str, dayfirst=dayfirst)
+        parsed_datetime = du_parse_with_warning(dt_str, dayfirst=dayfirst)
     except (ValueError, OverflowError):
         # In case the datetime can't be parsed, its format cannot be guessed
         return None

pandas/tests/frame/methods/test_reset_index.py

@@ -337,24 +337,19 @@ def test_reset_index_multiindex_nan(self):
         tm.assert_frame_equal(rs, df)
 
     @pytest.mark.parametrize(
-        "name",
+        "name, warn",
         [
-            None,
-            "foo",
-            2,
-            3.0,
-            pd.Timedelta(6),
-            Timestamp("2012-12-30", tz="UTC"),
-            "2012-12-31",
+            (None, UserWarning),
+            ("foo", UserWarning),
+            (2, None),
+            (3.0, None),
+            (pd.Timedelta(6), None),
+            (Timestamp("2012-12-30", tz="UTC"), FutureWarning),
+            ("2012-12-31", None),
         ],
     )
-    def test_reset_index_with_datetimeindex_cols(self, name):
+    def test_reset_index_with_datetimeindex_cols(self, name, warn):
         # GH#5818
-        warn = None
-        if isinstance(name, Timestamp) and name.tz is not None:
-            # _deprecate_mismatched_indexing
-            warn = FutureWarning
-
         df = DataFrame(
             [[1, 2], [3, 4]],
             columns=date_range("1/1/2013", "1/2/2013"),