Preserve formatting of reference time units under pandas 2.0.0 (#7441)

* [test-upstream] Preserve formatting of reference time units

Changes in pandas 2.0.0 interfere with the way we expect some times
to be formatted.

* Use strftime to force expected date format

Co-authored-by: Justus Magin <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* [test-upstream] Preserve formatting of reference time units

Changes in pandas 2.0.0 interfere with the way we expect some times
to be formatted.

* Use strftime to force expected date format

Co-authored-by: Justus Magin <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* [test-upstream] Remove old comment and move what's new entry to latest version

* [test-upstream] remove old comment and move what's new entry

* Update xarray/core/formatting.py

* Use isoformat() now that pandas zero-pads years

Co-authored-by: Justus Magin <[email protected]>

* [test-upstream] update what's new

* replace the `isoformat` separator with `" "` [test-upstream]

* use the `sep` parameter to `isoformat` instead

Co-authored-by: Spencer Clark <[email protected]>

* Update doc/whats-new.rst

---------

Co-authored-by: Deepak Cherian <[email protected]>
Co-authored-by: Justus Magin <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Justus Magin <[email protected]>

Author: 4 people

doc/whats-new.rst

@@ -47,6 +47,12 @@ Bug fixes
   By `Thomas Coleman <https://github.com/ColemanTom>`_.
 - Proper plotting when passing :py:class:`~matplotlib.colors.BoundaryNorm` type argument in :py:meth:`DataArray.plot`. (:issue:`4061`, :issue:`7014`,:pull:`7553`)
   By `Jelmer Veenstra <https://github.com/veenstrajelmer>`_.
+- Ensure the formatting of time encoding reference dates outside the range of
+  nanosecond-precision datetimes remains the same under pandas version 2.0.0
+  (:issue:`7420`, :pull:`7441`).
+  By `Justus Magin <https://github.com/keewis>`_ and
+  `Spencer Clark  <https://github.com/spencerkclark>`_.
+
 
 Documentation
 ~~~~~~~~~~~~~

xarray/core/formatting.py

@@ -114,9 +114,9 @@ def calc_max_rows_last(max_rows: int) -> int:
 
 def format_timestamp(t):
     """Cast given object to a Timestamp and return a nicely formatted string"""
-    # Timestamp is only valid for 1678 to 2262
     try:
-        datetime_str = str(pd.Timestamp(t))
+        timestamp = pd.Timestamp(t)
+        datetime_str = timestamp.isoformat(sep=" ")
     except OutOfBoundsDatetime:
         datetime_str = str(t)