Allow read_tmy3 to handle 00:00 timestamps (#1494)

* Update read_tmy3 date shift

* Add test coverage

* Add whatsnew entry

* Clean up TMY3 table

* Add comments to test files

* Fix stickler

* Update admonition

* Apply changes from kanderso-nrel

* Add encoding option to read_tmy3

* Fix stickler

* Update pvlib/iotools/tmy.py

Co-authored-by: Kevin Anderson <[email protected]>

* Remove encoding parameter and add try/except

* Fix incorrect parenthesis

* Add fbuf.close and first within try/except

* Add finally clause

* Utilize open and add _parse_tmy3

* Fix stickler

Co-authored-by: Kevin Anderson <[email protected]>

docs/sphinx/source/whatsnew/v0.9.2.rst

@@ -27,12 +27,15 @@ Bug fixes
   where passing localized timezones with large UTC offsets could return
   rise/set/transit times for the wrong day in recent versions of ``ephem``
   (:issue:`1449`, :pull:`1448`)
+* :py:func:`pvlib.iotools.read_tmy3` is now able to accept midnight
+  timestamps as either 24:00 (which is the standard) as well as 00:00.
+  Previously 00:00 timestamps would incorrectly be moved one day forward.
+  (:pull:`1494`)
 * :py:func:`pvlib.iotools.get_psm3` now raises a deprecation warning if
   the `leap_day` parameter is not specified in a single-year request.
   Starting in pvlib 0.11.0 `leap_day` will default to True instead of False.
   (:issue:`1481`, :pull:`1511`)
 
-
 Testing
 ~~~~~~~
 * Switched CI testing provider from Azure to GitHub Actions (:pull:`1306`)
@@ -46,7 +49,6 @@ Documentation
 * Add gallery example of simulating rearside irradiance for a fixed-tilt
   array with pvfactors (:pull:`1470`)
 
-
 Benchmarking
 ~~~~~~~~~~~~~
 * Updated version of numba in asv.conf from 0.36.1 to 0.40.0 to solve numba/numpy conflict. (:issue:`1439`, :pull:`1440`)