BUG: Convert output type in Excel for MultiIndex with period levels

[ZKaoChi]

• closes BUG: Inconsistent output type in Excel for PeriodIndex in Index or MultiIndex #60099
• Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• Added an entry in the latest doc/source/whatsnew/v3.0.0.rst file if fixing a bug or adding a new feature.

[rhshadrach]

Thanks for the PR! Please add tests whenever changing functionality.

[rhshadrach]

Looking good - can you also add one test in excel/test_writers.py for round-tripping (writes a DataFrame to an excel file, reads it, and asserts its the same as the original DataFrame).

[ZKaoChi]

Looking good - can you also add one test in excel/test_writers.py for round-tripping (writes a DataFrame to an excel file, reads it, and asserts its the same as the original DataFrame).

My test case gets called 8 times, 6 times the dtype of result is 'datetime64[us]', and the other two times is 'datetime64[s]'. I don't know why this is happening. I had to do unit conversions for result.

[ZKaoChi]

Looking good - can you also add one test in excel/test_writers.py for round-tripping (writes a DataFrame to an excel file, reads it, and asserts its the same as the original DataFrame).

My test case gets called 8 times, 6 times the dtype of result is 'datetime64[us]', and the other two times is 'datetime64[s]'. I don't know why this is happening. I had to do unit conversions for result.

I've found that the dtype of result depends on the suffix of tmp_excel, only in the case of .ods the dtype is datetime64[s] and in other cases it is datetime64[us]. The latest PR has been completed by changing the expected to match result, please check it out. Thanks a lot!!!

[ZKaoChi]

And maybe it can be a new issue to unify the units of datetime in different tmp_excel formats.

[rhshadrach]

Can you add a note in the whatsnew for v3.0.0 under the I/O section.

[ZKaoChi]

Can you add a note in the whatsnew for v3.0.0 under the I/O section.

Sure.