[Yaml-core] PyYaml serializing issues.

[Aniket P. <ma...@fe...>]

Hello PyYaml team!

Thanks a lot for the amazing tool/library. Great work!

Whilst using PyYaml, I had some queries on how the library serializes
different objects. For our purposes, we were using the `datetime`
library and were having some issues with it.

For example, we were dumping a `datetime.date` [0] object and the
resultant output showed that it was being serialized as a string,
whereas a similar object of type `datetime.time` [1] is being
serialized as a binary object ('!!python/object/apply:datetime.time\n-
!!binary |\n  DAwMAAAA\n').

I am interested in finding out how PyYaml serializes the two very
similar classes so differently. Both of these classes have implemented
almost the same methods (__str__, __repr__).

Logs for more information:

----------------------------------------------------------------------------
# I have used the default dumper, (yaml.dump()) method.
>>> d = datetime.date.today()
>>> d
datetime.date(2020, 1, 14)
>>> yaml.dump(d)
'2020-01-14\n...\n'
>>> datetime.time(12, 12, 12)
datetime.time(12, 12, 12)
>>> t = datetime.time(12, 12, 12)
>>> t
datetime.time(12, 12, 12)
>>> yaml.dump(t)
'!!python/object/apply:datetime.time\n- !!binary |\n  DAwMAAAA\n'
----------------------------------------------------------------------------

[0]: https://github.com/python/cpython/blob/3.8/Lib/datetime.py#L789
[1]:https://github.com/python/cpython/blob/3.8/Lib/datetime.py#L1211

-- 
Thanks
Regards

Aniket Pradhan
http://home.iiitd.edu.in/~aniket17133/
Aliases: MeWjOr/major

()  ascii ribbon campaign - against html e-mail
/\  www.asciiribbon.org   - against proprietary attachments

P.S. I know I can write a custom dumper for `datetime.time` objects,
but was wondering what is the reason that two almost similar classes
are being serialized so differently.