TimeDelta misleading 'precision' argument & v4 breaking change

[OJFord]

The docstring is correct:

:param precision: The time unit used for (de)serialization. Must be one of 'weeks',
   'days', 'hours', 'minutes', 'seconds', 'milliseconds' or 'microseconds'.

it's the units (de)serialised to; it doesn't effect precision.

That is, 123 microseconds with the default "seconds" precision will serialise to 0.000123, not 0.

This seems to be a breaking change in v4.0.0, perhaps somehow related to the date/time/isoformat change, I think in 0c90110 as previously it would return a timedelta(microseconds=123) in the _serialize method, but now returns a float via utils.timedelta_to_microseconds divided by the 'precision''s mapping value, i.e. 1000000 for seconds.

[sloria]

sorry, can you clarify the ask here? Is it the name of the argument that you're suggesting be changed?

[OJFord]

1. yes, I'd suggest units or denominator or scale or something rather than precision, if this is the way it's to behave;
2. was this breaking change intentional? It's not clear to me that it was (or how to fix it) from the release notes: what previously serialised as 0 now serialises as 0.000123 (for example), I don't know how to fix it without subclassing fields.TimeDelta to round it as in v3.