Pandas WARN unnecessary "can't infer datetime format" for out-of-bounds datetimes

pd.to_datetime(['9000-01-01'])

shows

<ipython-input-1-1af851f16643>:1: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.

and then also errors

OutOfBoundsDatetime: Out of bounds nanosecond timestamp: 9000-01-01, at position 0

However, the format could be inferred (%Y-%m-%d), it just can't be parsed as nanosecond timestamp.

The warning should probably not be shown in this case.

I'll make a PR, it's an easy change

EDIT: it's not quite as trivial as I initially thought - still, not a blocker for 2.0 by any stretch of the imagination

Comment From: MarcoGorelli

This isn't really worth it

Removing the check of whether the guessed format parses barely improves performance:

upstream/main:

In [1]: %%timeit
   ...: to_datetime(['2020-01-01'])
   ...: 
   ...: 
175 µs ± 3.33 µs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)

With the check removed:

In [1]: %%timeit
   ...: to_datetime(['2020-01-01'])
   ...: 
   ...: 
161 µs ± 2.16 µs per loop (mean ± std. dev. of 7 runs, 10,000 loops each)

And note that the check only needs doing once per to_datetime call, so if there's a handful of elements then the time difference becomes really negligible

Closing then, this probably isn't worth pursuing for now. May revisit later