BUG: to_datetime with errors coerce throws InvalidOperation decimal.DivisionImpossible

[gkep]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

import pandas as pd
x = pd.to_datetime('87156549591102612381000001219H5',infer_datetime_format=True, errors='coerce')

Issue Description

Function should return NaT in case of problem, not exception decimal.DivisionImpossible:
InvalidOperation: [<class 'decimal.DivisionImpossible'>]

Problem present also in former versions of pandas.

Expected Behavior

x = pd.NaT

Installed Versions

INSTALLED VERSIONS

commit : 87cfe4e
python : 3.10.4.final.0
python-bits : 64
OS : Windows
OS-release : 10
Version : 10.0.19044
machine : AMD64
processor : Intel64 Family 6 Model 154 Stepping 3, GenuineIntel
byteorder : little
LC_ALL : None
LANG : en
LOCALE : English_United States.1252

pandas : 1.5.0
numpy : 1.23.1
pytz : 2022.1
dateutil : 2.8.2
setuptools : 57.0.0
pip : 22.2.2
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : None
IPython : 7.31.1
pandas_datareader: None
bs4 : None
bottleneck : 1.3.5
brotli :
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : 3.5.3
numba : None
numexpr : 2.8.3
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : None
pyreadstat : None
pyxlsb : None
s3fs : None
scipy : 1.9.1
snappy : None
sqlalchemy : None
tables : None
tabulate : None
xarray : None
xlrd : None
xlwt : None
zstandard : None
tzdata : None

[MarcoGorelli]

thanks @gkep for the report

this comes from dateutil:

In [4]: from dateutil.parser import parse as du_parse

In [5]: du_parse('87156549591102612381000001219H5')
---------------------------------------------------------------------------
InvalidOperation                          Traceback (most recent call last)
Cell In[5], line 1
----> 1 du_parse('87156549591102612381000001219H5')

File ~/pandas-dev/.311venv/lib/python3.11/site-packages/dateutil/parser/_parser.py:1368, in parse(timestr, parserinfo, **kwargs)
   1366     return parser(parserinfo).parse(timestr, **kwargs)
   1367 else:
-> 1368     return DEFAULTPARSER.parse(timestr, **kwargs)

File ~/pandas-dev/.311venv/lib/python3.11/site-packages/dateutil/parser/_parser.py:640, in parser.parse(self, timestr, default, ignoretz, tzinfos, **kwargs)
    636 if default is None:
    637     default = datetime.datetime.now().replace(hour=0, minute=0,
    638                                               second=0, microsecond=0)
--> 640 res, skipped_tokens = self._parse(timestr, **kwargs)
    642 if res is None:
    643     raise ParserError("Unknown string format: %s", timestr)

File ~/pandas-dev/.311venv/lib/python3.11/site-packages/dateutil/parser/_parser.py:740, in parser._parse(self, timestr, dayfirst, yearfirst, fuzzy, fuzzy_with_tokens)
    736     value = None
    738 if value is not None:
    739     # Numeric token
--> 740     i = self._parse_numeric_token(l, i, info, ymd, res, fuzzy)
    742 # Check weekday
    743 elif info.weekday(l[i]) is not None:

File ~/pandas-dev/.311venv/lib/python3.11/site-packages/dateutil/parser/_parser.py:936, in parser._parse_numeric_token(self, tokens, idx, info, ymd, res, fuzzy)
    932     (idx, hms) = self._parse_hms(idx, tokens, info, hms_idx)
    933     if hms is not None:
    934         # TODO: checking that hour/minute/second are not
    935         # already set?
--> 936         self._assign_hms(res, value_repr, hms)
    938 elif idx + 2 < len_l and tokens[idx + 1] == ':':
    939     # HH:MM[:SS[.ss]]
    940     res.hour = int(value)

File ~/pandas-dev/.311venv/lib/python3.11/site-packages/dateutil/parser/_parser.py:1047, in parser._assign_hms(self, res, value_repr, hms)
   1044 if hms == 0:
   1045     # Hour
   1046     res.hour = int(value)
-> 1047     if value % 1:
   1048         res.minute = int(60*(value % 1))
   1050 elif hms == 1:

InvalidOperation: [<class 'decimal.DivisionImpossible'>]

I'd suggest reporting there

As for what to do on the pandas side, this should probably be caught, we just need to add this exception to

pandas/pandas/_libs/tslibs/parsing.pyx

Lines 892 to 896 in 5645847

	try:
	parsed_datetime = du_parse(dt_str, dayfirst=dayfirst)
	except (ValueError, OverflowError):
	# In case the datetime can't be parsed, its format cannot be guessed
	return None

, and add a test case for this in

pandas/pandas/tests/tslibs/test_parsing.py

Lines 238 to 254 in a23f101

	@pytest.mark.parametrize(
	"invalid_dt",
	[
	"01/2013",
	"12:00:00",
	"1/1/1/1",
	"this_is_not_a_datetime",
	"51a",
	"13/2019",
	"202001", # YYYYMM isn't ISO8601
	"2020/01", # YYYY/MM isn't ISO8601 either
	],
	)
	def test_guess_datetime_format_invalid_inputs(invalid_dt):
	# A datetime string must include a year, month and a day for it to be
	# guessable, in addition to being a string that looks like a datetime.
	assert parsing.guess_datetime_format(invalid_dt) is None

Interested in submitting a PR?

[kcaashish]

take