ERR: better error message on invalid timestamp parsing

[jreback]

In [2]: pd.Timestamp('2010011')
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-2-ccfd37d6ef5d> in <module>()
----> 1 pd.Timestamp('2010011')

/Users/jreback/pandas/pandas/tslib.pyx in pandas.tslib.Timestamp.__new__ (pandas/tslib.c:8958)()
    296         cdef _Timestamp ts_base
    297 
--> 298         ts = convert_to_tsobject(ts_input, tz, unit)
    299 
    300         if ts.value == NPY_NAT:

/Users/jreback/pandas/pandas/tslib.pyx in pandas.tslib.convert_to_tsobject (pandas/tslib.c:24122)()
   1232 
   1233     if util.is_string_object(ts):
-> 1234         return convert_str_to_tsobject(ts, tz, unit)
   1235 
   1236     if ts is None or ts is NaT:

/Users/jreback/pandas/pandas/tslib.pyx in pandas.tslib.convert_str_to_tsobject (pandas/tslib.c:26110)()
   1361                 ts = parse_datetime_string(ts, dayfirst=dayfirst, yearfirst=yearfirst)
   1362             except Exception:
-> 1363                 raise ValueError
   1364 
   1365     return convert_to_tsobject(ts, tz, unit)

ValueError: 

we are should have a nice error message here. This is normally caught for example in to_datetime, but if its only a single Timestamp no message.

[jreback]

xref #11976

[jreback]

can start by seeing the caught Exception has a nice message which I think it does (this is what is called in cython)

from pandas.compat import parse_date
In [5]: parse_date('2010011')
ValueError: year is out of range

[jreback]

#14684 makes this a bit better.

>>> pd.Timestamp('2010011')
Traceback (most recent call last):
  File "pandas/tslib.pyx", line 1587, in pandas.tslib.convert_str_to_tsobject (pandas/tslib.c:29386)
    _string_to_dts(ts, &obj.dts, &out_local, &out_tzoffset)
  File "pandas/src/datetime.pxd", line 138, in datetime._string_to_dts (pandas/tslib.c:95654)
    result = _cstring_to_dts(tmp, len(val), dts, out_local, out_tzoffset)
ValueError: Error parsing datetime string "2010011" at position 7

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "pandas/tslib.pyx", line 1607, in pandas.tslib.convert_str_to_tsobject (pandas/tslib.c:29811)
    ts = parse_datetime_string(
  File "pandas/tslib.pyx", line 1973, in pandas.tslib.parse_datetime_string (pandas/tslib.c:35412)
    dt = parse_date(date_string, default=_DEFAULT_DATETIME,
  File "/Users/jreback/miniconda3/envs/py3.6/lib/python3.6/site-packages/dateutil/parser.py", line 1168, in parse
    return DEFAULTPARSER.parse(timestr, **kwargs)
  File "/Users/jreback/miniconda3/envs/py3.6/lib/python3.6/site-packages/dateutil/parser.py", line 578, in parse
    if cday > monthrange(cyear, cmonth)[1]:
  File "/Users/jreback/miniconda3/envs/py3.6/lib/python3.6/calendar.py", line 124, in monthrange
    day1 = weekday(year, month, 1)
  File "/Users/jreback/miniconda3/envs/py3.6/lib/python3.6/calendar.py", line 116, in weekday
    return datetime.date(year, month, day).weekday()
ValueError: year is out of range

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "pandas/tslib.pyx", line 407, in pandas.tslib.Timestamp.__new__ (pandas/tslib.c:10434)
    ts = convert_to_tsobject(ts_input, tz, unit, 0, 0)
  File "pandas/tslib.pyx", line 1471, in pandas.tslib.convert_to_tsobject (pandas/tslib.c:27533)
    return convert_str_to_tsobject(ts, tz, unit, dayfirst, yearfirst)
  File "pandas/tslib.pyx", line 1610, in pandas.tslib.convert_str_to_tsobject (pandas/tslib.c:29864)
    raise ValueError("could not convert string to Timestamp")
ValueError: could not convert string to Timestamp