Timestamp AttributeError

[ktownv]

pandas 0.17.0, python 2.7.10:

import pandas as pd
p = pd.Timestamp(2015, 11, 12)
p
Traceback (most recent call last):
File "", line 1, in
File "pandas\tslib.pyx", line 339, in pandas.tslib.Timestamp.repr (pandas\tslib.c:9660)
AttributeError: 'int' object has no attribute 'freqstr'

[jreback]

You are not calling correct.y

In [40]: pd.Timestamp?
Type:        type
String form: <class 'pandas.tslib.Timestamp'>
File:        c:\miniconda\envs\neat\lib\site-packages\pandas\tslib.pyd
Docstring:
TimeStamp is the pandas equivalent of python's Datetime
and is interchangable with it in most cases. It's the type used
for the entries that make up a DatetimeIndex, and other timeseries
oriented data structures in pandas.

Parameters
----------
ts_input : datetime-like, str, int, float
    Value to be converted to Timestamp
offset : str, DateOffset
    Offset which Timestamp will have
tz : string, pytz.timezone, dateutil.tz.tzfile or None
    Time zone for time which Timestamp will have.
unit : string
    numpy unit used for conversion, if ts_input is int or float

[ktownv]

then it should fail at the creation of the object, not leave me with something in an inconsistent state that doesn't play nicely in the interpreter.

for example

str(p)
'1970-01-01 00:00:00.000002015+00:00'

works.

calls to repr should never raise an exception.

[jreback]

actually it should raise if the offset is not None or a string.

code is here

pull-requests to fix are welcome.

[IamGianluca]

I can work on this.

[IamGianluca]

I'm having some troubles solving this bug. I'm not familiar with cython and its workflow. Should I just fix the code in the .pyx file and then rebuild the C dependencies, right? Because when I try to run python setup.py build_ext --inplace I get lots of warnings and can't complete building the Timedelta C extension.

I've created a Question in StackOverflow with all the details.

[jreback]

yes, you need to fix in the .pyx and then run setup. if their are compilation errors then it won't import, warnings are ok.

SO in general is not helpful with this kind of thing.

[IamGianluca]

I don't understand what's wrong here. This is the log of the build. There are lots of warnings but nothing that help me identify the route cause of the issue. The build also doesn't fail, however when I try to run the nosetests immediately after I get this error message:

grossi-ml:pandas-iamgianluca grossi$ nosetests pandas/
E
======================================================================
ERROR: Failure: ImportError (C extension: Timedelta not built. If you want to import pandas from the source directory, you may need to run 'python setup.py build_ext --inplace' to build the C extensions first.)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/Users/grossi/anaconda/lib/python2.7/site-packages/nose/loader.py", line 418, in loadTestsFromName
    addr.filename, addr.module)
  File "/Users/grossi/anaconda/lib/python2.7/site-packages/nose/importer.py", line 47, in importFromPath
    return self.importFromDir(dir_path, fqname)
  File "/Users/grossi/anaconda/lib/python2.7/site-packages/nose/importer.py", line 94, in importFromDir
    mod = load_module(part_fqname, fh, filename, desc)
  File "/Users/grossi/PycharmProjects/pandas-iamgianluca/pandas/__init__.py", line 13, in <module>
    "extensions first.".format(module))
ImportError: C extension: Timedelta not built. If you want to import pandas from the source directory, you may need to run 'python setup.py build_ext --inplace' to build the C extensions first.

----------------------------------------------------------------------
Ran 1 test in 0.001s

FAILED (errors=1)

Interestingly if I remove my changes I can build and then run nosetests without problems.

My code however seems fine to me. I've just added an else statement and imported the DateOffset class:

        if util.is_string_object(offset):
            from pandas.tseries.frequencies import to_offset
            offset = to_offset(offset)
        else:
            from pandas.tseries.offsets import DateOffset
            if not isinstance(offset, DateOffset):
                raise ValueError

[jreback]

trying putting a debugging statement right before the import then you can see where its called. you have a circular import is the issue.

[gliptak]

Changes to isoformat in current HEAD corrected this.

File "pandas/tslib.pyx", line 567, in pandas.tslib.Timestamp.isoformat (pandas/tslib.c:13019)

[gliptak]

No longer sure that this works as expected, I will pull the change by @IamGianluca from above ...