Not all DateOffsets are comparable

[edrevo]

>>> from pandas.tseries.frequencies import to_offset as _to_offset
>>> _to_offset('D') < _to_offset('H')
False
>>> _to_offset('D') < _to_offset('W')
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-14-77799b586280> in <module>()
----> 1 _to_offset('D') < _to_offset('W')

C:\Anaconda\lib\site-packages\pandas\tseries\offsets.pyc in f(self, other)
   1873 def _tick_comp(op):
   1874     def f(self, other):
-> 1875         return op(self.delta, other.delta)
   1876 
   1877     return f

AttributeError: 'Week' object has no attribute 'delta'
>>> pd.show_versions()
INSTALLED VERSIONS
------------------
commit: None
python: 2.7.7.final.0
python-bits: 64
OS: Windows
OS-release: 8
machine: AMD64
processor: Intel64 Family 6 Model 42 Stepping 7, GenuineIntel
byteorder: little
LC_ALL: None
LANG: None

pandas: 0.14.0
nose: 1.3.3
Cython: 0.20.1
numpy: 1.8.1
scipy: 0.14.0
statsmodels: 0.5.0
IPython: 2.1.0
sphinx: 1.2.2
patsy: 0.2.1
scikits.timeseries: None
dateutil: 1.5
pytz: 2014.3
bottleneck: None
tables: 3.1.1
numexpr: 2.3.1
matplotlib: 1.3.1
openpyxl: 1.8.5
xlrd: 0.9.3
xlwt: 0.7.5
xlsxwriter: 0.5.5
lxml: 3.3.5
bs4: 4.3.1
html5lib: None
bq: None
apiclient: None
rpy2: None
sqlalchemy: 0.9.4
pymysql: None
psycopg2: None

[jreback]

I am not sure their is a guarantee of comparability between offsets anywhere. Have you seen and docs/code which indicate this?

e.g.

_to_offset('D') < _to_offset('30H') should be False but is True because its not comparing the correct things

so I suppose this is an issue, but what is the use case?

[jtratner]

at the minimum, should raise some kind of Type or Value error saying that
they are incompatible, no?

On Fri, Sep 26, 2014 at 7:38 AM, jreback [email protected] wrote:

I am not sure their is a guarantee of comparability between offsets
anywhere. Have you seen and docs/code which indicate this?

e.g.

_to_offset('D') < _to_offset('30H') should be False but is True because
its not comparing the correct things

so I suppose this is an issue, but what is the use case?

—
Reply to this email directly or view it on GitHub
#8386 (comment).

[jreback]

@jtratner yeh I think that they should raise for now

[edrevo]

The use case is the following: I have a metric that can be calculated on different frequencies (daily, hourly, monthly). The thing is that a different filtering needs to happen depending on the frequency: if the metric is calculated hourly, then all data points after 18:00 or before 8:00 for a given day need to be filtered out. This cannot happen if the metric has been calculated daily, since the timestamps will look like 2014-09-27 00:00, so the filtering would effectively wipe out all data.

So basically I need to know, given a datetimeindex, if its frequency ir more os less granular than a day.