issue_comments

@Geektrovert and @spencerkclark - any progress on this? It would be great to have it working in time for the CMIP6 hackathon in order to suppress all of the warning messages when opening zarr stores. For example, a zarr store using cftime.DatetimeGregorian (e.g., cftime.DatetimeGregorian(2349, 12, 16, 12, 0, 0, 0, 4, 350)) will always generate the warning: /usr/local/python/anaconda3/envs/pangeoJun2019/lib/python3.6/site-packages/xarray/coding/times.py:465: SerializationWarning: Unable to decode time axis into full numpy.datetime64 objects, continuing using cftime.datetime objects instead, reason: dates out of range dtype = _decode_cf_datetime_dtype(data, units, calendar, self.use_cftime) /usr/local/python/anaconda3/envs/pangeoJun2019/lib/python3.6/site-packages/xarray/coding/times.py:465: SerializationWarning: Unable to decode time axis into full numpy.datetime64 objects, continuing using cftime.datetime objects instead, reason: dates out of range dtype = _decode_cf_datetime_dtype(data, units, calendar, self.use_cftime) /usr/local/python/anaconda3/envs/pangeoJun2019/lib/python3.6/site-packages/numpy/core/numeric.py:538: SerializationWarning: Unable to decode time axis into full numpy.datetime64 objects, continuing using cftime.datetime objects instead, reason: dates out of range return array(a, dtype, copy=False, order=order) For the CMIP6 collection, I always use use_cftime=True in open_mfdataset, but then when I make the zarr store and read it back in, it generates these warnings - always a distraction for anyone not used to dealing with this data

Yes, when open_mfdataset decides to convert to CFTime this is much faster. When time is in datetime64, I get: ```

AttributeError Traceback (most recent call last) <ipython-input-72-a96fa0263d3e> in <module>() 9 dss = xr.open_mfdataset(files,decode_times=True,autoclose=True) 10 #month_start = [DatetimeNoLeap(date.dt.year, date.dt.month, 1) for date in dss.time] ---> 11 month_start = [DatetimeNoLeap(date.year, date.month, 1) for date in dss.time.values] 12 #month_start = [DatetimeNoLeap(yr, mon, 1) for yr,mon in zip(dss.time.dt.year,dss.time.dt.month)] 13 #break

<ipython-input-72-a96fa0263d3e> in <listcomp>(.0) 9 dss = xr.open_mfdataset(files,decode_times=True,autoclose=True) 10 #month_start = [DatetimeNoLeap(date.dt.year, date.dt.month, 1) for date in dss.time] ---> 11 month_start = [DatetimeNoLeap(date.year, date.month, 1) for date in dss.time.values] 12 #month_start = [DatetimeNoLeap(yr, mon, 1) for yr,mon in zip(dss.time.dt.year,dss.time.dt.month)] 13 #break

AttributeError: 'numpy.datetime64' object has no attribute 'year' ``` You can see I made a feeble attempt to fix it to work for all the CMIP5 calendars, but is just as slow. Any suggestions?

@spencerkclark thanks! I hadn't figured out that particular workaround, but it works, albeit quite slow. For now it will get me to the next step, but just changing to first-of-the-month takes longer than regridding all models to a common grid!

I am trying to combine the monthly CMIP5 rcp85 ts datasets (go past 2064AD) with the myriad calendars, so I love the new CFTimeIndex! But I need resample(time='MS') in order to force them all to start on the first of each month thanks!