Feather doesn't support a dataframe column of datetime.date objects in Python

I tried writing out a pandas dataframe to feather, and noticed that my column of datetime.date objects was producing the following error:

FeatherError                              Traceback (most recent call last)
<ipython-input-16-b9e7dd094217> in <module>()
----> 1 feather.write_dataframe(df, 'df.feather')

/Users/matthew/anaconda/lib/python2.7/site-packages/feather/api.pyc in write_dataframe(df, path)
     35             name = str(name)
     36 
---> 37         writer.write_array(name, col)
     38 
     39     writer.close()

/Users/matthew/anaconda/lib/python2.7/site-packages/feather/ext.pyx in feather.ext.FeatherWriter.write_array (feather/ext.cpp:2127)()
     86             self.write_timestamp(name, col, mask)
     87         else:
---> 88             self.write_primitive(name, col, mask)
     89 
     90     cdef write_category(self, name, col, mask):

/Users/matthew/anaconda/lib/python2.7/site-packages/feather/ext.pyx in feather.ext.FeatherWriter.write_primitive (feather/ext.cpp:2432)()
    108 
    109         col_values = _unbox_series(col)
--> 110         self.write_ndarray(col_values, mask, &values)
    111         check_status(self.writer.get().AppendPlain(c_name, values))
    112 

/Users/matthew/anaconda/lib/python2.7/site-packages/feather/ext.pyx in feather.ext.FeatherWriter.write_ndarray (feather/ext.cpp:2739)()
    133     cdef int write_ndarray(self, values, mask, PrimitiveArray* out) except -1:
    134         if mask is None:
--> 135             check_status(pandas_to_primitive(values, out))
    136         else:
    137             check_status(pandas_masked_to_primitive(values, mask, out))

/Users/matthew/anaconda/lib/python2.7/site-packages/feather/ext.pyx in feather.ext.check_status (feather/ext.cpp:1543)()
     52 
     53     cdef string c_message = status.ToString()
---> 54     raise FeatherError(frombytes(c_message))
     55 
     56 set_numpy_nan(np.nan)

FeatherError: Invalid: unhandled python type, index 0 

I’m going to look to using pandas timestamps in the interim.

[wesm]

This is supported in Feather V2 (coming in pyarrow / Arrow 0.17.0)

[ben519]

Experiencing similar pain with this issue. In my case I’m processing a large file using R, writing the result to a feather file and trying to read the file using Python. The dataset contains a Date type column which is generating an error when I try to read the file in Python (but not in R).

R Code

library(feather)
df <- data.frame(X=as.Date(c("2013-1-1", "2014-1-1", "2015-1-1")))
write_feather(df, "df.feather")
read_feather("df.feather")  # works

Python Code

import feather
feather.read_dataframe('df.feather')  # error

Error Message

NotImplementedErrorTraceback (most recent call last)
<ipython-input-39-4dfedb41c631> in <module>()
----> 1 feather.read_dataframe('df.feather')

//anaconda/envs/datasci/lib/python3.5/site-packages/feather/api.py in read_dataframe(path, columns)
     54     names = []
     55     for i in range(reader.num_columns):
---> 56         name, arr = reader.read_array(i)
     57         data[name] = arr
     58         names.append(name)

//anaconda/envs/datasci/lib/python3.5/site-packages/feather/ext.pyx in feather.ext.FeatherReader.read_array (feather/ext.cpp:3448)()

NotImplementedError: 3