Py3.5

[goodboy]

Got it mostly there just some outstanding oddness trying to get pandas to work with unicode strings.

pandas doesn't seem like it's supporting numpy dtypes fully an example being that I can't use the '<U' from Array-protocol type strings. On top of that shmarray doesn't seem to play well with passing built-in Python types to np.dtype.

Gonna hack on it some more.

[goodboy]

Couple of notes after digging into this:

• shmarry doen't support unicode as per the docs on multiprocessing.RawArray so if we want to keep the shared numpy array stuff it would seem we have to use Bytes.
• Turns out all the pandas "encoding" (type coercing?) problems are stemming from pydata/pandas#9712 where when a csv data store is written it hilariously keeps the b bytes prefix. This ends up causing problems with round tripping (which is done implicitly when reading the entire contents of a DataStorer: in mem + on disk) since pd.read_csv then parses the b as part of the data point.