On 30 Jun 2014 17:05, "Chris Barker" <chris.barker@noaa.gov> wrote:
>>
>> It's also an interesting
>> question whether they've fixed the unicode/binary issues, 
>
>
> Which brings up the "how do we handle text/strings in numpy? issue. We had a good thread going here about what the 'S' data type should be , what with py3 and all, but I don't think we ever really resolved that.
>
> IIRC, the key issue was whether we should have a "proper" one-byte-per-character text type -- after all, ASCI/ANSI text is pretty common in scientific data sets, and 4 bytes per char is a fair bit of overhead.
>
> Anyway, this all ties in with the text file parsing issues...

Only tangentially though :-) maybe start another thread if you want to talk about that?