Is it possible to use 'groupby' asynchronously?

[Ian Kelly]

On Thu, Oct 6, 2016 at 9:57 AM, Chris Angelico <rosuav at gmail.com> wrote:
> On Thu, Oct 6, 2016 at 11:09 PM, Frank Millman <frank at chagford.com> wrote:
>> Hi all
>>
>> I have used itertools.groupby before, and I love it. I used it to process a
>> csv file and 'break' on change of a particular field. It worked very well.
>>
>> Now I want to use it to process a database table. I can select the rows in
>> the desired sequence with no problem. However, I am using asyncio, so I am
>> reading the rows asynchronously.
>>
>> Before I spend hours trying to figure it out, can anyone confirm if this is
>> doable at all, or is groupby not designed for this.
>
> Most of itertools is going to assume synchronous iterables. However,
> the docs give pure-Python equivalents for quite a few:
>
> https://docs.python.org/3/library/itertools.html#itertools.groupby
>
> You may be able to tweak that into an actual "itertools.agroupby" or
> "aitertools.groupby".

There is already a third-party "aitertools" library, so there should
be no need to write one from scratch.

https://pypi.python.org/pypi/aitertools/0.1.0