After some testing I have found that the current implementation will not
work when called from applications that are using the asyncio event loop
(such as Jupyter). I plan to fix this with nest_asyncio, which will allow
the client to be used within an event loop. If the code is ever changed to
run
Thanks for the input, Dave. I suppose my thinking at this stage is that
given the window to 3.5.0 (in a few weeks) I'd be concerned about a major
shift in how gremlin-python works. Lyndon's solution is quite surgical and
seems to carry little risk. I'd be in favor of a pure aiohttp approach in a
fu
You guys should just switch to a real Asyncio (aiohttp) client
implementation that actually leverages the async/await syntax. As Stephen
mentioned I already did it with aiogremlin if you want to see an example. I
don't control the repo anymore but there is an example implementation there
with an as
This is an important body of work and I think it is being done in the right
fashion. I had thought that we would completely remove the Transport
abstraction but since "it works" it seems as though there is no need at
this point. I do seem to recall that David Brown had this abstraction in
play with
