Mark Dickinson

Thanks for the report. I confirm that I see the same results as you. I think what you're seeing is due to the default context having an `emin` of `-16493`...

It's possible that we want to change the way that the context treats `emin` here - that it should be automatically adjusted when the precision is increased. The `emin` exposed...

Marking as a bug: the code _is_ technically behaving as intended, but that behaviour is surprising; I think this is a design bug that we should try and fix.

Apologies for the (very) late response; you've probably already figured out that this library doesn't get as frequently maintained as it should, and moved on. But in case you haven't:...

Not closing this issue, because it _would_ make sense to have analogues of the various `get` and `set` methods that work with Python long integers; but it's unlikely that the...

At the same time, disable the use of `BigFloat.exact` for strings: it's misleading.

We need to support various cases: 1. Conversion from a decimal string, hex string, integer or float to the current context, taking into account the current rounding mode. This operation...

Moving this out of the 0.4.0 milestone.

Here's a minimal reproducer on my machine: ```python import zmq from jupyter_client.session import Session from ipykernel.iostream import OutStream, IOPubThread def test(): context = zmq.Context() pub_thread = IOPubThread(socket=context.socket(zmq.PUB)) pub_thread.start() stream =...

#868 may be related