How to access variable generated in a 32 bit python process/thread from a main 64 bit process?

Question

main.py runs on an infinite loop, the Conda environment is called py39_64 it's python 3.9 64 Bit.

data.py runs on periodically, the Conda environment is called py39_32 it's python 3.9 32 Bit.

data.py has a variable called date which is a datetime object.

I would like to access the date variable (format d/m/Y) from the 64 bit process / thread.

Is this possible?

Or would I need the 32 bit process to periodically output a json file that the 64 bit process could read?

Answer 1

a better approach would be to use a local socket between the two processes, which are sockets that are connected on your localhost.

1. have the 64 bit process as the server, and the 32 bit process as the client
2. have a thread in the 64 bit program. that will wait to be connected by the respective clients, using the select module, and will poll its clients when needed for this parameter.
3. have a thread in the 32 bit program that connects to the server and responds with the value of this variable when polled .. also using the select library.
4. the data has to be passed as a utf-8 encoded string since pickling won't help you here ... or just send the json data over the socket.
5. be sure to set a reasonable timeout on the select, and to handle disconnects and reconnects and timeouts when some time passes without sending any data.

setting up the connection is as simple as the socket chat example using select() method for client/ server chat in Python , except you need the server to run the sockets on a separate thread instead of the main thread using the threading module, so you need the entire code on the server and client side to be in a single function that runs in a separate thread, which is basically another infinite loop too.

you should be logging data to disk for long-term storage, but for sockets the advantage is that you reduce disk usage, and make it a "poll and get reply when needed" instead of "always save data to disk" and "always read data from disk and hope you get the most recent result and no error happens".