socket-client | streaming plaintext TCP/IP and secure TLS based connections | Encryption library

Likely the dictionary you receive from the web socket is a json object:

In most cases, instead of modifying the python installation in an existing image or installing a new one, I would recommend just using a docker image that already has python 3 installed. This bitbucket page talks about using different base images.

There's also a page about using a python base image spectifically.

Aside from this, depending on how you want to use SAM, you may be able to use the builtin Bitbucket "pipe" for deploying to SAM instead. Here is some info on pipes.. I think this would bypass the need to install SAM in your build.

I haven't used that particular pipe myself but the way I think it works is you build your deploy artifact file in your pipeline, caching it so it's available in future steps, then you set that file as the input for the SAM pipe to deploy it. The SAM pipe runs in a separate container, so it's already installed in that container. You just pass in the file.

Finally, I have never used SAM, so take this with a grain of salt, but I don't see a reference to installing the CLI with pip either on the official AWS Linux installation docs or on the github page. I did find this issue, from which I glean the following:

• pip installation is possible but not recommended
• Python 3.5 is not supported by aws sam cli
• If you use the installer in the instructions, it sounds like it will take care of the python versioning for you.
• They would like to target a single version of Python, so if you use pip, maybe the latest or 2nd latest Python version will work. Just a guess...

At the end, I ended up solving this by rolling back the changes I made using conda list --revisions to find out until which previous set up I had to roll back to, then afterwards I ran conda install --revision N (where N is the revision you want to trace back to). Suppose the changes you made are rev 4, you want to undo them, and sit back again under rev 3 (your previously "known and working" environment you had), so you run conda install --revision 3 for that case.

Afterwards I re-installed pandas_datareader with python -m pip install pandas-datareader and everything went good again.

Thanks anyways and I hope if someone else runs into this issue, can find this post valuable.

No, they should be the same. Spring should take care of the differences between them. That said, if there are any bugs you observe you should raise them.

It should not be possible for client code to observe whether the server is defined using Flux or Flow. Further to this, hopefully a client isn't aware of the implementation language of the server either.

I don't think your first example needs to be suspending either, as the Flow will normally be cold anyway. https://kotlin.github.io/kotlinx.coroutines/kotlinx-coroutines-core/kotlinx.coroutines.flow/-flow/

The Flow interface does not carry information whether a flow is a cold stream that can be collected repeatedly and triggers execution of the same code every time it is collected, or if it is a hot stream that emits different values from the same running source on each collection. Usually flows represent cold streams

Please read the Documentation on Events

You must declare function signatures that match that of the Event Delegate within your class for which a function pointer will be bound.

The example image above is from the Engine Source.

I don't think websocket-client provides an async API and, so, it's blocking the event loop.

You should use an async websocket client, such as the one Tornado provides:

From the list of libraries, it looks like you've installed CPU only version of the Pytorch.

You can only send String or Buffer data with WebSocket.prototype.send. Fortunately, JSON.parse + JSON.stringify quickly bridge the gap between structured data (Object, Array, etc.) and String.

Try:

So stupid bug... I found out that in one of my vuex store files I have this: