Use sanitised result of selected_subprotocol

[benz0li]

• Fix Handler subprotocol method fails to detect empty list #442
• Fix PDF preview of LaTeX Workshop (code-server) results in HTTP 400: Bad Request #445

[welcome]

Thanks for submitting your first pull request! You are awesome! 🤗

If you haven't done so already, check out Jupyter's Code of Conduct.

You can meet the other Jovyans by joining our Discourse forum. There is also a intro thread there where you can stop by and say Hi! 👋

Welcome to the Jupyter community! 🎉

[benz0li]

FYI @sylvaticus Fixed version deployed at https://coder.jupyter.b-data.ch.

---

cd /usr/local/lib/python3.11/site-packages/jupyter_server_proxy \
  && cp -a handlers.py handlers.py.orig \
  && sed -i 's/subprotocols=self\.subprotocols/subprotocols=self\.selected_subprotocol/g' handlers.py

ℹ️ /usr/local/lib/python3.11/site-packages/ is where Jupyter Server Proxy is installed in my setup. Adapt to your setup.

[benz0li]

Ping @consideRatio

FYI @duytnguyendtn

[sylvaticus]

I confirm that the sanitation applied to handlers.py solves the issue with the VSCode extension LaTex-Workshop, thank you very much...

[benz0li]

Alternatively subprotocols=self.selected_subprotcol as described in #442.

[duytnguyendtn]

Thanks for putting together this PR! Hopefully we can get some traction from the devs and get this merged

[consideRatio]

Thank you for digging into this topic @benz0li!!!

For a merge this needs to be vetted with regards to "is this a breaking change" and "is this secure" for example, things that i personally struggle with in this tech-context. It would be very helpful for me if you could provide your take on that @benz0li, it would be a great starting point for me to review this change proposal.

[benz0li]

It would be very helpful for me if you could provide your take on that @benz0li, it would be a great starting point for me to review this change proposal.

@consideRatio You find the technical details in #442.

self.select_subprotcol(self.subprotocols) did not work, so I deleted [override function] select_subprotocol in handlers.py and went for [tornado's] self.selected_subprotcol.

Technically, I am not up to such things either. Maybe @duytnguyendtn has more insight.

[duytnguyendtn]

Haha and you've discovered why I didn't feel brave enough to author the PR! Frankly, I don't have the background to evaluate the safety or the "breaking" nature of this change. I hoped to open a discussion in #442 with the maintainers to help evaluate it's safety and compatibility. Echoing @benz0li, @consideRatio please take a look at my investigation in #442 where I arrived at the two potential solutions at the bottom which @benz0li is basing this PR off of. I tried to make it as thorough as I could and explore the code traces.

Please let me know if I can be of any further assistance!

[benz0li]

Because Jupyter Server Proxy borrows a lot from Tornado Web Server, I dare to ping @bdarnell in this case.

IMHO self.subprotocols and self.selected_subprotcol return different results in one case only: When there is no subprotocol proposed.

Former vs latter:

1. Subprotocol: [] vs None
2. Header: Sec-Websocket-Protocol: vs 'not sent'
3. Connection: Aborted (in some cases) vs 'successfully established'

Maybe there is a simpler and saner way to prevent subprotocols=[] when there is no subprotocol proposed.

[benz0li]

Closed in favour of #448.