WebTCP: Making TCP Connections From Browser

http://artemyankov.com/tcp-client-for-browsers/

by yankoff • 14 years ago

50 31 14 years ago

artemyankov.com

chewxy4• 14 years ago

So, TCP over HTTP, which is a protocol built on top of TCP. Does anyone see that this is rather silly?

If you want a client-side app that uses TCP, why must you do it over the web? The appeal of write-once-run-everywhere?

yankoff4• 14 years ago

Well, you didn't read the post apparently and only a title :(

chewxy4• 14 years ago

I actually did. I was quite interested in the redis example. For about 10 seconds, which I then went "wtf is this useful for?"

We really need to stop reinventing the wheel

khakimov• 14 years ago

it's not about pure tcp, as i get it - it allows you communicate from browser directly (via webtcp server as a bridge) with any servers such as redis, mongodb, rabbitmq and so on.

WebBrowser --- data ---> WebTCP bridge (translate data to servers) --- data ---> redis/rabbitmq/any_tcp(and i think udp also possible?)_server_even_smtp

sneak2• 14 years ago

It's not TCP over HTTP, that would require a TCP stack. It's the io bytestream from a specific TCP connection, shuttled over HTTP.

It's not entirely bogus, just mostly so. I really dislike this whole "the internet == the web" way of thinking, though.

javajosh2• 14 years ago

On the contrary, this sort of things allows us to write clients in a uniform way. It will be nice to be able to debug the redis client in a jsfiddle. :)

icebraining• 14 years ago

It uses Websockets, which are only HTTP for the handshake: the rest of the connection is just a TCP stream.

So it's really more like a NAT bridge.

flux_w42• 14 years ago

I think this is a great idea. From the moment I heard of WebSockets I was disappointed it was only a thing between the sockets-capable-web-server and the client and not a real TCP connection to wherever you want. This solves stuff :) However, I think that you should keep an eye on the possible security implications (ex. fooling the same origin policy, using your websockets proxy as a hub to attack other systems on the internet, ...)

dwj2• 14 years ago

Doesn't websockify do this better?

yankoff4• 14 years ago

It looks like with websockify you have to run a server with their wrapper to make it work

0x0• 14 years ago

Isn't the major benefit of WebSockets that is actually has a fair chance to pass through all the crazy proxies and gateways installed in every office, home and coffe shop, that would otherwise filter any non-HTTP traffic?

lttlrck2• 14 years ago

So make a websocket proxy...

y0ghur7_xxx• 14 years ago

This is an awesome hack. We really need real TCP connections from the browser. Until then this looks like a great workaround. I eagerly await the day we can use the browser for everything we do (did?) with Desktop apps.

angersock• 14 years ago

We built a simple proof-of-concept library that lets you use a web app as the GUI for native code, if anyone would be interested.

lisper• 14 years ago

Yes please!

cstejerean• 14 years ago

This is definitely a cool hack, but please be mindful of the security considerations of deploying something like this on the internet. Running an open proxy that allows any client to perform arbitrary TCP connections bouncing through your servers is a bad idea.

First, you need to worry about the security implications of each internal server that you expose intentionally. Just how much damage can someone with direct access to your Redis instance cause? Better read http://redis.io/topics/security, and repeat this for every other service you want to expose.

Second, you are going to effectively need a whitelist of allowed connections, either in the proxy, or at the firewall level (which most people are going to forget to configure, if they even know how). Without that you are going to allow arbitrary bypass of your firewall for internal services, as well as effectively running an open proxy for anyone looking to bounce their malicious traffic through your server.

aidenn0• 14 years ago

Yeah, just look at all the havoc caused by anonymous FXP in the '90s

eungyu• 14 years ago

Nice hack (although I would put this in the category of 'workaround'). The idea is cool, but I can already imagine all sorts of debugging nightmares with this approach. To begin with, how do we find out where the connection is broken (Is it the client socket or the proxy socket)? Bad connections have different reasons (in C terms: ECONNRESET, EWOULDBLOCK, ENETUNREACH, etc); are these specific reasons propagated verbatim? What about flow control between proxy and client?

There is always a chasm between whipping up a cool hack and making something usable. This hack needs a long bridge in between.

drblast• 14 years ago

This is just what I need to complete the in-browser app I'm building to navigate and read these new hypertext pages I'm creating. I've developed a markup language that will let you reference any of these documents from any other. The app that lets you view them is going to work in any browser that supports WebTCP, too, for the ultimate in cross-platform compatibility!

seats• 14 years ago

If more web services supported CORS, pure client side apps would be way easier to do and less kludgy than via proxy.

http://en.wikipedia.org/wiki/Cross-origin_resource_sharing

smagch• 14 years ago

You may want to take a look at browserver

http://news.ycombinator.com/item?id=4366555 http://browserver.org/

saljam• 14 years ago

Chrome (and I believe Firefox too) has a TCP socket API, but it's only exposed to Chrome apps and extensions.

WebTCP: Making TCP Connections From Browser

Related Stories