Working proof of concept for Windows support from issue #2

[techtonik]

I didn't have time to polish this, but I tested this works

[joshvillbrandt]

Nice, @techtonik!

A few things need to happen before I can merge this:

• pass the TravisCI checks (currently failing the flake8 test)
• refactor to fit the class implementation of the existing wifi drivers in this project
• I'd like to test the implementation myself (once the first two items are fixed) on a Windows machine

I'd love it if you wanted to take another go at the first two items, otherwise I'm happy to finish off the work when I can make some time.

Thanks very much for this!

[techtonik]

To adopt it to class implementation, the password field on connect() method should be made optional.

[joshvillbrandt]

That is not the desired behavior. The desired behavior for this library is to connect to any wireless network with an ssid and a password. It should not have to rely an existing wireless "profile" with the password entered through a none-programmatic means.

[techtonik]

How about supporting open networks that don't require passwords?

[techtonik]

From what I can tell, there is no way to connect to network without creating a profile in Windows, so you either need to reuse existing one or create new one every time. Reusing exising profile is good, because it doesn't require you to store password in clear text. If profile for a given network does not exist, and network requires a password, then a password should be required to create and use the profile.

[GrimzEcho]

Has there been any progress on Windows support? I need to implement this in a project I'm working on and might look into completing it if there is still interest.

[joshvillbrandt]

Note to self: see #2 for background.