Add basic implementation of FtdiDeviceHandler

[jamesadevine]

Adds a non-functional implementation of UsbDeviceHandler for FTDI devices. An FTDI device now enumerates with FtdiDeviceHandler using the following code:

use usbip::ftdi::FtdiDeviceHandler;
...
let device_handler = Arc::new(Mutex::new(Box::new(FtdiDeviceHandler::new())
    as Box<dyn usbip::UsbDeviceHandler + Send>));

...

usbip::UsbDevice::new(0).with_interface(...).with_device_handler(device_handler);

Closes #7

[jiegec]

The ftdi code should go to examples.

[jamesadevine]

@jiegec - sure! I also just want to say that having no filter on new_from_host completely messed up USB on my machine. I would suggest it be removed and only allow new_from_host with a filter (user provided). There could be other wrapper functions that implement default filters: e.g. new_from_host_with_vid_pid

[jamesadevine]

@jiegec I'm not sure what you meant with your comment. I've added an example that spins up a fake 4 port FTDI device and echoes "hello" once a second. If you would like me to move the entirety of the FTDI code into examples, then I'd suggest that would be part of a bigger refactor. I mirrored the structure of the cdc/hid keyboard examples.

[jiegec]

@jiegec I'm not sure what you meant with your comment. I've added an example that spins up a fake 4 port FTDI device and echoes "hello" once a second.

Good job!

If you would like me to move the entirety of the FTDI code into examples, then I'd suggest that would be part of a bigger refactor. I mirrored the structure of the cdc/hid keyboard examples.

I put cdc/acm in src because it is a standard usb interface. But ftdi is not.

[coveralls]

Pull Request Test Coverage Report for Build 4105479928

• 17 of 164 (10.37%) changed or added relevant lines in 4 files are covered.
• 3 unchanged lines in 3 files lost coverage.
• Overall coverage decreased (-1.7%) to 40.03%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/host.rs	0	1	0.0%
src/device.rs	17	32	53.13%
src/ftdi.rs	0	42	0.0%
src/lib.rs	0	89	0.0%

Files with Coverage Reduction	New Missed Lines	%
../../../.cargo/registry/src/github.com-1ecc6299db9ec823/tokio-1.25.0/src/io/util/write_int.rs	1	72.22%
src/host.rs	1	0%
src/lib.rs	1	52.7%

Totals
Change from base Build 4061119800:	-1.7%
Covered Lines:	542
Relevant Lines:	1354

---

💛 - Coveralls

[jamesadevine]

I put cdc/acm in src because it is a standard usb interface. But ftdi is not.

I think that's a rather arbitrary distinction. A driver is a driver. Putting ftdi code in examples means it has less reuse for others. It does lessen the maintenance burden which might be your concern. If you provide me a concrete suggestion for a folder structure you'd like me to observe I will implement it.

I'm also going to break this PR up into smaller PRs since it has taken a while to get this one merged in and there are now other improvements and bugfixes contained in the commits.

[jiegec]

I put cdc/acm in src because it is a standard usb interface. But ftdi is not.

I think that's a rather arbitrary distinction. A driver is a driver. Putting ftdi code in examples means it has less reuse for others. It does lessen the maintenance burden which might be your concern. If you provide me a concrete suggestion for a folder structure you'd like me to observe I will implement it.

You are right, maybe we can take driver code to a separate crate e.g. usbip-xxx?

I'm also going to break this PR up into smaller PRs since it has taken a while to get this one merged in and there are now other improvements and bugfixes contained in the commits.

Happy to see that.