Giters

serialport / serialport-rs

A cross-platform serial port library in Rust. Provides a blocking I/O interface and port enumeration including USB device information.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Linux: `serialport::available_ports()` returns generic USB strings instead of actual descriptor values

makkarpov opened this issue · comments

commented

On Linux, manufacturer and product strings for USB serial ports are populated with generic strings from VID/PID database instead of actual string descriptors as expected.

For example, for a test device with VID 0x1209 (https://pid.codes) and PID 0x0002 (reserved test value) you would get the following (a bit useless) enumeration:

SerialPortInfo {
    port_name: "/dev/ttyACM0",
    port_type: UsbPort(
        UsbPortInfo {
            vid: 4617,
            pid: 2,
            serial_number: None,
            manufacturer: Some(
                "Generic",
            ),
            product: Some(
                "pid.codes Test PID",
            ),
        },
    ),
}

However, device itself contains much more detailed string data, which could be used to locate the device much more precisely:

[488989.469938] usb 11-4.1.4: new full-speed USB device number 48 using xhci_hcd
[488989.559135] usb 11-4.1.4: New USB device found, idVendor=1209, idProduct=0002, bcdDevice= 2.00
[488989.559143] usb 11-4.1.4: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[488989.559148] usb 11-4.1.4: Product: Your cool product
[488989.559151] usb 11-4.1.4: Manufacturer: Your manufacturer name

Alternate way of querying correct Linux device info could be found in pyserial repository: https://github.com/pyserial/pyserial/blob/master/serial/tools/list_ports_linux.py

udev also contains correct strings, but I'm not sure why serialport-rs extracts generic data instead.

commented

Thank you for bringing up this issue! I can confirm this behavior. We are using udev for retrieving this information, strange, that it does not report the actual value from iManufacturer and iProduct.

Thanks for the pointer to pySerial. I will have a look there.

commented

It looks like we are preferring the information from udevs hardware database over the one provided by the usb device itself. I'm working on #137 to sort this out. Could you please give these changes a try?

commented

Sure, I tested these. These changes indeed give a correct results.

Thank you for promt response!