Problem switching to version 1.0.0

[cbockermann]

Hi Matt, hi others,

thanks for this great library! I am using it for quite a long time and setup a Java based RFID kid's radio with it. Unfortunately, switching to 1.0.0 broke my code. I started digging into it, but failed.

I am using the MFRC522 Reader on a Raspberry PI Model B Rev 2, with the RST line connected to GPIO25. With version 0.14 everything works fine.

With version 1.0.0, I get the following error:

2021-01-09 04:52:29,596 INFO  main net.scinotes.Rfid - Creating RFID reader instance...
Exception in thread "main" java.lang.IllegalArgumentException: Chip not defined for pin PinInfo [keyPrefix=GPIO, header=DEFAULT, deviceNumber=25, physicalPin=22, name=GPIO25, chip=-1, lineOffset=-1, modes=[DIGITAL_INPUT, DIGITAL_OUTPUT]]
    at com.diozero.internal.provider.builtin.DefaultDeviceFactory.createDigitalOutputDevice(DefaultDeviceFactory.java:247)
    at com.diozero.internal.spi.GpioDeviceFactoryInterface.provisionDigitalOutputDevice(GpioDeviceFactoryInterface.java:75)
    at com.diozero.api.DigitalOutputDevice.<init>(DigitalOutputDevice.java:101)
    at com.diozero.api.DigitalOutputDevice.<init>(DigitalOutputDevice.java:85)
    at com.diozero.api.DigitalOutputDevice.<init>(DigitalOutputDevice.java:72)
    at com.diozero.devices.MFRC522.<init>(MFRC522.java:263)
    at net.scinotes.hbox.rfid.RfidReader.<init>(RfidReader.java:37)
    at net.scinotes.Rfid.run(Rfid.java:34)
    at net.scinotes.Main.run(Main.java:89)
    at net.scinotes.Main.main(Main.java:93)

I am actually not sure, I everything should continue running as with version 0.14 or if some defaults have changed, which I would need to adjust.

Any idea?

Best regards, Christian

[mattjlewis]

Hi Christian, sorry about this. I suspect I know what is going on here, could you let me know the output of these commands please:

uname -a
cat -v /proc/device-tree/compatible
grep Revision /proc/cpuinfo
ls -l /dev/gpio*

In the meantime could you please try run with -Ddiozero.gpio.chardev=false and let me know if that helps.

[cbockermann]

Hi Matt,

thanks for the reply. The diozero.gpio.chardev=false settings solved it.

As for the information you asked for:

chris@web-radio:~$ uname -a
Linux web-radio 5.4.79+ #1373 Mon Nov 23 13:18:15 GMT 2020 armv6l GNU/Linux

chris@web-radio:~$ cat -v /proc/device-tree/compatible
raspberrypi,model-b^@brcm,bcm2835^

chris@web-radio:~$ grep Revision /proc/cpuinfo 
Revision    : 000f

chris@web-radio:~$ ls -l /dev/gpio*
crw-rw---- 1 root gpio 254, 0 Jan  9 23:17 /dev/gpiochip0
crw-rw---- 1 root gpio 246, 0 Jan  9 23:17 /dev/gpiomem

Yes, it's an older PI, Java startup is sloooow, but RFID-Card reading seems to be much faster with MFRC522 using diozero than the Python libs I've tested... (+ I am a Java guy... :-))

Thanks for the library!

Regards, Christian

[mattjlewis]

I've just released 1.0.1 - hopefully with this release you won't need to run with -Ddiozero.gpio.chardev=false

[cbockermann]

Thanks! Will start testing right away tomorrow and give you feedback.

Christian

[cbockermann]

Version 1.0.1 works like charm (without -Ddiozero.gpio.chardev=false).

Thanks for the quick fix and new release!

[mattjlewis]

You're very welcome.