Fix PassThruConnect Flags for 11 bit CAN identifier

[kamilwierzbicki]

Issue: #287

Some of the PassThru devices report an error when an invalid bit is set in the flags parameter.
Currently the bit 6 is set but it should be zeroed.

[pylessard]

Not sure changing the definition of the constant is a proper fix. Should be zeroed at the usage point. Also, does this limitation apply to your use case only or all use case?

[kamilwierzbicki]

Constant value needs to be changed because it's invalid according to specification

Only bit 8 defines the length of CAN identifier so setting up the bit 7 for 11 bit length CAN identifiers caused the problem.

[pylessard]

Then why do you change the constant value to 0?
@kirya-dev, can you get involved? I never used J2534

[kamilwierzbicki]

This constant contains the value that needs to be set to get 11 bit CAN identifier. According to specification bit 8 specifies the CAN identifier length. If the CAN identifier is 29 bit then bit 8 should be set (0x400). Otherwise it should be cleared (0x000). This constant value can be deleted at all but I left it to keep it more readable.

[kirya-dev]

Otherwise it should be cleared (0x000)

No. Should be cleared only 8 bit. BUT i think better if const ISO15765_CAN_ID_11 will be deleted.
And modify flags assigns as shown here:

        self.txConnectFlags = 0
        # Determine mode ID29 or ID11
        if txid >> 11:
            self.txConnectFlags |= TxStatusFlag.ISO15765_CAN_ID_29.value
        self.txFlags = self.txConnectFlags | TxStatusFlag.ISO15765_FRAME_PAD.value

[pylessard]

@kirya-dev : It sure feels like adding the 29 bits flag if required makes more sense.
But looking at your solution, what if I want a 29bits address that has a values smaller than 0x7FF?

[kirya-dev]

what if I want a 29bits address that has a values smaller than 0x7FF?

I see topic start from another problem - bit 6 should be zerowed. No related with ID len. Related with FRAME_PAD flag.

[kamilwierzbicki]

Updated according to your comments.
@pylessard Your observation may be resolved in another CR. There is already an issue created for that.
#286

[pylessard]

@kirya-dev : I still find this PR weird. Will follow your call for merge.

[kirya-dev]

I can approve condition style shown in my message above.

[kirya-dev]

Currently the bit 6 is set but it should be zeroed.

Yes. Problem verifies when I using Autel VCMI with 11bit CAN_ID. This PR should fixit.

[kirya-dev]

I think will be better align with standard.

https://docs.rs/j2534/latest/src/j2534/lib.rs.html#603-775

There is no constant ISO15765_CAN_ID_11.

[kirya-dev]

Issue: #287

Some of the PassThru devices report an error when an invalid bit is set in the flags parameter. Currently the bit 6 is set but it should be zeroed.

Its resolved in:
#289