Re: [PATCH RFC 26/32] python//machine.py: use qmp.command

[John Snow]

[...]

>  
> -    def qmp(self, cmd, conv_keys=True, **args):
> -        """
> -        Invoke a QMP command and return the response dict
> -        """
> +    @classmethod
> +    def _qmp_args(cls, _conv_keys: bool = True, **args: Any) -> Dict[str, 
> Any]:
>          qmp_args = dict()
>          for key, value in args.items():
> -            if conv_keys:
> +            if _conv_keys:
>                  qmp_args[key.replace('_', '-')] = value
>              else:
>                  qmp_args[key] = value
> +        return qmp_args
>  
> +    def qmp(self, cmd: str,
> +            conv_keys: bool = True,
> +            **args: Any) -> QMPMessage:

This creates an interesting problem with iotests 297:


-Success: no issues found in 1 source file
+iotests.py:563: error: Argument 2 to "qmp" of "QEMUMachine" has
incompatible type "**Dict[str, str]"; expected "bool"
+Found 1 error in 1 file (checked 1 source file)


def hmp(self, command_line: str, use_log: bool = False) -> QMPResponse:
    cmd = 'human-monitor-command'
    kwargs = {'command-line': command_line}
    if use_log:
        return self.qmp_log(cmd, **kwargs)
    else:
        return self.qmp(cmd, **kwargs)

It seems like mypy is unable to understand that we are passing keyword
arguments, and instead believes we're passing something to the conv_keys
parameter.

(Is this a bug...?)

Even amending the function signature to indicate that conv_keys should
only ever appear as a keyword argument doesn't seem to help.

I'll have to think about a nice way to fix this; removing conv_keys out
of the argument namespace seems like the best approach.

qmp(cmd, foo=bar, hello=world)
qmp(cmd, **conv_keys(foo=bar, hello=world))

...but now this function looks really annoying to call.

Uh, I'll play around with this, but let me know if you have any cool ideas.

--js