Re: Notes on Generating Python signatures for QMP RPCs

2022-02-07 Thread Markus Armbruster
Daniel P. Berrangé writes: > On Thu, Feb 03, 2022 at 05:52:10PM -0500, John Snow wrote: >> On Thu, Feb 3, 2022 at 5:40 AM Daniel P. Berrangé >> wrote: >> > >> > On Wed, Jan 26, 2022 at 01:58:19PM -0500, John Snow wrote: >> > >> > As you mention though, bear in mind that a command returning >> >

Re: Notes on Generating Python signatures for QMP RPCs

2022-02-04 Thread Daniel P . Berrangé
On Thu, Feb 03, 2022 at 05:52:10PM -0500, John Snow wrote: > On Thu, Feb 3, 2022 at 5:40 AM Daniel P. Berrangé wrote: > > > > On Wed, Jan 26, 2022 at 01:58:19PM -0500, John Snow wrote: > > > > As you mention though, bear in mind that a command returning > > nothing today, might return something to

Re: Notes on Generating Python signatures for QMP RPCs

2022-02-04 Thread Markus Armbruster
John Snow writes: > On Thu, Feb 3, 2022 at 5:40 AM Daniel P. Berrangé wrote: >> >> On Wed, Jan 26, 2022 at 01:58:19PM -0500, John Snow wrote: [...] >> I think it is not unreasonable to expose the struct names >> on introspection though, and just accept that it ties our >> hands a little more t

Re: Notes on Generating Python signatures for QMP RPCs

2022-02-03 Thread Markus Armbruster
John Snow writes: > On Thu, Feb 3, 2022 at 5:04 AM Markus Armbruster wrote: >> John Snow writes: >> > On Thu, Jan 27, 2022 at 9:03 AM Markus Armbruster >> > wrote: >> >> John Snow writes: > >> >> > (7) I have no idea what to do about functions that "may not return". >> >> > The QGA stuff in

Re: Notes on Generating Python signatures for QMP RPCs

2022-02-03 Thread John Snow
On Thu, Feb 3, 2022 at 5:40 AM Daniel P. Berrangé wrote: > > On Wed, Jan 26, 2022 at 01:58:19PM -0500, John Snow wrote: > > (1) QAPI types the return of many commands as an empty object. That's > > literally indeed what happens on the wire, and it makes sense in that > > if these commands were eve

Re: Notes on Generating Python signatures for QMP RPCs

2022-02-03 Thread John Snow
On Thu, Feb 3, 2022 at 5:04 AM Markus Armbruster wrote: > John Snow writes: > > On Thu, Jan 27, 2022 at 9:03 AM Markus Armbruster wrote: > >> John Snow writes: > >> > (7) I have no idea what to do about functions that "may not return". > >> > The QGA stuff in particular, I believe, is prone to

Re: Notes on Generating Python signatures for QMP RPCs

2022-02-03 Thread Daniel P . Berrangé
On Wed, Jan 26, 2022 at 01:58:19PM -0500, John Snow wrote: > (1) QAPI types the return of many commands as an empty object. That's > literally indeed what happens on the wire, and it makes sense in that > if these commands were ever to return anything, it is a "compatible > evolution" to include ne

Re: Notes on Generating Python signatures for QMP RPCs

2022-02-03 Thread Markus Armbruster
John Snow writes: > On Thu, Jan 27, 2022 at 9:03 AM Markus Armbruster wrote: >> >> John Snow writes: >> >> > Hiya, I was experimenting with $subject and ran into a few points of >> > interest. This is basically an informal status report from me. I've >> > CC'd some of the usual suspects for peo

Re: Notes on Generating Python signatures for QMP RPCs

2022-02-02 Thread John Snow
On Thu, Jan 27, 2022 at 9:03 AM Markus Armbruster wrote: > > John Snow writes: > > > Hiya, I was experimenting with $subject and ran into a few points of > > interest. This is basically an informal status report from me. I've > > CC'd some of the usual suspects for people who care about SDKs and

Re: Notes on Generating Python signatures for QMP RPCs

2022-01-27 Thread Markus Armbruster
John Snow writes: > Hiya, I was experimenting with $subject and ran into a few points of > interest. This is basically an informal status report from me. I've > CC'd some of the usual suspects for people who care about SDKs and API > design and such. > > This is just a list of some observations I

Notes on Generating Python signatures for QMP RPCs

2022-01-26 Thread John Snow
Hiya, I was experimenting with $subject and ran into a few points of interest. This is basically an informal status report from me. I've CC'd some of the usual suspects for people who care about SDKs and API design and such. This is just a list of some observations I had, so not everything below i