Add Simple Command Sequencer - #178
Conversation
77a7208 to
fe41448
Compare
fe41448 to
e4e6861
Compare
e569b64 to
ea8530d
Compare
…it a keyword Also the record generation
…rgument serde not propogated yet)
aea2185 to
0ef12fa
Compare
| end case; | ||
| end record; | ||
|
|
||
| type Step_Array is array (Interfaces.Unsigned_32 range <>) of Step; |
There was a problem hiding this comment.
Make this capped at 16-bits Interfaces.Unsigned_16
|
|
||
| - name: Invalid_Command_Received | ||
| description: A command was received with invalid parameters. | ||
| param_type: Invalid_Command_Info.T No newline at end of file |
There was a problem hiding this comment.
We need data products. Frame running count, frame running HWM, num sequences started, num sequences finished success, num sequences failed, total commands sent, last sequence started ID, last sequence finished success, last sequence finished failure.
| # The id_bases parameter already holds this value, so | ||
| # the parameter-sync that set_id_base() also performs | ||
| # is a no-op here. | ||
| suite.id_base = int(par.value) |
There was a problem hiding this comment.
Seems like we need a public method to do this instead of hacking it. Maybe we just extend set_id_base with a second argument called assign_entity_ids which defaults to True, but we set it to false for this call.
| # setup, keeping ids identical for every other component. | ||
| for suite in suite_list: | ||
| if suite.id_base is not None: | ||
| suite._set_ids(suite.id_base) |
There was a problem hiding this comment.
Calling a _ function seems hacky. Maybe we need a new public interface called something like assign_entity_ids_with_id_base, since the id base is already stored within the suite. Seems weird to grab it from the suite and pass it back into a private function. Open to other ideas too, but calling _ function is a no go.
| - command: Component_A.Command_1 | ||
| - command: Component_A.Command_2 | ||
| arg: (Seconds => 3, Subseconds => 14) | ||
| - sleep: (Value => 3000) |
There was a problem hiding this comment.
Forcing the (Value => N) here seems unnecessary. Maybe we can just make it an integer in the schema?
| - command: Component_A.Command_1 | ||
| - command: Component_A.Command_2 | ||
| arg: (Seconds => 3, Subseconds => 14) | ||
| - sleep: (Value => 3000) |
There was a problem hiding this comment.
Can the sleep be dynamic (e.g. based on Arg?)
| description: Run a command sequence by ID. Per-sequence wrapper commands are the operator-friendly form; this is the underlying backbone they all dispatch through. | ||
| arg_type: Run_Sequence_Arg.T | ||
|
|
||
| - name: Kill_All_Sequences |
There was a problem hiding this comment.
Do we want to be able to just kill a specific frame too?
| description: Id of a Command Sequence Registered with the Simple Command Sequencer | ||
| type: Interfaces.Unsigned_16 | ||
| format: U16 | ||
| - name: Response_Behavior |
There was a problem hiding this comment.
This feature seems to be causing a lot of complexity. I am wondering if its worth considering removing it or changing it in some way. Let's discuss.
|
|
||
| sequence: | ||
| - command: Sequencer.Sequence_C | ||
| arg: (Response_Behavior => Sequence_Enums.Sequence_Response_Behavior.Send_After_Sequence_Start) |
There was a problem hiding this comment.
Yeah, not sure I love this. Do we really need this. Could it be statically configured for each sequence similar to wait_for_command_completion for each sequence.
2 participants
This component adds functionality to define command sequences in YAML that can be run through this component. It has support for simple time sleeping in sequences, a few useful flags to allow control of flow, and dynamic arguments that can be passed in when the sequence is run instead of at compile time.