add configurable homing timeout

[sgermanserrano]

Enables to have a configurable homing timeout for slower joints, e.g.

defaults: # optional, all defaults can be overwritten per node
  eds_pkg: my_config_package # optional package  name for relative path
  eds_file: "my_config.dcf" # path to EDS/DCF file
  motor_layer:
    homing_timeout: 60
  dcf_overlay: # "ObjectID": "ParameterValue" (both as strings)
    "1016sub1" : "0x7F0064" # heartbeat timeout of 100 ms for master at 127
    "1017": "100" # heartbeat producer

It defaults to 10s to keep the original behaviour

Signed-off-by: Servando German Serrano [email protected]

[mathias-luedtke]

Thanks! I had a similar patch in mind. Right now I am just not sure, if we should break the ABI like this.

[mathias-luedtke]

Why did you create a new function and did not pass it as a parameter to executeHoming?

[sgermanserrano]

@ipa-mdl tbh, I didn't know which option was better, I'll refactor to have it as:
executeHoming(canopen::LayerStatus &status, const boost::chrono::seconds &homing_timeout)
and pass it with the call to executeHoming

[mathias-luedtke]

Me neither ;)

That's why I wanted to know why you chose that option.

I'll refactor to have it as:
executeHoming(canopen::LayerStatus &status, const boost::chrono::seconds &homing_timeout)
and pass it with the call to executeHoming

👍

Please keep the function without the argument as well (stable API) and call the new one with a timeout of 10s (as before).
I will try to figure out, if we can avoid the ABI break in Motor402.

[sgermanserrano]

I've added the new function and replace the call to executeHoming with a default timeout of 10s

[ivan140]

Any updates or help needed on this pull request? I've tested it on a super slow joint and the homing was working really good, so looking forward for a merge :)

[amilcarlucas]

Ping. Please merge