methods to facilitate rule creation and debugging

[boc-tothefuture]

Two things I have found as I convert my rules.

1. I usually want debug/trace logging turned on for a specific rule, when I have it turned on everywhere there is too much logging for me to find what I want. Proposal, add a 'log_level' to the rule syntax.
2. I often have a set of things when I am building or debugging a rule that I want to know the state of/value of as the rule executes (or not due to a guard). Proposal, add a 'debug' which accepts an object or array of objects and logs the inspection of them.

[jimtng]

I like both ideas.

[ccutrer]

how is this better than the rule itself having a proper unique logger name, which you can set the log level from the openhab console? or alternately, is it possible to call logger.level = :DEBUG from inside the rule block to accomplish the same thing?

[boc-tothefuture]

is it possible to call logger.level = :DEBUG from inside the rule block to accomplish the same thing?

No because that only would impact the logging inside the run block - nothing about what happened leading up to the run block being called.

how is this better than the rule itself having a proper unique logger name, which you can set the log level from the openhab console?

Maybe my workflow is different than others.. but when I am creating the rule, I usually want to debug that rule and then I am happy to set that rule back to info logging. Going through the process of trying to convert the rule name (with space in it) into what you have to enter at the console and then debugging why a simple typo is not getting the logging level I want seems difficult compared to having a 'log_level' directive for each rule.

[jimtng]

Would this work:

rule 'mylog' do
  log_level = :warn
  on_start
  run do
    logger.send(log_level, 'test')
  end
end