Debugging, Naming, indexes and ids

[reconbot]

I was thinking about how we might improve the readability of the outputted automation. I came up with a small list of ideas/thoughts. My goal would be more debugable output.

• Allow naming nodes, because it's much easier to follow "action_turn_off" vs "action_1770147441877", you allow trigger Ids (and say they need to be unique) but they could be the same.
• I noticed that the long template string for the initial value checks the index of the trigger in the array. It could use the trigger's IDs which would be easier to read.
• The template string is a similar purpose as choose block, and it could actually be a choose block which would perhaps make multiple actions for a trigger easier to implement.
• Node IDs could be generated a little more friendly by using an integer vs a timestamp "action_1" and "action_2" are easier to reason about than a timestamp, bonus they could have data from their node "action_light_state_on_porch_light_1"

I haven't dug deep enough to know how easy or hard any of these could be, but if any are welcome (and maybe easy..) I'd try my hand at contributing a patch.

[FezVrasta]

I agree the transpiler could be improved, my initial focus was to make it work so I didn't spend too much time on these QoL improvements.

[reconbot]

Makes perfect sense. No worries.