Just a thought; isn’t the all-card the same as executing the condition cards in sequence? Except that using the all-card might be faster as you are executing the condition cards in parallell (but puting more load on the CPU).
My thoughts exactly. I cannot see any difference, other than maybe the all card solution is faster like you said. Just maybe in all card solution one can also make different combinations (any and all) with same cards, because in series the operation is always all.
Yes, it is, for an “all” sequence at least. I like the all card better because it gives me a better overview. But that is probably a matter of taste.
In combination with multiple all and any conditions it gets more value.
1 Like
Well I understand the thought. But for me it would not work as I do not want all of my flows to be startable manually.
I do however see something in a special kind of start card that includes a condition. To make the switch I create with 2 cards, in 1 card. But that is the only advantage I can see.
One other finding is that if you put two unit cards (e.g. swith on lamp1 and lamp2) in sequence expecting them to be triggered after each other, this will not happen if the first unit (here lamp1) fails to respond on the request due to e.g. mesh issue.
1 Like
Yes and no, in the Example below both family members get the push confirmation at the same time,
chaining them would make the response also sequential while in this case you don’t want that.
and it doesn’t have a real difference in CPU load.
In your case it works as the condition card (requesting for response) will always continue in some way.
Try make a flow where you turn on a switch which is not plugged in followed by a push message.
The message will not be sent.
If you trigger an action from error output of the switch it will be executed thou.
Exactly. I have tested it, works like a charm.
Or not putting them in sequence, or using all-card, in those cases.
Call them in parallell instead.
Are there more poeple missing zoom in/ zoom out on the canvas?
6 Likes
I am using the zoom in the browser, but it would be nice to be able to change the zoom of the canvas itself. But browser zoom does the trick for me.
3 Likes
I want to create a (advanced) flow which continues to run itself until a logic value is set to NO/FALSE but this seems to be impossible…
start flow → if logica_setting = YES → then start hue scene red → wait 5 → start hue scene blue → wait 5 → start hue scene green → start whole advanced flow again so it starts to loop until NO is set.
This seems to be impossible using advanced flow…
The (advanced) editor simply won’t allow to create a new connection line at the end to the beginning of the flow or an other card (in the flow) which start it again, it is not an infinity loop because there is a card which checks if a value it met or not.
Can’t you just start the flow again using a start flow card?
By the way, as far as I understand, as long as a flow is running, a value of a (logic) variable that is set outside the flow, will never reach the flow. As long as a flow is running, the values keep the value that it was started with. At least, that is what I understood. And then your loop would never stop as the NO would not reach the flow.
You can start it but it run only 1 time, is changes the light from blue, to red, green and then stops.
The last step would be to restart the whole flow so the lights constant change color until a specific flow_card is changed to FALSE and then it stops.
I have it working with ‘normal’ flow card but i need several flows to achieve it, so i was playing around to see if i could recreate the effect in the advanced editor but it is not possible.
I cannot find a way to restart the advanced flow itself, you need at least other flows to achieve it.
In the flow there is a logica card with a value, if TRUE the flow should run, if FALSE it should end, so when changing the value in theory the loop should end, this is the case when using simple flows…
If I understand your question correct than it should be possible with something like:
My flow (test) keeps running and repeating itself until the boolean (also called test…) is turned true. See result in notifications:
Maybe it helps. Otherwise I might have misunderstood the question.
1 Like
No, that is not how it works.
Ah, good to know!
That’s indeed the solution, but you don’t need the Any card. Just connect the output terminal and the error terminal to the second card input terminal. Either one of the signals will execute the second card.
Suggestion; An alternative to live montitoring AF. There could be a button recording the next x number of trigger events. And then “replay/show” the flow for those trigger events. Then we could se what actually happened in the flow.
3 Likes
I have the same issue and I’m on 8.0.1. Only doesn’t work in advanced flow, I now have the trigger moved to a ‘normal’ flow which starts the advanced flow
