Arie do we need to reselect capability ?
after re-selecting
Thank you.
No, not really, the ID in the background is still the same.
You only need to re-select it if you want the cool real property name visual in the floweditor 
2 Likes
Is it possible to get the zone name in a tag when using the card Get Insights from 0 minutes ago flowcard? This makes it easier to create a uniform logging card.
@Jeroen_W Sure! Please create a feature/support ticket (link at the top post).
Thanks!
1 Like
I have the problem that the order of the fields in the creation does not match the order on the map. I want to put the values (day, week) next to each other. The fields were created in the order I want them, but they are then out of order on the map. It would be helpful to be able to assign the order of the fields on the map.
iâm back:
this should work right? because right now it doesnât. either it does not allow text (because the dropdown contains stuff like âless thenâ and âgreater thenâ) or it needs to be exactly the value is âHoofdstraat 44â
if it is exact⌠could a dropdown option âcontainsâ be added ?
Yes, a way to sort is on the todo list and i have already started with it, but this one takes time to develop correctly, i hope it will be done in the next 2 weeks.
Sure! Please create a feature ticket. Thanks!
1 Like
Hi Arie,
Since some time now (at least since 05/06) my H23-system is quite unstable and I think it is related to my AVDâs. Either they are;
- too numerous (27 total) or
- too complex (some AVDâs have 12 populated fields, I have multiple reflections, BLL-calculations, and what not) or
- I made some stupid (looping) mistake somewhere
I am searching and searching to pinpoint the problem(s), but until now am not able to get this done. What is rather frustrating in the process is the fact that the app crashes randomly/continuously wile working with the different AVDâs to change some parameters. I think this is because of the Athom-imposed RAM-restriction concerning apps.
Attached are some screens of the system load.
I also have some concrete questions as I suspect the system-stress might be related to these AVDâs in particular.
- Is it okay to have nested reflections (for example AVD2 reflecting AVD1 reflecting Device0)?
- Is it okay to have reflections within an AVD (for example field2 reflecting max of field1 and field0)?
- Is it okay to have reflections of its own field (for example field1 reflecting min of field1 and field0)?
- Is it okay to log a certain value within a timeframe with reflection and to reset with a flow at chosen moments (for example a day max temperature being calculated on the fly reflecting the max of the prior logged temperature - thus the field itself - and the current temperature and to have the day max temperature reset at 00:00 to the current temperature)?
Attached are some examples.
I hope to get some leads as of where to continue my search. Thanks in advance.
If you need more info, let me know.
To all you questions: yes you can.
When the app crashes again, let me know when it happends, so i can see if there is a crash log.
This and next week i donât have much time at all, but after that, i hope to put more time in memory performance again for the DC app.
Simply provide debug log when/right after app crashes 
I am unsure what you mean by this?
Each up and down line is a crash?
And i am wondering why you app uses more than 100mb at all.
DC/AVD that uses reflection, uses a lot of memory. But i havenât seen that much memory usage at all.
I guess you indeed are crosing the MB limit set by Homey.
I still use the âoldâ athom-api, instead of the new homey-api. Once homey-api is working correct with the insights, i hope to migrate it. Hopefully, using capability-listeners (reflection) will use less memory than.
Yes they are, unfortunately. Or more precisely, I think strictly speaking they are not real app-crashes (sometimes I do, but most of the time I do not get the red crash alert in the app-list). Rather than that, the spikes correspond to the app getting unresponsive resulting in the devices getting offline. I guess at this point the app gets killed by Homey due to the high memory usage.
As you can imagine, I am tinkering on the high memory usage myself.
As stated, I really enjoy using the app to create elaborate AVDâs trying to both streamline and expand my automations wile at the same time - in the context of a lacking dashboard - improving overview by grouping related stuff.
But if reflection is indeed the reason for the high memory usage I should seriously reconsider the use of this killer-feature.
I doubt if its even possible to give a straightforward answer on such a broad question, but would you think using regular advanced flows to get the same outcome would result in a smaller memory footprint altogether? Than maybe I should revert back to a traditional structured approach and use reflection sparsely for some exceptions. (This would really be unfortunate as I put a lot of effort in the process already. And wile the memory issue is rather frustrating - to the point of getting inevitable - I was otherwise very pleased in the outcome.)
For now, I will at least pause the further conversion/rebuilding of my old devices/flows.
Yup, from what I know is, Pro 201x pauses/kills all apps using over 80MB for x minutes
For Pro 2023 Iâm not sure, probably 140, 150 MB, like your graphs show.
Hello,
I have some issues with the âListen cardâ, the User and Client always returns âblankâ, is this a known issue with HP2023? (FW.Rc133) It worked flawlessly with HP 2029, but now i can not set âfrom appâ or âfrom Webâ and so on, because it returns a blank tag.
Kind regards
Marius
I had the same. @Arie_J_Godschalk recommended to use the Get Insights from 0 minutes ago flowcard that worked perfect for me.
1 Like
Its an Advanced Flow only Then card.
I do have advanced flow, @Arie_J_Godschalk . Last picture.
Edit: Found it.
That card also returns tags which is blank.
