I’m working on it, but I don’t have a release date yet. Can you tell me more about your exact requirements? The concept I’m currently developing is more of a menu widget. It’s designed as a list of text buttons with optional icons that can be oriented horizontally or vertically. It can also be triggered by a (local) button so a user can create a dropdown, drop-up, or side-aligned menu from it. It’s still in development as I’m trying to streamline it with the lightgrid design. (I want to keep the options and style consistent with this widget)
30 menu items sounds like a lot, and 30 dashboards sounds like a lot too. What kind of dashboards are you producing? You might have told me already, and if so, I apologize, my head is a bit full at the moment.
I believe you also requested THEN flow cards to send data to specific dashboards. I’m still crunching on that. It should probably be based on the dashboard name, but it would be really annoying if renaming a dashboard (or doing things like V2, V3…) broke the link.
I also need to keep the communication universal and backwards compatible with MQTT. But these messages should have an extra dashboard ID parameter that the dashboard can ignore if it’s not the same. I’m afraid it might be best to keep all topics unique; if you don’t use that topic on other dashboards, it automatically gets ignored.
Yes, grouping is in the pipeline, but have not fully worked that one out:
Current leaning on the dirtier solution.
Copy and paste is also in the pipeline. It would just copy partial dashboard JSONs to the clipboard so you can transfer them between dashboards or share them with other users. Node-RED has a similar option.
But I have to be careful and keep a lot of things in mind. For example, what would happen if you paste an old version of a dashboard into a new version? It should go through the migration engine. Also, I need to solve grouping first so the copy/paste mechanism can handle groups as well. I should also think about how it handles pages. Should it paste with the pages already configured in the widgets? What happens if the user is on page 10 and pastes widgets that are configured for page 1? Nothing visible would happen. Should it automatically change the page to the current page, or should it show a dialog asking to change the page? Haha, all kinds of things I should overthink.
Anyway, good point about the breaking the pixel perfect adjustments when things are moved in a group. I shall try to fix that one also
Indeed, on purpose I left the x-axis switched on on both (stacked) graphs. Removing one of them leads to a “higher” graph (more room for the y-axis). So this results in a change of the scale of the y-axis.
However…
I did find out that reducing the height can be done in such a way that the x and y axis are aligned and the values have te same place as well. So this work, but then both graphs do have a differend height, which is less intuitive. But, if it works, it works.
I am trying to replace my current tailor made web based dashboard by Dashboard Studio (faster performance and (except some Homey scripts) no programming but configuring.
Close/exit app (not possible yet in Dashboard Studio I believe)
Clicking on a main menu icon with a submenu: the dropdown menu as shown above will appear. In the new dash a click on a main menu without sub, or click on submenu causes a switch to another dashboard + page, or to another page (if all pages are in one dash).
Until now I have not been renaming dash names. During export/backup I use the v2, v3… variants. I am afraid of loads of variables coming from Homey, but separate variables for each dash is a nice to have. And if the performance stays fast even sending a lot of script/flow variables to the dash, than it is no problem.
Short: it would be nice to define a submenu related to a main menu/icon:
defining multiple submenu lines
with an icon and submenu name
colors for selected, unselected and active menu (in my example color blue, white and orange)
changing screen to other dash and/or page
using external SVG icons
Some of the screens I posted before in this topic.
EDIT: do I have to define the menu structure 30 times for 30 dashes?
EDIT 2: Hopefully, you interpret all those wishes from demanding users, like me, as enthusiasm for your software. It remains a hobby, after all, and not an obligation.
Wow man, I applaud your determination in creating such an extreme dashboard. The amount of detail in all those menu options is insane! I still wonder, though, do you actually use all of it, or is this just for fun? For example, why not have a single dashboard page with all the weather info you need? Do you really have a page dedicated solely to “Air Quality”? It is probably a graph, but wouldn’t it be handier to have multiple graphs on a single page?
When I created my first dashboard years ago, I also crammed in as many options as I could fit. I wanted to control every light’s color, temperature, and brightness (and I have about 25 Hue RGB spots in the livingroom). To be honest, I think I used it maybe three times. The dashboard was full of unnecessary stuff, and I eventually discovered that lighting “moods” or “light-themes” were a much better way to control the home. I also wonder how many people are actually using the “light grid” widget, that widget to adjust all individual lights. I haven’t seen any bug reports or requests about it, so either I did an incredible job with that one or nobody is using it haha.
I am not sure what you mean by “exiting” the app since it is just a webpage. The button widget does have an “onclick” option that jumps to another URL. Like I did in the demo, when you press the Emergency Shutdown, it jumps to another page:
This discussion about performance also made me think about this button. Now you can choose 3 modes, topic, url or page. But I think it would be useful if the topic when filled in is also always send. This way a user can go to a different page of the dashboard, and in the same click send a message to homey to update the graph on that page.
Thanks for the screenshots. They remind me that I should also add an optional background blur for containers and menus.
Performance-wise, it wouldn’t make a difference. The data is sent anyway, and the dashboard just ignores it. I also don’t think the amount you send is an issue. I do not notice any slowdown even when I hammer it with a fast stream of image URLs. I do think you can make your Homey’s network more stable by using their direct Ethernet cable adapter.
If you put everything in a single dashboard, then no. You can set the same menu on all the pages, currently only with [1,2,3,4,5…] but I think I should update it to include ranges, like [1-5,7,8,12-15]. If you are going to create 30 separate dashboard files, well, then yes, you cannot share the same widget over multiple dashboards without copying.
Sure, I am happy to hear all requests. Most of your requests or reports made a lot of sense to integrate. Until now, I could spend a massive amount of time on this project, but that will be declining in the coming months. I am self-employed and currently between projects, but a big project is coming up. And since this is free software I don’t feel obligated at all
Within a configured group, only a single toggle switch or button can be active at any given time. When you switch a new button to the ON position, any other buttons in that same group will automatically toggle OFF. You also have the option to allow the button being turned OFF to report its updated value to Homey. Note: This group feature is disabled if the button “On click” behavior is set to the “Go to page” navigation option.
The Button now ALWAYS sends its value to the specified topic, provided there is one configured in the Output topic field. For example, a user can press a button to navigate to a dashboard page and simultaneously trigger an update for a graph on that page. This ensures that Homey only needs to broadcast new values when the page is actually open and active.
Fixed
Fixed. When widgets now move in a group their individual pixel perfect posistion wil not re-adjust to the grid again. (still working on the grouping option)
New Feature: A “When” card that triggers once the dashboard is saved and the user closes the success dialog. This allows Homey to trigger actions like the Fully Kiosk Browser card to reload the page automatically. It is incredibly handy for those times you are remotely editing a dashboard and need to refresh the display on a device that is hard to reach.
New optional setting that warns the user on the dashboard when the connection is lost to Homey. This way the use understands why controls are not working.
Both. Sometimes I ask myself: would it be possible to make a nice graph/list?
To much info that won’t fit on one screen.
Air quality per hour. Which brings me to the question: With Homey script or flow card I can create a json string with air quality, or solar energy values per day compared with previous years, or solar costs vs. revenues (payback period). How could I display these json “tables” in Dashboard Studio?
I don’t understand the meaning of this widget. Maybe a short explanation of an use case could help.
I use Fully Kiosk Browser free licence for using the dash on my tablet. A click on the dash Exit button could close/end the Fully Kiosk Browser app (if possible).
That sounds like a good idea.
I plan to make one flow per dashboard/page (otherwise the flows would be too big), and only activate that flow when the user navigates to that dashboard/page. Now: when I activate the dash tablet, a Macrodroid macro activates all Dash flows and start them. The opposite happens when I turn off the screen again.
I will restart Homey every night again to test if this improves stability.
With my “old” web dash I made a menu component. One definition of the menu structure and reuse it 30 times. Changes are made once. Could a group mechanism extension or template do the job? Also usable for a clock or message on every page.
Alternative: make with script and/or flow a json with menu structure once and read this json in the dash with a menu setup (one time on every page, but way better maintenance).
Big time nit picking:
The color gradient around the start (see red arrow) is identical to the start of the color gradient to the middle color. Is this me not setting up something right?
