@moritz31 said in [WIP] Pimatic iOS:
@leader21
thank you for your words, i’ll try to keep my work up as long as it is possible
sure, i didn’t mean to maintain forever ^^ then good luck for this nice project!
[WIP] Pimatic iOS
@moritz31 said in [WIP] Pimatic iOS:
@leader21
thank you for your words, i’ll try to keep my work up as long as it is possible
sure, i didn’t mean to maintain forever ^^ then good luck for this nice project!
pimatic v0.9 has been released!
Support Pimatic and get some free stickers
Like us on Facebook
make it so !
If you need beta testers let me know. I Started the post referenced above but I currently don’t find any time to create an app
One major issue I see in a native iOS app is how do you handle widget customizations that various plugin do - I would almost argue that its best to keep the core app HTML5 running from the web and just do the browser wrapper with extra features - a lot of stuff on the feature request page can be implemented this way.
The core pimatic UI is very mobile friendly and in my opinion good enough - we may want to add some ios UI Support features, stylesheet changes;
BTW I have been working on such an app -
http://i.imgur.com/q5ZDS4S.png?1
http://i.imgur.com/9QNpgI6.png
@moritz31 are you interested in combining our efforts? I am also a beginner - well not entirely but in this area - haven’t written iOS apps in a few years, but I do product management on mobile apps.
BTW I could use better quality artwork - where can I access the original artwork? I intend to release the copyright of this app to pimatic,org and will be putting V1 on github in a couple of weeks.
@Enis-Hoca
Wow quite nice work, really good idea, I already thought about handling plugins, but without an good result, maybe we can share and work together. Maybe write me an pm
@Enis-Hoca said in [WIP] Pimatic iOS:
where can I access the original artwork
https://github.com/pimatic/pimatic-artworks - To date, this contains the pimatic icon, only. If you need specific ressources please ask @sweetpi
"It always takes longer than you expect, even when you take into account Hofstadter's Law.", Hofstadter's Law
As I was developing this app, I discovered something quite accidentally - pimatic mobile front-end its already a very well behaved iOS aware web-app and doesn’t require any wrapping. Would have been good if it was documented better.
If you create a shortcut on the main screen from ios Safari menu, and use that to launch directly, it will load the pimatic app fullscreen without any Safari UI. Pimatic then behaves and runs independently of Safari and even shows up as a separate app, as you cycle through apps. In this mode the app even loads up offline.
This was pretty much my goal. my source code is here if anyone one want to use it https://github.com/enishoca/pimatic-ipimatic - I was going to add authentication next - but given this, i see no point in continuing.
@Enis-Hoca that web-app feature is not great. Its very slow compared to a real app.
The Web-app is nice, but it would be awesome to improve the socket connection on large configs.
Maybe parallel loading is an option.
Do only worse thing is the slow connection on large configs.
@sweebee the biggest challenge I see in a native app is handling the widget customization - If you ignore it then there is potential that the plugin would not be compatible - there probably needs to a higher level construct in the Pimatic base which lets the plugin developers influence the UI rather than extending the mobile front-end, that goes against the whole plugin model anyway.
The current model pretty much ensures that there can’t be any functional UIs other than the one out of box. Opehab’s model makes this much simpler - I have a kodi/xbmc front end for openhab. Looked at doing it for Pimatic, but gave up. Also the dimmer implemetation i.e - commingled dim level with on/off means there is information loss. I had to customize the widget and overwrite the dimmer in the Wink plugin, because of this
Cc: @sweetpi
@Enis-Hoca stated it right, very have to handle the plugins, so I see no other way then implementing all plugins by hand into the app.
Maybe there should be like an api which explain the client what the device should be like.
Don’t know if something like that is possible.
But as long as the app have to handle the plugins, I think EnisHoca’s webapp solution is much more comfortable.