Replies: 20 comments
-
|
here a short demo of the current version:
|
Beta Was this translation helpful? Give feedback.
-
|
please @fredlcore @1coderookie @LucekDev @dukess add your feedback |
Beta Was this translation helpful? Give feedback.
-
|
Thank you for your efforts and the demo, it really looks nice!
As for a general replacement of the current user interface, we have discussed the use of SD cards in the past already (somewhere down in the 5000+ post thread in the FHEM forum). There has been a design decision relatively early on to choose the Arduino in favor of a Raspi or other platform because of the fact it runs both without a operating system that would need to be maintained as well as any parts that are exposed to wear and tear (such as SD cards). The idea was to have a "fire and forget" kind of system that could run for years and decades without further interaction. Faulty SD cards can be a pain in the back to identify as origins of errors, as I myself have witnessed several times, so for all our functionality add-ons, we have always chosen the use of an SD card as optional. If your new web interface can save space for those users who are willing to use a SD-card, and we can make it optional, I'm all in, provided that we can clearly separate both approaches (the same way that Sergey has done it with the WEBSERVER definement), so that we can also split the workload. |
Beta Was this translation helpful? Give feedback.
-
|
@fredlcore thanks for your feedback. I understand your ideas around sd-card. What I see at the moment is, that if I compile it, that there are still 200kb of flash free, so if the person would not like to Would this a possible solution, or not? |
Beta Was this translation helpful? Give feedback.
-
|
It's true that for Latin (i.e. 8-Bit ASCI)I languages, we still have 200k left. The situation looks different for UTF-8 encoded languages like Russian or Greek where the language files are basically double the size - that was the reason why @dukess came up with the possibility to run the whole webinterface via SD-card. To add another 50k to that would bring us relatively soon to a point where we will be forced to make decisions which features to compile in. If we have a solution for this through the means that Sergey has provided, I would rather want us to make use of that. |
Beta Was this translation helpful? Give feedback.
-
|
Hi, have you ever thought to use something like https://github.com/siara-cc/Unishox_Arduino_Progmem_lib |
Beta Was this translation helpful? Give feedback.
-
|
I can have a look at it, but that won't be before beginning of January. I would prefer to check whether Sergey's approach is feasible because it would also compensate for all the other cons you have listed. |
Beta Was this translation helpful? Give feedback.
-
|
It's probably not the same, but didn't @dukess already wrote something like that..? https://github.com/dukess/bsb_lan_ajax |
Beta Was this translation helpful? Give feedback.
-
|
But in general, also in my opinion the usage should be possible without the (general) need of an SD card. |
Beta Was this translation helpful? Give feedback.
-
|
@1coderookie thx for the hint to dukess bsb_lan_ajax, my variant is completely different, I will also talk to @dukess to allign here. It doesn't make sense to develop thing multiple times. |
Beta Was this translation helpful? Give feedback.
-
|
As told Mao Zedong "lets blooms thousand flowers"! Yes, we can cut most part of native web interface for saving flash (with new definition like I_DO_NOT_NEED_NATIVE_WEB_INTERFACE). Yes, WEBSERVER module can deliver /index.html when user query / URL so internal interface can be replaced by HTML from SD card.
It is interesting idea. Possible it will be next stage of optimization. But I'm not ready to take it on, as at the moment for my purposes there is no need for it because all language resources placed on SD. |
Beta Was this translation helpful? Give feedback.
-
Great, so that sounds like an approach that combines best of both worlds. |
Beta Was this translation helpful? Give feedback.
-
|
@dukess #define I_DO_NOT_NEED_NATIVE_WEB_INTERFACE would be great if you can add this to the code and also the direct read for index.html from the sd card. |
Beta Was this translation helpful? Give feedback.
-
|
First we need to agree on what to cut. If we leave configure through URL, we can free ~12Kb (incl. D3 library). Not too much. |
Beta Was this translation helpful? Give feedback.
-
|
@dukess I can compile down my Vue App at the moment to around 48k but without d3. From my standpoint the real flash save has to come from the Unishox to reduce the strings. I think from the code the biggest advantage of a pure URL / JSON codebase would be the reduction of code and no mix between parsing and generation of Html. |
Beta Was this translation helpful? Give feedback.
-
|
@DukeS @fredlcore @1coderookie
bsbWEB{
"0": {
"name": "Datum/Zeit",
"error": 0,
"value": "02.01.2021 21:46:39",
"desc": "",
"dataType": 5,
"readonly": 0,
"unit": ""
},
"6": {
"name": "Sommerzeitende Tag/Monat",
"error": 0,
"value": "25.10.",
"desc": "",
"dataType": 6,
"readonly": 0,
"unit": ""
},
"500": {
"name": "Mo",
"error": 0,
"value": "1. 04:00 - 00:00 2. --:-- - --:-- 3. --:-- - --:--",
"desc": "",
"dataType": 5,
"readonly": 0,
"unit": ""
},
"600": {
"name": "Mo",
"error": 0,
"value": "1. 05:30 - 07:30 2. 10:30 - 12:30 3. 16:30 - 18:30",
"desc": "",
"dataType": 5,
"readonly": 0,
"unit": ""
},
"700": {
"name": "Betriebsart",
"error": 0,
"value": 1,
"desc": "Automatik",
"dataType": 1,
"readonly": 0,
"unit": ""
},
"701": {
"name": "Präsenztaste (temporäre Abwesenheit)",
"error": 7,
"value": "",
"desc": "",
"dataType": 1,
"readonly": 0,
"unit": ""
},
"710": {
"name": "Komfortsollwert",
"error": 0,
"value": "20.5",
"desc": "",
"dataType": 0,
"readonly": 0,
"unit": "°C"
},
"832": {
"name": "Antrieb Typ",
"error": 0,
"value": 255,
"desc": "3-Punkt",
"dataType": 1,
"readonly": 1,
"unit": ""
},
"1602": {
"name": "TWW Status",
"error": 0,
"value": "01000101",
"desc": "",
"dataType": 2,
"readonly": 0,
"unit": ""
},
"5011": {
"name": "Trinkwasser-Speicher Ladevorlegungszeit",
"error": 0,
"value": "01:00",
"desc": "",
"dataType": 4,
"readonly": 0,
"unit": ""
},
"6224": {
"name": "Geräte-Identifikation",
"error": 0,
"value": "LMS15.003A100",
"desc": "",
"dataType": 7,
"readonly": 0,
"unit": ""
}
}vs bsbLAN |
Beta Was this translation helpful? Give feedback.
-
|
Not sure to understand all ... but I'm new user of bsb_lan hard/software for now in testing mode. My target is to integrate it into my automation software so I think be closer to a "standard" JSON api is the way to go for this great and active project ! I subscribe to this post, maybe I can help you somehow ... @konne can I test your work ? |
Beta Was this translation helpful? Give feedback.
-
|
@rroblik I'm working at the moment on two main libraries:
bye |
Beta Was this translation helpful? Give feedback.
-
|
@konne : I'm going to test it once available |
Beta Was this translation helpful? Give feedback.
-
|
I'm moving this to the discussion area as this is a better environment where comments can be made directly to specific matters/questions. |
Beta Was this translation helpful? Give feedback.
-
hi,
I like to start a discussion about the idea to have a full new web interface that only communicates through the JSON API.
I'm starting at the moment to develop a bsbJS Lib that can run perfectly run on rapsi and any system that can run node.
The library uses the full C definition file of bsb_lan so that it
For that, I have to develop a full web interface that delivers the same results to the end-user.
My Idea now, why not have in the future only ONE web frontend that is used by everybody with Arduino and node user.
Here my Pro/Contra list. Please add missing points so that in the end you as a team can make your decisions.
Pro:
Contra:
Beta Was this translation helpful? Give feedback.
All reactions