Replacing Web UI framework

Markus Bonet

I know this will be a long shot and I didn't find something in the forum about the topic.
Has it been evaluated to use a different Web UI framework? The current one is based on ExtJS, if I'm not mistaken.

solar_flare

I didn't find something in the forum about the topic.

There is something on the forum, originally stemming from a discussion on the old issue tracker

Projects mentioned in the thread:
https://gitlab.com/bernhardberger/tvh-webui
PoulKalff/angularJS

@davep still seems to be working on his simple frontend for tvh dave-p/TVHadmin-JS that inspired others to create their own.

In the forum thread the developers were open to having a new Web UI accessible at a different path for testing purposes. But apparently nothing really happened towards that.

DeltaMikeCharlie

What follows is my personal opinion/recollection, it is neither official TVH policy nor strategy.

The developers have discussed this is the past. The version of ExtJS currently in use is VERY VERY old. Apparently, the changes between our current version and the most recent version are so extreme that a total re-write would be necessary. Also, I think that there was some mention of licensing changes, so ExtJS may no longer be suitable for our FOSS project (But don't quote me on this one).

One of the developers tried a quick AI prompt and it created something in either Angular or React, I don't remember. Unfortunately, from memory, the result was basically non-functional and totally unusable.

My personal preference would be to use Vue.js because it is independent and not linked to a big corporation bent on harvesting personal data. The bulk of TVH is based on 'idnode' objects. Once a single display table module and edit module are developed, then those modules can be reused for almost every object class. I would also really like to add an EPG grid display to replace/augment the existing table display.

If this were to happen, my personal opinion is that there should be 2 WebUIs available. The existing one would be the primary until the new one is 100% functional and then the roles can be swapped. After some time, the old WebUI could be retired.

The existing code would need to be evaluated. Code specifically dealing with the ExtJS framework would need to be re-written, however, some of the pure JavaScript functions may be reusable. The WebUI is also available in multiple languages, so it would be nice if we could re-use as much text as possible so that the whole WebUI does not need to be re-translated.

The fact of the matter is that the current WebUI works. Its look-and-feel is not particularly super modern, but it is functional nonetheless. A WebUI upgrade/replacement project would entail significant effort and there are not a lot of developers working on this project.

@Markus Bonet:
Do you have any programming skills?
Would you like to help develop a new WebUI?
Do you have a list of broken or missing features in the existing WebUI?

Flole

There was also one attempt to get AI to make a new Webinterface, but that wasn't very successful. It looked great but functionality was not really there. That was in the very early days though, with better AI models this might work out better now. Migrating to a newer ext.js version requires a complete rewrite, and their new licensing scheme is a pretty good reason to not do it.