Replies: 8 comments 4 replies
-
|
ping @roadriverrail Dunno if it auto notifies you of the discussion. |
Beta Was this translation helpful? Give feedback.
-
|
I indeed didn't get a notification until you pinged me. I have absolutely no attachment to anything in the current client. I grabbed some sample code and made it work so that my Roku could play my DJ streams reliably. Moreover, I don't actually have tons of time to devote to the Roku client, since I need to wait to have enough time off work to sit down and make more features and it competes for my free time with other projects (like how I'm currently trying to get Dropout.tv on Kodi). As such, I have no attachments to what exists, I'd prefer to maximize the ergonomics for a developer with lots of enthusiasm, and I'd prefer to defer to someone more experienced with me. |
Beta Was this translation helpful? Give feedback.
-
|
Cool cool. I don't mind creating a starter to match current capabilities that we can iterate on 🤘 |
Beta Was this translation helpful? Give feedback.
-
|
@roadriverrail @gabek Here's a quick demo video using the newly written Owncast Roku client: https://social.linux.pizza/@tgpo/113891069139681525 Zero SGDEX code 🤘 😆 |
Beta Was this translation helpful? Give feedback.
-
|
Fuckin' ship it! And here's the thing...if this means a code re-org that is more easy to reason about, it'll raise my ability to contribute, because probably 2/3 to 3/4 of my time working on the app went to figuring out how in the hell Roku's SGDEX stuff worked in the first place, leaving little time and mental bandwidth for thinking about doing things the way I wanted or they way they needed to be done. My proposal-- we've merged the favorites feature to main, so I can publish ASAP. We really ought to do that because a SNAFU in the original publishing process left the channel description in the store in a sorry state, and I can only update that when I publish. That'll give us breathing room to then publish the rewritten app after review, testing, polishing. Sound like a plan? |
Beta Was this translation helpful? Give feedback.
-
Yeah, the amount of extra code is bananas!
Sounds like a great play. @gabek I created a repo in GH to track my code. Do you want me to do a ownership transfer to the owncast org so the repo resides over here and not with me? |
Beta Was this translation helpful? Give feedback.
-
|
@1hitsong You could add this repo as a separate remote and then push your old repo's master into this repo: I think that would copy your repo as a branch. |
Beta Was this translation helpful? Give feedback.
-
|
Lovely. I'm away in the UK through Friday and I likely won't get a chance to look at it until I get back, especially as I lack a Roku on the road. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
I've spent a few hours mentally tracking the flow of the Roku channel's code, and there is a TON of scaffolding that abstracts almost all the component functionality so programmers need only to set required property values to have a working channel.
While this does make it easier/quicker to get up and running from scratch, it makes it a huge pain to change things from how they were originally setup by Roku.
As an example, I looked at adding the visitor count to the video, and it looks like even adding this simple value will be a big headache because of how Roku wrote the video component.
As we add more features and customizations to the channel, we would need to deviate further from the SGDEX abstractions, which would force us to poke holes in its logic to bend it into doing what we want.
Due to the hardships working with the SGDEX code, and the limiting of our future plans, I purpose we create a new channel using custom code and components now before we add so many features to the SGDEX code that it becomes impossible to convert it over.
This would allow us to do anything we want in the channel and not be held hostage by the SGDEX framework code.
We're all programmers, we all know the old lore of the new programmer who comes in and want to rewrite everything from scratch as their first order of business.
However, having spent so much time programming for Roku, I don't believe this is the case here, it's more of a situation for setting us up for success in the long term.
Thoughts? Questions? Counter-points?
Beta Was this translation helpful? Give feedback.
All reactions