|
Emacs actually comes with a built-in web browser called EWW (short for Emacs Web Wowser). It’s a lightweight, text-based browser that runs entirely inside Emacs, making it perfect if you want to stay within your editor without switching to an external browser. This is AI-generated list of key features of EWW:
I would like to have one in ecode. I think I will use it to browse lobste.rs or hackernews as these sites are mostly text. Yeah, online manuals like ss64.com and man7.org, too. |
Replies: 1 comment
|
It depends on the user expectations. What I'm working on is an HTML compatibility layer, the idea is to support a defined subset of the HTML+CSS specification which plans to be approximately on parity with something like litehtml or dillo or NetSurf. But this is a massive task even supporting an small subset of the specification. It's also orders of magnitude more complex than a text-based browser (they are not even remotely comparable in complexity). So I don't know, it depends, I'm currently experimenting and this is not something really serious, I might just leave it how it is right now or I might continue working on it, I'm unsure what will happen, what's for sure is the "upper limit" I plan to support, it won't be anything beyond any of the browser engines I mentioned, and it will not support JavaScript. Current state is not very capable, it can render some simple sites and some not so simple but most not perfectly at the pixel level. I started focusing on achieving perfect rendering of Hacker News and that has been working quite well for some time, you can even login and post from it: But as for now there are not plans of integrating a browser inside ecode, given that it would trigger requests for supporting some of the thousands features that need to be implemented to support some specific sites which I don't want to focus on. This is currently just experimentation. And, also, I'm currently constantly breaking things, since i'm iterating fast and I have not enough tests yet, so it's extremely unstable and very broken. A ton of work needs to be done to stabilize it. |


It depends on the user expectations. What I'm working on is an HTML compatibility layer, the idea is to support a defined subset of the HTML+CSS specification which plans to be approximately on parity with something like litehtml or dillo or NetSurf. But this is a massive task even supporting an small subset of the specification. It's also orders of magnitude more complex than a text-based browser (they are not even remotely comparable in complexity). So I don't know, it depends, I'm currently experimenting and this is not something really serious, I might just leave it how it is right now or I might continue working on it, I'm unsure what will happen, what's for sure is the "upper limit"…