chromiumoxide VS openmls

Re rust-headless-chrome: yes, being handwritten it probably has a nicer API/DX but is missing some features. I would give it a go if you are struggling with chromiumoxide. The flakoness could be due to chromiumoxide going faster than the browser, ie trying to get elements before they have loaded. Most web automation tools will have ways to slow down themselves down, eg wait for 1 second, or keep checking whether an element exists. This is what things like Playwright provide, not sure what chomiumoxide does. Also not sure if this is something that is part of the devtools spec and thus should already be in chomiumoxide, or features implemented on top, in which case chomiumoxide might not have it (though I suspect it probably does) eg see https://github.com/mattsse/chromiumoxide/issues/42 In hindsight chomiumoxide might not be the best choice for you if it is lacking these kinds of features. I'm just using it for executing Javascript, not acually interacting with the Dom. I prefer it because I think the codegen approach is best and is the project most likely to succeed long term.

Adding a ton of new features for the headless chrome api chromiumoxide, especially javascript evaluation functions. Also started implementing a test suite.

I case you refer to me as the MLS co-author, I'm not directly affiliated with the p2panda project. I think p2panda uses OpenMLS (OSS implementation of the MLS protocol, https://openmls.tech), hence the connection. I did however exchange with this friendly an motivated team!

Website: https://openmls.tech

They also published an implementation as Rust library: https://openmls.tech/

I just wanted to mention the development of a new protocol named Messaging Layer Security (MLS) which is theoretically supposed to make these messy encrypted group messages more prone to surveillance.

Two different projects: Arti and OpenMLS.