Support all browsers that support web extensions

[tombh]

The biggest hurdle is using a standardised RDP. This; https://remotedebug.org/ isn't yet supported by all browsers!

• Chrome doesn't support extensions in headless mode! So we need to find a way to stop Chrome stealing focus on things like window resizing. Only relevant if using Browsh locally of course.
• Chrome also has a different API for snapshoting the rendered tab.
• Firefox still doesn't have support for the standardised RDP. It'd be nice to not have to rewrite the RDP commands for every browser. Although theoretically the RDP commands should be minimal, eg; resizing the browser window and automatically installing the extension.

[tobimensch]

This is low priority IMO. First iron out all problems on the firefox platform and when browsh becomes popular, other people will adapt it for other platforms such as Chrome.

[msaffer]

It would be great if this supported Chrome.

[tombh]

@msaffer I agree. But could you just give me your own reasons? Thanks.

[MrAdam]

Just having acquired a Raspberry Pi 3 B+, I'd say the Chrome/Chromium support would be nice, as a version of firefox supported by Browsh is very hard to install on RPi3 due to an issue with armhf and the later versions of Firefox (https://bugs.launchpad.net/ubuntu/+source/firefox/+bug/1711337).

[kgfly]

+1 for chrome. Thanks. Chrome is the most popular. Even Microsoft has given up Edge/IE for chrome.

[niutech]

Can't we use any headless browser and inject a content script to every rendered web page?

[tombh]

In theory yes, but in reality there are some complicated hacks that make Browsh possible, like being able to screenshot the rendered page in order to detect the visibility of text.

[niutech]

Almost every headless browser allows to make a screenshot and access the DOM. Can't we use e.g. Playwright with multiple engines?

[niutech]

I was able to run Chromium (non-headless) with an extension using xvfb-run --server-args="-screen 0 1024x768x24" chromium-browser --remote-debugging-port=9222 --disable-gpu --load-extension=~/web-extension and chrome-remote-interface:

Next we could try loading a modified Browsh web extension using --load-extension.

[tombh]

Oh! Thanks for checking that out. So the screenshot needs to be available in code like this: https://github.com/browsh-org/browsh/blob/2f713f40fc87f5a3ca1ba15c469844c61f5f3fc6/webext/src/dom/graphics_builder.js

[niutech]

I can see you're using a deprecated Firefox-only drawWindow(), can't we use a standard captureVisibleTab() from background.js or html2canvas from the content script?

[tombh]

It's been a long time since I checked the latest Firefox API, I think current Browsh is pinned to some really old version, like v67!

[niutech]

Can we unpin it and use a standard way to capture a tab, so it works in both Firefox and Chromium?

[tombh]

Sure, but we'd need to test that

[MXP2095onetechguy]

I also want to see this with chromium, webextension can be used, but I think that browsh should support Selenium WebDriver, using that can add more browsers.

[MXP2095onetechguy]

Also I found this selenium webdriver client, it is for go. Browsh can use nodejs, but that would add more dependency, so here is the client.

I also found more dated webdriver clients.

Or Playwright with Playwright-Go.

Maybe that the browser can use selenium webdriver and if that is not available, then fall back playwright and then fall back to webextension.

[tombh]

I'm actually most interested in Chrome's DOMSnapshot API at the moment. It would restrict us to Chrome, but it could profoundly affect Browsh's rendering and performance. I made an issue for it: #390