Struggling to get dynamically loaded page (but scrolled and loaded completely) saved

[vb0]

I've noticed it with Pocket but this isn't the main use case (especially that they're closing and I have my data anyway), for example for https://getpocket.com/saves - it's one of the scroll-scroll-load-more pages, not excessively large in my case.

I changed all settings I could think of, but what seems to be happening is that during SingleFile saving there is some scrolling and content loading done while the page is partly/mostly white, no idea what's triggering it even with "dispatch "scroll" event" and "zoom out the page" disabled (and frankly with most settings flipped either way). In any case the saved file is missing all my entries that are otherwise visible in the page I'm looking at (these are three columns with web page thumbnails and descriptions).

Clearly, this is dynamic content generated from my list, and not some html that sits on the page, but I want somehow to harden this for other cases, if there is a way to make sure I'm getting the text I'm looking at and is already loaded in the browser.

[gildas-lormeau]

Could you confirm that just enabling "zoom out the page" and letting the other options to the default value does not fix the issue?

[vb0]

Thank you for your answer, yes I confirm this isn't fixing the issue, however I think now it isn't after all a SingleFile issue which works as well as it can in this scenario.

I was under the impression the page is loaded when you just scroll back to the bottom and I have a long page in my browser but I zoomed out to maximum and it appears that as I scroll the page is still dynamically loaded both backwards and forwards, so whatever number of rows SingleFile is catching is representing what the browser has anyway, and whatever is missing is some dynamic content that after all it would be expected to be missed.

[gildas-lormeau]

Sorry for the late reply, this is the same kind of issue as #1773 for example. I need to think of a way to detect this kind of problem, but it's not easy to do in a generic way, unfortunately.