having page renders in skpicture is very practical for various reasons, just to name a few:
1) lots of people still use qt webkit just because it allows them to render pages to SVG / pdf (such as
https://wkhtmltopdf.org/), qt webkit is very dead but there is no alternative, from skpicture it is trivial to save to true vector svg / pdf
this is something quite interesting - if you try to save to svg on chromium based browsers or on libwebkit-gtk (which uses the latest webkit codebase) you will get a SVG but it wont be vector just an embedded PNG, only qt webkit because it still uses webkit 1 can vector render, with skp it is possible to achieve the same
2) my personal interest in this is that i have used qt webkit in my toy "remote browser isolation" project, the interesting aspect was that it allowed me to render pages into SVG (which is a vector format so no loss of quality) - then i diffed subsequent frames and compressed the diff, streamed that to client browser where i use wasm to decompress/patch and display, works reasonably well / fast and is lossless yet qt webkit is very much outdated with "a lot" of critical security bugs thus I would like to port my toy project to something based on chromium / ideally CEF - i was thinking using SKP directly (serialized skpicture) diff / compress the same way but try to use canvaskit on the browser side to "replay the SKP"
sure there could be a ton more use cases, this is just what i have in mind
muchas gracias