3 ms·
My guess is that it's to optimize for ereader users, as the colored pages would be hard to read on eInk. It probably saves a LOT of bandwidth while improving th
by schwartzworld 3y ago
My guess is that it's to optimize for ereader users, as the colored pages would be hard to read on eInk. It probably saves a LOT of bandwidth while improving the legibility.
You could download the individual pages as images directly from the google books page, but the list is virtualized, so there are only ever a few pages being rendered at one time, which is why you can't just crtl-P the whole thing.
Looking at the URLs of the images, they seem to be sequentially numbered. I tried using CURL to fetch one of the urls, and it fetched the full color version as a PNG file. This means it should be trivial to write a script that iterates through the page numbers and downloads each page. Any programming language could do it, even BASH. These images could be combined into a PDF using the tool of your choice.
- jsx91 3y agoThanks for a great idea. If someone already has such a script (bash/zsh, AppleScript, ...) and post it here, I would be very appreciated.
- schwartzworld 3y agohttps://www.cyberciti.biz/faq/bash-for-loop/ https://www.cyberciti.biz/faq/bash-for-loop/