HACKER Q&A
📣 eimrine

Is there a browser plugin which replaces webpage with its screenshot?


I am being exhausted from observing interactivities of some webpages which are typically being fired while scrolling the webpage. Typically the interactivities are some pesky features like demonstrating me a video or some banner which wants some actions from me, typically expecting me to click "X" to get rid of the pesky demonstration. All I want from webpages is plain text, but I can agree with rich text - different fonts, with colors, with shadows, with stupid pseudoelements, you know what I mean. I have discovered some relief from reading links from HN as screenshots and all I want is a little bit optimization of obtaining screenshot instead of the webpage in a browser. Any plugins for any browser capable to do this?

  👤 HenryBemis Accepted Answer ✓
Not exactly screenshot, but for Firefox on Windows(10Pro) I have installed the "Open in Reader View" add-on. So when I see an article I want to open in a new tab, I right click on the link, on the menu that appears I choose the "Open in Reader View" and it opens a new tab with the article/page in Reader View. It is the cleanest way I access pages.
👤 keepamovin
This is a fascinating idea.

I think you could hack a way to do this with BrowserBox: which, funnily enough, already "replaces web pages with their screenshots" (we stream the screenshots or screencast from the headless browser and serve it).

However to get what you really need, I see some difficulties:

- you could scroll the entire page first, stitch the screenshots together and then not update any further, but this breaks in some weird edge cases:

  - dynamically loaded content which changes the scroll height as you move down, will not be loaded or displayed in this method reliably. 
  - this height change will then throw off any clicks that occur below it
  - this will still not prevent banners and the like from occurring
- you could use the --print-to-pdf command line flag for Chrom/Edge/Brav/ium browsers. Not sure if hyperlinks work from these. But you could set up a basic proxy to take all HTTP and HTTPs requests and pass them through your headless "to pdf" render process if they do.

- if hyperlinks don't work in these pdfs you could use Playwright (or CDP if you're braver), to evaluate "document.links" property (see: https://developer.mozilla.org/en-US/docs/Web/API/Document/li...), and just create a simple "list of links" HTML page templated from these and append that to the bottom of your PDF document (by using some command line tool in collab with your to-pdf proxy).

Anyway, fascinating idea, good luck with it! :)

👤 tedmiston
It's not exactly what you're asking for, but a means to a similar idea... I setup a "site search" in Chrome for archive.is so that I can just type `a ` before the URL in the location bar and have it auto-complete the archive.is URL for a page.

    Search engine: archive.is
    Shortcut: a
    URL with %s in place of query: https://archive.is/?run=1&url=%s&
This works to effectively make 99% of pages static.

Based on past experience, I believe the archiver allows JS during load but is disabling JS during runtime (but I haven't actually verified that).

There is also an unofficial Chrome extension with similar functionality, but it is closed source and requires entire browser history permission for some reason, so...

👤 huehehue
Other than reader mode, you can try disabling javascript. Lynx will get you plain text, but usability will vary from site to site.

If you truly want a screenshot, you can write a program to spin up a headless browser / selenium, load the page, and save the result somewhere. People do this all the time as part of UI and integration tests and tooling is decent enough.