[🚀 Feature]: Full page screenshots for Chrome

[lwassermann]

Feature and motivation

In the Ruby Selenium Driver, it is possible to provide the full_page-argument to save_screenshot (documented here).

That option only work when using Selenium with Firefox. Not Chrome or Edge.

4 years ago, there was already a discussion to add this feature: #8705
Unfortunately, that got closed as a duplicate of BiDi-work. That work referenced screenshots, but did not actually include that feature. And it is merged for some time now.

From other issues, I'm not quite sure whether that limitation is only in the ruby-bindings, or shared with other languages.

Or maybe I'm holding it wrong? If so, I'd be greatful if you could add the correct way to take screenshots to the method documentation 🙏

Usage example

I'm currently adding visual regression tests, creating screenshots of our website or something like Storybook. So I have some pages that are very long, and others that are less than a normal screen height. Taking long pictures with a lot of white-space of everything seems wasteful.

For now, we'll use Firefox. But since this was already closed as done, but actually isn't, I thought I'd re-raise this 😇

[github-actions]

@lwassermann, thank you for creating this issue. We will troubleshoot it as soon as we can.

---

Info for maintainers

Triage this issue by using labels.

If information is missing, add a helpful comment and then I-issue-template label.

If the issue is a question, add the I-question label.

If the issue is valid but there is no time to troubleshoot it, consider adding the help wanted label.

If the issue requires changes or fixes from an external project (e.g., ChromeDriver, GeckoDriver, MSEdgeDriver, W3C), add the applicable G-* label, and it will provide the correct link and auto-close the issue.

After troubleshooting the issue, please add the R-awaiting answer label.

Thank you!

[titusfortner]

1. There is a method for firefox because Mozilla created a specific endpoint for it that was easy to just use.
2. There is a way to do it currently in Chrome that uses Chrome Devtools Protocol, but we are not adding wrappers to CDP because we are looking to replace it with the BiDi spec.

This is a good request to add to a High Level API for BiDi, though, and we are no longer tracking that in another issue, so I'll leave this open.

@pujagani what do you think is the right API for this? driver.page.screenshot(params)? (does page make sense in this context vs the script context we're using with other methods? Let's discuss.

[pujagani]

Since "captureScreenshot" command comes under browsing context in the spec https://w3c.github.io/webdriver-bidi/#command-browsingContext-captureScreenshot. I think page looks good to me.
What do others think? @diemol @p0deje

[diemol]

page sounds good. But how do we bring together the current screenshot methods and this new one?

[pujagani]

Good point, Diego. I missed this. How do you suggest we do this?

[diemol]

Deprecate the existing ones and move them to page? Except the ones that take a screenshot of an element.

Maybe it is not necessary and we can leave them as it is?

[pujagani]

I was thinking in a similar direction. What do others think? @titusfortner @harsha509 @p0deje @nvborisenko

[nvborisenko]

Who is page? If it is new entity then it should include page.title as well... and page.url.

[pujagani]

Suggested API is "driver.page.screenshot(params)". Since that is the pattern similar to the other high-level BiDi APIs we are creating for use-cases that are not addressed by the WebDriver W3C spec (example: Log, Script). The rest of the existing commands for title and URL will be as it is and ported to use BiDi in the background Selenium 5 onwards.