This is a request for implementing getDisplayMedia, the Screen Capture API that is an extension to the Media Capture API. This allows a display surface to be used as a source for a media stream.
Examples of display surfaces that may be supported include: monitor, window, application (entire collection of windows for an application), or browser (single browser window). Having the ability to capture and share a monitor display would be a great start.
The specification is here: https://w3c.github.io/mediacapture-screen-share/
Microsoft Edge (EdgeHTML 17) added support for the Screen Capture API as part of the Windows 10 April 2018 update: https://blogs.windows.com/msedgedev/2018/05/02/bringing-screen-capture-to-microsoft-edge-media-capture-api/
Google Chrome and Mozilla Firefox have already had support for non-standard screen capture APIs for years. They have bugs tracking migrations to the standard getDisplayMedia API (Chrome: https://bugs.chromium.org/p/chromium/issues/detail?id=326740, Firefox: https://bugzilla.mozilla.org/show_bug.cgi?id=1321221).
getDisplayMedia is implemented (for displays only, not window, or application) on macOS behind the ScreenCapture runtime flag (off by default). See bugs 181070, 181291, and 181333.
That's great to hear! How do we go about enabling this ScreenCapture runtime flag? What (minimum) version of Safari is required?
There is no UI to enable this setting yet because the (complex) user prompt for getDisplayMedia hasn't been added to Safari yet.
Are there any plans to extend the implementation to include window or application? If possible, it would also be really valuable to get any sort of estimate about when the UI work for Safari is planned.
(In reply to Eric Carlson from comment #1)
> getDisplayMedia is implemented (for displays only, not window, or
> application) on macOS behind the ScreenCapture runtime flag (off by
> default). See bugs 181070, 181291, and 181333.
How to enable this `ScreenCapture` runtime flag to prepare code for this implementation and see how it works for now?
(In reply to Adam Szmyd from comment #5)
> How to enable this `ScreenCapture` runtime flag to prepare code for this
> implementation and see how it works for now?
It can only be enabled at build time currently.