Summary: | Provide device identification mechanism | ||
---|---|---|---|
Product: | Wayland | Reporter: | Peter Hutterer <peter.hutterer> |
Component: | libinput | Assignee: | Wayland bug list <wayland-bugs> |
Status: | RESOLVED WONTFIX | QA Contact: | |
Severity: | enhancement | ||
Priority: | medium | CC: | bugzilla, peter.hutterer |
Version: | unspecified | ||
Hardware: | Other | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Description
Peter Hutterer
2014-10-28 00:55:31 UTC
Does libinput "split" kernel devices? For example, would an external ThinkPad keyboard, or combined keyboard/touchpad device be one or two libinput devices? If it doesn't split them, a bitfield would be better. no, libinput just takes the kernel device as it is and sets the matching capabilities. lesson learned from X where splitting devices caused a couple of drawbacks, so the capabilities-based approach is much better. Closing this as WONTFIX. Every situation we have come up with so far was better solved without this, specifically by checking what sets of configuration options or keys/buttons are available. We also provide a handle to the udev device which let's the caller query information about it and then make decisions based on the context needed. libinput sits too low and has too little context to make the right decision here, any tagging system is likely going to fall short and require the caller to have a separate tagging system on top of it. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.