now detects the entire screen real-estate instead of a single screen #1854
Conversation
…hanges: ` Fix logging, scope mod to taskbar process, and optimize polling loop - Fixed incorrect use of Wh_Log by embedding format specifiers directly in the first argument - Restricted mod initialization to the explorer.exe process that owns the taskbar using EnumWindows and a proper CALLBACK - Replaced lambda with a static BOOL CALLBACK function to match WNDENUMPROC signature - Improved performance by: - Increasing polling interval to 100ms when idle - Skipping edge checks when mouse hasn't moved `
Co-authored-by: Michael Maltsev <[email protected]>
|
This means it will stop working on setups like this:
Making it work perfectly can be tricky, but perhaps it can at least be an option for either per-monitor edges or global edges. What do you think? |
|
Is there a way to detect this orientation(setup) like this? |
|
I don't think there's an easy way to detect it. But even if it was, I can imagine that some users might prefer to have these areas only:
And some might prefer these (e.g. if they mostly use monitor 1):
The ideal is probably to have this, but it can be tricky to support any monitor arrangement this way:
|
|
There can be three modes Global bounding‑rectangle edges (the leftmost/rightmost X of the entire virtual screen) Outline edges (only those monitor‑edges that actually lie on the outside of the union of all monitors) what do you think |
|
I think outline edges is the preferable behavior in most case (my third screenshot), but you'll have to iterate over all monitors and understand which edge is outline and which isn't. There can also be cases in which only part of an edge is outline:
|
|
Yes and there can be a function for detecting if there's a change in the orientation of the monitors. |
so no unnecessary triggers occus at the top and bottom edge.
|
I see that you pushed an update, let me know if it's ready for review/merge. |
2 participants
No description provided.