Adjust OS X MouseMove handling to use live cursor position. - #9123
Conversation
|
@brett19 Hey, looks like you have some conflicts. Would you mind resolving those? |
2608289 to
7347541
Compare
|
Hey @nbolton, I've updated the PR. Is there any chance there is someone around who has some more experience dealing with OS X who might be able to verify the logic of the change? It certainly fixes the issue I was encountering on my local system, but I'm not entirely convinced that the reasoning behind why it fixes it is sound. |
7347541 to
4417878
Compare
4417878 to
9e8cabf
Compare
|
I've rebased the PR onto the latest master along with some cleanup to the original. It looks like the CI check will need approval due to coming from a fork. |
nbolton
left a comment
nbolton
left a comment
There was a problem hiding this comment.
LGTM, I tried reproducing the drift but not sure if I'm seeing things or whether the mouse cursor was drifting. In any case it doesn't seem to have any negative effects!
9e8cabf to
1611ef3
Compare
nbolton
left a comment
nbolton
left a comment
There was a problem hiding this comment.
Oh, actually, could you change your commit message to the conventional commit format?
sithlord48
left a comment
sithlord48
left a comment
There was a problem hiding this comment.
Other then the lint and commit message should be fix: .... looks go to me
1611ef3 to
d973cab
Compare
|
I've updated the commit message and linting so it should be good now @nbolton |
d973cab to
81e7bf1
Compare
|
@nbolton When can I expect this change to make it into an upstream Synergy release? |
Downstream ;-) I've added this to the backports plan for the next update. We are aiming for the coming weeks. |
|
Hey @nbolton, any updates on this? I haven't seen anything to imply this is included in the latest changelogs. |
Decided to put it in the next beta: symless/synergy#190 |
4 participants
I wrote this patch many years ago so the comments / PR title are based on my recollection of why this is necessary. The demonstration of the original issue consisted of an OS X Server connected to a Windows 11 Client, you could then center your cursor on the remote system (Windows) and start performing perfect(ish) circles physically with the mouse. What you would observe is that on the remote system, while the circles would be circular-ish, your cursor would slowly drift away towards one side of the screen, demonstrating that the mouse position information was not being collected accurately.