On Tue, Sep 09, 2025 at 08:04:20PM -0400, Thomas Dickey wrote: > On Tue, Sep 09, 2025 at 11:12:48PM +0000, Robin Haberkorn wrote: > > Hello, > > > > here are the mouse-related fixes I was discussing earlier. > > See commit messages for more details. > > thanks - will review and see where we're disagreeing :-)
In a quick check, I don't see any apparent regressions. That's for the ABI 6 (no more than 5 buttons); I'll see about the ABI 7 stuff later, to see how that compares - takes time...). However: in my earlier response, I was checking with Debian/12. I have a functional FreeBSD 14.2 (whenever 15 is released, I'll check on that...). Interestingly, the mouse results are different for the FreeBSD machine, which I can see by extra position events. This happens with/without your patch. Also, with/without the patch, I see dropped press or release events. That's using an ssh connection to the Debian machine where I have the ncurses test-program at hand. I don't see any instance where xterm puts a press and release event without a delay, but proving that would require maintaining a timestamp to show the intervals (seems like work, and would only be useful for debugging the library). -- Thomas E. Dickey <[email protected]> https://invisible-island.net
signature.asc
Description: PGP signature
