| Weird case of click drag not doing the expected thing |
Tilps
Kwon-Tom Obsessive
Puzzles: 7116
Best Total: 18m 37s | Posted - 2025.09.06 05:07:14
I thought maybe this was a problem with variations puzzles, but I've found it exists in both so I'm posting here.
My understanding of the mouse logic is the following.
Click cycles (empty, filled, excluded)
Shift-click reverse cycles (empty, excluded, filled)
In either case, following up with a drag continues to apply whatever the last applied state was. So if click filled, then drag fills, if it excluded, more excluded are set, etc etc.
However, if I am very careful (for normal at least, seems easier to trigger in variations), there seems to be spots where a click doesn't immediately cycle (maybe only the outer areas near intersections on the very edge of the puzzle for normal) then if you drag, the last applied state is applied despite it coming from a completely different click occurrence.
I think that the mouse release event should be clearing out the prior state (and the dead zones fixed, but the clearing out seems most important.)
As an example, I double clicked the X in the middle of this puzzle, then carefully click dragged to create the top X. (Behaviour here is different to real puzzles, the border area means the click drag has to be off the intersection...)
In case it matters, using latest stable Chrome.
Last edited by Tilps - 2025.09.06 05:09:05 |
foilman
Kwon-Tom Admin
Puzzles: 3988
Best Total: 24m 6s | Posted - 2025.09.08 06:54:57
Interesting, thank you - I shall take a look into this when I have a chance. |
foilman
Kwon-Tom Admin
Puzzles: 3988
Best Total: 24m 6s | Posted - 2025.11.10 14:16:56
It took a while, but I finally got a chance to take a look at this today, and it's now fixed and working the way it should have originally. |
Tilps
Kwon-Tom Obsessive
Puzzles: 7116
Best Total: 18m 37s | Posted - 2025.11.15 21:07:55
Thanks for that, definitely can feel the improved experience on variations.
In case it is helpful, I think I've identified that it seems like the max distance to click near a line in a variation depends on how long that specific line is. This feels counter intuitive to me - and in some puzzles with short edges and long edges it means there is a dramatic difference in accuracy required near some edges compared to others.
I wonder if the implementation is somehow related to trying to support users zooming? (I use 150% zoom for variations, so I definitely don't want that to break) - but instead of scaling the size with the overall draw scale, its scaling each edge individually?
Or maybe it is intentional for reasons that aren't immediately obvious to me... |
foilman
Kwon-Tom Admin
Puzzles: 3988
Best Total: 24m 6s | Posted - 2025.11.18 09:06:17
I've tweaked the code on the variations to hopefully make it a little easier to get the line you want, let me know if you spot any weird issues!
Last edited by foilman - 2025.11.18 09:06:34 |
Tilps
Kwon-Tom Obsessive
Puzzles: 7116
Best Total: 18m 37s | Posted - 2025.11.19 01:13:19
Looks good to me (hopefully no one else thinks this is a regression...)
I did notice that clicking on the intersection dot still seems to do nothing (and replied as such on the variations forum), but that is a much smaller issue in my mind. |
