The New Sudoku Players' Forum

[1to9only]

I sent you a private message yesterday about your last version containing a bug report. I will check your new version later today.
Edit : new version contains the same bug.

I don't check the messages often, maybe every few weeks!!
As mentioned, the program has had limited testing - I usually fix things as I find them during use.
As for the bug you've reported, I'll try to reproduce and fix - it looks like something is not being reset...

I now think the 'bug' is just a feature of the program as I initially intended it to work.
If you start with JSB selected, the jigsaw solves, restart works, the jigsaw solves again, and no bug is observed.
If you start with JSB unselected, load the jigsaw, select JSB, and the jigsaw solves. Restarting appears to work but solving fails, because the puzzle was loaded/saved (for restart) as a jigsaw (no blocks), and some program options had changed in the meantime! A similar problem feature exists if you load the jigsaw and switch to another jigsaw layout - the givens remain, this is intentional.

An unselect/reselect JSB is needed - the program bug is that the reloaded jigsaw (JS) is now out of sync with the program displayed options (JSB). I'll fix this.
This will likely fix this specific issue with the restart of your jigsaw puzzle.

[1to9only]

An unselect/reselect JSB is needed - the program bug is that the reloaded jigsaw (JS) is now out of sync with the program displayed options (JSB). I'll fix this.

SudokuJigsawExplainer.jar - 1 Dec 2020 - PRE-RELEASE

[urhegyi]

An unselect/reselect JSB is needed - the program bug is that the reloaded jigsaw (JS) is now out of sync with the program displayed options (JSB). I'll fix this.

SudokuJigsawExplainer.jar - 1 Dec 2020 - PRE-RELEASE

I think a possible fix is to reset the JSB state to the initial state it was before importing the pattern on restart of program only as it's different, so that the user sees the correct state and can reselect JSB if needed. As you figured out I also think it is a sync problem, so the only thing to fix is making the current state visible in the program.

[1to9only]

It is 'working as designed' - it's unlikely I'll be making any changes to how it currently works, except for obvious bugs.

[urhegyi]

It is 'working as designed' - it's unlikely I'll be making any changes to how it currently works, except for obvious bugs.

I redownloaded it and it works fine!

[urhegyi]

examples with a specific pattern:

24167_254042.png (20.77 KiB) Viewed 716 times

[urhegyi]

solution:

irregular.png (18.68 KiB) Viewed 715 times

[urhegyi]

23747_413986.png (6.77 KiB) Viewed 715 times

[urhegyi]

solution:

irregular.png (18.81 KiB) Viewed 715 times

[urhegyi]

Is there a connection with jigsaw ? Are there some hidden constraints like in windoku? After solving a lot of them I haven't found the way they are generated yet.

[urhegyi]

23617_82091.png (7.25 KiB) Viewed 691 times

[urhegyi]

solution:

solution.png (19.04 KiB) Viewed 690 times

[Hajime]

This one has same layout (4 jigsaws) and is also point symmetric.
Generalized intersections between boxes and jigsaw (and with rows or columns) needed

Code: Select all

#1//B4,JSB
..1......3...1..............3.18.7.4....6....2.8.75.9..............4...8......1..
.....11.....1111..22..11...2222.1.3..22...33..2.4.3333...44..33..4444.....44.....


And the next one also needs some skyscrapers/2-string-kites

Code: Select all

#1//B4,JSB
4..1..9..5....3.1......5.4.....5....2.3...1.4....3.....1.4......4.5....7..7..8..1
.....11.....1111..22..11...2222.1.3..22...33..2.4.3333...44..33..4444.....44.....


All can be generated (and solved) with SiSeSuSo November version because it only has 1 grid.

[Hajime]

sensei_1_jsb.JPG (35.69 KiB) Viewed 667 times

Because the 4 jigsaws do not overlap in a Sensei-1 it produces nice puzzles.
Single only.

Code: Select all

#2//B4,JSB/H16,JSB
.......4..5..2......9...........6..4.....9.....7........8.6......2.7........3....
.....11.....1111..22..11...2222.1.3..22...33..2.4.3333...44..33..4444.....44.....
.....9........2..7.....5...6.....3.88.........9..3........6..2..1..8..5....5.....
.....11.....1111..22..11...2222.1.3..22...33..2.4.3333...44..33..4444.....44.....


[urhegyi]

Your second grid is solvable on it's own even without generalized intersection. (string-kite/skyscraper/turbotcrane/x-chain[4]) Is this coïncidence or designed so?