The New Sudoku Players' Forum

[m_b_metcalf]

Code: Select all

 . . . 1 2 3 . . .
 . . 4 . . . 5 . .
 . 2 . . . . . 6 .
 3 . . . . . . . 6
 1 . . . . . . . 7
 2 . . . . . . . 4
 . 8 . . . 4 . 7 .
 . . 9 . . . 8 . .
 . . . 5 7 8 . 3 .   moderate

 . . . 1 2 3 . . .
 . . 4 . . . 3 . .
 . 1 . . . . . 2 .
 5 . . . . . . . 2
 1 . . . . . . . 6
 7 . . . . . . . 4
 . 7 . . . 4 . 8 .
 . . 5 . . . 6 . .
 . . . 7 8 6 . 5 .   A nice Pearl

[Hajime]

found a diamond

Code: Select all

...123.....4...3...6.....2.5.......21.......67.......4.7...4.8...5...7.....736.5.   ED=2.6/2.6/2.6

[Hajime]

what to do?

Code: Select all

...742.....8...2...4.....6.8.......61.......54.......2.7...8.2...1...6.....136.5.   ED=20.0/7.9/2.6
...853.....8...5...4.....7.4.......27.......51.......6.2...9.6...9...2.....732.5.   ED=20.0/2.6/2.6

I use the SukakuExplainer from 1to9only .

[m_b_metcalf]

what to do?

Code: Select all

...742.....8...2...4.....6.8.......61.......54.......2.7...8.2...1...6.....136.5.   ED=20.0/7.9/2.6
...853.....8...5...4.....7.4.......27.......51.......6.2...9.6...9...2.....732.5.   ED=20.0/2.6/2.6

I use the SukakuExplainer from 1to9only .

With SukakuExplainer 1.17.7 I get, renumbered:

Code: Select all

...123.....4...3...2.....5.4.......56.......72.......3.1...4.3...6...5.....685.7.   ED=9.2/7.1/2.9 
...123.....1...2...4.....5.4.......65.......27.......8.6...9.8...9...6.....536.2.   ED=7.9/2.9/2.9 

Mike

[1to9only]

what to do?

SudokuMonster's SE rating:

Code: Select all

...742.....8...2...4.....6.8.......61.......54.......2.7...8.2...1...6.....136.5. ED=9.2/7.1/2.9
...853.....8...5...4.....7.4.......27.......51.......6.2...9.6...9...2.....732.5. ED=7.9/2.9/2.9


1to9only's SE rating:

Code: Select all

...742.....8...2...4.....6.8.......61.......54.......2.7...8.2...1...6.....136.5. ED=9.2/7.9/2.6
...853.....8...5...4.....7.4.......27.......51.......6.2...9.6...9...2.....732.5. ED=7.9/2.6/2.6


My version of SukakuExplainer appears to have issues with BUGs - I dont have time to look further into this right now, maybe later.

To rate the puzzles, start the GUI and use Options -> Solving techniques... to disable 'Bivalue Universal Grave' technique and exit GUI, then serate using the '-T' option, e.g.

Code: Select all

... serate puzzles.txt -X -T


[Hajime]

I use version 2021.4.7

[1to9only]

The Bivalue Universal Grave problem is fixed.
An updated SE (2023.3.10) can be downloaded from here: https://wormhole.app/bkpe8#Z3OQgCXD6oTaLROkln9PPA for the next 24 hours.
I'll update github and do a proper release some time later.

SukakuExplainer.jar
Size: 836,132 bytes
MD5: b7313a5de3f638893828094402e9c258

[yzfwsf]

Hi 1to9only:
How can I analyze sukaku by pressing F9 in a GUI program?

[1to9only]

How can I analyze sukaku by pressing F9 in a GUI program?

You'll need to turn off the sudoku checks (17 clues, at least 8 of digits 1-9), then F9 should work on sukakus.

The quick fix is to modify (comment out) the sudoku checks in analyse() function found towards the end of Solver.java:

Code: Select all

    public Hint analyse(Asker asker) {
        Grid copy = new Grid();
        grid.copyTo(copy);
        try {
            SingleHintAccumulator accu = new SingleHintAccumulator();
            try {
//              for (WarningHintProducer producer : validatorHintProducers)
//                  producer.getHints(grid, accu);
//              for (WarningHintProducer producer : warningHintProducers)
//                  producer.getHints(grid, accu);
                Analyser engine = new Analyser(this, asker);
                engine.getHints(grid, accu);
            } catch (InterruptedException willProbablyHappen) {}
            return accu.getHint();
        } finally {
            copy.copyTo(grid);
        }
    }



Rebuild SukakuExplainer.jar, F9 should now work on sukakus. I've tested this on a normal sukaku, not on sukaku-X.

[yzfwsf]

The quick fix is to modify (comment out) the sudoku checks in analyse() function found towards the end of Solver.java:

Code: Select all

    public Hint analyse(Asker asker) {
        Grid copy = new Grid();
        grid.copyTo(copy);
        try {
            SingleHintAccumulator accu = new SingleHintAccumulator();
            try {
//              for (WarningHintProducer producer : validatorHintProducers)
//                  producer.getHints(grid, accu);
//              for (WarningHintProducer producer : warningHintProducers)
//                  producer.getHints(grid, accu);
                Analyser engine = new Analyser(this, asker);
                engine.getHints(grid, accu);
            } catch (InterruptedException willProbablyHappen) {}
            return accu.getHint();
        } finally {
            copy.copyTo(grid);
        }
    }



Rebuild SukakuExplainer.jar, F9 should now work on sukakus. I've tested this on a normal sukaku, not on sukaku-X.

Thank you for providing the method, but I don't know Java and I haven't installed the compilation environment, so I won't bother with it. I originally thought that setting the startup parameters would be enough.

[1to9only]

You can download my latest SukakuExplainer (2025.10.8) here.

[m_b_metcalf]

Great. Many thanks.

Mike

[m_b_metcalf]

Great. Many thanks.

However, in batch mode I get:

Code: Select all

C:\sudoku\clean_s\clean_s>java -cp SukakuExplainer.jar diuf.sudoku.test.serate -X1 -G0 -U0 010023400000000000000000000230005006065000730400800029000000000000000000004290010    168   689395****160 1>>c:\sudoku\res.txt
Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: -1
        at diuf.sudoku.Grid$Custom.getCell(Unknown Source)
        at diuf.sudoku.solver.Solver.cancelBy(Unknown Source)
        at diuf.sudoku.solver.Solver.cancelPotentialValues(Unknown Source)
        at diuf.sudoku.solver.Solver.rebuildPotentialValues(Unknown Source)
        at diuf.sudoku.test.serate.main(Unknown Source)

('168 689395****160' is my codes).

[yzfwsf]

You can download my latest SukakuExplainer (2025.10.8) here.

Many thanks.

[1to9only]

Great. Many thanks.

However, in batch mode I get:

Code: Select all

C:\sudoku\clean_s\clean_s>java -cp SukakuExplainer.jar diuf.sudoku.test.serate -X1 -G0 -U0 010023400000000000000000000230005006065000730400800029000000000000000000004290010    168   689395****160 1>>c:\sudoku\res.txt
Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: -1
        at diuf.sudoku.Grid$Custom.getCell(Unknown Source)
        at diuf.sudoku.solver.Solver.cancelBy(Unknown Source)
        at diuf.sudoku.solver.Solver.cancelPotentialValues(Unknown Source)
        at diuf.sudoku.solver.Solver.rebuildPotentialValues(Unknown Source)
        at diuf.sudoku.test.serate.main(Unknown Source)

('168 689395****160' is my codes).

There are 2 versions of SukakuExplainer about, mine's (1to9only) latest version is 2025.10.8, and Tarek's (SudokuMonster) latest version is 1.18.1.

The serate usage (options) are different, the use of '-X1 -G0 -U0' indicate using SKE 1.18.1.

For the 2025.10.8 version, use: java.exe -cp SukakuExplainer.jar diuf.sudoku.test.serate --input=input.txt --output=output.txt -X
input.txt contains the sudoku
010023400000000000000000000230005006065000730400800029000000000000000000004290010
output.txt contains the sudoku + rating
010023400000000000000000000230005006065000730400800029000000000000000000004290010 ED=2.9/1.2/1.2

My SKE (2025.10.8) does not implement batch solving.