The New Sudoku Players' Forum

[denis_berthier]

.
From https://wiki.winehq.org/Download

Code: Select all

WINE-logo.png   Ubuntu - WineHQ binary packages for Ubuntu 18.04, 20.04, 22.04, and 22.10
Debian - WineHQ binary packages for Debian Bullseye and Bookworm
Fedora - WineHQ binary packages for Fedora 36 and 37
macOS - WineHQ binary packages for macOS 10.8 through 10.14

Current version of MacOS 13.2.1
All is said.

[ghfick]

You had posted this information earlier. It turns out that there are other tools like Wine that may provide a solution.

Again, as I said earlier, no one with Wine is currently acting as a maintainer for the Mac versions. Almost all of the current Wine development is for Linux.

Given that Linux and Unix and Mac are so similar, it is not out of the question for a Mac resolution

[denis_berthier]

.
Adding insults to hypothetical possibilities in an alternative world will not change the facts. I reposted the same real facts for every Mac user to know.
As I also already said before, MacOS has added security layers to Unix and this is a big problem for any software like Wine.

Indeed I have used Hodoku a reasonable number of times and I would have liked to try yzf_sudoku, be it only to see the evolution.

[ghfick]

A kind Sudoku friend has reminded me by PM of two options for Mac users. The first option is 'Boot Camp'. The second option is 'Parallels Desktop'.
I believe you do need an MSWindows licence for either of these options. But I believe an older MSWindows licence should suffice. Even an MSWindows XP licence should work as I understand that YZF_Sudoku will run on a machine with MSWindows XP.
Another Sudoku friend [ with way more programming background than me ] has begun to explore this issue too.

[ghfick]

Yet another friend notes that Mac users might want to wait before 'upgrading' to Ventura [ 13.2.1 ]. Apparently, many Mac users report major headaches with 13.2.1.

jason@enjoysudoku.com sends you an email when you get a PM but jason@enjoysudoku.com appears to be a bot.

the person with forum handle JasonLion appears to be inactive.

[yzfwsf]

Fixed some textual or grammatical errors as suggested by ghfick, and added the ability to list all minimal puzzles contained in non-minimal puzzles to a file. Of course, puzzles with a large number of redundant clues are slower to process, because the program uses the most primitive combination algorithm. If you want to handle puzzles with a large number of redundant clues, it is recommended to use a console program like gridchecker.

release 626V2
https://toi.teracloud.jp/share/11c1080bd2a84925
Extracted Code：enjoysudoku

[yzfwsf]

Release 626V3 on google drive

If you need to localize the gui, please modify the language file "Custom.lang", not the default language file "Default.lang". Please refer to the default language file when you modify the local language file, do not modify the text after the menu shortcut "\t", and do not remove the "&" from the menu.

[marek stefanik]

Very nice new features!

The RTs seem a little bugged, though. As I mentioned elsewhere, the eliminations writen below the grid are the ones of the TH itself, whether they have already been eliminated or not.
The new eliminations are shown in the grid but missing from the text.

In some cases, there is more than a visual bug, see the following grid:

Code: Select all

,--------------------,---------------------,--------------------,
| 1      2358  2379  | 4      359    6     | 27     3589  359   |
| 4      2356  23679 | 1      8      359   | 27     359   3569  |
| 68     3568  369   | 359    2      7     | 48     1     34569 |
:--------------------+---------------------+--------------------:
| 2359   236   236   | 359    4      1     | 359    7     8     |
| 359    4     1     | 7      359    8     | 6      359   2     |
| 359    7     8     | 2      6      359   | 1      4     359   |
:--------------------+---------------------+--------------------:
| 23678  168   4     | 35689  13579  23    | 3589   3589  1359  |
| 3678   9     5     | 368    137    34    | 348    2     134   |
| 238    18    23    | 3589   1359   23459 | 34589  6     7     |
'--------------------'---------------------'--------------------'

The solver claims to find the following step, but there are no eliminations:
Triplet Oddagon(RT) + Triplet ERI: 359r34c47,r1c59,r2c68,r5c58,r6c69 + 359b4 => r3c7<>

Autosolve from the above grid just causes the solver to crash if TH is checked for early enough (I assume it's because it gets stuck in a loop, finding the same step over and over, until it reaches a limit on the length of the path).

Added: The RTs visualisations look really nice and they should generally be quite clear, but they don't show up in krakens.
This appears to be a problem with TH relations in general (see the steps your solver finds in the following grid states).
(This could also be an intentional design choice, in which case I would still like to see them when viewing the specific branch where they're used and see 'with Triplet Oddagon' in the name of the chain.)

Hidden Text: Show

Code: Select all

,---------------------------------,---------------------------------,--------------------------------,
| 12356789   12356789   12356789  | 123        56789      56789     | 4         56789      12356789  |
| 23456789   123456789  123456789 | 456789     123        456789    | 12356789  123        12356789  |
| 23456789   123456789  123456789 | 456789     456789     123       | 12356789  12356789   123       |
:---------------------------------+---------------------------------+--------------------------------:
| 23456789   123456789  123456789 | 23         456789     456789    | 123       456789     456789    |
| 23456789   123456789  123456789 | 456789     123        456789    | 56789     456789     123       |
| 23456789   123456789  123456789 | 456789     456789     123       | 56789     123        456789    |
:---------------------------------+---------------------------------+--------------------------------:
| 123456789  123456789  123456789 | 123456789  123456789  123456789 | 12356789  123456789  123456789 |
| 123456789  123456789  123456789 | 123456789  123456789  123456789 | 12356789  123456789  123456789 |
| 23456789   123456789  123456789 | 23456789   123456789  123456789 | 12356789  23456789   23456789  |
'---------------------------------'---------------------------------'--------------------------------'


Code: Select all

,---------------------------------,---------------------------------,---------------------------------,
| 123456789  123456789  123456789 | 123        123456789  123456789 | 123        456789     456789    |
| 123456789  123456789  123456789 | 123456789  1234       123456789 | 456789     123        456789    |
| 123456789  123456789  123456789 | 123456789  123456789  123       | 456789     456789     123       |
:---------------------------------+---------------------------------+---------------------------------:
| 123456789  12346789   12346789  | 123        456789     456789    | 123        123456789  123456789 |
| 123456789  12346789   12346789  | 456789     123        456789    | 123456789  123456789  123       |
| 123456789  123456789  123456789 | 456789     46789      123       | 123456789  1235       123456789 |
:---------------------------------+---------------------------------+---------------------------------:
| 12346789   123456789  123456789 | 123456789  45         123456789 | 123456789  12346789   123456789 |
| 123456789  123456789  123456789 | 123456789  123456789  123456789 | 123456789  123456789  123456789 |
| 123456789  12346789   12346789  | 123456789  123456789  123456789 | 12346789   123456789  12346789  |
'---------------------------------'---------------------------------'---------------------------------'

Marek

[yzfwsf]

Release 626V4 on google drive
Fix the mentioned bugs and visualize the krakens embedded in Triplet Oddagon.

Autosolve from the above grid just causes the solver to crash if TH is checked for early enough (I assume it's because it gets stuck in a loop, finding the same step over and over, until it reaches a limit on the length of the path).
Marek

You are right, the limit is 300.

BTW:AIC and the Kraken Chain Techniques implement the strong links formed by Almost Fish and FW,
now they are processed into X-Links, no good visual representation has been thought of for the
time being, so they are not mentioned in the text description either.

[shye]

there is an MSLS in this puzzle from hajime that the solver doesnt seem to pick up

Code: Select all

,--------------,--------------------,-----------------,
| 4    1   2   |#357   9      368-5 | 78  67-5 #567   |
| 8    5   9   |#247  #247    6-2   | 3   1    #2467  |
| 3    7   6   | 1     248    258   | 28  59    2459  |
:--------------+--------------------+-----------------:
| 259  39  35  | 6     238    7     | 1   4    #38    |
| 16   8   347 | 49-3  134    139   | 5   2    #367   |
| 126  46  347 | 245-3 12348  12358 | 9   367  #3678  |
:--------------+--------------------+-----------------:
| 59   39  8   | 279-3 6      239   | 4   3579  1     |
| 569  46  345 | 8     137    139   | 27  3579  259-37|
| 7    2   1   |#39    5      4     | 6   8    #39    |
'--------------'--------------------'-----------------'

10 cells (r129c4, r2c5, r124569c9)
10 links (5r1, 24r2, 9r9, 3c4, 3678c9, 7b2)
all candidates covered, links become truths

Xsudo Input: Show

10 Truths = {129N4 2N5 124569N9}
10 Links = {5r1 24r2 9r9 3c4 3678c9 7b2}
8 Eliminations --> r567c4<>3, r1c68<>5, r8c9<>37, r2c6<>2,

[yzfwsf]

Since there are so many cases of Cells and Links combinations in MSLS, there is no way to traverse all of them as time permits. Thanks for the sample, after modifying the code, the solver can find the MSLS you described.
Release a small modified version on google drive, V627

[shye]

i figured that might be the case, but thought it would be worth sending your way regardless

[stan ellison]

Hi. I've just downloaded YZF-Sudoku and it runs fine on my Linux PC using Wine. I like the simple basic appearance of the GUI however I have a problem. The given numbers are black but the those I enter are blue.

Basic Sudoku, I'm not talking Killer here, is a pattern puzzle rather than a number puzzle. For me different colours introduce a pattern element that interferes with solving.

Is there a way to make entered numbers look like given numbers?

If not is there a software package that does this?

[yzfwsf]

You can modify the configuration yourself to achieve what you need. Configuration modification is entered from the Edit menu, and then click Preferences...

[stan ellison]

Thanks for the reply. I've just tried to access preferences but nothing happens. Perhaps it's just something to do with Wine. It seems that other click down options work, just not that one. Thanks anyway.