fiveham Sudoku_Solver issues

fiveham / Sudoku_Solver

Solve sudoku puzzles by representing the puzzle as a bipartite graph of truth-claims about cells' values and rule-statements that have exactly one true neighbor and removing edges as information is added

MIT License

0 stars 0 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

WhatIf searches through its Puzzle's entire population of Facts too often

#27 fiveham opened 7 years ago
0
BackedSet's size() method is O(universe size)

#26 fiveham opened 7 years ago
0
IndexValue's nature as a mock-enum should be made more clear in code

#25 fiveham opened 7 years ago
0
IndexValue should be named Coordinate

#24 fiveham opened 7 years ago
0
Techniques using constant Universe from Puzzle have less efficient BackedSets

#23 fiveham opened 7 years ago
0
Technique's extension of Function<Sudoku,Technique> doesn't guarantee that apply()'s return type is the type implementing Technique

#22 fiveham opened 7 years ago
0
Solver's private constructor tests each time whether the threadgroup needs its root Solver set

#21 fiveham closed 7 years ago
1
BackedSet's iterator's remove method can be called multiple times between calls to next()

#20 fiveham closed 7 years ago
1
Traversing a BackedSet is O(universe size), which is inefficient for small sets

#19 fiveham closed 7 years ago
2
Previous exploration of a Fact should factor into the Fact's popularity rating

#18 fiveham closed 8 years ago
1
ColorChain doesn't detect Sledgehammer scenarios until it accounts for all possible solutions of the network

#17 fiveham opened 8 years ago
1
Looking for the smallest Fact by sorting the whole list may be inefficient

#16 fiveham closed 7 years ago
1
WhatIf's lookup for its smallest affected Fact is inefficient

#15 fiveham opened 8 years ago
0
WhatIf can assume contradictory things.

#14 fiveham closed 8 years ago
2
WhatIf should be an inner class of Logic, which should be an inner class of ColorChain

#13 fiveham closed 8 years ago
1
Sets.massIntersectionCollector() can't take advantage of BackedSet's bulk-op efficiency

#12 fiveham closed 8 years ago
1
Size-2 Sledgehammer analysis is a particular case of XYChain analysis

#11 fiveham closed 8 years ago
4
Child Times should add themselves as children of their parent during construction

#10 fiveham closed 8 years ago
0
Solver's call to wait(100) is arbitrary, should call wait() (no args) instead

#9 fiveham opened 8 years ago
0
Copying Sets in Sledgehammer.exploreSourceCombos() less efficient than using internal BigInteger masks

#8 fiveham closed 8 years ago
3
PuzzleVizApp.treeSize() is unnecessary and hard to read

#7 fiveham closed 8 years ago
1
NodeSet.visible() duplicates code from ConnectedComponent

#6 fiveham opened 8 years ago
0
Claim-falsification code is duplicated

#5 fiveham closed 8 years ago
1
Xor-chain pairs (in ColorChain) are tested by brute force

#4 fiveham closed 8 years ago
1
Support for different charsets

#3 fiveham closed 8 years ago
1
Wrap.wrap() should wrap using arbitrary WrapVertexs

#2 fiveham closed 8 years ago
1
Formalize the need for FalsifiedTime args in retrofitted Collection methods in NodeSet

#1 fiveham closed 8 years ago
1