Merge pull request #16 from dmbates/sortedtree

Sort siblings in the tree representation by tree size

NEWS.md

@@ -1,4 +1,10 @@
-## MixedModels v0.2.0 Release Notes
+## Wordlegames v0.3.0 Release Notes
+
+- Sort siblings in tree by size of tree rooted at a node.
+- Drop the index from the output of print_node
+- add targets and validguesses for Greek game from https://github.com/gvarnavi/wordle-greek
+
+## Wordlegames v0.2.0 Release Notes
 
 - Bugfix release Issue #8, *Incorrect scoring when characters are repeated in guess*
 - Fixing this led to slightly better mean number of guesses to solution.

Project.toml

@@ -1,7 +1,7 @@
 name = "Wordlegames"
 uuid = "1cb69566-e1cf-455f-a587-fd79a2e00f5a"
 authors = ["Douglas Bates <[email protected]> and contributors"]
-version = "0.2.0"
+version = "0.3.0"
 
 [deps]
 AbstractTrees = "1520ce14-60c1-5f80-bbc7-55ef81b5835c"

README.md

@@ -351,38 +351,38 @@ For illustration, consider just a portion of the tree of Wordle games using the
 Games with targets `["super", "hobby", "mince", "goner", "watch"]` are shown above.
 They can be combined into a tree as
 
-```jl
-julia> print_tree(tree(wordle, ["super", "hobby", "mince", "goner", "watch"]), maxdepth=8)
-missing, raise, 1535, 2315, 5.87791, 61.0009
-├─ 🟫🟫🟨🟫🟩, binge, 198, 25, 3.28386, 3.64
-│  └─ 🟫🟩🟩🟫🟩, mince, 1234, 2, 1.0, 1.0
-├─ 🟨🟫🟫🟫🟨, deter, 546, 102, 4.37007, 9.23529
-│  └─ 🟫🟫🟫🟩🟩, cower, 454, 26, 2.74682, 5.23077
-│     └─ 🟫🟩🟫🟩🟩, hover, 999, 9, 1.65774, 3.44444
-│        └─ 🟫🟩🟫🟩🟩, joker, 1059, 5, 1.37095, 2.2
-│           └─ 🟫🟩🟫🟩🟩, boxer, 258, 3, 0.918296, 1.66667
-│              └─ 🟫🟩🟫🟩🟩, foyer, 800, 2, 1.0, 1.0
-│                 └─ 🟫🟩🟫🟩🟩, goner, 884, 1, -0.0, 1.0
-├─ 🟫🟫🟫🟫🟫, mulch, 1275, 168, 5.21165, 6.85714
-│  └─ 🟫🟫🟫🟫🟨, whoop, 2262, 6, 2.58496, 1.0
-│     └─ 🟫🟨🟨🟫🟫, hobby, 985, 1, -0.0, 1.0
-├─ 🟨🟫🟫🟨🟨, sheer, 1720, 18, 3.28104, 2.11111
-│  └─ 🟩🟫🟫🟩🟩, sober, 1835, 4, 1.5, 1.5
-│     └─ 🟩🟫🟫🟩🟩, super, 1969, 2, 1.0, 1.0
-└─ 🟫🟩🟫🟫🟫, tangy, 2012, 91, 4.03061, 7.48352
-   └─ 🟨🟩🟫🟫🟫, caput, 334, 13, 2.4997, 2.84615
-      └─ 🟨🟩🟫🟫🟨, batch, 160, 5, 0.721928, 3.4
-         └─ 🟫🟩🟩🟩🟩, hatch, 959, 4, 0.811278, 2.5
-            └─ 🟫🟩🟩🟩🟩, latch, 1102, 3, 0.918296, 1.66667
-               └─ 🟫🟩🟩🟩🟩, match, 1206, 2, 1.0, 1.0
-                  └─ 🟫🟩🟩🟩🟩, watch, 2233, 1, -0.0, 1.0
+```julia
+julia> print_tree(tree(wordle, ["super","hobby","mince","goner","watch"]); maxdepth=8)
+missing, raise, 2315, 5.87791, 61.0009
+├─ 🟨🟫🟫🟫🟨, deter, 102, 4.37007, 9.23529
+│  └─ 🟫🟫🟫🟩🟩, cower, 26, 2.74682, 5.23077
+│     └─ 🟫🟩🟫🟩🟩, hover, 9, 1.65774, 3.44444
+│        └─ 🟫🟩🟫🟩🟩, joker, 5, 1.37095, 2.2
+│           └─ 🟫🟩🟫🟩🟩, boxer, 3, 0.918296, 1.66667
+│              └─ 🟫🟩🟫🟩🟩, foyer, 2, 1.0, 1.0
+│                 └─ 🟫🟩🟫🟩🟩, goner, 1, -0.0, 1.0
+├─ 🟫🟩🟫🟫🟫, tangy, 91, 4.03061, 7.48352
+│  └─ 🟨🟩🟫🟫🟫, caput, 13, 2.4997, 2.84615
+│     └─ 🟨🟩🟫🟫🟨, batch, 5, 0.721928, 3.4
+│        └─ 🟫🟩🟩🟩🟩, hatch, 4, 0.811278, 2.5
+│           └─ 🟫🟩🟩🟩🟩, latch, 3, 0.918296, 1.66667
+│              └─ 🟫🟩🟩🟩🟩, match, 2, 1.0, 1.0
+│                 └─ 🟫🟩🟩🟩🟩, watch, 1, -0.0, 1.0
+├─ 🟫🟫🟫🟫🟫, mulch, 168, 5.21165, 6.85714
+│  └─ 🟫🟫🟫🟫🟨, whoop, 6, 2.58496, 1.0
+│     └─ 🟫🟨🟨🟫🟫, hobby, 1, -0.0, 1.0
+├─ 🟨🟫🟫🟨🟨, sheer, 18, 3.28104, 2.11111
+│  └─ 🟩🟫🟫🟩🟩, sober, 4, 1.5, 1.5
+│     └─ 🟩🟫🟫🟩🟩, super, 2, 1.0, 1.0
+└─ 🟫🟫🟨🟫🟩, binge, 25, 3.28386, 3.64
+   └─ 🟫🟩🟩🟫🟩, mince, 2, 1.0, 1.0
 ```
 
 Although this is not a particularly interesting tree, it serves to illustrate some of the properties.
 The first node, called the "root" node, is the first guess in all the games.
-The guess is "raise" at index 1535 with pool size 2315, an entropy of 5.88 and an expected pool size of 61.00 after scoring.
+The guess is "raise" with a pool size of 2315, an entropy of 5.88 and an expected pool size of 61.00 after scoring.
 
-If the score for "raise" is `🟫🟫🟨🟫🟩`, the next guess will be "binge", with the characteristics shown.
+If the score for "raise" is `🟨🟫🟫🟫🟨`, the next guess will be "deter", with the characteristics shown.
 If the score is `🟫🟫🟫🟫🟫`, which is the most likely score for the first guess, the next guess is "mulch", and so on.
 
 Note that in the tree the score is associated with the guess that it will produce next, whereas in the summary of the game the score is associated with the guess that produced it.
@@ -391,41 +391,41 @@ The reason that this tree is not very interesting is that it simply reproduces t
 
 It is more interesting to play a random selection of games
 
-```jl
+```julia
 julia> print_tree(tree(wordle, Random.seed!(1234321), 12))
-missing, raise, 1535, 2315, 5.87791, 61.0009
-├─ 🟫🟫🟫🟫🟨, betel, 189, 121, 5.06266, 4.95041
-│  └─ 🟫🟩🟫🟫🟨, cello, 349, 9, 2.9477, 1.22222
-│     └─ 🟫🟩🟩🟫🟨, felon, 714, 2, 1.0, 1.0
-│        └─ 🟫🟩🟩🟩🟩, melon, 1220, 1, -0.0, 1.0
-├─ 🟨🟩🟩🟫🟫, dairy, 515, 4, 1.5, 1.5
-│  └─ 🟫🟩🟩🟩🟩, fairy, 699, 2, 1.0, 1.0
-│     └─ 🟫🟩🟩🟩🟩, hairy, 948, 1, -0.0, 1.0
-├─ 🟨🟫🟫🟫🟨, deter, 546, 102, 4.37007, 9.23529
-│  └─ 🟩🟩🟫🟫🟩, decor, 530, 2, 1.0, 1.0
-│     └─ 🟩🟩🟫🟫🟩, demur, 540, 1, -0.0, 1.0
-├─ 🟫🟫🟫🟫🟫, mulch, 1275, 168, 5.21165, 6.85714
-│  ├─ 🟫🟩🟩🟫🟫, bully, 302, 6, 1.79248, 2.0
-│  │  └─ 🟫🟩🟩🟫🟩, pulpy, 1492, 1, -0.0, 1.0
-│  ├─ 🟫🟨🟨🟨🟫, cloud, 419, 4, 2.0, 1.0
-│  │  └─ 🟩🟩🟩🟩🟫, clout, 420, 1, -0.0, 1.0
-│  └─ 🟫🟫🟫🟫🟨, whoop, 2262, 6, 2.58496, 1.0
-│     └─ 🟫🟨🟨🟫🟫, hobby, 985, 1, -0.0, 1.0
-├─ 🟨🟩🟫🟫🟫, party, 1377, 26, 3.12276, 3.84615
-│  └─ 🟫🟩🟩🟫🟩, carry, 338, 4, 1.5, 1.5
-│     └─ 🟫🟩🟩🟩🟩, harry, 955, 2, 1.0, 1.0
-├─ 🟫🟫🟨🟫🟫, pilot, 1413, 107, 4.69342, 6.38318
-│  ├─ 🟫🟨🟫🟨🟫, comic, 435, 4, 2.0, 1.0
-│  │  └─ 🟩🟩🟫🟩🟩, conic, 439, 1, -0.0, 1.0
-│  ├─ 🟫🟩🟫🟫🟨, width, 2267, 13, 2.93121, 2.07692
-│  │  └─ 🟫🟩🟫🟩🟫, bitty, 204, 4, 1.5, 1.5
-│  │     └─ 🟫🟩🟫🟩🟩, fifty, 733, 2, 1.0, 1.0
-│  └─ 🟫🟩🟫🟫🟫, windy, 2274, 16, 3.20282, 1.875
-│     └─ 🟫🟩🟫🟫🟩, fizzy, 746, 2, 1.0, 1.0
-│        └─ 🟨🟩🟫🟫🟩, jiffy, 1056, 1, -0.0, 1.0
-├─ 🟨🟩🟫🟨🟫, satyr, 1648, 2, 1.0, 1.0
-└─ 🟨🟫🟫🟨🟫, short, 1739, 24, 3.60539, 2.25
-   └─ 🟨🟫🟨🟨🟨, torus, 2085, 1, -0.0, 1.0
+missing, raise, 2315, 5.87791, 61.0009
+├─ 🟫🟫🟨🟫🟫, pilot, 107, 4.69342, 6.38318
+│  ├─ 🟫🟩🟫🟫🟨, width, 13, 2.93121, 2.07692
+│  │  └─ 🟫🟩🟫🟩🟫, bitty, 4, 1.5, 1.5
+│  │     └─ 🟫🟩🟫🟩🟩, fifty, 2, 1.0, 1.0
+│  ├─ 🟫🟩🟫🟫🟫, windy, 16, 3.20282, 1.875
+│  │  └─ 🟫🟩🟫🟫🟩, fizzy, 2, 1.0, 1.0
+│  │     └─ 🟨🟩🟫🟫🟩, jiffy, 1, -0.0, 1.0
+│  └─ 🟫🟨🟫🟨🟫, comic, 4, 2.0, 1.0
+│     └─ 🟩🟩🟫🟩🟩, conic, 1, -0.0, 1.0
+├─ 🟫🟫🟫🟫🟫, mulch, 168, 5.21165, 6.85714
+│  ├─ 🟫🟩🟩🟫🟫, bully, 6, 1.79248, 2.0
+│  │  └─ 🟫🟩🟩🟫🟩, pulpy, 1, -0.0, 1.0
+│  ├─ 🟫🟨🟨🟨🟫, cloud, 4, 2.0, 1.0
+│  │  └─ 🟩🟩🟩🟩🟫, clout, 1, -0.0, 1.0
+│  └─ 🟫🟫🟫🟫🟨, whoop, 6, 2.58496, 1.0
+│     └─ 🟫🟨🟨🟫🟫, hobby, 1, -0.0, 1.0
+├─ 🟫🟫🟫🟫🟨, betel, 121, 5.06266, 4.95041
+│  └─ 🟫🟩🟫🟫🟨, cello, 9, 2.9477, 1.22222
+│     └─ 🟫🟩🟩🟫🟨, felon, 2, 1.0, 1.0
+│        └─ 🟫🟩🟩🟩🟩, melon, 1, -0.0, 1.0
+├─ 🟨🟩🟩🟫🟫, dairy, 4, 1.5, 1.5
+│  └─ 🟫🟩🟩🟩🟩, fairy, 2, 1.0, 1.0
+│     └─ 🟫🟩🟩🟩🟩, hairy, 1, -0.0, 1.0
+├─ 🟨🟫🟫🟫🟨, deter, 102, 4.37007, 9.23529
+│  └─ 🟩🟩🟫🟫🟩, decor, 2, 1.0, 1.0
+│     └─ 🟩🟩🟫🟫🟩, demur, 1, -0.0, 1.0
+├─ 🟨🟩🟫🟫🟫, party, 26, 3.12276, 3.84615
+│  └─ 🟫🟩🟩🟫🟩, carry, 4, 1.5, 1.5
+│     └─ 🟫🟩🟩🟩🟩, harry, 2, 1.0, 1.0
+├─ 🟨🟫🟫🟨🟫, short, 24, 3.60539, 2.25
+│  └─ 🟨🟫🟨🟨🟨, torus, 1, -0.0, 1.0
+└─ 🟨🟩🟫🟨🟫, satyr, 2, 1.0, 1.0
 ```
 
 Again, the root is "raise", which is the first guess in any game using the `MaximumEntropy` strategy, and if the first score is `🟫🟫🟫🟫🟫` then the second guess will be "mulch".

data/Greek/README.md

@@ -0,0 +1 @@
+These files are derived from  https://github.com/gvarnavi/wordle-greek/raw/main/src/constants/validGuesses.ts and https://github.com/gvarnavi/wordle-greek/raw/main/src/constants/wordlist.ts