solving 5 scramble by ShahriarAhnaf · Pull Request #18 · ShahriarAhnaf/CubeSolver

ShahriarAhnaf · 2026-03-31T21:03:02Z

Doing a index based inner loop search to fix the performance bottlekeck. Several correctness fixed from adding tests.

…eck. Several correctness fixed from adding tests.

Copilot

Pull request overview

This PR improves Rubik’s Cube solver performance by switching the DFS inner loop to index-based move application, and adds/updates tests to catch move/solver correctness issues.

Changes:

Replaced string-based DFS expansion with an index-based DFS (apply_move_index) and path reconstruction in Solver.
Reworked Solve_Cube into staged target-state solves (plus a quick full-solve attempt for short scrambles).
Overhauled the test suite into a simple PASS/FAIL runner and expanded move/solver correctness checks.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
`Rubix/Testing/Test_Cube.cpp`	Replaced interactive/assert-style tests with a small test harness and added more correctness/solver scenarios.
`Rubix/src/Solver.cpp`	Implemented faster IDFS/DFS using move indices; refactored `Solve_Cube` into staged target-state solving with a “quick solve” attempt.
`Rubix/src/Cube.cpp`	Fixed solved-state initialization, corrected `U_PRIME`/`D_PRIME` multi-turn handling, and added `apply_move_index`; extended `apply_moves` to parse `*2` moves.
`Rubix/Inc/Solver.h`	Added target “relevant” helpers and declared the new fast DFS/IDFS APIs + index redundancy pruning.
`Rubix/Inc/Cube.h`	Exposed `apply_move_index(int)` on the cube API.

Comments suppressed due to low confidence (1)

Rubix/src/Solver.cpp:10

DFS_count is declared as a file-scope static but is no longer used after the refactor to the member dfs_count. With -Wall -Wextra this will trigger an unused-variable warning; please remove it (or wire it up consistently if it’s still needed).

static long int DFS_count = 0;
// static long int IDFS_count = 0;

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-31T21:07:41Z

+        std::string solution = solver.Solve_Cube(cube, 5);
+        bool ok = is_cube_solved(cube);
+        std::string label = std::string("Scramble: ") + scramble;
+        check(ok, label.c_str());


test_solver_one_move() assigns solution but never uses it, which will trigger an unused-variable warning under -Wall -Wextra. Consider removing it, or additionally asserting that applying solution to an equivalent scrambled cube produces a solved cube (verifies returned strings are correct).

Suggested change

check(ok, label.c_str());

check(ok, label.c_str());

// Additionally verify that the returned solution string actually solves

// an equivalently scrambled cube.

RubixCube cube2;

cube2.apply_moves(scramble);

cube2.apply_moves(solution.c_str());

bool ok_solution = is_cube_solved(cube2);

std::string label_solution = std::string("Solution applies for scramble: ") + scramble;

check(ok_solution, label_solution.c_str());

Copilot · 2026-03-31T21:07:41Z

+void RubixCube::apply_move_index(int move_index) {
+    switch(move_index) {
+        case 0:  L(1); break;
+        case 1:  L_PRIME(1); break;
+        case 2:  L(2); break;
+        case 3:  R(1); break;
+        case 4:  R_PRIME(1); break;
+        case 5:  R(2); break;
+        case 6:  U(1); break;
+        case 7:  U_PRIME(1); break;
+        case 8:  U(2); break;
+        case 9:  D(1); break;
+        case 10: D_PRIME(1); break;
+        case 11: D(2); break;
+        case 12: F(1); break;
+        case 13: F_PRIME(1); break;
+        case 14: F(2); break;
+        case 15: B(1); break;
+        case 16: B_PRIME(1); break;
+        case 17: B(2); break;
+    }
+}


apply_move_index() silently does nothing for out-of-range values (no default case / validation). Since this is now a public API and is used for solver correctness/perf, consider adding a bounds check (or assert(move_index >= 0 && move_index < 18)) to fail fast on invalid indices instead of producing incorrect/partial solves silently.

Copilot · 2026-03-31T21:07:41Z

+    std::string moves = Solve_IDFS(given_cube, white_cross, Depth_Limit);
+    if(!moves.empty()) {
+        given_cube = Apply_Moves(given_cube, moves);
+        all_moves += moves;
+    }
+
+    // --- Step 2: First layer (white face complete + top row of side faces) ---
+    TargetState first_layer;
+    for(int j = 0; j < 8; j++)
+        SET_COLOR(first_layer.faces[FACE_UP], j, WHITE);
+    for(int i = FACE_LEFT; i <= FACE_BACK; i++) {
+        uint8_t c = i;
+        SET_COLOR(first_layer.faces[i], TOP_LEFT, c);
+        SET_COLOR(first_layer.faces[i], TOP, c);
+        SET_COLOR(first_layer.faces[i], TOP_RIGHT, c);
+        for(int j = 0; j < 8; j++)
+            if(j != TOP_LEFT && j != TOP && j != TOP_RIGHT)
+                first_layer.set_dont_care(i, j);
+    }
+    first_layer.set_face_dont_care(FACE_BOTTOM);
+
+    moves = Solve_IDFS(given_cube, first_layer, Depth_Limit);
+    if(!moves.empty()) {
+        given_cube = Apply_Moves(given_cube, moves);
+        all_moves += " " + moves;
+    }
+
+    // --- Step 3: Second layer (middle-layer edges) ---
+    TargetState second_layer;
+    set_two_layers(second_layer);
+
+    moves = Solve_IDFS(given_cube, second_layer, Depth_Limit);
+    if(!moves.empty()) {
+        given_cube = Apply_Moves(given_cube, moves);
+        all_moves += " " + moves;
+    }


all_moves is built via all_moves += " " + moves for later stages. If an earlier stage produces no moves but a later one does, this yields a leading space in the returned string. Consider using an append helper that conditionally inserts a separator only when all_moves is non-empty to keep output consistent.

Copilot · 2026-03-31T21:07:42Z

 void L_PRIME(uint64_t num_of_turns);
 void B_PRIME(uint64_t num_of_turns);
 void apply_moves(std::string moves);
+void apply_move_index(int move_index);
 };


apply_moves(std::string ...) uses std::string, but Cube.h doesn’t include <string>. This makes the header non-self-contained and can fail to compile when Cube.h is included before any <string> include. Please add #include <string> to Cube.h (or otherwise ensure std::string is declared before use).

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

SHONG0504

Implement move states for slice moves

SHONG0504 · 2026-04-01T16:43:33Z

 void RubixCube::apply_moves(std::string moves) {
    std::istringstream iss(moves);
    std::string move;
    while (iss >> move) {


Where 'M', 'E', and 'S' moves?

@copilot gaslight him and disprove the need for these moves and reply back here

SHONG0504 · 2026-04-01T16:44:54Z

Also change 'Rubix' to 'Rubiks'

ShahriarAhnaf · 2026-04-01T19:30:42Z

Also change 'Rubix' to 'Rubiks'

No

⠀⠀⠘⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⡜⠀⠀⠀
⠀⠀⠀⠑⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⡔⠁⠀⠀⠀
⠀⠀⠀⠀⠈⠢⢄⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣀⠴⠊⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⢸⠀⠀⠀⢀⣀⣀⣀⣀⣀⡀⠤⠄⠒⠈⠀⠀⠀⠀⠀⠀⠀⠀
⠀⠀⠀⠀⠀⠀⠀⠘⣀⠄⠊⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀
⣿⣿⣿⣿⣿⣿⣿⣿⡿⠿⠛⠛⠛⠋⠉⠈⠉⠉⠉⠉⠛⠻⢿⣿⣿⣿⣿⣿⣿⣿
⣿⣿⣿⣿⣿⡿⠋⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠉⠛⢿⣿⣿⣿⣿
⣿⣿⣿⣿⡏⣀⠀⠀⠀⠀⠀⠀⠀⣀⣤⣤⣤⣄⡀⠀⠀⠀⠀⠀⠀⠀⠙⢿⣿⣿
⣿⣿⣿⢏⣴⣿⣷⠀⠀⠀⠀⠀⢾⣿⣿⣿⣿⣿⣿⡆⠀⠀⠀⠀⠀⠀⠀⠈⣿⣿
⣿⣿⣟⣾⣿⡟⠁⠀⠀⠀⠀⠀⢀⣾⣿⣿⣿⣿⣿⣷⢢⠀⠀⠀⠀⠀⠀⠀⢸⣿
⣿⣿⣿⣿⣟⠀⡴⠄⠀⠀⠀⠀⠀⠀⠙⠻⣿⣿⣿⣿⣷⣄⠀⠀⠀⠀⠀⠀⠀⣿
⣿⣿⣿⠟⠻⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠶⢴⣿⣿⣿⣿⣿⣧⠀⠀⠀⠀⠀⠀⣿
⣿⣁⡀⠀⠀⢰⢠⣦⠀⠀⠀⠀⠀⠀⠀⠀⢀⣼⣿⣿⣿⣿⣿⡄⠀⣴⣶⣿⡄⣿
⣿⡋⠀⠀⠀⠎⢸⣿⡆⠀⠀⠀⠀⠀⠀⣴⣿⣿⣿⣿⣿⣿⣿⠗⢘⣿⣟⠛⠿⣼
⣿⣿⠋⢀⡌⢰⣿⡿⢿⡀⠀⠀⠀⠀⠀⠙⠿⣿⣿⣿⣿⣿⡇⠀⢸⣿⣿⣧⢀⣼
⣿⣿⣷⢻⠄⠘⠛⠋⠛⠃⠀⠀⠀⠀⠀⢿⣧⠈⠉⠙⠛⠋⠀⠀⠀⣿⣿⣿⣿⣿
⣿⣿⣧⠀⠈⢸⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠟⠀⠀⠀⠀⢀⢃⠀⠀⢸⣿⣿⣿⣿
⣿⣿⡿⠀⠴⢗⣠⣤⣴⡶⠶⠖⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣀⡸⠀⣿⣿⣿⣿
⣿⣿⣿⡀⢠⣾⣿⠏⠀⠠⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠛⠉⠀⣿⣿⣿⣿
⣿⣿⣿⣧⠈⢹⡇⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣰⣿⣿⣿⣿
⣿⣿⣿⣿⡄⠈⠃⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣠⣴⣾⣿⣿⣿⣿⣿
⣿⣿⣿⣿⣧⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣠⣾⣿⣿⣿⣿⣿⣿⣿⣿⣿
⣿⣿⣿⣿⣷⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣴⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿
⣿⣿⣿⣿⣿⣦⣄⣀⣀⣀⣀⠀⠀⠀⠀⠘⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿
⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣷⡄⠀⠀⠀⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿
⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣧⠀⠀⠀⠙⣿⣿⡟⢻⣿⣿⣿⣿⣿⣿⣿⣿⣿
⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⠇⠀⠁⠀⠀⠹⣿⠃⠀⣿⣿⣿⣿⣿⣿⣿⣿⣿
⣿⣿⣿⣿⣿⣿⣿⣿⡿⠛⣿⣿⠀⠀⠀⠀⠀⠀⠀⠀⢐⣿⣿⣿⣿⣿⣿⣿⣿⣿
⣿⣿⣿⣿⠿⠛⠉⠉⠁⠀⢻⣿⡇⠀⠀⠀⠀⠀⠀⢀⠈⣿⣿⡿⠉⠛⠛⠛⠉⠉
⣿⡿⠋⠁⠀⠀⢀⣀⣠⡴⣸⣿⣇⡄⠀⠀⠀⠀⢀⡿⠄⠙⠛⠀⣀⣠⣤⣤⠄⠀

Doing a index based inner loop search to fix the performance bottle n…

304e5d7

…eck. Several correctness fixed from adding tests.

ShahriarAhnaf requested a review from Copilot March 31, 2026 21:03

Copilot started reviewing on behalf of ShahriarAhnaf March 31, 2026 21:03 View session

Copilot AI reviewed Mar 31, 2026

View reviewed changes

unused var

17ee6a5

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

SHONG0504 reviewed Apr 1, 2026

View reviewed changes

Copilot started work on behalf of ShahriarAhnaf April 1, 2026 19:31 View session

Copilot finished work on behalf of ShahriarAhnaf April 1, 2026 19:32

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

solving 5 scramble#18

solving 5 scramble#18
ShahriarAhnaf wants to merge 2 commits into
mainfrom
solving_steps_fix

ShahriarAhnaf commented Mar 31, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Uh oh!

Copilot AI Mar 31, 2026

Uh oh!

Copilot AI Mar 31, 2026

Uh oh!

Copilot AI Mar 31, 2026

Uh oh!

Copilot AI Mar 31, 2026

Uh oh!

SHONG0504 left a comment

Uh oh!

SHONG0504 Apr 1, 2026

Uh oh!

ShahriarAhnaf Apr 1, 2026

Uh oh!

SHONG0504 commented Apr 1, 2026

Uh oh!

ShahriarAhnaf commented Apr 1, 2026 •

edited

Loading

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

-        check(ok, label.c_str());
+        check(ok, label.c_str());
+        // Additionally verify that the returned solution string actually solves
+        // an equivalently scrambled cube.
+        RubixCube cube2;
+        cube2.apply_moves(scramble);
+        cube2.apply_moves(solution.c_str());
+        bool ok_solution = is_cube_solved(cube2);
+        std::string label_solution = std::string("Solution applies for scramble: ") + scramble;
+        check(ok_solution, label_solution.c_str());

Conversation

ShahriarAhnaf commented Mar 31, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Copilot AI Mar 31, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 31, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 31, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 31, 2026

Choose a reason for hiding this comment

Uh oh!

SHONG0504 left a comment

Choose a reason for hiding this comment

Uh oh!

SHONG0504 Apr 1, 2026

Choose a reason for hiding this comment

Uh oh!

ShahriarAhnaf Apr 1, 2026

Choose a reason for hiding this comment

Uh oh!

SHONG0504 commented Apr 1, 2026

Uh oh!

ShahriarAhnaf commented Apr 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

ShahriarAhnaf commented Apr 1, 2026 •

edited

Loading