jules-orchestrator-kit 0.41.1 → 0.51.0

package/src/mutation.mjs

@@ -0,0 +1,739 @@
+import { readFileSync, writeFileSync, existsSync } from "node:fs";
+import { resolve, isAbsolute } from "node:path";
+import { execSync } from "node:child_process";
+import { createHash } from "node:crypto";
+import { diffText } from "./git.mjs";
+
+/**
+ * @typedef {Object} MutationCandidate
+ * @property {string} id - Unique mutant identifier (file:line:operator)
+ * @property {string} file - Relative path of target file
+ * @property {number} line - 1-indexed line number
+ * @property {string} originalLine - Original line content
+ * @property {string} mutatedLine - Mutated line content
+ * @property {string} mutationType - Category of mutation (e.g., 'EQUALITY', 'LOGICAL', 'RELATIONAL', 'ARITHMETIC', 'BOOLEAN', 'RETURN')
+ * @property {string} description - Human-readable explanation of mutation
+ */
+
+/**
+ * @typedef {Object} MutantResult
+ * @property {MutationCandidate} mutant - The evaluated mutant
+ * @property {"KILLED" | "SURVIVED" | "TIMEOUT" | "ERROR"} status - Outcome of mutant execution
+ * @property {number} exitCode - Exit code from test command
+ * @property {number} durationMs - Execution time in milliseconds
+ * @property {string} [stdout] - Test command stdout
+ * @property {string} [stderr] - Test command stderr
+ */
+
+/**
+ * @typedef {Object} MutationReport
+ * @property {boolean} ok - True if mutation score meets or exceeds minScore
+ * @property {number} totalMutants - Total candidate mutants tested
+ * @property {number} killedMutants - Number of mutants killed (test failed as expected)
+ * @property {number} survivedMutants - Number of mutants survived (test passed despite mutation)
+ * @property {number} errorMutants - Number of mutants with execution errors/timeouts
+ * @property {number} mutationScore - Percentage of killed mutants (0 - 100)
+ * @property {number} minScore - Required threshold score
+ * @property {MutantResult[]} results - Detailed results for all mutants
+ * @property {MutantResult[]} survivors - List of survived mutants for remediation
+ * @property {number} durationMs - Total run duration in milliseconds
+ */
+
+/**
+ * Operator mutation replacement rules.
+ * Sorted to prevent premature substring collision (e.g. `===` before `==`).
+ */
+export const MUTATION_RULES = [
+  // 1. Strict & Loose Equality
+  {
+    type: "EQUALITY",
+    pattern: /===/g,
+    replace: "!==",
+    desc: "Inverted strict equality (=== -> !==)",
+  },
+  {
+    type: "EQUALITY",
+    pattern: /!==/g,
+    replace: "===",
+    desc: "Inverted strict inequality (!== -> ===)",
+  },
+  {
+    type: "EQUALITY",
+    // Ensure we don't match === or !==
+    pattern: /(?<![=!])==(?![=])/g,
+    replace: "!=",
+    desc: "Inverted loose equality (== -> !=)",
+  },
+  {
+    type: "EQUALITY",
+    pattern: /(?<![=!])!=(?![=])/g,
+    replace: "==",
+    desc: "Inverted loose inequality (!= -> ==)",
+  },
+
+  // 2. Relational Operators
+  {
+    type: "RELATIONAL",
+    pattern: />=/g,
+    replace: "<",
+    desc: "Inverted relational operator (>= -> <)",
+  },
+  {
+    type: "RELATIONAL",
+    // Exclude arrow functions (<= is relational, => is arrow)
+    pattern: /<=(?!>)/g,
+    replace: ">",
+    desc: "Inverted relational operator (<= -> >)",
+  },
+  {
+    type: "RELATIONAL",
+    // Exclude =>, >>, ->, and closing XML/HTML tags
+    pattern: /(?<![=-])>(?![=>])/g,
+    replace: "<=",
+    desc: "Inverted comparison (> -> <=)",
+  },
+  {
+    type: "RELATIONAL",
+    // Exclude <<, <-, and HTML/XML tag openings (<foo)
+    pattern: /(?<![<])<(?![<=/a-zA-Z!])/g,
+    replace: ">=",
+    desc: "Inverted comparison (< -> >=)",
+  },
+
+  // 3. Logical Operators
+  {
+    type: "LOGICAL",
+    pattern: /&&/g,
+    replace: "||",
+    desc: "Swapped logical AND with OR (&& -> ||)",
+  },
+  {
+    type: "LOGICAL",
+    pattern: /\|\|/g,
+    replace: "&&",
+    desc: "Swapped logical OR with AND (|| -> &&)",
+  },
+
+  // 4. Boolean Literals
+  {
+    type: "BOOLEAN",
+    pattern: /\btrue\b/g,
+    replace: "false",
+    desc: "Inverted boolean literal (true -> false)",
+  },
+  {
+    type: "BOOLEAN",
+    pattern: /\bfalse\b/g,
+    replace: "true",
+    desc: "Inverted boolean literal (false -> true)",
+  },
+
+  // 5. Unary & Arithmetic
+  {
+    type: "ARITHMETIC",
+    pattern: /\+\+/g,
+    replace: "--",
+    desc: "Swapped increment with decrement (++ -> --)",
+  },
+  {
+    type: "ARITHMETIC",
+    pattern: /--/g,
+    replace: "++",
+    desc: "Swapped decrement with increment (-- -> ++)",
+  },
+  {
+    type: "ARITHMETIC",
+    pattern: /(\s)\+(\s)/g,
+    replace: "$1-$2",
+    desc: "Swapped addition with subtraction (+ -> -)",
+  },
+  {
+    type: "ARITHMETIC",
+    pattern: /(\s)-(?![\d>])(\s)/g,
+    replace: "$1+$2",
+    desc: "Swapped subtraction with addition (- -> +)",
+  },
+  {
+    type: "ARITHMETIC",
+    pattern: /(\s)\*(\s)/g,
+    replace: "$1/$2",
+    desc: "Swapped multiplication with division (* -> /)",
+  },
+  {
+    type: "ARITHMETIC",
+    pattern: /(\s)\/(\s)/g,
+    replace: "$1*$2",
+    desc: "Swapped division with multiplication (/ -> *)",
+  },
+
+  // 6. Return Statements
+  {
+    type: "RETURN",
+    pattern: /\breturn\s+true\b/g,
+    replace: "return false",
+    desc: "Inverted return boolean (return true -> return false)",
+  },
+  {
+    type: "RETURN",
+    pattern: /\breturn\s+false\b/g,
+    replace: "return true",
+    desc: "Inverted return boolean (return false -> return true)",
+  },
+  {
+    type: "RETURN",
+    pattern: /\breturn\s+null\b/g,
+    replace: "return undefined",
+    desc: "Altered return null (return null -> return undefined)",
+  },
+];
+
+/**
+ * Checks if a file path is a test or non-implementation file that should be excluded from mutation.
+ * @param {string} filePath
+ * @returns {boolean}
+ */
+export function isExcludedFromMutation(filePath = "") {
+  const normalized = filePath.replace(/\\/g, "/").toLowerCase();
+
+  // Exclude test files
+  if (
+    normalized.includes(".test.") ||
+    normalized.includes(".spec.") ||
+    normalized.includes("_test.") ||
+    normalized.includes("/test/") ||
+    normalized.includes("/tests/") ||
+    normalized.includes("/__tests__/")
+  ) {
+    return true;
+  }
+
+  // Exclude non-executable / config / documentation formats
+  const nonExecExtensions = [
+    ".md", ".markdown", ".json", ".yml", ".yaml", ".toml", ".txt",
+    ".svg", ".png", ".jpg", ".jpeg", ".gif", ".ico", ".woff", ".woff2",
+    ".lock", ".lockb", ".css", ".scss", ".less", ".html",
+  ];
+
+  if (nonExecExtensions.some((ext) => normalized.endsWith(ext))) {
+    return true;
+  }
+
+  // Exclude package manifests and agent rules
+  if (
+    normalized.endsWith("package.json") ||
+    normalized.endsWith("package-lock.json") ||
+    normalized.includes(".agent/") ||
+    normalized.includes(".github/")
+  ) {
+    return true;
+  }
+
+  return false;
+}
+
+/**
+ * Returns a map of line number (1-indexed) to array of character index ranges [startCol, endCol]
+ * that are part of string literals (including multiline template literals and block comments).
+ * @param {string} sourceCode
+ * @returns {Map<number, Array<[number, number]>>}
+ */
+export function getFileStringLiteralLineMap(sourceCode = "") {
+  const lines = sourceCode.split("\n");
+  const lineMap = new Map();
+  let inTemplate = false;
+  let inBlockComment = false;
+
+  for (let l = 0; l < lines.length; l++) {
+    const line = lines[l];
+    const ranges = [];
+    let inQuote = null;
+    let quoteStart = -1;
+    let escaped = false;
+
+    for (let i = 0; i < line.length; i++) {
+      const char = line[i];
+      if (escaped) {
+        escaped = false;
+        continue;
+      }
+      if (char === "\\") {
+        escaped = true;
+        continue;
+      }
+
+      if (inBlockComment) {
+        if (char === "*" && line[i + 1] === "/") {
+          inBlockComment = false;
+          i++;
+        }
+        continue;
+      }
+
+      if (inTemplate) {
+        if (char === "`") {
+          inTemplate = false;
+          ranges.push([0, i]);
+        }
+        continue;
+      }
+
+      if (inQuote) {
+        if (char === inQuote) {
+          ranges.push([quoteStart, i]);
+          inQuote = null;
+        }
+      } else {
+        if (char === "/" && line[i + 1] === "*") {
+          inBlockComment = true;
+          quoteStart = i;
+          i++;
+          continue;
+        }
+        if (char === "/" && line[i + 1] === "/") {
+          ranges.push([i, line.length - 1]);
+          break;
+        }
+        if (char === '"' || char === "'") {
+          inQuote = char;
+          quoteStart = i;
+        } else if (char === "`") {
+          inTemplate = true;
+          quoteStart = i;
+        }
+      }
+    }
+
+    if (inTemplate) {
+      ranges.push([quoteStart >= 0 ? quoteStart : 0, line.length - 1]);
+      quoteStart = 0;
+    } else if (inBlockComment) {
+      ranges.push([quoteStart >= 0 ? quoteStart : 0, line.length - 1]);
+      quoteStart = 0;
+    } else if (inQuote && quoteStart >= 0) {
+      ranges.push([quoteStart, line.length - 1]);
+    }
+
+    lineMap.set(l + 1, ranges);
+  }
+  return lineMap;
+}
+
+/**
+ * Returns character index ranges [start, end] for all string literals on a line.
+ * @param {string} line
+ * @returns {Array<[number, number]>}
+ */
+export function getStringLiteralRanges(line) {
+  const ranges = [];
+  let inQuote = null;
+  let start = -1;
+  let escaped = false;
+
+  for (let i = 0; i < line.length; i++) {
+    const char = line[i];
+    if (escaped) {
+      escaped = false;
+      continue;
+    }
+    if (char === "\\") {
+      escaped = true;
+      continue;
+    }
+    if (inQuote) {
+      if (char === inQuote) {
+        ranges.push([start, i]);
+        inQuote = null;
+      }
+    } else {
+      if (char === '"' || char === "'" || char === "`") {
+        inQuote = char;
+        start = i;
+      }
+    }
+  }
+  if (inQuote && start !== -1) {
+    ranges.push([start, line.length - 1]);
+  }
+  return ranges;
+}
+
+/**
+ * Generates mutation candidates for a given single line of code.
+ * @param {string} line - Line text
+ * @param {number} lineNo - 1-indexed line number
+ * @param {string} filePath - Target file path
+ * @param {Array<[number, number]>} [explicitStringRanges] - Pre-calculated string ranges
+ * @returns {MutationCandidate[]}
+ */
+export function generateLineMutants(line = "", lineNo = 1, filePath = "", explicitStringRanges = null) {
+  const trimmed = line.trim();
+  // Skip comments, imports, exports, and empty lines
+  if (
+    !trimmed ||
+    trimmed.startsWith("//") ||
+    trimmed.startsWith("/*") ||
+    trimmed.startsWith("*") ||
+    trimmed.startsWith("import ") ||
+    trimmed.startsWith("#") ||
+    trimmed.startsWith("from ") ||
+    trimmed.startsWith("package ")
+  ) {
+    return [];
+  }
+
+  const mutants = [];
+  const lineHash = createHash("sha256").update(line).digest("hex").slice(0, 6);
+  const stringRanges = Array.isArray(explicitStringRanges) ? explicitStringRanges : getStringLiteralRanges(line);
+
+  for (let ruleIdx = 0; ruleIdx < MUTATION_RULES.length; ruleIdx++) {
+    const rule = MUTATION_RULES[ruleIdx];
+    const regex = new RegExp(rule.pattern.source, rule.pattern.flags);
+
+    let match;
+    while ((match = regex.exec(line)) !== null) {
+      const matchIndex = match.index;
+      const matchedText = match[0];
+      if (matchedText.length === 0) {
+        regex.lastIndex++;
+        continue;
+      }
+
+      // Do not mutate operators inside string literals or comments
+      if (rule.type !== "BOOLEAN" && rule.type !== "RETURN") {
+        const isInsideString = stringRanges.some(([start, end]) => matchIndex >= start && matchIndex <= end);
+        if (isInsideString) {
+          if (!rule.pattern.global) break;
+          continue;
+        }
+      }
+
+      let replacement = rule.replace;
+      if (typeof replacement === "string" && replacement.includes("$")) {
+        const singleRegex = new RegExp(rule.pattern.source);
+        replacement = matchedText.replace(singleRegex, rule.replace);
+      }
+
+      const mutatedLine = line.slice(0, matchIndex) + replacement + line.slice(matchIndex + matchedText.length);
+
+      if (mutatedLine !== line) {
+        mutants.push({
+          id: `${filePath}:${lineNo}:m${ruleIdx}-${matchIndex}-${lineHash}`,
+          file: filePath,
+          line: lineNo,
+          originalLine: line,
+          mutatedLine,
+          mutationType: rule.type,
+          description: rule.desc,
+        });
+      }
+
+      if (!rule.pattern.global) break;
+    }
+  }
+
+  return mutants;
+}
+
+/**
+ * Generates mutation candidates for an entire file content.
+ * @param {string} sourceCode
+ * @param {string} filePath
+ * @returns {MutationCandidate[]}
+ */
+export function generateMutants(sourceCode = "", filePath = "source.js") {
+  if (isExcludedFromMutation(filePath)) return [];
+
+  const lines = sourceCode.split("\n");
+  const stringMap = getFileStringLiteralLineMap(sourceCode);
+  const allMutants = [];
+
+  for (let i = 0; i < lines.length; i++) {
+    const lineRanges = stringMap.get(i + 1) || [];
+    const lineMutants = generateLineMutants(lines[i], i + 1, filePath, lineRanges);
+    allMutants.push(...lineMutants);
+  }
+
+  return allMutants;
+}
+
+/**
+ * Parses a unified diff string and extracts mutation candidates strictly from added lines (`+` hunks).
+ * @param {string} diffStr - Unified git diff
+ * @param {string} [root=process.cwd()] - Project root
+ * @returns {MutationCandidate[]}
+ */
+export function generateDiffMutants(diffStr = "", root = process.cwd()) {
+  if (!diffStr) return [];
+
+  const candidates = [];
+  let currentFile = null;
+  let currentLineNo = null;
+  let currentFileStringMap = null;
+
+  const lines = diffStr.split("\n");
+  for (let i = 0; i < lines.length; i++) {
+    const line = lines[i];
+
+    if ((line.startsWith("+++ ") || line.startsWith("+++ b/") || line.startsWith("+++ /dev/null")) && !line.startsWith("++++")) {
+      const target = line.slice(3).split("\t")[0].trim().replace(/^b\//, "");
+      currentFile = target && target !== "/dev/null" ? target : null;
+      currentLineNo = null;
+      currentFileStringMap = null;
+
+      if (currentFile && root) {
+        const fullPath = isAbsolute(currentFile) ? currentFile : resolve(root, currentFile);
+        if (existsSync(fullPath)) {
+          try {
+            const content = readFileSync(fullPath, "utf-8");
+            currentFileStringMap = getFileStringLiteralLineMap(content);
+          } catch (_) {}
+        }
+      }
+      continue;
+    }
+
+    const hunkMatch = /^@@ -\d+(?:,\d+)? \+(\d+)/.exec(line);
+    if (hunkMatch) {
+      currentLineNo = Number(hunkMatch[1]);
+      continue;
+    }
+
+    if (line.startsWith("+") && !line.startsWith("+++")) {
+      if (currentFile && !isExcludedFromMutation(currentFile) && currentLineNo !== null) {
+        const addedText = line.slice(1);
+        const lineRanges = currentFileStringMap ? currentFileStringMap.get(currentLineNo) : null;
+        const lineMutants = generateLineMutants(addedText, currentLineNo, currentFile, lineRanges);
+        candidates.push(...lineMutants);
+      }
+      if (currentLineNo !== null) currentLineNo++;
+    } else if (currentLineNo !== null && !line.startsWith("-") && !line.startsWith("\\")) {
+      currentLineNo++;
+    }
+  }
+
+  return candidates;
+}
+
+/**
+ * Applies a candidate mutation to a target file, executes the test command, and safely rolls back.
+ *
+ * @param {MutationCandidate} mutant
+ * @param {Object} [options]
+ * @param {string} [options.root=process.cwd()]
+ * @param {string} [options.testCmd="npm test"]
+ * @param {number} [options.timeoutMs=15000]
+ * @param {Function} [options.executor] - Optional custom execution function for mocking / speed
+ * @returns {MutantResult}
+ */
+export function executeMutant(mutant, options = {}) {
+  const root = options.root || process.cwd();
+  const testCmd = options.testCmd || "npm test";
+  const timeoutMs = options.timeoutMs || 15000;
+  const absPath = isAbsolute(mutant.file) ? mutant.file : resolve(root, mutant.file);
+
+  if (!existsSync(absPath)) {
+    return {
+      mutant,
+      status: "ERROR",
+      exitCode: 1,
+      durationMs: 0,
+      stderr: `Target file does not exist: ${mutant.file}`,
+    };
+  }
+
+  const originalContent = readFileSync(absPath, "utf-8");
+  const lines = originalContent.split("\n");
+
+  if (mutant.line < 1 || mutant.line > lines.length) {
+    return {
+      mutant,
+      status: "ERROR",
+      exitCode: 1,
+      durationMs: 0,
+      stderr: `Mutant line ${mutant.line} out of range (1..${lines.length})`,
+    };
+  }
+
+  // Replace target line
+  lines[mutant.line - 1] = mutant.mutatedLine;
+  const mutatedContent = lines.join("\n");
+
+  const startTime = Date.now();
+  let exitCode = 0;
+  let stdout = "";
+  let stderr = "";
+  let status = "SURVIVED";
+
+  try {
+    // Write mutant to disk
+    writeFileSync(absPath, mutatedContent, "utf-8");
+
+    if (typeof options.executor === "function") {
+      const execResult = options.executor({ mutant, testCmd, absPath });
+      exitCode = typeof execResult.exitCode === "number" ? execResult.exitCode : execResult.status || 0;
+      stdout = execResult.stdout || "";
+      stderr = execResult.stderr || "";
+    } else {
+      try {
+        stdout = execSync(testCmd, {
+          cwd: root,
+          timeout: timeoutMs,
+          stdio: ["ignore", "pipe", "pipe"],
+          env: { ...process.env, CI: "true", JULES_MUTATION_RUN: "true" },
+          encoding: "utf-8",
+        });
+        exitCode = 0;
+      } catch (execErr) {
+        exitCode = execErr.status || execErr.statusCode || 1;
+        stdout = execErr.stdout ? execErr.stdout.toString() : "";
+        stderr = execErr.stderr ? execErr.stderr.toString() : execErr.message;
+        if (execErr.code === "ETIMEDOUT" || execErr.killed) {
+          status = "TIMEOUT";
+        }
+      }
+    }
+
+    const durationMs = Date.now() - startTime;
+
+    // If test failed (non-zero exit code) or timed out, mutant was KILLED
+    if (status === "TIMEOUT") {
+      return { mutant, status: "KILLED", exitCode, durationMs, stdout, stderr };
+    }
+
+    if (exitCode !== 0) {
+      status = "KILLED";
+    } else {
+      status = "SURVIVED";
+    }
+
+    return {
+      mutant,
+      status,
+      exitCode,
+      durationMs,
+      stdout: stdout.slice(0, 500),
+      stderr: stderr.slice(0, 500),
+    };
+  } catch (err) {
+    return {
+      mutant,
+      status: "ERROR",
+      exitCode: 1,
+      durationMs: Date.now() - startTime,
+      stderr: err.message,
+    };
+  } finally {
+    // GUARANTEED SAFE ROLLBACK
+    try {
+      writeFileSync(absPath, originalContent, "utf-8");
+    } catch (_) {}
+  }
+}
+
+/**
+ * Runs a complete mutation test sweep across candidate mutants.
+ *
+ * @param {Object} [options]
+ * @param {string} [options.root=process.cwd()]
+ * @param {string} [options.diffStr] - Optional unified diff to target (defaults to git working tree diff)
+ * @param {string} [options.base="main"] - Base branch if computing diff
+ * @param {string} [options.mode="working-tree"] - Diff mode ("working-tree" | "staged" | "committed")
+ * @param {string} [options.testCmd="npm test"] - Test command
+ * @param {number} [options.minScore=80] - Minimum required mutation score (default 80%)
+ * @param {number} [options.maxMutants=20] - Maximum mutants to evaluate to prevent excessive CI duration
+ * @param {number} [options.timeoutMs=15000] - Per-mutant test timeout
+ * @param {Function} [options.executor] - Custom runner for unit testing
+ * @returns {MutationReport}
+ */
+export function runMutationTest(options = {}) {
+  const root = options.root || process.cwd();
+  const base = options.base || "main";
+  const mode = options.mode || "working-tree";
+  const minScore = typeof options.minScore === "number" ? options.minScore : 80;
+  const maxMutants = typeof options.maxMutants === "number" ? options.maxMutants : 20;
+  const testCmd = options.testCmd || "npm test";
+
+  const startTime = Date.now();
+
+  let candidates = [];
+  if (Array.isArray(options.files) && options.files.length > 0) {
+    for (const relFile of options.files) {
+      const abs = resolve(root, relFile);
+      if (existsSync(abs)) {
+        const content = readFileSync(abs, "utf-8");
+        candidates.push(...generateMutants(content, relFile));
+      }
+    }
+  } else {
+    let diffContent = options.diffStr;
+    if (!diffContent) {
+      try {
+        diffContent = diffText(root, base, mode);
+      } catch (_) {
+        diffContent = "";
+      }
+    }
+    if (diffContent) {
+      candidates = generateDiffMutants(diffContent, root);
+    }
+  }
+
+  // Cap candidates to maxMutants
+  const selectedMutants = candidates.slice(0, maxMutants);
+
+  if (selectedMutants.length === 0) {
+    return {
+      ok: true,
+      totalMutants: 0,
+      killedMutants: 0,
+      survivedMutants: 0,
+      errorMutants: 0,
+      mutationScore: 100,
+      minScore,
+      results: [],
+      survivors: [],
+      durationMs: Date.now() - startTime,
+    };
+  }
+
+  const results = [];
+  let killedCount = 0;
+  let survivedCount = 0;
+  let errorCount = 0;
+
+  for (const mutant of selectedMutants) {
+    const res = executeMutant(mutant, {
+      root,
+      testCmd,
+      timeoutMs: options.timeoutMs,
+      executor: options.executor,
+    });
+
+    results.push(res);
+    if (res.status === "KILLED") {
+      killedCount++;
+    } else if (res.status === "SURVIVED") {
+      survivedCount++;
+    } else {
+      errorCount++;
+    }
+  }
+
+  const evaluatedCount = killedCount + survivedCount;
+  const mutationScore = evaluatedCount > 0 ? Math.round((killedCount / evaluatedCount) * 100) : 100;
+  const ok = mutationScore >= minScore;
+  const survivors = results.filter((r) => r.status === "SURVIVED");
+
+  return {
+    ok,
+    totalMutants: selectedMutants.length,
+    killedMutants: killedCount,
+    survivedMutants: survivedCount,
+    errorMutants: errorCount,
+    mutationScore,
+    minScore,
+    results,
+    survivors,
+    durationMs: Date.now() - startTime,
+  };
+}