@aiready/pattern-detect 0.17.8 → 0.17.12

package/dist/chunk-YLVV6YZ5.mjs

@@ -0,0 +1,143 @@
+import {
+  calculateSeverity
+} from "./chunk-TIBF7KST.mjs";
+
+// src/detector.ts
+import {
+  calculateStringSimilarity,
+  calculateHeuristicConfidence,
+  extractCodeBlocks
+} from "@aiready/core";
+
+// src/core/normalizer.ts
+function normalizeCode(code, isPython = false) {
+  if (!code) return "";
+  let normalized = code;
+  if (isPython) {
+    normalized = normalized.replace(/#.*/g, "");
+  } else {
+    normalized = normalized.replace(/\/\/.*$/gm, "").replace(/\/\*[\s\S]*?\*\//g, "");
+  }
+  return normalized.replace(/"[^"]*"/g, '"STR"').replace(/'[^']*'/g, "'STR'").replace(/`[^`]*`/g, "`STR`").replace(/\b\d+\b/g, "NUM").replace(/\s+/g, " ").trim().toLowerCase();
+}
+
+// src/detector.ts
+function extractBlocks(file, content) {
+  return extractCodeBlocks(file, content);
+}
+function calculateSimilarity(a, b) {
+  return calculateStringSimilarity(a, b);
+}
+function calculateConfidence(similarity, tokens, lines) {
+  return calculateHeuristicConfidence(similarity, tokens, lines);
+}
+async function detectDuplicatePatterns(fileContents, options) {
+  const {
+    minSimilarity,
+    minLines,
+    streamResults,
+    onProgress,
+    excludePatterns = [],
+    confidenceThreshold = 0,
+    ignoreWhitelist = []
+  } = options;
+  const allBlocks = [];
+  const excludeRegexes = excludePatterns.map((p) => new RegExp(p, "i"));
+  for (const { file, content } of fileContents) {
+    const blocks = extractBlocks(file, content);
+    for (const b of blocks) {
+      if (b.endLine - b.startLine + 1 < minLines) continue;
+      const isExcluded = excludeRegexes.some((regex) => regex.test(b.code));
+      if (isExcluded) continue;
+      allBlocks.push(b);
+    }
+  }
+  const duplicates = [];
+  const totalBlocks = allBlocks.length;
+  let comparisons = 0;
+  const totalComparisons = totalBlocks * (totalBlocks - 1) / 2;
+  if (onProgress) {
+    onProgress(
+      0,
+      totalComparisons,
+      `Starting duplicate detection on ${totalBlocks} blocks...`
+    );
+  }
+  for (let i = 0; i < allBlocks.length; i++) {
+    if (i % 50 === 0 && i > 0) {
+      await new Promise((resolve) => setImmediate(resolve));
+      if (onProgress) {
+        onProgress(
+          comparisons,
+          totalComparisons,
+          `Analyzing blocks (${i}/${totalBlocks})...`
+        );
+      }
+    }
+    const b1 = allBlocks[i];
+    const isPython1 = b1.file.toLowerCase().endsWith(".py");
+    const norm1 = normalizeCode(b1.code, isPython1);
+    for (let j = i + 1; j < allBlocks.length; j++) {
+      comparisons++;
+      const b2 = allBlocks[j];
+      if (b1.file === b2.file) continue;
+      const isWhitelisted = ignoreWhitelist.some((pattern) => {
+        return b1.file.includes(pattern) && b2.file.includes(pattern) || pattern === `${b1.file}::${b2.file}` || pattern === `${b2.file}::${b1.file}`;
+      });
+      if (isWhitelisted) continue;
+      const isPython2 = b2.file.toLowerCase().endsWith(".py");
+      const norm2 = normalizeCode(b2.code, isPython2);
+      const sim = calculateSimilarity(norm1, norm2);
+      if (sim >= minSimilarity) {
+        const confidence = calculateConfidence(
+          sim,
+          b1.tokens,
+          b1.endLine - b1.startLine + 1
+        );
+        if (confidence < confidenceThreshold) continue;
+        const { severity, reason, suggestion, matchedRule } = calculateSeverity(
+          b1.file,
+          b2.file,
+          b1.code,
+          sim,
+          b1.endLine - b1.startLine + 1
+        );
+        const dup = {
+          file1: b1.file,
+          line1: b1.startLine,
+          endLine1: b1.endLine,
+          file2: b2.file,
+          line2: b2.startLine,
+          endLine2: b2.endLine,
+          code1: b1.code,
+          code2: b2.code,
+          similarity: sim,
+          confidence,
+          patternType: b1.patternType,
+          tokenCost: b1.tokens + b2.tokens,
+          severity,
+          reason,
+          suggestion,
+          matchedRule
+        };
+        duplicates.push(dup);
+        if (streamResults)
+          console.log(
+            `[DUPLICATE] ${dup.file1}:${dup.line1} <-> ${dup.file2}:${dup.line2} (${Math.round(sim * 100)}%, conf: ${Math.round(confidence * 100)}%)`
+          );
+      }
+    }
+  }
+  if (onProgress) {
+    onProgress(
+      totalComparisons,
+      totalComparisons,
+      `Duplicate detection complete. Found ${duplicates.length} patterns.`
+    );
+  }
+  return duplicates.sort((a, b) => b.similarity - a.similarity);
+}
+
+export {
+  detectDuplicatePatterns
+};

package/dist/chunk-ZUWPFVJV.mjs

@@ -0,0 +1,115 @@
+// src/scoring.ts
+import {
+  calculateMonthlyCost,
+  calculateProductivityImpact,
+  DEFAULT_COST_CONFIG,
+  ToolName
+} from "@aiready/core";
+function calculatePatternScore(duplicates, totalFilesAnalyzed, costConfig) {
+  const actionableDuplicates = duplicates.filter(
+    (d) => d.severity !== "info"
+  );
+  const totalDuplicates = actionableDuplicates.length;
+  const totalTokenCost = actionableDuplicates.reduce((sum, d) => sum + d.tokenCost, 0);
+  const highImpactDuplicates = actionableDuplicates.filter(
+    (d) => d.tokenCost > 1e3 || d.similarity > 0.7
+  ).length;
+  if (totalFilesAnalyzed === 0) {
+    return {
+      toolName: ToolName.PatternDetect,
+      score: 100,
+      rawMetrics: {
+        totalDuplicates: 0,
+        totalTokenCost: 0,
+        highImpactDuplicates: 0,
+        totalFilesAnalyzed: 0
+      },
+      factors: [],
+      recommendations: []
+    };
+  }
+  const duplicatesPerFile = totalDuplicates / totalFilesAnalyzed * 100;
+  const tokenWastePerFile = totalTokenCost / totalFilesAnalyzed;
+  const duplicatesPenalty = Math.min(60, duplicatesPerFile * 0.6);
+  const tokenPenalty = Math.min(40, tokenWastePerFile / 125);
+  const highImpactPenalty = highImpactDuplicates > 0 ? Math.min(15, highImpactDuplicates * 2 - 5) : -5;
+  const score = 100 - duplicatesPenalty - tokenPenalty - highImpactPenalty;
+  const finalScore = Math.max(0, Math.min(100, Math.round(score)));
+  const factors = [
+    {
+      name: "Duplication Density",
+      impact: -Math.round(duplicatesPenalty),
+      description: `${duplicatesPerFile.toFixed(1)} duplicates per 100 files`
+    },
+    {
+      name: "Token Waste",
+      impact: -Math.round(tokenPenalty),
+      description: `${Math.round(tokenWastePerFile)} tokens wasted per file`
+    }
+  ];
+  if (highImpactDuplicates > 0) {
+    factors.push({
+      name: "High-Impact Patterns",
+      impact: -Math.round(highImpactPenalty),
+      description: `${highImpactDuplicates} high-impact duplicates (>1000 tokens or >70% similar)`
+    });
+  } else {
+    factors.push({
+      name: "No High-Impact Patterns",
+      impact: 5,
+      description: "No severe duplicates detected"
+    });
+  }
+  const recommendations = [];
+  if (highImpactDuplicates > 0) {
+    const estimatedImpact = Math.min(15, highImpactDuplicates * 3);
+    recommendations.push({
+      action: `Deduplicate ${highImpactDuplicates} high-impact pattern${highImpactDuplicates > 1 ? "s" : ""}`,
+      estimatedImpact,
+      priority: "high"
+    });
+  }
+  if (totalDuplicates > 10 && duplicatesPerFile > 20) {
+    const estimatedImpact = Math.min(10, Math.round(duplicatesPenalty * 0.3));
+    recommendations.push({
+      action: "Extract common patterns into shared utilities",
+      estimatedImpact,
+      priority: "medium"
+    });
+  }
+  if (tokenWastePerFile > 2e3) {
+    const estimatedImpact = Math.min(8, Math.round(tokenPenalty * 0.4));
+    recommendations.push({
+      action: "Consolidate duplicated logic to reduce AI context waste",
+      estimatedImpact,
+      priority: totalTokenCost > 1e4 ? "high" : "medium"
+    });
+  }
+  const cfg = { ...DEFAULT_COST_CONFIG, ...costConfig };
+  const estimatedMonthlyCost = calculateMonthlyCost(totalTokenCost, cfg);
+  const issues = duplicates.map((d) => ({
+    severity: d.severity === "critical" ? "critical" : d.severity === "major" ? "major" : "minor"
+  }));
+  const productivityImpact = calculateProductivityImpact(issues);
+  return {
+    toolName: "pattern-detect",
+    score: finalScore,
+    rawMetrics: {
+      totalDuplicates,
+      totalTokenCost,
+      highImpactDuplicates,
+      totalFilesAnalyzed,
+      duplicatesPerFile: Math.round(duplicatesPerFile * 10) / 10,
+      tokenWastePerFile: Math.round(tokenWastePerFile),
+      // Business value metrics
+      estimatedMonthlyCost,
+      estimatedDeveloperHours: productivityImpact.totalHours
+    },
+    factors,
+    recommendations
+  };
+}
+
+export {
+  calculatePatternScore
+};