@aiready/pattern-detect 0.17.15 → 0.17.16

package/dist/chunk-KMAOEVRS.mjs

@@ -0,0 +1,150 @@
+import {
+  calculateSeverity
+} from "./chunk-FSXOU23F.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(/\/\/.*$/gm, "").replace(/\/\*[\s\S]*?\*\//g, "");
+  }
+  return normalized.replace(/"[^"]*"/g, '"STR"').replace(/'[^']*'/g, "'STR'").replace(/`[^`]*`/g, "`STR`").replace(/\b\d+\b/g, "NUM").replace(/\s+/g, " ").trim().toLowerCase();
+}
+
+// src/detector.ts
+function extractBlocks(file, content) {
+  return extractCodeBlocks(file, content);
+}
+function calculateSimilarity(a, b) {
+  return calculateStringSimilarity(a, b);
+}
+function calculateConfidence(similarity, tokens, lines) {
+  return calculateHeuristicConfidence(similarity, tokens, lines);
+}
+async function detectDuplicatePatterns(fileContents, options) {
+  const {
+    minSimilarity,
+    minLines,
+    streamResults,
+    onProgress,
+    excludePatterns = [],
+    excludeFiles = [],
+    confidenceThreshold = 0,
+    ignoreWhitelist = []
+  } = options;
+  const allBlocks = [];
+  const excludeRegexes = (excludePatterns || []).map((p) => new RegExp(p, "i"));
+  const excludeFileRegexes = (options.excludeFiles || []).map((f) => {
+    const escaped = f.replace(/[.+^${}()|[\]\\]/g, "\\$&");
+    const regexStr = escaped.replace(/\*\*/g, ".*").replace(/\*/g, "[^/]*");
+    return new RegExp(`^${regexStr}$|${regexStr}`, "i");
+  });
+  for (const { file, content } of fileContents) {
+    if (excludeFileRegexes.some((regex) => regex.test(file))) continue;
+    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-NWG2ZIGX.mjs

@@ -0,0 +1,146 @@
+import {
+  calculateSeverity
+} from "./chunk-FSXOU23F.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(/\/\/.*$/gm, "").replace(/\/\*[\s\S]*?\*\//g, "");
+  }
+  return normalized.replace(/"[^"]*"/g, '"STR"').replace(/'[^']*'/g, "'STR'").replace(/`[^`]*`/g, "`STR`").replace(/\b\d+\b/g, "NUM").replace(/\s+/g, " ").trim().toLowerCase();
+}
+
+// src/detector.ts
+function extractBlocks(file, content) {
+  return extractCodeBlocks(file, content);
+}
+function calculateSimilarity(a, b) {
+  return calculateStringSimilarity(a, b);
+}
+function calculateConfidence(similarity, tokens, lines) {
+  return calculateHeuristicConfidence(similarity, tokens, lines);
+}
+async function detectDuplicatePatterns(fileContents, options) {
+  const {
+    minSimilarity,
+    minLines,
+    streamResults,
+    onProgress,
+    excludePatterns = [],
+    confidenceThreshold = 0,
+    ignoreWhitelist = []
+  } = options;
+  const allBlocks = [];
+  const regexStr = (f) => f.replace(/[.+^${}()|[\]\\]/g, "\\$&").replace(/\\\*/g, "[^/]*").replace(/\[\^\/\]\*\[\^\/\]\*/g, ".*");
+  const excludeRegexes = excludePatterns.map(
+    (p) => new RegExp(`${regexStr(p)}$`, "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
+};