@goshenkata/dryscan-core 1.0.0 → 1.0.8

package/README.md

@@ -0,0 +1,96 @@
+# @goshenkata/dryscan-core
+
+Core library for DryScan - semantic code duplication analyzer using embeddings.
+
+## Installation
+
+```bash
+npm install @goshenkata/dryscan-core
+```
+
+## Usage
+
+### Basic Example
+
+```typescript
+import { DryScan } from '@goshenkata/dryscan-core';
+
+const scanner = new DryScan('/path/to/your/repository');
+
+// Initialize repository index
+await scanner.init();
+
+// Build duplicate analysis report
+const report = await scanner.buildDuplicateReport();
+
+console.log(`Duplication Score: ${report.score.score}`);
+console.log(`Found ${report.duplicates.length} duplicate pairs`);
+```
+
+### API
+
+Initializes the repository with a 3-phase analysis:
+1. Extract and index all code units (functions, methods, blocks)
+2. Resolve internal dependencies
+3. Compute semantic embeddings
+
+#### `async updateIndex(): Promise<void>`
+
+Incrementally updates the index by detecting changed, new, and deleted files. Only reprocesses modified units for efficiency.
+
+```typescript
+await scanner.updateIndex();
+```
+
+#### `async buildDuplicateReport(): Promise<DuplicateReport>`
+
+Runs duplicate detection and returns a comprehensive report with similarity scores and duplication metrics.
+
+```typescript
+const report = await scanner.buildDuplicateReport();
+
+report.duplicates.forEach(dup => {
+  console.log(`${dup.similarity.toFixed(2)} - ${dup.left.name} ↔ ${dup.right.name}`);
+});
+```
+
+#### `async cleanExclusions(): Promise<{ removed: number; kept: number }>`
+
+Removes stale exclusion rules that no longer match any indexed units.
+
+```typescript
+const result = await scanner.cleanExclusions();
+console.log(`Cleaned ${result.removed} stale exclusions`);
+```
+
+### Configuration
+
+Place a `dryconfig.json` in your repository root:
+
+```json
+{
+  "threshold": 0.88,
+  "minLines": 5,
+  "minBlockLines": 8,
+  "embeddingModel": "embeddinggemma",
+  "embeddingSource": "http://localhost:11434",
+  "excludedPaths": ["**/test/**", "**/node_modules/**"],
+  "excludedPairs": []
+}
+```
+
+**Supported Embedding Providers:**
+- **Ollama** (default): Set `embeddingSource` to `"http://localhost:11434"` and `embeddingModel` to `"embeddinggemma"`
+- **Google Gemini**: Set `embeddingSource` to `"google"` and `embeddingModel` to `"gemini-embedding-001"` (requires `GOOGLE_API_KEY` env var)
+
+## Requirements
+
+- Node.js >= 18.0.0
+- Ollama running locally (default) or Google API key for embeddings
+
+## Supported languages**
+
+Just java for now
+## License
+
+MIT

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@goshenkata/dryscan-core",
-  "version": "1.0.0",
+  "version": "1.0.8",
   "description": "Core library for DryScan - semantic code duplication analyzer",
   "type": "module",
   "main": "./dist/index.js",
@@ -27,6 +27,13 @@
   ],
   "author": "Goshenkata",
   "license": "MIT",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/Goshenkata/DryScan.git"
+  },
+  "publishConfig": {
+    "access": "public"
+  },
   "devDependencies": {
     "@types/better-sqlite3": "^7.6.13",
     "@types/debug": "^4.1.12",
@@ -56,8 +63,5 @@
     "tree-sitter-python": "^0.25.0",
     "typeorm": "^0.3.28",
     "upath": "^2.0.1"
-  },
-  "publishConfig": {
-    "access": "public"
   }
 }

package/src/DryScan.ts

@@ -108,6 +108,7 @@ export class DryScan {
       version: 1,
       generatedAt: new Date().toISOString(),
       threshold: config.threshold,
+      grade: analysis.score.grade,
       score: analysis.score,
       duplicates: analysis.duplicates,
     };

package/src/types.ts

@@ -30,6 +30,7 @@ export interface DuplicateReport {
   version: number;
   generatedAt: string;
   threshold: number;
+  grade: DuplicationScore["grade"];
   score: DuplicationScore;
   duplicates: DuplicateGroup[];
 }