From be725ce21f450c67c30b5fda6ecab6a73f5a624a Mon Sep 17 00:00:00 2001 From: catlog22 Date: Sat, 20 Dec 2025 17:34:26 +0800 Subject: [PATCH] =?UTF-8?q?chore:=20=E5=88=A0=E9=99=A4=E4=B8=8D=E5=86=8D?= =?UTF-8?q?=E4=BD=BF=E7=94=A8=E7=9A=84=20reindex=20=E5=92=8C=E6=B5=8B?= =?UTF-8?q?=E8=AF=95=E8=84=9A=E6=9C=AC?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- scripts/reindex-with-relationships.sh | 141 ------------------------ scripts/test-graph-analyzer.py | 153 -------------------------- 2 files changed, 294 deletions(-) delete mode 100644 scripts/reindex-with-relationships.sh delete mode 100644 scripts/test-graph-analyzer.py diff --git a/scripts/reindex-with-relationships.sh b/scripts/reindex-with-relationships.sh deleted file mode 100644 index 4f8e1937..00000000 --- a/scripts/reindex-with-relationships.sh +++ /dev/null @@ -1,141 +0,0 @@ -#!/bin/bash -# 重新索引项目以提取代码关系数据 -# 用于解决 Graph Explorer 显示为空的问题 - -set -e - -PROJECT_PATH="${1:-D:/Claude_dms3}" -INDEX_DIR="$HOME/.codexlens/indexes" - -# 规范化路径用于索引目录 -NORMALIZED_PATH=$(echo "$PROJECT_PATH" | sed 's|^/\([a-z]\)/|\U\1/|' | sed 's|^/||') -INDEX_DB_DIR="$INDEX_DIR/$NORMALIZED_PATH" -INDEX_DB="$INDEX_DB_DIR/_index.db" - -echo "==========================================" -echo "CodexLens 重新索引工具" -echo "==========================================" -echo "项目路径: $PROJECT_PATH" -echo "索引路径: $INDEX_DB" -echo "" - -# 检查数据库是否存在 -if [ ! -f "$INDEX_DB" ]; then - echo "❌ 索引数据库不存在: $INDEX_DB" - echo "请先运行: codex init $PROJECT_PATH" - exit 1 -fi - -# 检查当前数据统计 -echo "📊 当前数据统计：" -sqlite3 "$INDEX_DB" " -SELECT - '文件数: ' || (SELECT COUNT(*) FROM files) || - ' | 符号数: ' || (SELECT COUNT(*) FROM symbols) || - ' | 关系数: ' || (SELECT COUNT(*) FROM code_relationships); -" - -RELATIONSHIPS_COUNT=$(sqlite3 "$INDEX_DB" "SELECT COUNT(*) FROM code_relationships;") - -if [ "$RELATIONSHIPS_COUNT" -gt 0 ]; then - echo "" - echo "✅ 数据库已包含 $RELATIONSHIPS_COUNT 个代码关系" - echo "如果 Graph Explorer 仍然显示为空，请检查前端控制台错误" - exit 0 -fi - -echo "" -echo "⚠️ 检测到 code_relationships 表为空" -echo "" -echo "解决方案：" -echo "1. 备份现有索引（推荐）" -echo "2. 删除旧索引" -echo "3. 重新索引项目" -echo "" - -read -p "是否继续？这将删除并重建索引。(y/N) " -n 1 -r -echo -if [[ ! $REPLY =~ ^[Yy]$ ]]; then - echo "已取消" - exit 0 -fi - -# 1. 备份现有索引 -BACKUP_DIR="$INDEX_DB_DIR/backup_$(date +%Y%m%d_%H%M%S)" -echo "" -echo "📦 备份现有索引到: $BACKUP_DIR" -mkdir -p "$BACKUP_DIR" -cp "$INDEX_DB" "$BACKUP_DIR/" -echo "✓ 备份完成" - -# 2. 删除旧索引 -echo "" -echo "🗑️ 删除旧索引..." -rm -f "$INDEX_DB" -echo "✓ 已删除" - -# 3. 重新索引 -echo "" -echo "🔍 重新索引项目（这可能需要几分钟）..." -cd "$PROJECT_PATH" - -# 使用 CodexLens CLI 重新索引 -if command -v codex &> /dev/null; then - codex init . -else - echo "❌ 未找到 codex 命令" - echo "请先安装 CodexLens:" - echo " cd codex-lens" - echo " pip install -e ." - exit 1 -fi - -# 4. 验证结果 -echo "" -echo "📊 重新索引后的数据统计：" -sqlite3 "$INDEX_DB" " -SELECT - '文件数: ' || (SELECT COUNT(*) FROM files) || - ' | 符号数: ' || (SELECT COUNT(*) FROM symbols) || - ' | 关系数: ' || (SELECT COUNT(*) FROM code_relationships); -" - -RELATIONSHIPS_AFTER=$(sqlite3 "$INDEX_DB" "SELECT COUNT(*) FROM code_relationships;") - -echo "" -if [ "$RELATIONSHIPS_AFTER" -gt 0 ]; then - echo "✅ 成功！已提取 $RELATIONSHIPS_AFTER 个代码关系" - echo "" - echo "📋 示例关系：" - sqlite3 "$INDEX_DB" " - SELECT - s.name || ' --[' || r.relationship_type || ']--> ' || r.target_qualified_name - FROM code_relationships r - JOIN symbols s ON r.source_symbol_id = s.id - LIMIT 5; - " | head -5 - echo "" - echo "下一步：" - echo "1. 启动 CCW Dashboard: ccw view" - echo "2. 点击左侧边栏的 Graph 图标" - echo "3. 应该能看到代码关系图谱" -else - echo "⚠️ 警告：关系数据仍然为 0" - echo "" - echo "可能原因：" - echo "1. 项目中没有 Python/JavaScript/TypeScript 文件" - echo "2. TreeSitter 解析器未正确安装" - echo "3. 文件语法错误导致解析失败" - echo "" - echo "调试步骤：" - echo "1. 检查项目语言：" - sqlite3 "$INDEX_DB" "SELECT DISTINCT language FROM files LIMIT 10;" - echo "" - echo "2. 测试 GraphAnalyzer：" - echo " python -c 'from codexlens.semantic.graph_analyzer import GraphAnalyzer; print(GraphAnalyzer(\"python\").is_available())'" -fi - -echo "" -echo "==========================================" -echo "完成" -echo "==========================================" diff --git a/scripts/test-graph-analyzer.py b/scripts/test-graph-analyzer.py deleted file mode 100644 index 1714f937..00000000 --- a/scripts/test-graph-analyzer.py +++ /dev/null @@ -1,153 +0,0 @@ -#!/usr/bin/env python3 -""" -Test script to verify GraphAnalyzer is working correctly. -Checks if TreeSitter is available and can extract relationships from sample files. -""" -import sys -from pathlib import Path - -# Add codex-lens to path -sys.path.insert(0, str(Path(__file__).parent.parent / "codex-lens" / "src")) - -from codexlens.semantic.graph_analyzer import GraphAnalyzer -from codexlens.parsers.treesitter_parser import TreeSitterSymbolParser - -def test_graph_analyzer_availability(): - """Test if GraphAnalyzer is available for different languages.""" - print("=" * 60) - print("Testing GraphAnalyzer Availability") - print("=" * 60) - - languages = ["python", "javascript", "typescript"] - for lang in languages: - try: - analyzer = GraphAnalyzer(lang) - available = analyzer.is_available() - parser = TreeSitterSymbolParser(lang) - parser_available = parser.is_available() - - print(f"\n{lang.upper()}:") - print(f" GraphAnalyzer available: {available}") - print(f" TreeSitter parser available: {parser_available}") - - if not available: - print(f" [X] GraphAnalyzer NOT available for {lang}") - else: - print(f" [OK] GraphAnalyzer ready for {lang}") - except Exception as e: - print(f"\n{lang.upper()}:") - print(f" [ERROR] Error: {e}") - -def test_sample_file_analysis(file_path: Path): - """Test relationship extraction on a real file.""" - print("\n" + "=" * 60) - print(f"Testing File: {file_path.name}") - print("=" * 60) - - if not file_path.exists(): - print(f"[X] File not found: {file_path}") - return - - # Determine language - suffix = file_path.suffix - lang_map = { - '.py': 'python', - '.js': 'javascript', - '.ts': 'typescript', - '.tsx': 'typescript' - } - language = lang_map.get(suffix) - - if not language: - print(f"[X] Unsupported file type: {suffix}") - return - - print(f"Language: {language}") - - # Read file content - try: - content = file_path.read_text(encoding='utf-8') - print(f"File size: {len(content)} characters") - except Exception as e: - print(f"[X] Failed to read file: {e}") - return - - # Test parser first - try: - parser = TreeSitterSymbolParser(language) - if not parser.is_available(): - print(f"[X] TreeSitter parser not available for {language}") - return - - indexed_file = parser.parse(content, file_path) - symbols = indexed_file.symbols if indexed_file else [] - print(f"[OK] Parsed {len(symbols)} symbols") - - if symbols: - print("\nSample symbols:") - for i, sym in enumerate(symbols[:5], 1): - print(f" {i}. {sym.kind:10s} {sym.name:30s}") - except Exception as e: - print(f"[X] Symbol parsing failed: {e}") - import traceback - traceback.print_exc() - return - - # Test relationship extraction - try: - analyzer = GraphAnalyzer(language) - if not analyzer.is_available(): - print(f"[X] GraphAnalyzer not available for {language}") - return - - relationships = analyzer.analyze_with_symbols(content, file_path, symbols) - print(f"\n{'[OK]' if relationships else '[WARN]'} Extracted {len(relationships)} relationships") - - if relationships: - print("\nSample relationships:") - for i, rel in enumerate(relationships[:10], 1): - print(f" {i}. {rel.source_symbol:20s} --[{rel.relationship_type}]--> {rel.target_symbol} (line {rel.source_line})") - else: - print("\n[WARN] No relationships found") - print(" This could be normal if the file has no function calls") - print(" or if all calls are to external modules") - except Exception as e: - print(f"[X] Relationship extraction failed: {e}") - import traceback - traceback.print_exc() - -def main(): - """Run all tests.""" - # Test availability - test_graph_analyzer_availability() - - # Test on sample files from project - project_root = Path(__file__).parent.parent - - sample_files = [ - project_root / "ccw" / "src" / "core" / "routes" / "graph-routes.ts", - project_root / "codex-lens" / "src" / "codexlens" / "storage" / "dir_index.py", - project_root / "ccw" / "src" / "templates" / "dashboard-js" / "views" / "graph-explorer.js", - ] - - for sample_file in sample_files: - if sample_file.exists(): - test_sample_file_analysis(sample_file) - else: - print(f"\nSkipping non-existent file: {sample_file}") - - print("\n" + "=" * 60) - print("Test Summary") - print("=" * 60) - print("\nIf all tests passed:") - print(" [OK] GraphAnalyzer is working correctly") - print(" [OK] TreeSitter parsers are installed") - print("\nIf relationships were found:") - print(" [OK] Relationship extraction is functional") - print("\nNext steps:") - print(" 1. If no relationships found, check if files have function calls") - print(" 2. Re-run 'codex init' to re-index with relationship extraction") - print(" 3. Check database: sqlite3 ~/.codexlens/indexes/.../\\_index.db 'SELECT COUNT(*) FROM code_relationships;'") - -if __name__ == "__main__": - main()