@hupan56/wlkj 2.0.0

package/templates/qoder/scripts/add_session.py

@@ -0,0 +1,245 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+QODER Pipeline - 会话日志记录
+
+每次 AI 会话结束后, 记录一条日志到开发者的工作空间。
+日志文件按行数自动轮转 (默认 2000 行)。
+
+Usage:
+    python add_session.py --title "Title" --commit "hash" --summary "Summary"
+    python add_session.py --title "Title" --branch "feat/my-branch"
+
+日志文件: .qoder/workspace/<developer>/journal-N.md
+索引文件: .qoder/workspace/<developer>/index.md
+
+参考: Trellis 的 add_session.py 设计
+"""
+
+from __future__ import annotations
+
+import argparse
+import re
+import subprocess
+import sys
+import os
+from datetime import datetime
+from pathlib import Path
+
+sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
+
+from common.paths import (
+    FILE_JOURNAL_PREFIX,
+    get_repo_root,
+    get_developer,
+    get_workspace_dir,
+    get_active_journal_file,
+    count_lines,
+)
+from common.developer import ensure_developer
+
+
+# =============================================================================
+# 默认配置
+# =============================================================================
+
+MAX_JOURNAL_LINES = 2000  # 每个日志文件最大行数
+
+
+# =============================================================================
+# 核心函数
+# =============================================================================
+
+def get_latest_journal_info(dev_dir: Path) -> tuple[Path | None, int, int]:
+    """获取最新日志文件信息。
+
+    Returns:
+        (文件路径, 文件编号, 行数)
+    """
+    latest_file = None
+    latest_num = 0
+
+    for f in dev_dir.glob(f"{FILE_JOURNAL_PREFIX}*.md"):
+        if not f.is_file():
+            continue
+        match = re.search(r"(\d+)$", f.stem)
+        if match:
+            num = int(match.group(1))
+            if num > latest_num:
+                latest_num = num
+                latest_file = f
+
+    if latest_file:
+        lines = count_lines(latest_file)
+        return latest_file, latest_num, lines
+
+    return None, 0, 0
+
+
+def get_current_session(index_file: Path) -> int:
+    """从 index.md 读取当前会话编号。"""
+    if not index_file.is_file():
+        return 0
+
+    try:
+        content = index_file.read_text(encoding="utf-8")
+        for line in content.splitlines():
+            if "Total Sessions" in line:
+                match = re.search(r":\s*(\d+)", line)
+                if match:
+                    return int(match.group(1))
+    except (OSError, IOError):
+        pass
+    return 0
+
+
+def create_new_journal_file(
+    dev_dir: Path, num: int, developer: str, max_lines: int = MAX_JOURNAL_LINES
+) -> Path:
+    """创建新的日志文件。"""
+    prev_num = num - 1
+    new_file = dev_dir / f"{FILE_JOURNAL_PREFIX}{num}.md"
+    today = datetime.now().strftime("%Y-%m-%d")
+
+    content = f"""# Journal - {developer} (Part {num})
+
+> Continuation from `{FILE_JOURNAL_PREFIX}{prev_num}.md` (archived at ~{max_lines} lines)
+> Started: {today}
+
+---
+
+"""
+    new_file.write_text(content, encoding="utf-8")
+    return new_file
+
+
+def update_index(
+    index_file: Path,
+    developer: str,
+    session_num: int,
+    journal_file: str,
+    max_lines: int = MAX_JOURNAL_LINES,
+) -> None:
+    """更新工作空间索引文件 index.md。"""
+    today = datetime.now().strftime("%Y-%m-%d %H:%M")
+
+    # 生成日志表格
+    dev_dir = index_file.parent
+    table_rows = []
+    for f in sorted(dev_dir.glob(f"{FILE_JOURNAL_PREFIX}*.md")):
+        lines = count_lines(f)
+        status = "Active" if f.name == journal_file else "Archived"
+        table_rows.append(f"| `{f.name}` | ~{lines} | {status} |")
+
+    journals_table = "\n".join(table_rows)
+
+    index_content = f"""# Workspace Index - {developer}
+
+> Journal tracking for AI development sessions.
+
+---
+
+## Current Status
+
+<!-- @@@auto:current-status -->
+- **Active File**: `{journal_file}`
+- **Total Sessions**: {session_num}
+- **Last Active**: {today}
+<!-- @@@/auto:current-status -->
+
+---
+
+## Journals
+
+| File | Lines | Status |
+|------|-------|--------|
+<!-- @@@auto:journals-table -->
+{journals_table}
+<!-- @@@/auto:journals-table -->
+
+---
+
+*Auto-managed by QODER Pipeline*
+"""
+    index_file.write_text(index_content, encoding="utf-8")
+
+
+# =============================================================================
+# Main
+# =============================================================================
+
+def main() -> int:
+    """CLI 入口。"""
+    parser = argparse.ArgumentParser(description="Record AI session to journal")
+    parser.add_argument("--title", required=True, help="Session title")
+    parser.add_argument("--summary", default="", help="Session summary")
+    parser.add_argument("--commit", default="", help="Git commit hash")
+    parser.add_argument("--branch", default="", help="Git branch (auto-detected if omitted)")
+    args = parser.parse_args()
+
+    repo_root = get_repo_root()
+    developer = ensure_developer(repo_root)
+
+    workspace = get_workspace_dir(repo_root)
+    if not workspace:
+        print("Error: No workspace directory", file=sys.stderr)
+        return 1
+
+    workspace.mkdir(parents=True, exist_ok=True)
+
+    # 获取 Git 分支
+    branch = args.branch
+    if not branch:
+        try:
+            result = subprocess.run(
+                ["git", "branch", "--show-current"],
+                capture_output=True, text=True, cwd=str(repo_root)
+            )
+            branch = result.stdout.strip()
+        except Exception:
+            branch = "unknown"
+
+    # 获取最新日志文件
+    journal_file, journal_num, lines = get_latest_journal_info(workspace)
+
+    # 如果没有日志文件或已满, 创建新的
+    if not journal_file or lines >= MAX_JOURNAL_LINES:
+        journal_num = journal_num + 1 if journal_file else 1
+        journal_file = create_new_journal_file(workspace, journal_num, developer)
+        lines = 0
+
+    # 读取当前会话编号
+    index_file = workspace / "index.md"
+    session_num = get_current_session(index_file) + 1
+
+    # 追加会话记录
+    today = datetime.now().strftime("%Y-%m-%d %H:%M")
+    commit_line = f"- **Commit**: `{args.commit}`\n" if args.commit else ""
+    branch_line = f"- **Branch**: `{branch}`\n" if branch else ""
+
+    entry = f"""### Session {session_num}: {args.title}
+
+- **Date**: {today}
+- **Developer**: {developer}
+{commit_line}{branch_line}
+**Summary**: {args.summary or 'No summary provided.'}
+
+---
+
+"""
+    try:
+        with open(journal_file, "a", encoding="utf-8") as f:
+            f.write(entry)
+    except (OSError, IOError) as e:
+        print(f"Error: Failed to append to journal: {e}", file=sys.stderr)
+        return 1
+
+    # 更新索引
+    update_index(index_file, developer, session_num, journal_file.name)
+
+    print(f"Session {session_num} recorded in {journal_file.name}")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

package/templates/qoder/scripts/benchmark.py

@@ -0,0 +1,209 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+benchmark.py - QODER 流水线性能基准 (性能优化 A7)
+
+测量关键操作的延迟和内存, 供优化前后对比。
+
+Usage:
+    python benchmark.py                    # 跑全部基准
+    python benchmark.py --search           # 只测搜索
+    python benchmark.py --build            # 只测构建 (耗时)
+    python benchmark.py --json             # 输出 JSON (供对比)
+
+指标:
+    - search_cold: 冷启动搜索 (Python 启动 + JSON 加载 + 扫描)
+    - search_warm: 热搜索 (倒排索引缓存命中)
+    - index_load: keyword-index.json 加载时间
+    - inverted_build: 倒排索引构建时间
+    - inverted_cache_load: 缓存加载时间
+    - cnmap_lookup: CN_MAP 查询 (含 auto-expand)
+"""
+
+import argparse
+import json
+import os
+import sys
+import time
+from datetime import datetime
+
+try:
+    sys.stdout.reconfigure(encoding='utf-8', errors='replace')
+except (AttributeError, TypeError, OSError):
+    pass
+
+THIS_DIR = os.path.dirname(os.path.abspath(__file__))  # .qoder/scripts
+sys.path.insert(0, THIS_DIR)
+# BASE = repo root = scripts 的上两级 (.qoder/scripts -> .qoder -> repo)
+BASE = os.path.dirname(os.path.dirname(THIS_DIR))
+
+INDEX_DIR = os.path.join(BASE, 'data', 'index')
+
+
+def measure(fn, repeat=1):
+    """测函数执行时间, 返回 (avg_ms, min_ms, max_ms)。"""
+    times = []
+    for _ in range(repeat):
+        t0 = time.perf_counter()
+        fn()
+        times.append((time.perf_counter() - t0) * 1000)
+    return sum(times) / len(times), min(times), max(times)
+
+
+def bench_search():
+    """搜索基准: 冷启动 + 热搜索对比。"""
+    results = {}
+
+    # 1. JSON 加载 (冷)
+    ki_path = os.path.join(INDEX_DIR, 'keyword-index.json')
+    def load_ki():
+        with open(ki_path, encoding='utf-8') as f:
+            return json.load(f)
+    results['index_load_ms'] = measure(load_ki)[0]
+    ki = load_ki()
+    results['keyword_count'] = len(ki)
+
+    # 2. 全扫描 (旧逻辑)
+    from search_index import _match_keyword
+    queries = ['insurance', 'attendance', 'salary', 'vehicle', 'user']
+    def full_scan():
+        for q in queries:
+            for kw in ki:
+                if _match_keyword(q, kw):
+                    pass
+    results['full_scan_5q_ms'] = measure(full_scan, repeat=5)[0] / 5
+
+    # 3. 倒排索引构建 (首次)
+    from common.search_engine import get_inverted_index, search_keywords_fast, build_inverted_index
+    # 清缓存测构建
+    cache_path = os.path.join(INDEX_DIR, '.inverted-cache.json')
+    if os.path.exists(cache_path):
+        os.remove(cache_path)
+    results['inverted_build_ms'] = measure(lambda: get_inverted_index(BASE))[0]
+
+    # 4. 缓存加载 (热)
+    results['inverted_cache_load_ms'] = measure(lambda: get_inverted_index(BASE), repeat=5)[0]
+
+    # 5. 倒排搜索 (热)
+    inverted = get_inverted_index(BASE)
+    def inverted_search():
+        for q in queries:
+            search_keywords_fast([q], inverted, ki)
+    results['inverted_scan_5q_ms'] = measure(inverted_search, repeat=5)[0] / 5
+
+    # 6. 提速比
+    if results['inverted_scan_5q_ms'] > 0:
+        results['speedup'] = round(results['full_scan_5q_ms'] / results['inverted_scan_5q_ms'], 1)
+    else:
+        results['speedup'] = float('inf')
+
+    return results
+
+
+def bench_cnmap():
+    """CN_MAP 查询基准。"""
+    from common.terms import CN_MAP, BUSINESS_PATH_MAP, get_cn_map_with_auto
+    results = {}
+    results['cn_map_static_count'] = len(CN_MAP)
+    results['business_path_map_count'] = len(BUSINESS_PATH_MAP)
+
+    # auto-expand (含 PRD 扫描)
+    def auto_expand():
+        return get_cn_map_with_auto()
+    results['auto_expand_ms'] = measure(auto_expand)[0]
+    results['cn_map_total_count'] = len(get_cn_map_with_auto())
+
+    # 查询扩展
+    test_terms = ['考勤', '保险', '库存', '采购', '发票']
+    def expand_query():
+        out = []
+        for t in test_terms:
+            if t in CN_MAP:
+                out.extend(CN_MAP[t].split())
+        return out
+    results['query_expand_5q_ms'] = measure(expand_query, repeat=10)[0] / 10
+
+    return results
+
+
+def bench_index_sizes():
+    """索引文件大小。"""
+    results = {}
+    for name in ['keyword-index.json', 'api-index.json', 'style-index.json',
+                 'prd-index.json', 'module-map.json', 'prd-code-map.json']:
+        p = os.path.join(INDEX_DIR, name)
+        if os.path.isfile(p):
+            results[name] = os.path.getsize(p)
+    # 缓存文件
+    for name in ['.inverted-cache.json', '.file-keys.json']:
+        p = os.path.join(INDEX_DIR, name)
+        if os.path.isfile(p):
+            results[name] = os.path.getsize(p)
+    return results
+
+
+def main():
+    parser = argparse.ArgumentParser(description='QODER 性能基准')
+    parser.add_argument('--search', action='store_true', help='只测搜索')
+    parser.add_argument('--build', action='store_true', help='测构建 (耗时)')
+    parser.add_argument('--json', action='store_true', help='JSON 输出')
+    args = parser.parse_args()
+
+    report = {
+        'timestamp': datetime.now().isoformat(timespec='seconds'),
+        'python': sys.version.split()[0],
+        'platform': sys.platform,
+        'base': os.path.basename(BASE),
+    }
+
+    if not args.build:
+        print('=' * 50)
+        print('QODER 性能基准')
+        print('=' * 50)
+
+    if not args.build or args.search:
+        if not args.json:
+            print('\n--- 搜索性能 ---')
+        sr = bench_search()
+        report['search'] = sr
+        if not args.json:
+            print('  keyword-index 加载: %.0f ms (%d keys)' % (sr['index_load_ms'], sr['keyword_count']))
+            print('  全扫描 (5 query): %.2f ms' % sr['full_scan_5q_ms'])
+            print('  倒排索引构建: %.0f ms (首次)' % sr['inverted_build_ms'])
+            print('  缓存加载: %.2f ms (热)' % sr['inverted_cache_load_ms'])
+            print('  倒排搜索 (5 query): %.2f ms' % sr['inverted_scan_5q_ms'])
+            print('  提速: %.1fx' % sr['speedup'])
+
+        if not args.json:
+            print('\n--- CN_MAP ---')
+        cr = bench_cnmap()
+        report['cnmap'] = cr
+        if not args.json:
+            print('  静态词: %d, 路径映射: %d' % (cr['cn_map_static_count'], cr['business_path_map_count']))
+            print('  auto-expand: %.0f ms, 总计 %d 词' % (cr['auto_expand_ms'], cr['cn_map_total_count']))
+
+    if not args.build:
+        if not args.json:
+            print('\n--- 索引大小 ---')
+        sz = bench_index_sizes()
+        report['index_sizes'] = sz
+        if not args.json:
+            for name, size in sorted(sz.items(), key=lambda x: -x[1]):
+                print('  %-30s %s' % (name, format_size(size)))
+
+    if args.json:
+        print(json.dumps(report, indent=2, ensure_ascii=False))
+    else:
+        print('\n' + '=' * 50)
+
+
+def format_size(n):
+    for unit in ['B', 'KB', 'MB']:
+        if n < 1024:
+            return '%.1f %s' % (n, unit)
+        n /= 1024
+    return '%.1f GB' % n
+
+
+if __name__ == '__main__':
+    main()