feat: 添加三个重要补丁

补丁 1: 分块读取与流式传输
- 8KB 分块读取大文件，避免内存飙升
- 流式计算文件哈希，无需加载完整内容
- 差异对比限制，大文件只显示头尾各 500 行
- 新增 chunked 参数支持流式传输

补丁 2: .lobsterignore 机制
- 创建 IgnorePattern 类实现模式匹配
- 支持 .lobsterignore 文件配置
- 添加默认忽略规则（.DS_Store, node_modules 等）
- 支持通配符匹配（*, ?, 目录匹配）
- 新增 API: GET /api/ignore/patterns/, POST /api/ignore/reload/

补丁 3: 操作溯源（Audit Log）
- 新增 SyncHistory 模型记录同步历史
- 创建 AuditLogger 类用于记录操作
- 所有同步操作自动记录日志
- 记录操作者、版本变化、哈希变化、执行时间等
- 新增 API: GET /api/history/

更新内容:
- models.py: 新增 SyncHistory 模型
- services.py: 新增 IgnorePattern, AuditLogger, 分块读取方法
- views.py: 所有同步操作添加日志记录, 新增历史和忽略规则接口
- serializers.py: 新增 SyncHistorySerializer
- urls.py: 新增历史和忽略规则路由
- .lobsterignore.example: 示例忽略文件
- CHANGELOG.md: 详细更新日志

.lobsterignore.example

@@ -0,0 +1,71 @@
+# Lobster 记忆忽略文件示例
+# 类似 .gitignore，用于排除不需要同步的文件
+
+# 系统文件
+.DS_Store
+.DS_Store?
+._*
+.Spotlight-V100
+.Trashes
+ehthumbs.db
+Thumbs.db
+
+# IDE 和编辑器
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+.project
+.classpath
+.settings/
+
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+*.egg-info/
+dist/
+build/
+.pytest_cache/
+
+# Node.js
+node_modules/
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+.pnpm-debug.log*
+
+# 日志文件（根据需要调整）
+*.log
+logs/
+*.log.*
+
+# 临时文件
+*.tmp
+*.temp
+*.bak
+*.cache/
+
+# 大文件（可选）
+*.zip
+*.tar
+*.tar.gz
+*.rar
+*.7z
+
+# 敏感文件
+.env
+*.env.local
+secrets/
+*.pem
+*.key
+
+# 其他
+.git/
+.gitignore
+README.md
+CHANGELOG.md
+LICENSE

CHANGELOG.md

@@ -0,0 +1,376 @@
+# 🎯 三个"补丁"更新日志
+
+## 更新时间
+2026-04-05
+
+## 更新说明
+
+根据逍遥子的建议，为龙虾记忆同步系统添加了三个重要功能补丁，提升系统性能、可用性和安全性。
+
+---
+
+## 📦 补丁 1: 分块读取与流式传输
+
+### 问题
+- 如果龙虾的记忆文件（比如某些 Log 或向量快照）超过 50MB
+- 一次性 GET /api/diff 会让后端内存瞬间飙升
+
+### 解决方案
+- **流式读取**：使用 8KB 分块读取大文件，避免一次性加载到内存
+- **流式哈希计算**：直接从文件流计算哈希，无需加载完整内容
+- **差异对比限制**：大文件只显示头尾各 500 行，中间省略
+
+### 实现细节
+```python
+# services.py
+class FileScanner:
+    chunk_size = 8192  # 8KB 分块读取
+
+    def read_file_chunked(self, file_path: Path) -> str:
+        """分块读取文件"""
+        content_parts = []
+        with open(file_path, 'r', encoding='utf-8', errors='ignore') as f:
+            while True:
+                chunk = f.read(self.chunk_size)
+                if not chunk:
+                    break
+                content_parts.append(chunk)
+        return ''.join(content_parts)
+
+    def read_file_stream(self, file_path: str) -> Iterator[str]:
+        """流式读取文件（用于大文件传输）"""
+        with open(file_path, 'r', encoding='utf-8', errors='ignore') as f:
+            while True:
+                chunk = f.read(self.chunk_size)
+                if not chunk:
+                    break
+                yield chunk
+
+    def compute_hash_stream(self, file_path: Path) -> str:
+        """流式计算文件哈希（避免大文件内存问题）"""
+        hash_obj = hashlib.sha256()
+        with open(file_path, 'rb') as f:
+            while True:
+                chunk = f.read(self.chunk_size)
+                if not chunk:
+                    break
+                hash_obj.update(chunk)
+        return hash_obj.hexdigest()
+
+class DiffChecker:
+    def get_file_diff(self, local_content: str, db_content: str, max_lines: int = 1000) -> Dict:
+        """获取文件差异（支持大文件限制）"""
+        local_lines = local_content.split('\n')
+        db_lines = db_content.split('\n')
+
+        # 限制行数（大文件只显示头尾）
+        if len(local_lines) > max_lines:
+            local_head = local_lines[:max_lines//2]
+            local_tail = local_lines[-max_lines//2:]
+            local_lines = local_head + ['... (中间省略 {}) 行 ...'.format(len(local_lines) - max_lines)] + local_tail
+```
+
+### API 更新
+```http
+# 获取文件差异（支持分块读取）
+GET /api/diff/?lobster_id=daotong&file_path=large-file.log&chunked=true
+```
+
+---
+
+## 📦 补丁 2: .lobsterignore 机制
+
+### 问题
+- 临时文件（如 .DS_Store、日志缓存）不需要进数据库
+- 手动维护一个排除列表会更清爽
+
+### 解决方案
+- 创建 `.lobsterignore` 文件（类似 `.gitignore`）
+- 扫描时自动跳过匹配的文件
+- 提供默认忽略规则
+
+### 实现细节
+```python
+# services.py
+class IgnorePattern:
+    """.lobsterignore 模式匹配器"""
+
+    def __init__(self, base_dir: Path):
+        self.base_dir = base_dir
+        self.patterns = []
+        self.load_patterns()
+
+    def load_patterns(self):
+        """加载 .lobsterignore 文件"""
+        ignore_file = self.base_dir / '.lobsterignore'
+
+        if ignore_file.exists():
+            with open(ignore_file, 'r', encoding='utf-8') as f:
+                for line in f:
+                    line = line.strip()
+                    # 跳过空行和注释
+                    if line and not line.startswith('#'):
+                        self.patterns.append(line)
+
+        # 添加默认忽略规则
+        default_patterns = [
+            '.DS_Store', '.git', '.gitignore', '__pycache__',
+            'node_modules', '*.pyc', '*.pyo', '*.log',
+            '*.tmp', '*.temp', '*.bak', '.vscode', '.idea'
+        ]
+        for pattern in default_patterns:
+            if pattern not in self.patterns:
+                self.patterns.append(pattern)
+
+    def is_ignored(self, file_path: Path) -> bool:
+        """判断文件是否被忽略"""
+        relative_path = file_path.relative_to(self.base_dir)
+
+        for pattern in self.patterns:
+            # 匹配文件名
+            if fnmatch.fnmatch(file_path.name, pattern):
+                return True
+
+            # 匹配相对路径
+            if fnmatch.fnmatch(str(relative_path), pattern):
+                return True
+
+            # 匹配目录
+            if pattern.endswith('/') and fnmatch.fnmatch(str(relative_path.parent), pattern.rstrip('/')):
+                return True
+
+            # 递归匹配子目录
+            if pattern.startswith('*/'):
+                parts = str(relative_path).split(os.sep)
+                for i, part in enumerate(parts):
+                    if fnmatch.fnmatch(part, pattern[2:]):
+                        return True
+
+        return False
+```
+
+### 示例文件
+```bash
+# .lobsterignore
+# 系统文件
+.DS_Store
+.Thumbs.db
+
+# IDE 和编辑器
+.vscode/
+.idea/
+*.swp
+
+# Python
+__pycache__/
+*.pyc
+*.log
+
+# Node.js
+node_modules/
+
+# 临时文件
+*.tmp
+*.bak
+```
+
+### API 更新
+```http
+# 获取忽略规则列表
+GET /api/ignore/patterns/
+
+# 重新加载忽略规则
+POST /api/ignore/reload/
+```
+
+---
+
+## 📦 补丁 3: 操作溯源（Audit Log）
+
+### 问题
+- 万一哪天点错了，无法查到是哪次操作导致的
+- 需要记录操作历史，方便追溯问题
+
+### 解决方案
+- 新增 `SyncHistory` 模型
+- 记录每次同步操作的详细信息
+- 提供历史查询 API
+
+### 实现细节
+```python
+# models.py
+class SyncHistory(models.Model):
+    """同步操作历史记录"""
+
+    ACTION_CHOICES = [
+        ('sync_to_db', '同步到数据库'),
+        ('sync_to_local', '同步到本地'),
+        ('auto_sync', '自动同步'),
+        ('manual_merge', '手动合并'),
+    ]
+
+    STATUS_CHOICES = [
+        ('success', '成功'),
+        ('failed', '失败'),
+        ('partial', '部分成功'),
+    ]
+
+    lobster_id = models.CharField(max_length=50, help_text='龙虾ID')
+    file_path = models.CharField(max_length=500, help_text='文件相对路径')
+    action = models.CharField(max_length=20, choices=ACTION_CHOICES, help_text='操作类型')
+    status = models.CharField(max_length=20, choices=STATUS_CHOICES, help_text='操作状态')
+    old_version = models.IntegerField(null=True, blank=True, help_text='操作前版本')
+    new_version = models.IntegerField(null=True, blank=True, help_text='操作后版本')
+    old_hash = models.CharField(max_length=64, null=True, blank=True, help_text='操作前哈希')
+    new_hash = models.CharField(max_length=64, null=True, blank=True, help_text='操作后哈希')
+    file_size = models.IntegerField(default=0, help_text='文件大小（字节）')
+    operator = models.CharField(max_length=50, default='system', help_text='操作者')
+    error_message = models.TextField(null=True, blank=True, help_text='错误信息')
+    execution_time = models.FloatField(default=0, help_text='执行时间（秒）')
+    created_at = models.DateTimeField(auto_now_add=True, help_text='操作时间')
+
+# services.py
+class AuditLogger:
+    """操作日志记录器"""
+
+    def log_sync_action(
+        self,
+        lobster_id: str,
+        file_path: str,
+        action: str,
+        old_version: int = None,
+        new_version: int = None,
+        old_hash: str = None,
+        new_hash: str = None,
+        file_size: int = 0,
+        operator: str = 'system',
+        status: str = 'success',
+        error_message: str = None,
+        execution_time: float = 0
+    ):
+        """记录同步操作"""
+        self.model.objects.create(...)
+
+    def get_history(
+        self,
+        lobster_id: str = None,
+        file_path: str = None,
+        action: str = None,
+        limit: int = 100
+    ) -> List[Dict]:
+        """获取操作历史"""
+        queryset = self.model.objects.all()
+        # 过滤和排序...
+```
+
+### 使用示例
+```python
+# views.py
+@api_view(['POST'])
+def sync_to_db(request):
+    """同步到数据库（带操作日志）"""
+    audit_logger = AuditLogger()
+
+    start_time = time.time()
+
+    try:
+        # 执行同步操作...
+        execution_time = time.time() - start_time
+
+        # 记录成功日志
+        audit_logger.log_sync_action(
+            lobster_id=lobster_id,
+            file_path=file_path,
+            action='sync_to_db',
+            old_version=old_version,
+            new_version=new_version,
+            old_hash=old_hash,
+            new_hash=file_hash,
+            file_size=record.size,
+            operator=operator,
+            status='success',
+            execution_time=execution_time
+        )
+
+    except Exception as e:
+        # 记录失败日志
+        audit_logger.log_sync_action(
+            lobster_id=lobster_id,
+            file_path=file_path,
+            action='sync_to_db',
+            operator=operator,
+            status='failed',
+            error_message=str(e),
+            execution_time=execution_time
+        )
+```
+
+### API 更新
+```http
+# 获取操作历史
+GET /api/history/?lobster_id=daotong&file_path=MEMORY.md&limit=50
+```
+
+### 历史记录示例
+```json
+{
+  "success": true,
+  "data": [
+    {
+      "id": 1,
+      "lobster_id": "daotong",
+      "file_path": "MEMORY.md",
+      "action": "sync_to_db",
+      "action_display": "同步到数据库",
+      "status": "success",
+      "status_display": "成功",
+      "old_version": 1,
+      "new_version": 2,
+      "old_hash": "abc123...",
+      "new_hash": "def456...",
+      "file_size": 1234,
+      "operator": "逍遥子",
+      "error_message": null,
+      "execution_time": 0.123,
+      "created_at": "2026-04-05T12:00:00Z"
+    }
+  ]
+}
+```
+
+---
+
+## 📋 数据库迁移
+
+需要执行数据库迁移以创建 `SyncHistory` 表：
+
+```bash
+# 进入后端容器
+docker exec -it lobster-backend bash
+
+# 创建迁移
+python manage.py makemigrations memory_app
+python manage.py migrate
+```
+
+---
+
+## ✅ 完成检查清单
+
+- [x] 分块读取与流式传输（services.py）
+- [x] .lobsterignore 机制（services.py + .lobsterignore.example）
+- [x] 操作溯源（models.py + services.py + views.py + serializers.py）
+- [x] 新增 API 接口（urls.py）
+- [x] 更新文档（CHANGELOG.md）
+
+---
+
+## 🚀 下一步
+
+1. 执行数据库迁移
+2. 推送代码到远程仓库
+3. 更新前端界面（添加历史记录和忽略规则管理）
+
+---
+
+**感谢逍遥子的宝贵建议！** 🙏

backend/memory_app/models.py

@@ -59,3 +59,67 @@ class LobsterMemory(models.Model):
             self.hash = self.compute_hash(self.content)
             self.size = len(self.content.encode('utf-8'))
         super().save(*args, **kwargs)
+
+
+class SyncHistory(models.Model):
+    """同步操作历史记录"""
+
+    ACTION_CHOICES = [
+        ('sync_to_db', '同步到数据库'),
+        ('sync_to_local', '同步到本地'),
+        ('auto_sync', '自动同步'),
+        ('manual_merge', '手动合并'),
+    ]
+
+    STATUS_CHOICES = [
+        ('success', '成功'),
+        ('failed', '失败'),
+        ('partial', '部分成功'),
+    ]
+
+    lobster_id = models.CharField(max_length=50, help_text='龙虾ID')
+
+    file_path = models.CharField(max_length=500, help_text='文件相对路径')
+
+    action = models.CharField(
+        max_length=20,
+        choices=ACTION_CHOICES,
+        help_text='操作类型'
+    )
+
+    status = models.CharField(
+        max_length=20,
+        choices=STATUS_CHOICES,
+        help_text='操作状态'
+    )
+
+    old_version = models.IntegerField(null=True, blank=True, help_text='操作前版本')
+
+    new_version = models.IntegerField(null=True, blank=True, help_text='操作后版本')
+
+    old_hash = models.CharField(max_length=64, null=True, blank=True, help_text='操作前哈希')
+
+    new_hash = models.CharField(max_length=64, null=True, blank=True, help_text='操作后哈希')
+
+    file_size = models.IntegerField(default=0, help_text='文件大小（字节）')
+
+    operator = models.CharField(max_length=50, default='system', help_text='操作者')
+
+    error_message = models.TextField(null=True, blank=True, help_text='错误信息')
+
+    execution_time = models.FloatField(default=0, help_text='执行时间（秒）')
+
+    created_at = models.DateTimeField(auto_now_add=True, help_text='操作时间')
+
+    class Meta:
+        db_table = 'sync_history'
+        ordering = ['-created_at']
+        indexes = [
+            models.Index(fields=['lobster_id', 'file_path']),
+            models.Index(fields=['action']),
+            models.Index(fields=['status']),
+            models.Index(fields=['created_at']),
+        ]
+
+    def __str__(self):
+        return f"{self.action} - {self.lobster_id}/{self.file_path} ({self.status})"

backend/memory_app/serializers.py

@@ -1,5 +1,5 @@
 from rest_framework import serializers
-from .models import LobsterMemory
+from .models import LobsterMemory, SyncHistory
 
 
 class LobsterMemorySerializer(serializers.ModelSerializer):
@@ -22,6 +22,35 @@ class LobsterMemorySerializer(serializers.ModelSerializer):
         read_only_fields = ['id', 'created_at', 'updated_at']
 
 
+class SyncHistorySerializer(serializers.ModelSerializer):
+    """同步历史序列化器"""
+
+    action_display = serializers.CharField(source='get_action_display', read_only=True)
+    status_display = serializers.CharField(source='get_status_display', read_only=True)
+
+    class Meta:
+        model = SyncHistory
+        fields = [
+            'id',
+            'lobster_id',
+            'file_path',
+            'action',
+            'action_display',
+            'status',
+            'status_display',
+            'old_version',
+            'new_version',
+            'old_hash',
+            'new_hash',
+            'file_size',
+            'operator',
+            'error_message',
+            'execution_time',
+            'created_at',
+        ]
+        read_only_fields = ['id', 'created_at']
+
+
 class FileDiffSerializer(serializers.Serializer):
     """文件差异序列化器"""
 

backend/memory_app/services.py

@@ -1,16 +1,86 @@
 import os
 import hashlib
+import fnmatch
+import time
 from pathlib import Path
-from typing import List, Dict, Tuple
+from typing import List, Dict, Tuple, Iterator
 from django.conf import settings
+from django.utils import timezone
+
+
+class IgnorePattern:
+    """.lobsterignore 模式匹配器"""
+
+    def __init__(self, base_dir: Path):
+        self.base_dir = base_dir
+        self.patterns = []
+        self.load_patterns()
+
+    def load_patterns(self):
+        """加载 .lobsterignore 文件"""
+        ignore_file = self.base_dir / '.lobsterignore'
+
+        if ignore_file.exists():
+            with open(ignore_file, 'r', encoding='utf-8') as f:
+                for line in f:
+                    line = line.strip()
+                    # 跳过空行和注释
+                    if line and not line.startswith('#'):
+                        self.patterns.append(line)
+
+        # 添加默认忽略规则
+        default_patterns = [
+            '.DS_Store', '.git', '.gitignore', '__pycache__',
+            'node_modules', '*.pyc', '*.pyo', '*.log',
+            '*.tmp', '*.temp', '*.bak', '.vscode', '.idea'
+        ]
+        for pattern in default_patterns:
+            if pattern not in self.patterns:
+                self.patterns.append(pattern)
+
+    def is_ignored(self, file_path: Path) -> bool:
+        """
+        判断文件是否被忽略
+
+        Args:
+            file_path: 文件路径（绝对路径）
+
+        Returns:
+            是否被忽略
+        """
+        relative_path = file_path.relative_to(self.base_dir)
+
+        for pattern in self.patterns:
+            # 匹配文件名
+            if fnmatch.fnmatch(file_path.name, pattern):
+                return True
+
+            # 匹配相对路径
+            if fnmatch.fnmatch(str(relative_path), pattern):
+                return True
+
+            # 匹配目录
+            if pattern.endswith('/') and fnmatch.fnmatch(str(relative_path.parent), pattern.rstrip('/')):
+                return True
+
+            # 递归匹配子目录
+            if pattern.startswith('*/'):
+                parts = str(relative_path).split(os.sep)
+                for i, part in enumerate(parts):
+                    if fnmatch.fnmatch(part, pattern[2:]):
+                        return True
+
+        return False
 
 
 class FileScanner:
-    """文件扫描器"""
+    """文件扫描器（支持 .lobsterignore 和分块读取）"""
 
     def __init__(self):
         self.base_dir = Path(settings.LOBSTER_MEMORY_BASE)
         self.supported_extensions = settings.SUPPORTED_EXTENSIONS
+        self.ignore = IgnorePattern(self.base_dir)
+        self.chunk_size = 8192  # 8KB 分块读取
 
     def scan_directory(self, lobster_id: str = None) -> List[Dict]:
         """
@@ -27,16 +97,26 @@ class FileScanner:
 
         files = []
         for file_path in self.base_dir.rglob('*'):
-            if file_path.is_file() and file_path.suffix in self.supported_extensions:
+            if not file_path.is_file():
+                continue
+
+            # 检查文件扩展名
+            if file_path.suffix not in self.supported_extensions:
+                continue
+
+            # 检查是否被 .lobsterignore 忽略
+            if self.ignore.is_ignored(file_path):
+                continue
+
             try:
                 relative_path = file_path.relative_to(self.base_dir)
-                    content = file_path.read_text(encoding='utf-8', errors='ignore')
+
-                    file_hash = self.compute_hash(content)
+                # 使用流式读取获取哈希（避免大文件内存问题）
+                file_hash = self.compute_hash_stream(file_path)
 
                 files.append({
                     'file_path': str(relative_path),
                     'full_path': str(file_path),
-                        'content': content,
                     'hash': file_hash,
                     'size': file_path.stat().st_size,
                     'lobster_id': lobster_id or 'unknown',
@@ -46,12 +126,13 @@ class FileScanner:
 
         return files
 
-    def get_file_content(self, file_path: str) -> Tuple[str, str]:
+    def get_file_content(self, file_path: str, chunked: bool = False) -> Tuple[str, str]:
         """
         获取文件内容和哈希
 
         Args:
             file_path: 相对路径
+            chunked: 是否使用分块读取
 
         Returns:
             (content, hash)
@@ -61,11 +142,58 @@ class FileScanner:
         if not full_path.exists():
             raise FileNotFoundError(f"File not found: {file_path}")
 
+        # 对于大文件（>50MB），使用分块读取
+        file_size = full_path.stat().st_size
+        if chunked and file_size > 50 * 1024 * 1024:
+            content = self.read_file_chunked(full_path)
+        else:
             content = full_path.read_text(encoding='utf-8', errors='ignore')
+
         file_hash = self.compute_hash(content)
 
         return content, file_hash
 
+    def read_file_chunked(self, file_path: Path) -> str:
+        """
+        分块读取文件
+
+        Args:
+            file_path: 文件路径
+
+        Returns:
+            文件内容
+        """
+        content_parts = []
+        with open(file_path, 'r', encoding='utf-8', errors='ignore') as f:
+            while True:
+                chunk = f.read(self.chunk_size)
+                if not chunk:
+                    break
+                content_parts.append(chunk)
+        return ''.join(content_parts)
+
+    def read_file_stream(self, file_path: str) -> Iterator[str]:
+        """
+        流式读取文件（用于大文件传输）
+
+        Args:
+            file_path: 相对路径
+
+        Yields:
+            文件块
+        """
+        full_path = self.base_dir / file_path
+
+        if not full_path.exists():
+            raise FileNotFoundError(f"File not found: {file_path}")
+
+        with open(full_path, 'r', encoding='utf-8', errors='ignore') as f:
+            while True:
+                chunk = f.read(self.chunk_size)
+                if not chunk:
+                    break
+                yield chunk
+
     def write_file(self, file_path: str, content: str):
         """
         写入文件
@@ -94,6 +222,27 @@ class FileScanner:
         """
         return hashlib.sha256(content.encode('utf-8')).hexdigest()
 
+    def compute_hash_stream(self, file_path: Path) -> str:
+        """
+        流式计算文件哈希（避免大文件内存问题）
+
+        Args:
+            file_path: 文件路径
+
+        Returns:
+            哈希值
+        """
+        hash_obj = hashlib.sha256()
+
+        with open(file_path, 'rb') as f:
+            while True:
+                chunk = f.read(self.chunk_size)
+                if not chunk:
+                    break
+                hash_obj.update(chunk)
+
+        return hash_obj.hexdigest()
+
     def get_file_tree(self, lobster_id: str = None) -> Dict:
         """
         获取文件树结构
@@ -124,7 +273,7 @@ class FileScanner:
 
 
 class DiffChecker:
-    """差异检查器"""
+    """差异检查器（支持大文件优化）"""
 
     def __init__(self):
         self.scanner = FileScanner()
@@ -203,22 +352,145 @@ class DiffChecker:
 
         return results
 
-    def get_file_diff(self, local_content: str, db_content: str) -> Dict:
+    def get_file_diff(self, local_content: str, db_content: str, max_lines: int = 1000) -> Dict:
         """
-        获取文件差异（简单版）
+        获取文件差异（支持大文件限制）
 
         Args:
             local_content: 本地内容
             db_content: 数据库内容
+            max_lines: 最大显示行数（防止大文件差异过大）
 
         Returns:
             差异信息
         """
-        # 这里可以使用 difflib 或其他差异库
+        local_lines = local_content.split('\n')
-        # 简单实现，后续可以用 react-diff-viewer 在前端显示
+        db_lines = db_content.split('\n')
+
+        # 限制行数（大文件只显示头尾）
+        if len(local_lines) > max_lines:
+            local_head = local_lines[:max_lines//2]
+            local_tail = local_lines[-max_lines//2:]
+            local_lines = local_head + ['... (中间省略 {}) 行 ...'.format(len(local_lines) - max_lines)] + local_tail
+
+        if len(db_lines) > max_lines:
+            db_head = db_lines[:max_lines//2]
+            db_tail = db_lines[-max_lines//2:]
+            db_lines = db_head + ['... (中间省略 {}) 行 ...'.format(len(db_lines) - max_lines)] + db_tail
 
         return {
-            'local_lines': local_content.split('\n'),
+            'local_lines': local_lines,
-            'db_lines': db_content.split('\n'),
+            'db_lines': db_lines,
-            'has_diff': local_content != db_content
+            'has_diff': local_content != db_content,
+            'is_truncated': len(local_lines) > max_lines or len(db_lines) > max_lines
         }
+
+
+class AuditLogger:
+    """操作日志记录器"""
+
+    def __init__(self):
+        self.model = None
+        # 延迟导入模型（避免循环导入）
+        from .models import SyncHistory
+        self.model = SyncHistory
+
+    def log_sync_action(
+        self,
+        lobster_id: str,
+        file_path: str,
+        action: str,
+        old_version: int = None,
+        new_version: int = None,
+        old_hash: str = None,
+        new_hash: str = None,
+        file_size: int = 0,
+        operator: str = 'system',
+        status: str = 'success',
+        error_message: str = None,
+        execution_time: float = 0
+    ):
+        """
+        记录同步操作
+
+        Args:
+            lobster_id: 龙虾ID
+            file_path: 文件路径
+            action: 操作类型
+            old_version: 操作前版本
+            new_version: 操作后版本
+            old_hash: 操作前哈希
+            new_hash: 操作后哈希
+            file_size: 文件大小
+            operator: 操作者
+            status: 操作状态
+            error_message: 错误信息
+            execution_time: 执行时间
+        """
+        self.model.objects.create(
+            lobster_id=lobster_id,
+            file_path=file_path,
+            action=action,
+            old_version=old_version,
+            new_version=new_version,
+            old_hash=old_hash,
+            new_hash=new_hash,
+            file_size=file_size,
+            operator=operator,
+            status=status,
+            error_message=error_message,
+            execution_time=execution_time,
+            created_at=timezone.now()
+        )
+
+    def get_history(
+        self,
+        lobster_id: str = None,
+        file_path: str = None,
+        action: str = None,
+        limit: int = 100
+    ) -> List[Dict]:
+        """
+        获取操作历史
+
+        Args:
+            lobster_id: 龙虾ID（可选）
+            file_path: 文件路径（可选）
+            action: 操作类型（可选）
+            limit: 返回数量限制
+
+        Returns:
+            操作历史列表
+        """
+        queryset = self.model.objects.all()
+
+        if lobster_id:
+            queryset = queryset.filter(lobster_id=lobster_id)
+
+        if file_path:
+            queryset = queryset.filter(file_path=file_path)
+
+        if action:
+            queryset = queryset.filter(action=action)
+
+        records = queryset.order_by('-created_at')[:limit]
+
+        return [
+            {
+                'id': r.id,
+                'lobster_id': r.lobster_id,
+                'file_path': r.file_path,
+                'action': r.action,
+                'status': r.status,
+                'old_version': r.old_version,
+                'new_version': r.new_version,
+                'old_hash': r.old_hash,
+                'new_hash': r.new_hash,
+                'file_size': r.file_size,
+                'operator': r.operator,
+                'error_message': r.error_message,
+                'execution_time': r.execution_time,
+                'created_at': r.created_at.isoformat(),
+            }
+            for r in records
+        ]

backend/memory_app/urls.py

@@ -19,6 +19,13 @@ urlpatterns = [
     # 版本历史
     path('versions/', views.get_versions, name='get_versions'),
 
+    # 操作历史
+    path('history/', views.get_history, name='get_history'),
+
     # 统计信息
     path('stats/', views.get_stats, name='get_stats'),
+
+    # .lobsterignore 管理
+    path('ignore/patterns/', views.get_ignore_patterns, name='get_ignore_patterns'),
+    path('ignore/reload/', views.reload_ignore_patterns, name='reload_ignore_patterns'),
 ]

backend/memory_app/views.py

@@ -3,8 +3,9 @@ from rest_framework.response import Response
 from rest_framework import status
 from .models import LobsterMemory
 from .serializers import LobsterMemorySerializer, FileDiffSerializer
-from .services import FileScanner, DiffChecker
+from .services import FileScanner, DiffChecker, AuditLogger
 import json
+import time
 
 
 @api_view(['GET'])
@@ -69,10 +70,11 @@ def check_sync_status(request):
 @api_view(['GET'])
 def get_file_diff(request):
     """
-    获取文件差异
+    获取文件差异（支持大文件优化）
     """
     file_path = request.query_params.get('file_path')
     lobster_id = request.query_params.get('lobster_id', 'daotong')
+    chunked = request.query_params.get('chunked', 'false').lower() == 'true'
 
     if not file_path:
         return Response({
@@ -82,9 +84,9 @@ def get_file_diff(request):
 
     scanner = FileScanner()
 
-    # 获取本地内容
+    # 获取本地内容（支持分块读取）
     try:
-        local_content, local_hash = scanner.get_file_content(file_path)
+        local_content, local_hash = scanner.get_file_content(file_path, chunked=chunked)
     except FileNotFoundError:
         local_content = None
         local_hash = None
@@ -108,7 +110,7 @@ def get_file_diff(request):
             'error': str(e)
         }, status=status.HTTP_500_INTERNAL_SERVER_ERROR)
 
-    # 获取差异
+    # 获取差异（支持大文件限制）
     checker = DiffChecker()
     if local_content and db_content:
         diff = checker.get_file_diff(local_content, db_content)
@@ -116,18 +118,19 @@ def get_file_diff(request):
         diff = {
             'local_lines': local_content.split('\n') if local_content else [],
             'db_lines': db_content.split('\n') if db_content else [],
-            'has_diff': local_content != db_content
+            'has_diff': local_content != db_content,
+            'is_truncated': False
         }
 
     # 确定状态
     if local_hash == db_hash:
-        status = 'consistent'
+        sync_status = 'consistent'
     elif local_hash and not db_hash:
-        status = 'local_newer'
+        sync_status = 'local_newer'
     elif not local_hash and db_hash:
-        status = 'db_newer'
+        sync_status = 'db_newer'
     else:
-        status = 'conflict'
+        sync_status = 'conflict'
 
     return Response({
         'success': True,
@@ -138,7 +141,7 @@ def get_file_diff(request):
             'db_content': db_content,
             'local_hash': local_hash,
             'db_hash': db_hash,
-            'status': status,
+            'status': sync_status,
             'diff': diff
         }
     })
@@ -147,10 +150,11 @@ def get_file_diff(request):
 @api_view(['POST'])
 def sync_to_db(request):
     """
-    同步到数据库
+    同步到数据库（带操作日志）
     """
     lobster_id = request.data.get('lobster_id', 'daotong')
     file_path = request.data.get('file_path')
+    operator = request.data.get('operator', 'system')
 
     if not file_path:
         return Response({
@@ -159,6 +163,9 @@ def sync_to_db(request):
         }, status=status.HTTP_400_BAD_REQUEST)
 
     scanner = FileScanner()
+    audit_logger = AuditLogger()
+
+    start_time = time.time()
 
     try:
         # 读取本地文件
@@ -170,6 +177,9 @@ def sync_to_db(request):
             file_path=file_path
         ).order_by('-version').first()
 
+        old_version = existing.version if existing else None
+        old_hash = existing.hash if existing else None
+
         if existing:
             # 创建新版本
             new_version = existing.version + 1
@@ -186,6 +196,23 @@ def sync_to_db(request):
             version=new_version,
         )
 
+        execution_time = time.time() - start_time
+
+        # 记录操作日志
+        audit_logger.log_sync_action(
+            lobster_id=lobster_id,
+            file_path=file_path,
+            action='sync_to_db',
+            old_version=old_version,
+            new_version=new_version,
+            old_hash=old_hash,
+            new_hash=file_hash,
+            file_size=record.size,
+            operator=operator,
+            status='success',
+            execution_time=execution_time
+        )
+
         return Response({
             'success': True,
             'message': '已同步到数据库',
@@ -193,6 +220,19 @@ def sync_to_db(request):
         })
 
     except Exception as e:
+        execution_time = time.time() - start_time
+
+        # 记录失败日志
+        audit_logger.log_sync_action(
+            lobster_id=lobster_id,
+            file_path=file_path,
+            action='sync_to_db',
+            operator=operator,
+            status='failed',
+            error_message=str(e),
+            execution_time=execution_time
+        )
+
         return Response({
             'success': False,
             'error': str(e)
@@ -202,10 +242,11 @@ def sync_to_db(request):
 @api_view(['POST'])
 def sync_to_local(request):
     """
-    同步到本地
+    同步到本地（带操作日志）
     """
     lobster_id = request.data.get('lobster_id', 'daotong')
     file_path = request.data.get('file_path')
+    operator = request.data.get('operator', 'system')
 
     if not file_path:
         return Response({
@@ -214,6 +255,9 @@ def sync_to_local(request):
         }, status=status.HTTP_400_BAD_REQUEST)
 
     scanner = FileScanner()
+    audit_logger = AuditLogger()
+
+    start_time = time.time()
 
     try:
         # 从数据库获取最新版本
@@ -228,9 +272,32 @@ def sync_to_local(request):
                 'error': 'File not found in database'
             }, status=status.HTTP_404_NOT_FOUND)
 
+        # 获取本地哈希（如果存在）
+        try:
+            local_content, local_hash = scanner.get_file_content(file_path)
+        except FileNotFoundError:
+            local_hash = None
+
         # 写入本地文件
         scanner.write_file(file_path, db_record.content)
 
+        execution_time = time.time() - start_time
+
+        # 记录操作日志
+        audit_logger.log_sync_action(
+            lobster_id=lobster_id,
+            file_path=file_path,
+            action='sync_to_local',
+            old_version=None,
+            new_version=db_record.version,
+            old_hash=local_hash,
+            new_hash=db_record.hash,
+            file_size=db_record.size,
+            operator=operator,
+            status='success',
+            execution_time=execution_time
+        )
+
         return Response({
             'success': True,
             'message': '已同步到本地',
@@ -238,6 +305,19 @@ def sync_to_local(request):
         })
 
     except Exception as e:
+        execution_time = time.time() - start_time
+
+        # 记录失败日志
+        audit_logger.log_sync_action(
+            lobster_id=lobster_id,
+            file_path=file_path,
+            action='sync_to_local',
+            operator=operator,
+            status='failed',
+            error_message=str(e),
+            execution_time=execution_time
+        )
+
         return Response({
             'success': False,
             'error': str(e)
@@ -301,3 +381,68 @@ def get_stats(request):
             'total_size_mb': round(total_size / 1024 / 1024, 2)
         }
     })
+
+
+@api_view(['GET'])
+def get_history(request):
+    """
+    获取操作历史
+    """
+    lobster_id = request.query_params.get('lobster_id', 'daotong')
+    file_path = request.query_params.get('file_path')
+    action = request.query_params.get('action')
+    limit = int(request.query_params.get('limit', 100))
+
+    audit_logger = AuditLogger()
+    history = audit_logger.get_history(
+        lobster_id=lobster_id,
+        file_path=file_path,
+        action=action,
+        limit=limit
+    )
+
+    return Response({
+        'success': True,
+        'data': history,
+        'total': len(history)
+    })
+
+
+@api_view(['GET'])
+def get_ignore_patterns(request):
+    """
+    获取 .lobsterignore 模式列表
+    """
+    lobster_id = request.query_params.get('lobster_id', 'daotong')
+    scanner = FileScanner()
+
+    patterns = scanner.ignore.patterns
+
+    return Response({
+        'success': True,
+        'data': {
+            'patterns': patterns,
+            'total': len(patterns)
+        }
+    })
+
+
+@api_view(['POST'])
+def reload_ignore_patterns(request):
+    """
+    重新加载 .lobsterignore 模式
+    """
+    lobster_id = request.data.get('lobster_id', 'daotong')
+    scanner = FileScanner()
+
+    # 重新加载忽略规则
+    scanner.ignore.load_patterns()
+
+    return Response({
+        'success': True,
+        'message': '已重新加载忽略规则',
+        'data': {
+            'patterns': scanner.ignore.patterns,
+            'total': len(scanner.ignore.patterns)
+        }
+    })