openclaw-memory/backend/memory_app/views.py

"""
龙虾记忆同步系统 - API 视图模块

集成所有核心功能：
- 分块与流式处理
- .lobsterignore 支持
- 审计日志
- 语义摘要
- 完善的冲突判定
"""

from rest_framework.decorators import api_view
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, AuditLogger, SemanticSummaryGenerator
)
import time


@api_view(['GET'])
def scan_files(request):
    """
    扫描本地文件

    自动应用 .lobsterignore 规则过滤不需要同步的文件
    使用流式哈希计算，避免大文件内存问题
    """
    lobster_id = request.query_params.get('lobster_id', 'daotong')
    scanner = FileScanner()

    files = scanner.scan_directory(lobster_id)

    return Response({
        'success': True,
        'data': files,
        'total': len(files)
    })


@api_view(['GET'])
def get_file_tree(request):
    """
    获取文件树结构

    展示所有未被 .lobsterignore 过滤的文件
    """
    lobster_id = request.query_params.get('lobster_id', 'daotong')
    scanner = FileScanner()

    tree = scanner.get_file_tree(lobster_id)

    return Response({
        'success': True,
        'data': tree
    })


@api_view(['GET'])
def check_sync_status(request):
    """
    检查同步状态（完善冲突判定）

    支持的状态：
    - consistent: 内容一致
    - local_newer: 只有本地存在
    - db_newer: 只有数据库存在
    - conflict: 两边都存在但哈希不同
    - hard_conflict: 严重冲突（版本 > 1 且 1 小时内更新）
    - local_only: 仅本地
    - db_only: 仅数据库
    """
    lobster_id = request.query_params.get('lobster_id', 'daotong')

    # 获取本地文件（应用 .lobsterignore）
    scanner = FileScanner()
    local_files = scanner.scan_directory(lobster_id)

    # 获取数据库文件
    db_files = list(LobsterMemory.objects.filter(
        lobster_id=lobster_id
    ).values('file_path', 'hash', 'version', 'updated_at'))

    # 检查同步状态（包含 HARD_CONFLICT 判定）
    checker = DiffChecker()
    sync_status = checker.check_sync_status(local_files, db_files)

    return Response({
        'success': True,
        'data': sync_status
    })


@api_view(['GET'])
def get_file_diff(request):
    """
    获取文件差异（支持大文件优化）

    使用 8KB 分块读取，计算变动行数
    """
    file_path = request.query_params.get('file_path')
    lobster_id = request.query_params.get('lobster_id', 'daotong')
    chunked = request.query_params.get('chunked', 'true').lower() == 'true'

    if not file_path:
        return Response({
            'success': False,
            'error': 'file_path is required'
        }, status=status.HTTP_400_BAD_REQUEST)

    scanner = FileScanner()

    # 获取本地内容（强制使用分块读取）
    try:
        local_content, local_hash = scanner.get_file_content(file_path, chunked=chunked)
    except FileNotFoundError:
        local_content = None
        local_hash = None

    # 获取数据库内容
    try:
        db_record = LobsterMemory.objects.filter(
            lobster_id=lobster_id,
            file_path=file_path
        ).order_by('-version').first()

        if db_record:
            db_content = db_record.content
            db_hash = db_record.hash
        else:
            db_content = None
            db_hash = None
    except Exception as e:
        return Response({
            'success': False,
            'error': str(e)
        }, status=status.HTTP_500_INTERNAL_SERVER_ERROR)

    # 获取差异（支持大文件限制，计算变动行数）
    checker = DiffChecker()
    if local_content and db_content:
        diff = checker.get_file_diff(local_content, db_content)
    else:
        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,
            'is_truncated': False,
            'lines_changed': 0
        }

    # 确定状态
    if local_hash == db_hash:
        sync_status = 'consistent'
    elif local_hash and not db_hash:
        sync_status = 'local_newer'
    elif not local_hash and db_hash:
        sync_status = 'db_newer'
    else:
        sync_status = 'conflict'

    return Response({
        'success': True,
        'data': {
            'file_path': file_path,
            'lobster_id': lobster_id,
            'local_content': local_content,
            'db_content': db_content,
            'local_hash': local_hash,
            'db_hash': db_hash,
            'status': sync_status,
            'diff': diff
        }
    })


@api_view(['POST'])
def sync_to_db(request):
    """
    同步到数据库（带完整审计日志）

    功能：
    - 使用分块读取文件
    - 生成语义摘要
    - 记录变动行数
    - 记录数据源、操作人、执行时间
    """
    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({
            'success': False,
            'error': 'file_path is required'
        }, status=status.HTTP_400_BAD_REQUEST)

    scanner = FileScanner()
    audit_logger = AuditLogger()
    summary_generator = SemanticSummaryGenerator()

    start_time = time.time()

    try:
        # 读取本地文件（使用分块读取）
        content, file_hash = scanner.get_file_content(file_path, chunked=True)

        # 查找现有记录
        existing = LobsterMemory.objects.filter(
            lobster_id=lobster_id,
            file_path=file_path
        ).order_by('-version').first()

        old_version = existing.version if existing else None
        old_hash = existing.hash if existing else None
        old_content = existing.content if existing else None

        # 计算变动行数
        lines_changed = 0
        if old_content:
            checker = DiffChecker()
            lines_changed = checker.calculate_lines_changed(old_content, content)

        if existing:
            # 创建新版本
            new_version = existing.version + 1
        else:
            new_version = 1

        # 生成语义摘要
        summary = summary_generator.generate_summary(content)

        # 创建新记录
        record = LobsterMemory.objects.create(
            lobster_id=lobster_id,
            file_path=file_path,
            content=content,
            hash=file_hash,
            status='consistent',
            version=new_version,
            summary=summary,
        )

        execution_time = time.time() - start_time

        # 记录操作日志（包含变动行数和数据源）
        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,
            lines_changed=lines_changed,
            source='local',
            operator=operator,
            status='success',
            execution_time=execution_time
        )

        return Response({
            'success': True,
            'message': '已同步到数据库',
            'data': LobsterMemorySerializer(record).data
        })

    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',
            source='local',
            operator=operator,
            status='failed',
            error_message=str(e),
            execution_time=execution_time
        )

        return Response({
            'success': False,
            'error': str(e)
        }, status=status.HTTP_500_INTERNAL_SERVER_ERROR)


@api_view(['POST'])
def sync_to_local(request):
    """
    同步到本地（带完整审计日志）

    功能：
    - 记录变动行数
    - 记录数据源、操作人、执行时间
    """
    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({
            'success': False,
            'error': 'file_path is required'
        }, 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
        db_record = LobsterMemory.objects.filter(
            lobster_id=lobster_id,
            file_path=file_path
        ).order_by('-version').first()

        if not db_record:
            return Response({
                'success': False,
                '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, chunked=True)
        except FileNotFoundError:
            local_content = None
            local_hash = None

        # 计算变动行数
        lines_changed = 0
        if local_content:
            checker = DiffChecker()
            lines_changed = checker.calculate_lines_changed(local_content, db_record.content)

        # 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（包含变动行数和数据源）
        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,
            lines_changed=lines_changed,
            source='database',
            operator=operator,
            status='success',
            execution_time=execution_time
        )

        return Response({
            'success': True,
            'message': '已同步到本地',
            'data': LobsterMemorySerializer(db_record).data
        })

    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',
            source='database',
            operator=operator,
            status='failed',
            error_message=str(e),
            execution_time=execution_time
        )

        return Response({
            'success': False,
            'error': str(e)
        }, status=status.HTTP_500_INTERNAL_SERVER_ERROR)


@api_view(['GET'])
def get_versions(request):
    """
    获取文件的所有版本（包含摘要）
    """
    file_path = request.query_params.get('file_path')
    lobster_id = request.query_params.get('lobster_id', 'daotong')

    if not file_path:
        return Response({
            'success': False,
            'error': 'file_path is required'
        }, status=status.HTTP_400_BAD_REQUEST)

    versions = LobsterMemory.objects.filter(
        lobster_id=lobster_id,
        file_path=file_path
    ).order_by('-version')

    return Response({
        'success': True,
        'data': LobsterMemorySerializer(versions, many=True).data
    })


@api_view(['GET'])
def get_stats(request):
    """
    获取统计信息（包含 hard_conflict 状态）
    """
    lobster_id = request.query_params.get('lobster_id', 'daotong')

    total_files = LobsterMemory.objects.filter(lobster_id=lobster_id).count()

    status_counts = {}
    for status_choice, _ in LobsterMemory.STATUS_CHOICES:
        count = LobsterMemory.objects.filter(
            lobster_id=lobster_id,
            status=status_choice
        ).count()
        status_counts[status_choice] = count

    # 获取总大小
    from django.db.models import Sum
    total_size = LobsterMemory.objects.filter(
        lobster_id=lobster_id
    ).aggregate(total=Sum('size'))['total'] or 0

    return Response({
        'success': True,
        'data': {
            'total_files': total_files,
            'status_counts': status_counts,
            'total_size': total_size,
            'total_size_mb': round(total_size / 1024 / 1024, 2)
        }
    })


@api_view(['GET'])
def get_history(request):
    """
    获取操作历史（包含变动行数和数据源）
    """
    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

    显示所有生效的忽略规则，包括：
    - 通配符模式 (*.pyc)
    - 正则表达式模式 (re:.*\\.log\$)
    - 默认规则
    """
    lobster_id = request.query_params.get('lobster_id', 'daotong')
    scanner = FileScanner()

    patterns = []
    for pattern_type, pattern, _ in scanner.ignore.patterns:
        patterns.append({
            'type': pattern_type,
            'pattern': pattern
        })

    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

    当修改 .lobsterignore 文件后调用此接口
    """
    lobster_id = request.data.get('lobster_id', 'daotong')
    scanner = FileScanner()

    # 重新加载忽略规则
    scanner.ignore.load_patterns()

    patterns = []
    for pattern_type, pattern, _ in scanner.ignore.patterns:
        patterns.append({
            'type': pattern_type,
            'pattern': pattern
        })

    return Response({
        'success': True,
        'message': '已重新加载忽略规则',
        'data': {
            'patterns': patterns,
            'total': len(patterns)
        }
    })