auto-coder 0.1.397__py3-none-any.whl → 0.1.398__py3-none-any.whl

autocoder/common/conversations/models.py

@@ -0,0 +1,154 @@
+"""
+PersistConversationManager 数据模型定义
+"""
+
+import time
+import uuid
+from typing import Union, Dict, List, Optional, Any
+from dataclasses import dataclass, field
+
+
+@dataclass
+class ConversationMessage:
+    """对话消息数据模型"""
+    
+    role: str
+    content: Union[str, dict, list]
+    timestamp: float = field(default_factory=time.time)
+    message_id: str = field(default_factory=lambda: str(uuid.uuid4()))
+    metadata: Optional[dict] = None
+    
+    def __post_init__(self):
+        """数据验证"""
+        self._validate()
+    
+    def _validate(self):
+        """验证消息数据"""
+        # 验证角色
+        valid_roles = ["system", "user", "assistant"]
+        if not self.role or self.role not in valid_roles:
+            raise ValueError(f"无效的消息角色: {self.role}，有效角色: {valid_roles}")
+        
+        # 验证内容
+        if self.content is None or (isinstance(self.content, str) and len(self.content) == 0):
+            raise ValueError("消息内容不能为空")
+        
+        # 验证时间戳
+        if not isinstance(self.timestamp, (int, float)) or self.timestamp <= 0:
+            raise ValueError("无效的时间戳")
+        
+        # 验证消息ID
+        if not isinstance(self.message_id, str) or len(self.message_id) == 0:
+            raise ValueError("消息ID不能为空")
+    
+    def to_dict(self) -> dict:
+        """序列化为字典"""
+        return {
+            "role": self.role,
+            "content": self.content,
+            "timestamp": self.timestamp,
+            "message_id": self.message_id,
+            "metadata": self.metadata
+        }
+    
+    @classmethod
+    def from_dict(cls, data: dict) -> "ConversationMessage":
+        """从字典反序列化"""
+        return cls(
+            role=data["role"],
+            content=data["content"],
+            timestamp=data["timestamp"],
+            message_id=data["message_id"],
+            metadata=data.get("metadata")
+        )
+
+
+@dataclass
+class Conversation:
+    """对话数据模型"""
+    
+    name: str
+    conversation_id: str = field(default_factory=lambda: str(uuid.uuid4()))
+    description: Optional[str] = None
+    created_at: float = field(default_factory=time.time)
+    updated_at: float = field(default_factory=time.time)
+    messages: List[dict] = field(default_factory=list)
+    metadata: Optional[dict] = None
+    version: int = 1
+    
+    def __post_init__(self):
+        """数据验证"""
+        self._validate()
+    
+    def _validate(self):
+        """验证对话数据"""
+        # 验证名称
+        if not self.name or (isinstance(self.name, str) and len(self.name) == 0):
+            raise ValueError("对话名称不能为空")
+        
+        # 验证版本号
+        if not isinstance(self.version, int) or self.version <= 0:
+            raise ValueError("版本号必须是正整数")
+        
+        # 验证时间戳
+        if not isinstance(self.created_at, (int, float)) or self.created_at <= 0:
+            raise ValueError("无效的创建时间戳")
+        
+        if not isinstance(self.updated_at, (int, float)) or self.updated_at < self.created_at:
+            raise ValueError("无效的更新时间戳")
+        
+        # 验证对话ID
+        if not isinstance(self.conversation_id, str) or len(self.conversation_id) == 0:
+            raise ValueError("对话ID不能为空")
+        
+        # 验证消息列表
+        if not isinstance(self.messages, list):
+            raise ValueError("消息列表必须是列表类型")
+    
+    def add_message(self, message: ConversationMessage):
+        """添加消息到对话"""
+        self.messages.append(message.to_dict())
+        self.updated_at = time.time()
+    
+    def remove_message(self, message_id: str) -> bool:
+        """从对话中删除消息"""
+        for i, msg in enumerate(self.messages):
+            if msg.get("message_id") == message_id:
+                del self.messages[i]
+                self.updated_at = time.time()
+                return True
+        return False
+    
+    def get_message(self, message_id: str) -> Optional[dict]:
+        """从对话中获取消息"""
+        for msg in self.messages:
+            if msg.get("message_id") == message_id:
+                return msg
+        return None
+    
+    def to_dict(self) -> dict:
+        """序列化为字典"""
+        return {
+            "conversation_id": self.conversation_id,
+            "name": self.name,
+            "description": self.description,
+            "created_at": self.created_at,
+            "updated_at": self.updated_at,
+            "messages": self.messages,
+            "metadata": self.metadata,
+            "version": self.version
+        }
+    
+    @classmethod
+    def from_dict(cls, data: dict) -> "Conversation":
+        """从字典反序列化"""
+        return cls(
+            conversation_id=data["conversation_id"],
+            name=data["name"],
+            description=data.get("description"),
+            created_at=data["created_at"],
+            updated_at=data["updated_at"],
+            messages=data.get("messages", []),
+            metadata=data.get("metadata"),
+            version=data.get("version", 1)
+        ) 

autocoder/common/conversations/search/__init__.py

@@ -0,0 +1,15 @@
+"""
+Search and filtering module for conversations.
+
+This module provides text search and filtering capabilities for conversations
+and messages, supporting full-text search, keyword matching, and complex
+filtering operations.
+"""
+
+from .text_searcher import TextSearcher
+from .filter_manager import FilterManager
+
+__all__ = [
+    'TextSearcher',
+    'FilterManager'
+] 

autocoder/common/conversations/search/filter_manager.py

@@ -0,0 +1,431 @@
+"""
+Filter management for conversations and messages.
+
+This module provides comprehensive filtering capabilities including
+time-based filters, content filters, and complex query combinations.
+"""
+
+import re
+from datetime import datetime, timedelta
+from typing import List, Dict, Any, Optional, Union, Callable
+from dataclasses import dataclass
+from enum import Enum
+
+
+class FilterOperator(Enum):
+    """Enumeration of filter operators."""
+    EQUALS = "eq"
+    NOT_EQUALS = "ne"
+    GREATER_THAN = "gt"
+    GREATER_THAN_OR_EQUAL = "gte"
+    LESS_THAN = "lt"
+    LESS_THAN_OR_EQUAL = "lte"
+    CONTAINS = "contains"
+    NOT_CONTAINS = "not_contains"
+    STARTS_WITH = "starts_with"
+    ENDS_WITH = "ends_with"
+    REGEX = "regex"
+    IN = "in"
+    NOT_IN = "not_in"
+
+
+class LogicalOperator(Enum):
+    """Enumeration of logical operators for combining filters."""
+    AND = "and"
+    OR = "or"
+    NOT = "not"
+
+
+@dataclass
+class Filter:
+    """Represents a single filter condition."""
+    field: str
+    operator: FilterOperator
+    value: Any
+    case_sensitive: bool = False
+    
+    def apply(self, item: Dict[str, Any]) -> bool:
+        """Apply this filter to an item."""
+        field_value = self._get_field_value(item, self.field)
+        
+        if field_value is None:
+            return False
+        
+        return self._compare_values(field_value, self.value, self.operator)
+    
+    def _get_field_value(self, item: Dict[str, Any], field_path: str) -> Any:
+        """Get field value supporting nested field access (e.g., 'metadata.tags')."""
+        value = item
+        
+        for field_part in field_path.split('.'):
+            if isinstance(value, dict) and field_part in value:
+                value = value[field_part]
+            else:
+                return None
+        
+        return value
+    
+    def _compare_values(self, field_value: Any, filter_value: Any, operator: FilterOperator) -> bool:
+        """Compare field value with filter value using the specified operator."""
+        try:
+            if operator == FilterOperator.EQUALS:
+                return self._normalize_for_comparison(field_value) == self._normalize_for_comparison(filter_value)
+            
+            elif operator == FilterOperator.NOT_EQUALS:
+                return self._normalize_for_comparison(field_value) != self._normalize_for_comparison(filter_value)
+            
+            elif operator == FilterOperator.GREATER_THAN:
+                return field_value > filter_value
+            
+            elif operator == FilterOperator.GREATER_THAN_OR_EQUAL:
+                return field_value >= filter_value
+            
+            elif operator == FilterOperator.LESS_THAN:
+                return field_value < filter_value
+            
+            elif operator == FilterOperator.LESS_THAN_OR_EQUAL:
+                return field_value <= filter_value
+            
+            elif operator == FilterOperator.CONTAINS:
+                field_str = self._normalize_for_comparison(str(field_value))
+                filter_str = self._normalize_for_comparison(str(filter_value))
+                return filter_str in field_str
+            
+            elif operator == FilterOperator.NOT_CONTAINS:
+                field_str = self._normalize_for_comparison(str(field_value))
+                filter_str = self._normalize_for_comparison(str(filter_value))
+                return filter_str not in field_str
+            
+            elif operator == FilterOperator.STARTS_WITH:
+                field_str = self._normalize_for_comparison(str(field_value))
+                filter_str = self._normalize_for_comparison(str(filter_value))
+                return field_str.startswith(filter_str)
+            
+            elif operator == FilterOperator.ENDS_WITH:
+                field_str = self._normalize_for_comparison(str(field_value))
+                filter_str = self._normalize_for_comparison(str(filter_value))
+                return field_str.endswith(filter_str)
+            
+            elif operator == FilterOperator.REGEX:
+                field_str = str(field_value)
+                flags = 0 if self.case_sensitive else re.IGNORECASE
+                return bool(re.search(filter_value, field_str, flags))
+            
+            elif operator == FilterOperator.IN:
+                if not isinstance(filter_value, (list, tuple, set)):
+                    return False
+                return field_value in filter_value
+            
+            elif operator == FilterOperator.NOT_IN:
+                if not isinstance(filter_value, (list, tuple, set)):
+                    return True
+                return field_value not in filter_value
+            
+        except (TypeError, ValueError, AttributeError):
+            return False
+        
+        return False
+    
+    def _normalize_for_comparison(self, value: Any) -> Any:
+        """Normalize value for comparison based on case sensitivity."""
+        if isinstance(value, str) and not self.case_sensitive:
+            return value.lower()
+        return value
+
+
+@dataclass 
+class FilterGroup:
+    """Represents a group of filters combined with logical operators."""
+    filters: List[Union[Filter, 'FilterGroup']]
+    operator: LogicalOperator = LogicalOperator.AND
+    
+    def apply(self, item: Dict[str, Any]) -> bool:
+        """Apply this filter group to an item."""
+        if not self.filters:
+            return True
+            
+        if self.operator == LogicalOperator.AND:
+            return all(f.apply(item) for f in self.filters)
+        
+        elif self.operator == LogicalOperator.OR:
+            return any(f.apply(item) for f in self.filters)
+        
+        elif self.operator == LogicalOperator.NOT:
+            # For NOT operator, apply AND logic and negate the result
+            return not all(f.apply(item) for f in self.filters)
+        
+        return True
+
+
+class FilterManager:
+    """Manager for building and applying complex filters."""
+    
+    def __init__(self):
+        """Initialize filter manager."""
+        self.predefined_filters = {}
+        self._register_predefined_filters()
+    
+    def _register_predefined_filters(self):
+        """Register commonly used predefined filters."""
+        # Time-based filters
+        now = datetime.now().timestamp()
+        
+        self.predefined_filters.update({
+            'today': Filter('created_at', FilterOperator.GREATER_THAN_OR_EQUAL, 
+                          now - 24 * 3600),
+            'this_week': Filter('created_at', FilterOperator.GREATER_THAN_OR_EQUAL, 
+                              now - 7 * 24 * 3600),
+            'this_month': Filter('created_at', FilterOperator.GREATER_THAN_OR_EQUAL, 
+                               now - 30 * 24 * 3600),
+            'has_messages': Filter('message_count', FilterOperator.GREATER_THAN, 0),
+            'no_messages': Filter('message_count', FilterOperator.EQUALS, 0),
+        })
+    
+    def create_filter(
+        self,
+        field: str,
+        operator: Union[FilterOperator, str],
+        value: Any,
+        case_sensitive: bool = False
+    ) -> Filter:
+        """
+        Create a single filter.
+        
+        Args:
+            field: Field name to filter on
+            operator: Filter operator
+            value: Value to compare against
+            case_sensitive: Whether comparison should be case sensitive
+            
+        Returns:
+            Filter instance
+        """
+        if isinstance(operator, str):
+            operator = FilterOperator(operator)
+        
+        return Filter(field, operator, value, case_sensitive)
+    
+    def create_filter_group(
+        self,
+        filters: List[Union[Filter, FilterGroup]],
+        operator: Union[LogicalOperator, str] = LogicalOperator.AND
+    ) -> FilterGroup:
+        """
+        Create a filter group.
+        
+        Args:
+            filters: List of filters or filter groups
+            operator: Logical operator to combine filters
+            
+        Returns:
+            FilterGroup instance
+        """
+        if isinstance(operator, str):
+            operator = LogicalOperator(operator)
+        
+        return FilterGroup(filters, operator)
+    
+    def create_time_range_filter(
+        self,
+        field: str,
+        start_time: Optional[Union[datetime, float]] = None,
+        end_time: Optional[Union[datetime, float]] = None
+    ) -> FilterGroup:
+        """
+        Create a time range filter.
+        
+        Args:
+            field: Time field name (e.g., 'created_at', 'updated_at')
+            start_time: Start of time range
+            end_time: End of time range
+            
+        Returns:
+            FilterGroup with time range filters
+        """
+        filters = []
+        
+        if start_time is not None:
+            if isinstance(start_time, datetime):
+                start_time = start_time.timestamp()
+            filters.append(Filter(field, FilterOperator.GREATER_THAN_OR_EQUAL, start_time))
+        
+        if end_time is not None:
+            if isinstance(end_time, datetime):
+                end_time = end_time.timestamp()
+            filters.append(Filter(field, FilterOperator.LESS_THAN_OR_EQUAL, end_time))
+        
+        return FilterGroup(filters, LogicalOperator.AND)
+    
+    def create_text_search_filter(
+        self,
+        fields: List[str],
+        query: str,
+        case_sensitive: bool = False
+    ) -> FilterGroup:
+        """
+        Create a text search filter across multiple fields.
+        
+        Args:
+            fields: List of field names to search in
+            query: Search query
+            case_sensitive: Whether search should be case sensitive
+            
+        Returns:
+            FilterGroup with text search filters
+        """
+        filters = []
+        
+        for field in fields:
+            filters.append(Filter(field, FilterOperator.CONTAINS, query, case_sensitive))
+        
+        return FilterGroup(filters, LogicalOperator.OR)
+    
+    def create_role_filter(self, roles: List[str]) -> Filter:
+        """Create a filter for message roles."""
+        return Filter('role', FilterOperator.IN, roles)
+    
+    def create_content_type_filter(self, content_types: List[str]) -> FilterGroup:
+        """
+        Create a filter for content types (string, dict, list).
+        
+        Args:
+            content_types: List of content types ('string', 'dict', 'list')
+            
+        Returns:
+            FilterGroup for content type filtering
+        """
+        filters = []
+        
+        for content_type in content_types:
+            if content_type == 'string':
+                # Filter for string content
+                filters.append(Filter('content', FilterOperator.REGEX, r'^[^{\[].*', case_sensitive=True))
+            elif content_type == 'dict':
+                # Filter for dict content (starts with {)
+                filters.append(Filter('content', FilterOperator.REGEX, r'^\{.*', case_sensitive=True))
+            elif content_type == 'list':
+                # Filter for list content (starts with [)
+                filters.append(Filter('content', FilterOperator.REGEX, r'^\[.*', case_sensitive=True))
+        
+        return FilterGroup(filters, LogicalOperator.OR)
+    
+    def apply_filters(
+        self,
+        items: List[Dict[str, Any]],
+        filter_spec: Union[Filter, FilterGroup, str, Dict[str, Any]]
+    ) -> List[Dict[str, Any]]:
+        """
+        Apply filters to a list of items.
+        
+        Args:
+            items: List of items to filter
+            filter_spec: Filter specification (Filter, FilterGroup, predefined name, or dict)
+            
+        Returns:
+            Filtered list of items
+        """
+        if not items:
+            return []
+        
+        # Convert filter_spec to Filter or FilterGroup
+        filter_obj = self._parse_filter_spec(filter_spec)
+        
+        if filter_obj is None:
+            return items
+        
+        # Apply filter
+        return [item for item in items if filter_obj.apply(item)]
+    
+    def _parse_filter_spec(self, filter_spec: Union[Filter, FilterGroup, str, Dict[str, Any]]) -> Optional[Union[Filter, FilterGroup]]:
+        """Parse filter specification into Filter or FilterGroup."""
+        if isinstance(filter_spec, (Filter, FilterGroup)):
+            return filter_spec
+        
+        elif isinstance(filter_spec, str):
+            # Predefined filter name
+            return self.predefined_filters.get(filter_spec)
+        
+        elif isinstance(filter_spec, dict):
+            # Dictionary specification
+            return self._parse_dict_filter(filter_spec)
+        
+        return None
+    
+    def _parse_dict_filter(self, filter_dict: Dict[str, Any]) -> Optional[Union[Filter, FilterGroup]]:
+        """Parse dictionary filter specification."""
+        if 'filters' in filter_dict:
+            # FilterGroup specification
+            filters = []
+            for f in filter_dict['filters']:
+                parsed_filter = self._parse_filter_spec(f)
+                if parsed_filter:
+                    filters.append(parsed_filter)
+            
+            operator = LogicalOperator(filter_dict.get('operator', 'and'))
+            return FilterGroup(filters, operator)
+        
+        elif 'field' in filter_dict and 'operator' in filter_dict and 'value' in filter_dict:
+            # Single Filter specification
+            return Filter(
+                field=filter_dict['field'],
+                operator=FilterOperator(filter_dict['operator']),
+                value=filter_dict['value'],
+                case_sensitive=filter_dict.get('case_sensitive', False)
+            )
+        
+        return None
+    
+    def create_conversation_filters(self) -> Dict[str, Union[Filter, FilterGroup]]:
+        """Create common conversation filters."""
+        return {
+            'active': Filter('message_count', FilterOperator.GREATER_THAN, 0),
+            'empty': Filter('message_count', FilterOperator.EQUALS, 0),
+            'recent': Filter('updated_at', FilterOperator.GREATER_THAN_OR_EQUAL, 
+                           datetime.now().timestamp() - 7 * 24 * 3600),
+            'old': Filter('updated_at', FilterOperator.LESS_THAN, 
+                         datetime.now().timestamp() - 30 * 24 * 3600),
+        }
+    
+    def create_message_filters(self) -> Dict[str, Union[Filter, FilterGroup]]:
+        """Create common message filters."""
+        return {
+            'user_messages': Filter('role', FilterOperator.EQUALS, 'user'),
+            'assistant_messages': Filter('role', FilterOperator.EQUALS, 'assistant'),
+            'system_messages': Filter('role', FilterOperator.EQUALS, 'system'),
+            'recent_messages': Filter('timestamp', FilterOperator.GREATER_THAN_OR_EQUAL,
+                                    datetime.now().timestamp() - 24 * 3600),
+            'has_metadata': Filter('metadata', FilterOperator.NOT_EQUALS, None),
+        }
+    
+    def combine_filters(
+        self,
+        filters: List[Union[Filter, FilterGroup]],
+        operator: LogicalOperator = LogicalOperator.AND
+    ) -> FilterGroup:
+        """
+        Combine multiple filters with a logical operator.
+        
+        Args:
+            filters: List of filters to combine
+            operator: Logical operator (AND, OR, NOT)
+            
+        Returns:
+            Combined FilterGroup
+        """
+        return FilterGroup(filters, operator)
+    
+    def validate_filter_spec(self, filter_spec: Dict[str, Any]) -> bool:
+        """
+        Validate a filter specification dictionary.
+        
+        Args:
+            filter_spec: Filter specification to validate
+            
+        Returns:
+            True if valid, False otherwise
+        """
+        try:
+            parsed = self._parse_dict_filter(filter_spec)
+            return parsed is not None
+        except (ValueError, KeyError, TypeError):
+            return False