agno 1.8.1__py3-none-any.whl → 2.0.0__py3-none-any.whl

agno/memory/v2/schema.py

@@ -1,55 +0,0 @@
-from dataclasses import dataclass
-from datetime import datetime
-from typing import Any, Dict, List, Optional
-
-
-@dataclass
-class UserMemory:
-    """Model for User Memories"""
-
-    memory: str
-    topics: Optional[List[str]] = None
-    input: Optional[str] = None
-    last_updated: Optional[datetime] = None
-    memory_id: Optional[str] = None
-
-    def to_dict(self) -> Dict[str, Any]:
-        _dict = {
-            "memory_id": self.memory_id,
-            "memory": self.memory,
-            "topics": self.topics,
-            "last_updated": self.last_updated.isoformat() if self.last_updated else None,
-            "input": self.input,
-        }
-        return {k: v for k, v in _dict.items() if v is not None}
-
-    @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> "UserMemory":
-        last_updated = data.get("last_updated")
-        if last_updated:
-            data["last_updated"] = datetime.fromisoformat(last_updated)
-        return cls(**data)
-
-
-@dataclass
-class SessionSummary:
-    """Model for Session Summary."""
-
-    summary: str
-    topics: Optional[List[str]] = None
-    last_updated: Optional[datetime] = None
-
-    def to_dict(self) -> Dict[str, Any]:
-        _dict = {
-            "summary": self.summary,
-            "topics": self.topics,
-            "last_updated": self.last_updated.isoformat() if self.last_updated else None,
-        }
-        return {k: v for k, v in _dict.items() if v is not None}
-
-    @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> "SessionSummary":
-        last_updated = data.get("last_updated")
-        if last_updated:
-            data["last_updated"] = datetime.fromisoformat(last_updated)
-        return cls(**data)

agno/memory/v2/summarizer.py

@@ -1,215 +0,0 @@
-from copy import deepcopy
-from dataclasses import dataclass
-from textwrap import dedent
-from typing import Any, Dict, List, Optional, Type, Union
-
-from pydantic import BaseModel, Field
-
-from agno.models.base import Model
-from agno.models.message import Message
-from agno.utils.log import log_debug, log_error, log_info, log_warning
-from agno.utils.prompts import get_json_output_prompt
-from agno.utils.string import parse_response_model_str
-
-
-class SessionSummaryResponse(BaseModel):
-    """Model for Session Summary."""
-
-    summary: str = Field(
-        ...,
-        description="Summary of the session. Be concise and focus on only important information. Do not make anything up.",
-    )
-    topics: Optional[List[str]] = Field(None, description="Topics discussed in the session.")
-
-    def to_dict(self) -> Dict[str, Any]:
-        return self.model_dump(exclude_none=True)
-
-    def to_json(self) -> str:
-        return self.model_dump_json(exclude_none=True, indent=2)
-
-
-@dataclass
-class SessionSummarizer:
-    # Model used for summarization
-    model: Optional[Model] = None
-
-    # System message for the summarizer. If not provided, a default prompt will be used.
-    system_message: Optional[str] = None
-
-    # Additional instructions for the summarizer. If not provided, a default prompt will be used.
-    additional_instructions: Optional[str] = None
-
-    # Whether the summarizer has created a summary
-    summary_updated: bool = False
-
-    def __init__(
-        self,
-        model: Optional[Model] = None,
-        system_message: Optional[str] = None,
-        additional_instructions: Optional[str] = None,
-    ):
-        self.model = model
-        if self.model is not None and isinstance(self.model, str):
-            raise ValueError("Model must be a Model object, not a string")
-        self.system_message = system_message
-        self.additional_instructions = additional_instructions
-
-    def get_response_format(self, model: Model) -> Union[Dict[str, Any], Type[BaseModel]]:
-        if model.supports_native_structured_outputs:
-            return SessionSummaryResponse
-
-        elif model.supports_json_schema_outputs:
-            return {
-                "type": "json_schema",
-                "json_schema": {
-                    "name": SessionSummaryResponse.__name__,
-                    "schema": SessionSummaryResponse.model_json_schema(),
-                },
-            }
-        else:
-            return {"type": "json_object"}
-
-    def get_system_message(
-        self, conversation: List[Message], response_format: Union[Dict[str, Any], Type[BaseModel]]
-    ) -> Message:
-        if self.system_message is not None:
-            return Message(role="system", content=self.system_message)
-
-        # -*- Return a system message for summarization
-        system_prompt = dedent("""\
-        Analyze the following conversation between a user and an assistant, and extract the following details:
-          - Summary (str): Provide a concise summary of the session, focusing on important information that would be helpful for future interactions.
-          - Topics (Optional[List[str]]): List the topics discussed in the session.
-        Keep the summary concise and to the point. Only include relevant information.
-
-        <conversation>
-        """)
-        conversation_messages = []
-        for message in conversation:
-            if message.role == "user":
-                conversation_messages.append(f"User: {message.content}")
-            elif message.role in ["assistant", "model"]:
-                conversation_messages.append(f"Assistant: {message.content}\n")
-        system_prompt += "\n".join(conversation_messages)
-        system_prompt += "</conversation>"
-
-        if self.additional_instructions:
-            system_prompt += "\n" + self.additional_instructions
-
-        if response_format == {"type": "json_object"}:
-            system_prompt += "\n" + get_json_output_prompt(SessionSummaryResponse)  # type: ignore
-
-        return Message(role="system", content=system_prompt)
-
-    def run(
-        self,
-        conversation: List[Message],
-    ) -> Optional[SessionSummaryResponse]:
-        if self.model is None:
-            log_error("No model provided for summary_manager")
-            return None
-
-        log_debug("SessionSummarizer Start", center=True)
-
-        if conversation is None or len(conversation) == 0:
-            log_info("No conversation provided for summarization.")
-            return None
-
-        model_copy = deepcopy(self.model)
-        response_format = self.get_response_format(model_copy)
-
-        # Prepare the List of messages to send to the Model
-        messages_for_model: List[Message] = [
-            self.get_system_message(conversation, response_format=response_format),
-            # For models that require a non-system message
-            Message(role="user", content="Provide the summary of the conversation."),
-        ]
-
-        # Generate a response from the Model (includes running function calls)
-        response = model_copy.response(messages=messages_for_model, response_format=response_format)
-
-        if response.content is not None:
-            self.summary_updated = True
-
-        log_debug("SessionSummarizer End", center=True)
-
-        # If the model natively supports structured outputs, the parsed value is already in the structured format
-        if (
-            model_copy.supports_native_structured_outputs
-            and response.parsed is not None
-            and isinstance(response.parsed, SessionSummaryResponse)
-        ):
-            return response.parsed
-
-        # Otherwise convert the response to the structured format
-        if isinstance(response.content, str):
-            try:
-                session_summary: Optional[SessionSummaryResponse] = parse_response_model_str(  # type: ignore
-                    response.content, SessionSummaryResponse
-                )
-
-                # Update RunResponse
-                if session_summary is not None:
-                    return session_summary
-                else:
-                    log_warning("Failed to convert session_summary response to SessionSummaryResponse")
-            except Exception as e:
-                log_warning(f"Failed to convert session_summary response to SessionSummaryResponse: {e}")
-
-        return None
-
-    async def arun(
-        self,
-        conversation: List[Message],
-    ) -> Optional[SessionSummaryResponse]:
-        if self.model is None:
-            log_error("No model provided for summary_manager")
-            return None
-
-        log_debug("SessionSummarizer Start", center=True)
-
-        if conversation is None or len(conversation) == 0:
-            log_info("No conversation provided for summarization.")
-            return None
-
-        model_copy = deepcopy(self.model)
-        response_format = self.get_response_format(model_copy)
-
-        # Prepare the List of messages to send to the Model
-        messages_for_model: List[Message] = [
-            self.get_system_message(conversation, response_format=response_format),
-            # For models that require a non-system message
-            Message(role="user", content="Provide the summary of the conversation."),
-        ]
-
-        # Generate a response from the Model (includes running function calls)
-        response = await model_copy.aresponse(messages=messages_for_model, response_format=response_format)
-
-        if response.content is not None:
-            self.summary_updated = True
-
-        log_debug("SessionSummarizer End", center=True)
-
-        # If the model natively supports structured outputs, the parsed value is already in the structured format
-        if (
-            model_copy.supports_native_structured_outputs
-            and response.parsed is not None
-            and isinstance(response.parsed, SessionSummaryResponse)
-        ):
-            return response.parsed
-
-        # Otherwise convert the response to the structured format
-        if isinstance(response.content, str):
-            try:
-                session_summary: Optional[SessionSummaryResponse] = parse_response_model_str(  # type: ignore
-                    response.content, SessionSummaryResponse
-                )
-
-                # Update RunResponse
-                if session_summary is not None:
-                    return session_summary
-                else:
-                    log_warning("Failed to convert session_summary response to SessionSummaryResponse")
-            except Exception as e:
-                log_warning(f"Failed to convert session_summary response to SessionSummaryResponse: {e}")
-        return None

agno/memory/workflow.py

@@ -1,38 +0,0 @@
-from typing import Any, Dict, List, Optional
-
-from pydantic import BaseModel, ConfigDict
-
-from agno.run.response import RunResponse
-from agno.utils.log import log_debug
-
-
-class WorkflowRun(BaseModel):
-    input: Optional[Dict[str, Any]] = None
-    response: Optional[RunResponse] = None
-
-    model_config = ConfigDict(arbitrary_types_allowed=True)
-
-
-class WorkflowMemory(BaseModel):
-    runs: List[WorkflowRun] = []
-
-    model_config = ConfigDict(arbitrary_types_allowed=True)
-
-    def to_dict(self) -> Dict[str, Any]:
-        return self.model_dump(exclude_none=True)
-
-    def add_run(self, workflow_run: WorkflowRun) -> None:
-        """Adds a WorkflowRun to the runs list."""
-        self.runs.append(workflow_run)
-        log_debug("Added WorkflowRun to WorkflowMemory")
-
-    def clear(self) -> None:
-        """Clear the WorkflowMemory"""
-
-        self.runs = []
-
-    def deep_copy(self, *, update: Optional[Dict[str, Any]] = None) -> "WorkflowMemory":
-        new_memory = self.model_copy(deep=True, update=update)
-        # clear the new memory to remove any references to the old memory
-        new_memory.clear()
-        return new_memory