agno 1.8.2__py3-none-any.whl → 2.0.0__py3-none-any.whl

agno/db/gcs_json/gcs_json_db.py

@@ -0,0 +1,1001 @@
+import json
+import time
+from datetime import date, datetime, timedelta, timezone
+from typing import Any, Dict, List, Optional, Tuple, Union
+from uuid import uuid4
+
+from agno.db.base import BaseDb, SessionType
+from agno.db.gcs_json.utils import (
+    apply_sorting,
+    calculate_date_metrics,
+    fetch_all_sessions_data,
+    get_dates_to_calculate_metrics_for,
+)
+from agno.db.schemas.evals import EvalFilterType, EvalRunRecord, EvalType
+from agno.db.schemas.knowledge import KnowledgeRow
+from agno.db.schemas.memory import UserMemory
+from agno.session import AgentSession, Session, TeamSession, WorkflowSession
+from agno.utils.log import log_debug, log_error, log_info, log_warning
+
+try:
+    from google.cloud import storage as gcs  # type: ignore
+except ImportError:
+    raise ImportError("`google-cloud-storage` not installed. Please install it with `pip install google-cloud-storage`")
+
+
+class GcsJsonDb(BaseDb):
+    def __init__(
+        self,
+        bucket_name: str,
+        prefix: Optional[str] = None,
+        session_table: Optional[str] = None,
+        memory_table: Optional[str] = None,
+        metrics_table: Optional[str] = None,
+        eval_table: Optional[str] = None,
+        knowledge_table: Optional[str] = None,
+        project: Optional[str] = None,
+        credentials: Optional[Any] = None,
+    ):
+        """
+        Interface for interacting with JSON files stored in Google Cloud Storage as database.
+
+        Args:
+            bucket_name (str): Name of the GCS bucket where JSON files will be stored.
+            prefix (Optional[str]): Path prefix for organizing files in the bucket. Defaults to "agno/".
+            session_table (Optional[str]): Name of the JSON file to store sessions (without .json extension).
+            memory_table (Optional[str]): Name of the JSON file to store user memories.
+            metrics_table (Optional[str]): Name of the JSON file to store metrics.
+            eval_table (Optional[str]): Name of the JSON file to store evaluation runs.
+            knowledge_table (Optional[str]): Name of the JSON file to store knowledge content.
+            project (Optional[str]): GCP project ID. If None, uses default project.
+            location (Optional[str]): GCS bucket location. If None, uses default location.
+            credentials (Optional[Any]): GCP credentials. If None, uses default credentials.
+        """
+        super().__init__(
+            session_table=session_table,
+            memory_table=memory_table,
+            metrics_table=metrics_table,
+            eval_table=eval_table,
+            knowledge_table=knowledge_table,
+        )
+
+        self.bucket_name = bucket_name
+        self.prefix = prefix or "agno/"
+        if self.prefix and not self.prefix.endswith("/"):
+            self.prefix += "/"
+
+        # Initialize GCS client and bucket
+        self.client = gcs.Client(project=project, credentials=credentials)
+        self.bucket = self.client.bucket(self.bucket_name)
+
+    def _get_blob_name(self, filename: str) -> str:
+        """Get the full blob name including prefix for a given filename."""
+        return f"{self.prefix}{filename}.json"
+
+    def _read_json_file(self, filename: str, create_table_if_not_found: Optional[bool] = False) -> List[Dict[str, Any]]:
+        """Read data from a JSON file in GCS, creating it if it doesn't exist.
+
+        Args:
+            filename (str): The name of the JSON file to read.
+
+        Returns:
+            List[Dict[str, Any]]: The data from the JSON file.
+
+        Raises:
+            json.JSONDecodeError: If the JSON file is not valid.
+        """
+        blob_name = self._get_blob_name(filename)
+        blob = self.bucket.blob(blob_name)
+
+        try:
+            data_str = blob.download_as_bytes().decode("utf-8")
+            return json.loads(data_str)
+
+        except Exception as e:
+            # Check if it's a 404 (file not found) error
+            if "404" in str(e) or "Not Found" in str(e):
+                if create_table_if_not_found:
+                    log_debug(f"Creating new GCS JSON file: {blob_name}")
+                    blob.upload_from_string("[]", content_type="application/json")
+                return []
+            else:
+                log_error(f"Error reading the {blob_name} JSON file from GCS: {e}")
+                raise json.JSONDecodeError(f"Error reading {blob_name}", "", 0)
+
+    def _write_json_file(self, filename: str, data: List[Dict[str, Any]]) -> None:
+        """Write data to a JSON file in GCS.
+
+        Args:
+            filename (str): The name of the JSON file to write.
+            data (List[Dict[str, Any]]): The data to write to the JSON file.
+
+        Raises:
+            Exception: If an error occurs while writing to the JSON file.
+        """
+        blob_name = self._get_blob_name(filename)
+        blob = self.bucket.blob(blob_name)
+
+        try:
+            json_data = json.dumps(data, indent=2, default=str)
+            blob.upload_from_string(json_data, content_type="application/json")
+
+        except Exception as e:
+            log_error(f"Error writing to the {blob_name} JSON file in GCS: {e}")
+            return
+
+    # -- Session methods --
+
+    def delete_session(self, session_id: str) -> bool:
+        """Delete a session from the GCS JSON file.
+
+        Args:
+            session_id (str): The ID of the session to delete.
+
+        Returns:
+            bool: True if the session was deleted, False otherwise.
+
+        Raises:
+            Exception: If an error occurs during deletion.
+        """
+        try:
+            sessions = self._read_json_file(self.session_table_name)
+            original_count = len(sessions)
+            sessions = [s for s in sessions if s.get("session_id") != session_id]
+
+            if len(sessions) < original_count:
+                self._write_json_file(self.session_table_name, sessions)
+                log_debug(f"Successfully deleted session with session_id: {session_id}")
+                return True
+
+            else:
+                log_debug(f"No session found to delete with session_id: {session_id}")
+                return False
+
+        except Exception as e:
+            log_warning(f"Error deleting session: {e}")
+            return False
+
+    def delete_sessions(self, session_ids: List[str]) -> None:
+        """Delete multiple sessions from the GCS JSON file.
+
+        Args:
+            session_ids (List[str]): The IDs of the sessions to delete.
+
+        Raises:
+            Exception: If an error occurs during deletion.
+        """
+        try:
+            sessions = self._read_json_file(self.session_table_name)
+            sessions = [s for s in sessions if s.get("session_id") not in session_ids]
+            self._write_json_file(self.session_table_name, sessions)
+            log_debug(f"Successfully deleted sessions with ids: {session_ids}")
+
+        except Exception as e:
+            log_warning(f"Error deleting sessions: {e}")
+
+    def get_session(
+        self,
+        session_id: str,
+        session_type: Optional[SessionType] = None,
+        user_id: Optional[str] = None,
+        deserialize: Optional[bool] = True,
+    ) -> Optional[Union[AgentSession, TeamSession, WorkflowSession, Dict[str, Any]]]:
+        """Read a session from the GCS JSON file.
+
+        Args:
+            session_id (str): The ID of the session to read.
+            session_type (Optional[SessionType]): The type of the session to read.
+            user_id (Optional[str]): The ID of the user to read the session for.
+            deserialize (Optional[bool]): Whether to deserialize the session.
+
+        Returns:
+            Union[Session, Dict[str, Any], None]:
+                - When deserialize=True: Session object
+                - When deserialize=False: Session dictionary
+
+        Raises:
+            Exception: If an error occurs while reading the session.
+        """
+        try:
+            sessions = self._read_json_file(self.session_table_name)
+
+            for session_data in sessions:
+                if session_data.get("session_id") == session_id:
+                    if user_id is not None and session_data.get("user_id") != user_id:
+                        continue
+
+                    session_type_value = session_type.value if isinstance(session_type, SessionType) else session_type
+                    if session_data.get("session_type") != session_type_value:
+                        continue
+
+                    if not deserialize:
+                        return session_data
+
+                    if session_type == SessionType.AGENT:
+                        return AgentSession.from_dict(session_data)
+                    elif session_type == SessionType.TEAM:
+                        return TeamSession.from_dict(session_data)
+                    elif session_type == SessionType.WORKFLOW:
+                        return WorkflowSession.from_dict(session_data)
+
+            return None
+
+        except Exception as e:
+            log_warning(f"Exception reading from session file: {e}")
+            return None
+
+    def get_sessions(
+        self,
+        session_type: Optional[SessionType] = None,
+        user_id: Optional[str] = None,
+        component_id: Optional[str] = None,
+        session_name: Optional[str] = None,
+        start_timestamp: Optional[int] = None,
+        end_timestamp: Optional[int] = None,
+        limit: Optional[int] = None,
+        page: Optional[int] = None,
+        sort_by: Optional[str] = None,
+        sort_order: Optional[str] = None,
+        deserialize: Optional[bool] = True,
+    ) -> Union[List[Session], Tuple[List[Dict[str, Any]], int]]:
+        """Get all sessions from the GCS JSON file with filtering and pagination.
+
+        Args:
+            session_type (Optional[SessionType]): The type of the sessions to read.
+            user_id (Optional[str]): The ID of the user to read the sessions for.
+            component_id (Optional[str]): The ID of the component to read the sessions for.
+            session_name (Optional[str]): The name of the session to read.
+            start_timestamp (Optional[int]): The start timestamp of the sessions to read.
+            end_timestamp (Optional[int]): The end timestamp of the sessions to read.
+            limit (Optional[int]): The limit of the sessions to read.
+            page (Optional[int]): The page of the sessions to read.
+            sort_by (Optional[str]): The field to sort the sessions by.
+            sort_order (Optional[str]): The order to sort the sessions by.
+            deserialize (Optional[bool]): Whether to deserialize the sessions.
+            create_table_if_not_found (Optional[bool]): Whether to create a file to track sessions if it doesn't exist.
+
+        Returns:
+            Union[List[AgentSession], List[TeamSession], List[WorkflowSession], Tuple[List[Dict[str, Any]], int]]:
+                - When deserialize=True: List of sessions
+                - When deserialize=False: Tuple with list of sessions and total count
+
+        Raises:
+            Exception: If an error occurs while reading the sessions.
+        """
+        try:
+            sessions = self._read_json_file(self.session_table_name)
+
+            # Apply filters
+            filtered_sessions = []
+            for session_data in sessions:
+                if user_id is not None and session_data.get("user_id") != user_id:
+                    continue
+                if component_id is not None:
+                    if session_type == SessionType.AGENT and session_data.get("agent_id") != component_id:
+                        continue
+                    elif session_type == SessionType.TEAM and session_data.get("team_id") != component_id:
+                        continue
+                    elif session_type == SessionType.WORKFLOW and session_data.get("workflow_id") != component_id:
+                        continue
+                if start_timestamp is not None and session_data.get("created_at", 0) < start_timestamp:
+                    continue
+                if end_timestamp is not None and session_data.get("created_at", 0) > end_timestamp:
+                    continue
+                if session_name is not None:
+                    stored_name = session_data.get("session_data", {}).get("session_name", "")
+                    if session_name.lower() not in stored_name.lower():
+                        continue
+                session_type_value = session_type.value if isinstance(session_type, SessionType) else session_type
+                if session_data.get("session_type") != session_type_value:
+                    continue
+
+                filtered_sessions.append(session_data)
+
+            total_count = len(filtered_sessions)
+
+            # Apply sorting
+            filtered_sessions = apply_sorting(filtered_sessions, sort_by, sort_order)
+
+            # Apply pagination
+            if limit is not None:
+                start_idx = 0
+                if page is not None:
+                    start_idx = (page - 1) * limit
+                filtered_sessions = filtered_sessions[start_idx : start_idx + limit]
+
+            if not deserialize:
+                return filtered_sessions, total_count
+
+            if session_type == SessionType.AGENT:
+                return [AgentSession.from_dict(session) for session in filtered_sessions]  # type: ignore
+            elif session_type == SessionType.TEAM:
+                return [TeamSession.from_dict(session) for session in filtered_sessions]  # type: ignore
+            elif session_type == SessionType.WORKFLOW:
+                return [WorkflowSession.from_dict(session) for session in filtered_sessions]  # type: ignore
+            else:
+                raise ValueError(f"Invalid session type: {session_type}")
+
+        except Exception as e:
+            log_warning(f"Exception reading from session file: {e}")
+            return [] if deserialize else ([], 0)
+
+    def rename_session(
+        self, session_id: str, session_type: SessionType, session_name: str, deserialize: Optional[bool] = True
+    ) -> Optional[Union[Session, Dict[str, Any]]]:
+        """Rename a session in the GCS JSON file."""
+        try:
+            sessions = self._read_json_file(self.session_table_name)
+
+            for i, session_data in enumerate(sessions):
+                if (
+                    session_data.get("session_id") == session_id
+                    and session_data.get("session_type") == session_type.value
+                ):
+                    # Update session name in session_data
+                    if "session_data" not in session_data:
+                        session_data["session_data"] = {}
+                    session_data["session_data"]["session_name"] = session_name
+
+                    sessions[i] = session_data
+                    self._write_json_file(self.session_table_name, sessions)
+
+                    if not deserialize:
+                        return session_data
+
+                    if session_type == SessionType.AGENT:
+                        return AgentSession.from_dict(session_data)
+                    elif session_type == SessionType.TEAM:
+                        return TeamSession.from_dict(session_data)
+                    elif session_type == SessionType.WORKFLOW:
+                        return WorkflowSession.from_dict(session_data)
+
+            return None
+        except Exception as e:
+            log_warning(f"Exception renaming session: {e}")
+            return None
+
+    def upsert_session(
+        self, session: Session, deserialize: Optional[bool] = True
+    ) -> Optional[Union[Session, Dict[str, Any]]]:
+        """Insert or update a session in the GCS JSON file."""
+        try:
+            sessions = self._read_json_file(self.session_table_name, create_table_if_not_found=True)
+            session_dict = session.to_dict()
+
+            # Add session_type based on session instance type
+            if isinstance(session, AgentSession):
+                session_dict["session_type"] = SessionType.AGENT.value
+            elif isinstance(session, TeamSession):
+                session_dict["session_type"] = SessionType.TEAM.value
+            elif isinstance(session, WorkflowSession):
+                session_dict["session_type"] = SessionType.WORKFLOW.value
+
+            # Find existing session to update
+            session_updated = False
+            for i, existing_session in enumerate(sessions):
+                if existing_session.get("session_id") == session_dict.get("session_id") and self._matches_session_key(
+                    existing_session, session
+                ):
+                    # Update existing session
+                    session_dict["updated_at"] = int(time.time())
+                    sessions[i] = session_dict
+                    session_updated = True
+                    break
+
+            if not session_updated:
+                # Add new session
+                session_dict["created_at"] = session_dict.get("created_at", int(time.time()))
+                session_dict["updated_at"] = session_dict.get("created_at")
+                sessions.append(session_dict)
+
+            self._write_json_file(self.session_table_name, sessions)
+
+            if not deserialize:
+                return session_dict
+
+            return session
+
+        except Exception as e:
+            log_warning(f"Exception upserting session: {e}")
+            return None
+
+    def _matches_session_key(self, existing_session: Dict[str, Any], session: Session) -> bool:
+        """Check if existing session matches the key for the session type."""
+        if isinstance(session, AgentSession):
+            return existing_session.get("agent_id") == session.agent_id
+        elif isinstance(session, TeamSession):
+            return existing_session.get("team_id") == session.team_id
+        elif isinstance(session, WorkflowSession):
+            return existing_session.get("workflow_id") == session.workflow_id
+        return False
+
+    # -- Memory methods --
+    def delete_user_memory(self, memory_id: str) -> None:
+        """Delete a user memory from the GCS JSON file."""
+        try:
+            memories = self._read_json_file(self.memory_table_name)
+            original_count = len(memories)
+            memories = [m for m in memories if m.get("memory_id") != memory_id]
+
+            if len(memories) < original_count:
+                self._write_json_file(self.memory_table_name, memories)
+                log_debug(f"Successfully deleted user memory id: {memory_id}")
+
+            else:
+                log_debug(f"No user memory found with id: {memory_id}")
+
+        except Exception as e:
+            log_warning(f"Error deleting user memory: {e}")
+
+    def delete_user_memories(self, memory_ids: List[str]) -> None:
+        """Delete multiple user memories from the GCS JSON file."""
+        try:
+            memories = self._read_json_file(self.memory_table_name)
+            memories = [m for m in memories if m.get("memory_id") not in memory_ids]
+            self._write_json_file(self.memory_table_name, memories)
+            log_debug(f"Successfully deleted user memories with ids: {memory_ids}")
+        except Exception as e:
+            log_warning(f"Error deleting user memories: {e}")
+
+    def get_all_memory_topics(self) -> List[str]:
+        """Get all memory topics from the GCS JSON file."""
+        try:
+            memories = self._read_json_file(self.memory_table_name)
+            topics = set()
+            for memory in memories:
+                memory_topics = memory.get("topics", [])
+                if isinstance(memory_topics, list):
+                    topics.update(memory_topics)
+            return list(topics)
+
+        except Exception as e:
+            log_warning(f"Exception reading from memory file: {e}")
+            return []
+
+    def get_user_memory(
+        self, memory_id: str, deserialize: Optional[bool] = True
+    ) -> Optional[Union[UserMemory, Dict[str, Any]]]:
+        """Get a memory from the GCS JSON file."""
+        try:
+            memories = self._read_json_file(self.memory_table_name)
+
+            for memory_data in memories:
+                if memory_data.get("memory_id") == memory_id:
+                    if not deserialize:
+                        return memory_data
+
+                    return UserMemory.from_dict(memory_data)
+
+            return None
+        except Exception as e:
+            log_warning(f"Exception reading from memory file: {e}")
+            return None
+
+    def get_user_memories(
+        self,
+        user_id: Optional[str] = None,
+        agent_id: Optional[str] = None,
+        team_id: Optional[str] = None,
+        topics: Optional[List[str]] = None,
+        search_content: Optional[str] = None,
+        limit: Optional[int] = None,
+        page: Optional[int] = None,
+        sort_by: Optional[str] = None,
+        sort_order: Optional[str] = None,
+        deserialize: Optional[bool] = True,
+    ) -> Union[List[UserMemory], Tuple[List[Dict[str, Any]], int]]:
+        """Get all memories from the GCS JSON file with filtering and pagination."""
+        try:
+            memories = self._read_json_file(self.memory_table_name)
+
+            # Apply filters
+            filtered_memories = []
+            for memory_data in memories:
+                if user_id is not None and memory_data.get("user_id") != user_id:
+                    continue
+                if agent_id is not None and memory_data.get("agent_id") != agent_id:
+                    continue
+                if team_id is not None and memory_data.get("team_id") != team_id:
+                    continue
+                if topics is not None:
+                    memory_topics = memory_data.get("topics", [])
+                    if not any(topic in memory_topics for topic in topics):
+                        continue
+                if search_content is not None:
+                    memory_content = str(memory_data.get("memory", ""))
+                    if search_content.lower() not in memory_content.lower():
+                        continue
+
+                filtered_memories.append(memory_data)
+
+            total_count = len(filtered_memories)
+
+            # Apply sorting
+            filtered_memories = apply_sorting(filtered_memories, sort_by, sort_order)
+
+            # Apply pagination
+            if limit is not None:
+                start_idx = 0
+                if page is not None:
+                    start_idx = (page - 1) * limit
+                filtered_memories = filtered_memories[start_idx : start_idx + limit]
+
+            if not deserialize:
+                return filtered_memories, total_count
+
+            return [UserMemory.from_dict(memory) for memory in filtered_memories]
+
+        except Exception as e:
+            log_warning(f"Exception reading from memory file: {e}")
+            return [] if deserialize else ([], 0)
+
+    def get_user_memory_stats(
+        self, limit: Optional[int] = None, page: Optional[int] = None
+    ) -> Tuple[List[Dict[str, Any]], int]:
+        """Get user memory statistics."""
+        try:
+            memories = self._read_json_file(self.memory_table_name)
+            user_stats = {}
+
+            for memory in memories:
+                user_id = memory.get("user_id")
+                if user_id:
+                    if user_id not in user_stats:
+                        user_stats[user_id] = {"user_id": user_id, "total_memories": 0, "last_memory_updated_at": 0}
+                    user_stats[user_id]["total_memories"] += 1
+                    updated_at = memory.get("updated_at", 0)
+                    if updated_at > user_stats[user_id]["last_memory_updated_at"]:
+                        user_stats[user_id]["last_memory_updated_at"] = updated_at
+
+            stats_list = list(user_stats.values())
+            stats_list.sort(key=lambda x: x["last_memory_updated_at"], reverse=True)
+
+            total_count = len(stats_list)
+
+            # Apply pagination
+            if limit is not None:
+                start_idx = 0
+                if page is not None:
+                    start_idx = (page - 1) * limit
+                stats_list = stats_list[start_idx : start_idx + limit]
+
+            return stats_list, total_count
+
+        except Exception as e:
+            log_warning(f"Exception getting user memory stats: {e}")
+            return [], 0
+
+    def upsert_user_memory(
+        self, memory: UserMemory, deserialize: Optional[bool] = True
+    ) -> Optional[Union[UserMemory, Dict[str, Any]]]:
+        """Upsert a user memory in the GCS JSON file."""
+        try:
+            memories = self._read_json_file(self.memory_table_name, create_table_if_not_found=True)
+
+            if memory.memory_id is None:
+                memory.memory_id = str(uuid4())
+
+            memory_dict = memory.to_dict() if hasattr(memory, "to_dict") else memory.__dict__
+            memory_dict["updated_at"] = int(time.time())
+
+            # Find existing memory to update
+            memory_updated = False
+            for i, existing_memory in enumerate(memories):
+                if existing_memory.get("memory_id") == memory.memory_id:
+                    memories[i] = memory_dict
+                    memory_updated = True
+                    break
+
+            if not memory_updated:
+                memories.append(memory_dict)
+
+            self._write_json_file(self.memory_table_name, memories)
+
+            if not deserialize:
+                return memory_dict
+            return UserMemory.from_dict(memory_dict)
+
+        except Exception as e:
+            log_error(f"Exception upserting user memory: {e}")
+            return None
+
+    def clear_memories(self) -> None:
+        """Delete all memories from the database.
+
+        Raises:
+            Exception: If an error occurs during deletion.
+        """
+        try:
+            # Simply write an empty list to the memory JSON file
+            self._write_json_file(self.memory_table_name, [])
+
+        except Exception as e:
+            log_warning(f"Exception deleting all memories: {e}")
+
+    # -- Metrics methods --
+    def calculate_metrics(self) -> Optional[list[dict]]:
+        """Calculate metrics for all dates without complete metrics."""
+        try:
+            metrics = self._read_json_file(self.metrics_table_name, create_table_if_not_found=True)
+
+            starting_date = self._get_metrics_calculation_starting_date(metrics)
+            if starting_date is None:
+                log_info("No session data found. Won't calculate metrics.")
+                return None
+
+            dates_to_process = get_dates_to_calculate_metrics_for(starting_date)
+            if not dates_to_process:
+                log_info("Metrics already calculated for all relevant dates.")
+                return None
+
+            start_timestamp = int(datetime.combine(dates_to_process[0], datetime.min.time()).timestamp())
+            end_timestamp = int(
+                datetime.combine(dates_to_process[-1] + timedelta(days=1), datetime.min.time()).timestamp()
+            )
+
+            sessions = self._get_all_sessions_for_metrics_calculation(start_timestamp, end_timestamp)
+            all_sessions_data = fetch_all_sessions_data(
+                sessions=sessions, dates_to_process=dates_to_process, start_timestamp=start_timestamp
+            )
+            if not all_sessions_data:
+                log_info("No new session data found. Won't calculate metrics.")
+                return None
+
+            results = []
+
+            for date_to_process in dates_to_process:
+                date_key = date_to_process.isoformat()
+                sessions_for_date = all_sessions_data.get(date_key, {})
+
+                # Skip dates with no sessions
+                if not any(len(sessions) > 0 for sessions in sessions_for_date.values()):
+                    continue
+
+                metrics_record = calculate_date_metrics(date_to_process, sessions_for_date)
+
+                # Upsert metrics record
+                existing_record_idx = None
+                for i, existing_metric in enumerate(metrics):
+                    if (
+                        existing_metric.get("date") == str(date_to_process)
+                        and existing_metric.get("aggregation_period") == "daily"
+                    ):
+                        existing_record_idx = i
+                        break
+
+                if existing_record_idx is not None:
+                    metrics[existing_record_idx] = metrics_record
+                else:
+                    metrics.append(metrics_record)
+
+                results.append(metrics_record)
+
+            if results:
+                self._write_json_file(self.metrics_table_name, metrics)
+
+            return results
+
+        except Exception as e:
+            log_warning(f"Exception refreshing metrics: {e}")
+            return None
+
+    def _get_metrics_calculation_starting_date(self, metrics: List[Dict[str, Any]]) -> Optional[date]:
+        """Get the first date for which metrics calculation is needed."""
+        if metrics:
+            # Sort by date in descending order
+            sorted_metrics = sorted(metrics, key=lambda x: x.get("date", ""), reverse=True)
+            latest_metric = sorted_metrics[0]
+
+            if latest_metric.get("completed", False):
+                latest_date = datetime.strptime(latest_metric["date"], "%Y-%m-%d").date()
+                return latest_date + timedelta(days=1)
+            else:
+                return datetime.strptime(latest_metric["date"], "%Y-%m-%d").date()
+
+        # No metrics records. Return the date of the first recorded session.
+        # We need to get sessions of all types, so we'll read directly from the file
+        all_sessions = self._read_json_file(self.session_table_name)
+        if all_sessions:
+            # Sort by created_at
+            all_sessions.sort(key=lambda x: x.get("created_at", 0))
+            first_session_date = all_sessions[0]["created_at"]
+            return datetime.fromtimestamp(first_session_date, tz=timezone.utc).date()
+
+        return None
+
+    def _get_all_sessions_for_metrics_calculation(
+        self, start_timestamp: Optional[int] = None, end_timestamp: Optional[int] = None
+    ) -> List[Dict[str, Any]]:
+        """Get all sessions for metrics calculation."""
+        try:
+            sessions = self._read_json_file(self.session_table_name)
+
+            filtered_sessions = []
+            for session in sessions:
+                created_at = session.get("created_at", 0)
+                if start_timestamp is not None and created_at < start_timestamp:
+                    continue
+                if end_timestamp is not None and created_at >= end_timestamp:
+                    continue
+
+                # Only include necessary fields for metrics
+                filtered_session = {
+                    "user_id": session.get("user_id"),
+                    "session_data": session.get("session_data"),
+                    "runs": session.get("runs"),
+                    "created_at": session.get("created_at"),
+                    "session_type": session.get("session_type"),
+                }
+                filtered_sessions.append(filtered_session)
+
+            return filtered_sessions
+
+        except Exception as e:
+            log_warning(f"Exception reading sessions for metrics: {e}")
+            return []
+
+    def get_metrics(
+        self,
+        starting_date: Optional[date] = None,
+        ending_date: Optional[date] = None,
+    ) -> Tuple[List[dict], Optional[int]]:
+        """Get all metrics matching the given date range."""
+        try:
+            metrics = self._read_json_file(self.metrics_table_name)
+
+            filtered_metrics = []
+            latest_updated_at = None
+
+            for metric in metrics:
+                metric_date = datetime.strptime(metric.get("date", ""), "%Y-%m-%d").date()
+
+                if starting_date and metric_date < starting_date:
+                    continue
+                if ending_date and metric_date > ending_date:
+                    continue
+
+                filtered_metrics.append(metric)
+
+                updated_at = metric.get("updated_at")
+                if updated_at and (latest_updated_at is None or updated_at > latest_updated_at):
+                    latest_updated_at = updated_at
+
+            return filtered_metrics, latest_updated_at
+
+        except Exception as e:
+            log_warning(f"Exception getting metrics: {e}")
+            return [], None
+
+    # -- Knowledge methods --
+    def delete_knowledge_content(self, id: str):
+        """Delete knowledge content by ID."""
+        try:
+            knowledge_items = self._read_json_file(self.knowledge_table_name)
+            knowledge_items = [item for item in knowledge_items if item.get("id") != id]
+            self._write_json_file(self.knowledge_table_name, knowledge_items)
+        except Exception as e:
+            log_warning(f"Error deleting knowledge content: {e}")
+
+    def get_knowledge_content(self, id: str) -> Optional[KnowledgeRow]:
+        """Get knowledge content by ID."""
+        try:
+            knowledge_items = self._read_json_file(self.knowledge_table_name)
+
+            for item in knowledge_items:
+                if item.get("id") == id:
+                    return KnowledgeRow.model_validate(item)
+
+            return None
+        except Exception as e:
+            log_warning(f"Error getting knowledge content: {e}")
+            return None
+
+    def get_knowledge_contents(
+        self,
+        limit: Optional[int] = None,
+        page: Optional[int] = None,
+        sort_by: Optional[str] = None,
+        sort_order: Optional[str] = None,
+    ) -> Tuple[List[KnowledgeRow], int]:
+        """Get all knowledge contents from the GCS JSON file."""
+        try:
+            knowledge_items = self._read_json_file(self.knowledge_table_name)
+
+            total_count = len(knowledge_items)
+
+            # Apply sorting
+            knowledge_items = apply_sorting(knowledge_items, sort_by, sort_order)
+
+            # Apply pagination
+            if limit is not None:
+                start_idx = 0
+                if page is not None:
+                    start_idx = (page - 1) * limit
+                knowledge_items = knowledge_items[start_idx : start_idx + limit]
+
+            return [KnowledgeRow.model_validate(item) for item in knowledge_items], total_count
+
+        except Exception as e:
+            log_warning(f"Error getting knowledge contents: {e}")
+            return [], 0
+
+    def upsert_knowledge_content(self, knowledge_row: KnowledgeRow):
+        """Upsert knowledge content in the GCS JSON file."""
+        try:
+            knowledge_items = self._read_json_file(self.knowledge_table_name, create_table_if_not_found=True)
+            knowledge_dict = knowledge_row.model_dump()
+
+            # Find existing item to update
+            item_updated = False
+            for i, existing_item in enumerate(knowledge_items):
+                if existing_item.get("id") == knowledge_row.id:
+                    knowledge_items[i] = knowledge_dict
+                    item_updated = True
+                    break
+
+            if not item_updated:
+                knowledge_items.append(knowledge_dict)
+
+            self._write_json_file(self.knowledge_table_name, knowledge_items)
+            return knowledge_row
+
+        except Exception as e:
+            log_warning(f"Error upserting knowledge row: {e}")
+            return None
+
+    # -- Eval methods --
+    def create_eval_run(self, eval_run: EvalRunRecord) -> Optional[EvalRunRecord]:
+        """Create an EvalRunRecord in the GCS JSON file."""
+        try:
+            eval_runs = self._read_json_file(self.eval_table_name, create_table_if_not_found=True)
+
+            current_time = int(time.time())
+            eval_dict = eval_run.model_dump()
+            eval_dict["created_at"] = current_time
+            eval_dict["updated_at"] = current_time
+
+            eval_runs.append(eval_dict)
+            self._write_json_file(self.eval_table_name, eval_runs)
+
+            return eval_run
+        except Exception as e:
+            log_warning(f"Error creating eval run: {e}")
+            return None
+
+    def delete_eval_run(self, eval_run_id: str) -> None:
+        """Delete an eval run from the GCS JSON file."""
+        try:
+            eval_runs = self._read_json_file(self.eval_table_name)
+            original_count = len(eval_runs)
+            eval_runs = [run for run in eval_runs if run.get("run_id") != eval_run_id]
+
+            if len(eval_runs) < original_count:
+                self._write_json_file(self.eval_table_name, eval_runs)
+                log_debug(f"Deleted eval run with ID: {eval_run_id}")
+            else:
+                log_warning(f"No eval run found with ID: {eval_run_id}")
+        except Exception as e:
+            log_warning(f"Error deleting eval run {eval_run_id}: {e}")
+
+    def delete_eval_runs(self, eval_run_ids: List[str]) -> None:
+        """Delete multiple eval runs from the GCS JSON file."""
+        try:
+            eval_runs = self._read_json_file(self.eval_table_name)
+            original_count = len(eval_runs)
+            eval_runs = [run for run in eval_runs if run.get("run_id") not in eval_run_ids]
+
+            deleted_count = original_count - len(eval_runs)
+            if deleted_count > 0:
+                self._write_json_file(self.eval_table_name, eval_runs)
+                log_debug(f"Deleted {deleted_count} eval runs")
+            else:
+                log_warning(f"No eval runs found with IDs: {eval_run_ids}")
+        except Exception as e:
+            log_warning(f"Error deleting eval runs {eval_run_ids}: {e}")
+
+    def get_eval_run(
+        self, eval_run_id: str, deserialize: Optional[bool] = True
+    ) -> Optional[Union[EvalRunRecord, Dict[str, Any]]]:
+        """Get an eval run from the GCS JSON file."""
+        try:
+            eval_runs = self._read_json_file(self.eval_table_name)
+
+            for run_data in eval_runs:
+                if run_data.get("run_id") == eval_run_id:
+                    if not deserialize:
+                        return run_data
+                    return EvalRunRecord.model_validate(run_data)
+
+            return None
+        except Exception as e:
+            log_warning(f"Exception getting eval run {eval_run_id}: {e}")
+            return None
+
+    def get_eval_runs(
+        self,
+        limit: Optional[int] = None,
+        page: Optional[int] = None,
+        sort_by: Optional[str] = None,
+        sort_order: Optional[str] = None,
+        agent_id: Optional[str] = None,
+        team_id: Optional[str] = None,
+        workflow_id: Optional[str] = None,
+        model_id: Optional[str] = None,
+        filter_type: Optional[EvalFilterType] = None,
+        eval_type: Optional[List[EvalType]] = None,
+        deserialize: Optional[bool] = True,
+    ) -> Union[List[EvalRunRecord], Tuple[List[Dict[str, Any]], int]]:
+        """Get all eval runs from the GCS JSON file with filtering and pagination."""
+        try:
+            eval_runs = self._read_json_file(self.eval_table_name)
+
+            # Apply filters
+            filtered_runs = []
+            for run_data in eval_runs:
+                if agent_id is not None and run_data.get("agent_id") != agent_id:
+                    continue
+                if team_id is not None and run_data.get("team_id") != team_id:
+                    continue
+                if workflow_id is not None and run_data.get("workflow_id") != workflow_id:
+                    continue
+                if model_id is not None and run_data.get("model_id") != model_id:
+                    continue
+                if eval_type is not None and len(eval_type) > 0:
+                    if run_data.get("eval_type") not in eval_type:
+                        continue
+                if filter_type is not None:
+                    if filter_type == EvalFilterType.AGENT and run_data.get("agent_id") is None:
+                        continue
+                    elif filter_type == EvalFilterType.TEAM and run_data.get("team_id") is None:
+                        continue
+                    elif filter_type == EvalFilterType.WORKFLOW and run_data.get("workflow_id") is None:
+                        continue
+
+                filtered_runs.append(run_data)
+
+            total_count = len(filtered_runs)
+
+            # Apply sorting (default by created_at desc)
+            if sort_by is None:
+                filtered_runs.sort(key=lambda x: x.get("created_at", 0), reverse=True)
+            else:
+                filtered_runs = apply_sorting(filtered_runs, sort_by, sort_order)
+
+            # Apply pagination
+            if limit is not None:
+                start_idx = 0
+                if page is not None:
+                    start_idx = (page - 1) * limit
+                filtered_runs = filtered_runs[start_idx : start_idx + limit]
+
+            if not deserialize:
+                return filtered_runs, total_count
+
+            return [EvalRunRecord.model_validate(run) for run in filtered_runs]
+
+        except Exception as e:
+            log_warning(f"Exception getting eval runs: {e}")
+            return [] if deserialize else ([], 0)
+
+    def rename_eval_run(
+        self, eval_run_id: str, name: str, deserialize: Optional[bool] = True
+    ) -> Optional[Union[EvalRunRecord, Dict[str, Any]]]:
+        """Rename an eval run in the GCS JSON file."""
+        try:
+            eval_runs = self._read_json_file(self.eval_table_name)
+
+            for i, run_data in enumerate(eval_runs):
+                if run_data.get("run_id") == eval_run_id:
+                    run_data["name"] = name
+                    run_data["updated_at"] = int(time.time())
+                    eval_runs[i] = run_data
+                    self._write_json_file(self.eval_table_name, eval_runs)
+
+                    if not deserialize:
+                        return run_data
+                    return EvalRunRecord.model_validate(run_data)
+
+            return None
+        except Exception as e:
+            log_warning(f"Error renaming eval run {eval_run_id}: {e}")
+            return None