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

agno/db/mysql/schemas.py

@@ -0,0 +1,124 @@
+"""Table schemas and related utils used by the MySQLDb class"""
+
+from typing import Any
+
+try:
+    from sqlalchemy.types import JSON, BigInteger, Boolean, Date, String, Text
+except ImportError:
+    raise ImportError("`sqlalchemy` not installed. Please install it using `pip install sqlalchemy`")
+
+SESSION_TABLE_SCHEMA = {
+    "session_id": {"type": String, "nullable": False},
+    "session_type": {"type": String, "nullable": False, "index": True},
+    "agent_id": {"type": String, "nullable": True},
+    "team_id": {"type": String, "nullable": True},
+    "workflow_id": {"type": String, "nullable": True},
+    "user_id": {"type": String, "nullable": True},
+    "session_data": {"type": JSON, "nullable": True},
+    "agent_data": {"type": JSON, "nullable": True},
+    "team_data": {"type": JSON, "nullable": True},
+    "workflow_data": {"type": JSON, "nullable": True},
+    "metadata": {"type": JSON, "nullable": True},
+    "runs": {"type": JSON, "nullable": True},
+    "summary": {"type": JSON, "nullable": True},
+    "created_at": {"type": BigInteger, "nullable": False, "index": True},
+    "updated_at": {"type": BigInteger, "nullable": True},
+    "_unique_constraints": [
+        {
+            "name": "uq_session_id",
+            "columns": ["session_id"],
+        },
+    ],
+}
+
+USER_MEMORY_TABLE_SCHEMA = {
+    "memory_id": {"type": String, "primary_key": True, "nullable": False},
+    "memory": {"type": JSON, "nullable": False},
+    "input": {"type": Text, "nullable": True},
+    "agent_id": {"type": String, "nullable": True},
+    "team_id": {"type": String, "nullable": True},
+    "user_id": {"type": String, "nullable": True, "index": True},
+    "topics": {"type": JSON, "nullable": True},
+    "updated_at": {"type": BigInteger, "nullable": True, "index": True},
+}
+
+EVAL_TABLE_SCHEMA = {
+    "run_id": {"type": String, "primary_key": True, "nullable": False},
+    "eval_type": {"type": String, "nullable": False},
+    "eval_data": {"type": JSON, "nullable": False},
+    "eval_input": {"type": JSON, "nullable": False},
+    "name": {"type": String, "nullable": True},
+    "agent_id": {"type": String, "nullable": True},
+    "team_id": {"type": String, "nullable": True},
+    "workflow_id": {"type": String, "nullable": True},
+    "model_id": {"type": String, "nullable": True},
+    "model_provider": {"type": String, "nullable": True},
+    "evaluated_component_name": {"type": String, "nullable": True},
+    "created_at": {"type": BigInteger, "nullable": False, "index": True},
+    "updated_at": {"type": BigInteger, "nullable": True},
+}
+
+KNOWLEDGE_TABLE_SCHEMA = {
+    "id": {"type": String, "primary_key": True, "nullable": False},
+    "name": {"type": String, "nullable": False},
+    "description": {"type": Text, "nullable": False},
+    "metadata": {"type": JSON, "nullable": True},
+    "type": {"type": String, "nullable": True},
+    "size": {"type": BigInteger, "nullable": True},
+    "linked_to": {"type": String, "nullable": True},
+    "access_count": {"type": BigInteger, "nullable": True},
+    "created_at": {"type": BigInteger, "nullable": True},
+    "updated_at": {"type": BigInteger, "nullable": True},
+    "status": {"type": String, "nullable": True},
+    "status_message": {"type": Text, "nullable": True},
+    "external_id": {"type": String, "nullable": True},
+}
+
+METRICS_TABLE_SCHEMA = {
+    "id": {"type": String, "primary_key": True, "nullable": False},
+    "agent_runs_count": {"type": BigInteger, "nullable": False},
+    "team_runs_count": {"type": BigInteger, "nullable": False},
+    "workflow_runs_count": {"type": BigInteger, "nullable": False},
+    "agent_sessions_count": {"type": BigInteger, "nullable": False},
+    "team_sessions_count": {"type": BigInteger, "nullable": False},
+    "workflow_sessions_count": {"type": BigInteger, "nullable": False},
+    "users_count": {"type": BigInteger, "nullable": False},
+    "token_metrics": {"type": JSON, "nullable": False},
+    "model_metrics": {"type": JSON, "nullable": False},
+    "date": {"type": Date, "nullable": False, "index": True},
+    "aggregation_period": {"type": String, "nullable": False},
+    "created_at": {"type": BigInteger, "nullable": False},
+    "updated_at": {"type": BigInteger, "nullable": True},
+    "completed": {"type": Boolean, "nullable": False},
+    "_unique_constraints": [
+        {
+            "name": "uq_metrics_date_period",
+            "columns": ["date", "aggregation_period"],
+        }
+    ],
+}
+
+
+def get_table_schema_definition(table_type: str) -> dict[str, Any]:
+    """
+    Get the expected schema definition for the given table.
+
+    Args:
+        table_type (str): The type of table to get the schema for.
+
+    Returns:
+        Dict[str, Any]: Dictionary containing column definitions for the table
+    """
+    schemas = {
+        "sessions": SESSION_TABLE_SCHEMA,
+        "evals": EVAL_TABLE_SCHEMA,
+        "metrics": METRICS_TABLE_SCHEMA,
+        "memories": USER_MEMORY_TABLE_SCHEMA,
+        "knowledge": KNOWLEDGE_TABLE_SCHEMA,
+    }
+
+    schema = schemas.get(table_type, {})
+    if not schema:
+        raise ValueError(f"Unknown table type: {table_type}")
+
+    return schema  # type: ignore[return-value]

agno/db/mysql/utils.py

@@ -0,0 +1,297 @@
+"""Utility functions for the MySQL database class."""
+
+import time
+from datetime import date, datetime, timedelta, timezone
+from typing import Any, Dict, List, Optional
+from uuid import uuid4
+
+from sqlalchemy import Engine
+
+from agno.db.mysql.schemas import get_table_schema_definition
+from agno.utils.log import log_debug, log_error, log_warning
+
+try:
+    from sqlalchemy import Table
+    from sqlalchemy.dialects import mysql
+    from sqlalchemy.inspection import inspect
+    from sqlalchemy.orm import Session
+    from sqlalchemy.sql.expression import text
+except ImportError:
+    raise ImportError("`sqlalchemy` not installed. Please install it using `pip install sqlalchemy`")
+
+
+# -- DB util methods --
+def apply_sorting(stmt, table: Table, sort_by: Optional[str] = None, sort_order: Optional[str] = None):
+    """Apply sorting to the given SQLAlchemy statement.
+
+    Args:
+        stmt: The SQLAlchemy statement to modify
+        table: The table being queried
+        sort_by: The field to sort by
+        sort_order: The sort order ('asc' or 'desc')
+
+    Returns:
+        The modified statement with sorting applied
+    """
+    if sort_by is None:
+        return stmt
+
+    if not hasattr(table.c, sort_by):
+        log_debug(f"Invalid sort field: '{sort_by}'. Will not apply any sorting.")
+        return stmt
+
+    # Apply the given sorting
+    sort_column = getattr(table.c, sort_by)
+    if sort_order and sort_order == "asc":
+        return stmt.order_by(sort_column.asc())
+    else:
+        return stmt.order_by(sort_column.desc())
+
+
+def create_schema(session: Session, db_schema: str) -> None:
+    """Create the database schema if it doesn't exist.
+
+    Args:
+        session: The SQLAlchemy session to use
+        db_schema (str): The definition of the database schema to create
+    """
+    try:
+        log_debug(f"Creating database if not exists: {db_schema}")
+        # MySQL uses CREATE DATABASE instead of CREATE SCHEMA
+        session.execute(text(f"CREATE DATABASE IF NOT EXISTS {db_schema};"))
+    except Exception as e:
+        log_warning(f"Could not create database {db_schema}: {e}")
+
+
+def is_table_available(session: Session, table_name: str, db_schema: str) -> bool:
+    """
+    Check if a table with the given name exists in the given schema.
+
+    Returns:
+        bool: True if the table exists, False otherwise.
+    """
+    try:
+        exists_query = text(
+            "SELECT 1 FROM information_schema.tables WHERE table_schema = :schema AND table_name = :table"
+        )
+        exists = session.execute(exists_query, {"schema": db_schema, "table": table_name}).scalar() is not None
+        if not exists:
+            log_debug(f"Table {db_schema}.{table_name} {'exists' if exists else 'does not exist'}")
+
+        return exists
+
+    except Exception as e:
+        log_error(f"Error checking if table exists: {e}")
+        return False
+
+
+def is_valid_table(db_engine: Engine, table_name: str, table_type: str, db_schema: str) -> bool:
+    """
+    Check if the existing table has the expected column names.
+
+    Args:
+        table_name (str): Name of the table to validate
+        schema (str): Database schema name
+
+    Returns:
+        bool: True if table has all expected columns, False otherwise
+    """
+    try:
+        expected_table_schema = get_table_schema_definition(table_type)
+        expected_columns = {col_name for col_name in expected_table_schema.keys() if not col_name.startswith("_")}
+
+        # Get existing columns
+        inspector = inspect(db_engine)
+        existing_columns_info = inspector.get_columns(table_name, schema=db_schema)
+        existing_columns = set(col["name"] for col in existing_columns_info)
+
+        # Check if all expected columns exist
+        missing_columns = expected_columns - existing_columns
+        if missing_columns:
+            log_warning(f"Missing columns {missing_columns} in table {db_schema}.{table_name}")
+            return False
+
+        return True
+    except Exception as e:
+        log_error(f"Error validating table schema for {db_schema}.{table_name}: {e}")
+        return False
+
+
+# -- Metrics util methods --
+def bulk_upsert_metrics(session: Session, table: Table, metrics_records: list[dict]) -> list[dict]:
+    """Bulk upsert metrics into the database.
+
+    Args:
+        table (Table): The table to upsert into.
+        metrics_records (list[dict]): The metrics records to upsert.
+
+    Returns:
+        list[dict]: The upserted metrics records.
+    """
+    if not metrics_records:
+        return []
+
+    results = []
+
+    # MySQL doesn't support returning in the same way as PostgreSQL
+    # We'll need to insert/update and then fetch the records
+    for record in metrics_records:
+        stmt = mysql.insert(table).values(record)
+
+        # Columns to update in case of conflict
+        update_dict = {
+            col.name: record.get(col.name)
+            for col in table.columns
+            if col.name not in ["id", "date", "created_at", "aggregation_period"] and col.name in record
+        }
+
+        stmt = stmt.on_duplicate_key_update(**update_dict)
+        session.execute(stmt)
+
+    session.commit()
+
+    # Fetch the updated records
+    from sqlalchemy import and_, select
+
+    for record in metrics_records:
+        select_stmt = select(table).where(
+            and_(table.c.date == record["date"], table.c.aggregation_period == record["aggregation_period"])
+        )
+        result = session.execute(select_stmt).fetchone()
+        if result:
+            results.append(result._mapping)
+
+    return results  # type: ignore
+
+
+def calculate_date_metrics(date_to_process: date, sessions_data: dict) -> dict:
+    """Calculate metrics for the given single date.
+
+    Args:
+        date_to_process (date): The date to calculate metrics for.
+        sessions_data (dict): The sessions data to calculate metrics for.
+
+    Returns:
+        dict: The calculated metrics.
+    """
+    metrics = {
+        "users_count": 0,
+        "agent_sessions_count": 0,
+        "team_sessions_count": 0,
+        "workflow_sessions_count": 0,
+        "agent_runs_count": 0,
+        "team_runs_count": 0,
+        "workflow_runs_count": 0,
+    }
+    token_metrics = {
+        "input_tokens": 0,
+        "output_tokens": 0,
+        "total_tokens": 0,
+        "audio_total_tokens": 0,
+        "audio_input_tokens": 0,
+        "audio_output_tokens": 0,
+        "cache_read_tokens": 0,
+        "cache_write_tokens": 0,
+        "reasoning_tokens": 0,
+    }
+    model_counts: Dict[str, int] = {}
+
+    session_types = [
+        ("agent", "agent_sessions_count", "agent_runs_count"),
+        ("team", "team_sessions_count", "team_runs_count"),
+        ("workflow", "workflow_sessions_count", "workflow_runs_count"),
+    ]
+    all_user_ids = set()
+
+    for session_type, sessions_count_key, runs_count_key in session_types:
+        sessions = sessions_data.get(session_type, [])
+        metrics[sessions_count_key] = len(sessions)
+
+        for session in sessions:
+            if session.get("user_id"):
+                all_user_ids.add(session["user_id"])
+            metrics[runs_count_key] += len(session.get("runs", []))
+            if runs := session.get("runs", []):
+                for run in runs:
+                    if model_id := run.get("model"):
+                        model_provider = run.get("model_provider", "")
+                        model_counts[f"{model_id}:{model_provider}"] = (
+                            model_counts.get(f"{model_id}:{model_provider}", 0) + 1
+                        )
+
+            session_metrics = session.get("session_data", {}).get("session_metrics", {})
+            for field in token_metrics:
+                token_metrics[field] += session_metrics.get(field, 0)
+
+    model_metrics = []
+    for model, count in model_counts.items():
+        model_id, model_provider = model.split(":")
+        model_metrics.append({"model_id": model_id, "model_provider": model_provider, "count": count})
+
+    metrics["users_count"] = len(all_user_ids)
+    current_time = int(time.time())
+
+    return {
+        "id": str(uuid4()),
+        "date": date_to_process,
+        "completed": date_to_process < datetime.now(timezone.utc).date(),
+        "token_metrics": token_metrics,
+        "model_metrics": model_metrics,
+        "created_at": current_time,
+        "updated_at": current_time,
+        "aggregation_period": "daily",
+        **metrics,
+    }
+
+
+def fetch_all_sessions_data(
+    sessions: List[Dict[str, Any]], dates_to_process: list[date], start_timestamp: int
+) -> Optional[dict]:
+    """Return all session data for the given dates, for all session types.
+
+    Args:
+        dates_to_process (list[date]): The dates to fetch session data for.
+
+    Returns:
+        dict: A dictionary with dates as keys and session data as values, for all session types.
+
+    Example:
+    {
+        "2000-01-01": {
+            "agent": [<session1>, <session2>, ...],
+            "team": [...],
+            "workflow": [...],
+        }
+    }
+    """
+    if not dates_to_process:
+        return None
+
+    all_sessions_data: Dict[str, Dict[str, List[Dict[str, Any]]]] = {
+        date_to_process.isoformat(): {"agent": [], "team": [], "workflow": []} for date_to_process in dates_to_process
+    }
+
+    for session in sessions:
+        session_date = (
+            datetime.fromtimestamp(session.get("created_at", start_timestamp), tz=timezone.utc).date().isoformat()
+        )
+        if session_date in all_sessions_data:
+            all_sessions_data[session_date][session["session_type"]].append(session)
+
+    return all_sessions_data
+
+
+def get_dates_to_calculate_metrics_for(starting_date: date) -> list[date]:
+    """Return the list of dates to calculate metrics for.
+
+    Args:
+        starting_date (date): The starting date to calculate metrics for.
+
+    Returns:
+        list[date]: The list of dates to calculate metrics for.
+    """
+    today = datetime.now(timezone.utc).date()
+    days_diff = (today - starting_date).days + 1
+    if days_diff <= 0:
+        return []
+    return [starting_date + timedelta(days=x) for x in range(days_diff)]

agno/db/postgres/__init__.py

@@ -0,0 +1,3 @@
+from agno.db.postgres.postgres import PostgresDb
+
+__all__ = ["PostgresDb"]