MindsDB 25.4.5.0__py3-none-any.whl → 25.5.4.0__py3-none-any.whl

mindsdb/interfaces/storage/fs.py

@@ -37,6 +37,7 @@ class RESOURCE_GROUP:
     PREDICTOR = 'predictor'
     INTEGRATION = 'integration'
     TAB = 'tab'
+    SYSTEM = 'system'
 
 
 RESOURCE_GROUP = RESOURCE_GROUP()

mindsdb/interfaces/variables/variables_controller.py

@@ -0,0 +1,97 @@
+import os
+from typing import Callable
+
+from mindsdb_sql_parser import Function, Constant, Variable
+
+from mindsdb.utilities import log
+from mindsdb.interfaces.storage.fs import RESOURCE_GROUP
+from mindsdb.interfaces.storage.json import get_json_storage
+
+
+logger = log.getLogger(__name__)
+
+
+ENV_VAR_PREFIX = "MDB_"
+
+
+class VariablesController:
+
+    def __init__(self) -> None:
+        self._storage = get_json_storage(
+            resource_id=0,
+            resource_group=RESOURCE_GROUP.SYSTEM
+        )
+        self._store_key = 'variables'
+        self._data = None
+
+    def _get_data(self) -> dict:
+        if self._data is None:
+            self._data = self._storage.get(self._store_key)
+            if self._data is None:
+                self._data = {}
+        return self._data
+
+    def get_value(self, name: str):
+        data = self._get_data()
+        if name not in data:
+            raise ValueError(f"Variable {name} is not defined")
+        return data[name]
+
+    def set_value(self, name: str, value):
+        data = self._get_data()
+        data[name] = value
+        self._storage.set(self._store_key, data)
+
+    def _from_env(self, name: Constant) -> str:
+        # gets variable value from environment.
+        # available names are restricted by ENV_VAR_PREFIX to don't provide access to arbitrary venv variable
+
+        var_name = name.value
+        if not var_name.startswith(ENV_VAR_PREFIX):
+            raise ValueError(f"Can access only to variable names starting with {ENV_VAR_PREFIX}")
+        if var_name not in os.environ:
+            raise ValueError(f"Environment variable {var_name} is not defined")
+        return os.environ[var_name]
+
+    def _get_function(self, name: str) -> Callable:
+        if name == 'from_env':
+            return self._from_env
+        raise ValueError(f"Function {name} is not found")
+
+    def set_variable(self, name: str, value):
+        # store new value for variable in database
+        #   if value is a function - extract value using this function
+
+        name = name.lower()
+        if isinstance(value, Function):
+            fnc = self._get_function(value.op)
+            value = fnc(*value.args)
+
+        elif isinstance(value, Constant):
+            value = value.value
+
+        else:
+            # ignore
+            return
+
+        self.set_value(name, value)
+
+    def fill_parameters(self, var):
+        # recursively check input and fill Variables if they exist there
+
+        if isinstance(var, Variable):
+            return self.get_value(var.value.lower())
+        elif isinstance(var, dict):
+            return {
+                key: self.fill_parameters(value)
+                for key, value in var.items()
+            }
+        elif isinstance(var, list):
+            return [
+                self.fill_parameters(value)
+                for value in var
+            ]
+        return var
+
+
+variables_controller = VariablesController()

mindsdb/migrations/env.py

@@ -1,5 +1,9 @@
 from alembic import context
 from sqlalchemy import engine_from_config, pool
+import os
+
+from mindsdb.interfaces.storage import db
+from mindsdb.utilities.config import config as app_config
 
 # this is the Alembic Config object, which provides
 # access to the values within the .ini file in use.
@@ -10,12 +14,6 @@ config = context.config
 # from myapp import mymodel
 # target_metadata = mymodel.Base.metadata
 
-
-import os
-
-from mindsdb.interfaces.storage import db
-from mindsdb.utilities.config import config as app_config
-
 # initialize
 
 if "MINDSDB_CONFIG_PATH" not in os.environ:
@@ -25,7 +23,7 @@ db.init()
 
 target_metadata = db.Base.metadata
 
-config.set_main_option("sqlalchemy.url", app_config['storage_db'])
+config.set_main_option("sqlalchemy.url", app_config["storage_db"])
 
 
 # other values from the config, defined by the needs of env.py,

mindsdb/migrations/migrate.py

@@ -4,13 +4,52 @@ from alembic.command import upgrade, autogen    # noqa
 from alembic.config import Config
 from alembic.script import ScriptDirectory
 from alembic.script.revision import ResolutionError
+from alembic.operations import Operations
 from alembic.migration import MigrationContext
+from alembic import util
 
 import mindsdb.interfaces.storage.db as db
 from mindsdb.utilities import log
 
 logger = log.getLogger(__name__)
 
+# This is a migration that is like a 'base version'. Applying only this
+# migration to a fresh DB is equivalent to applying all previous migrations.
+current_checkpoint = '9f150e4f9a05'
+
+
+def apply_checkpoint_migration(script) -> None:
+    """Apply the checkpoint migration to the database.
+    """
+    with db.engine.begin() as connection:
+        context = MigrationContext.configure(
+            connection,
+            opts={
+                'as_sql': False,
+                'starting_rev': None,  # ignore current version
+                'destination_rev': current_checkpoint,
+            }
+        )
+        revision = script.get_revision(current_checkpoint)
+        if not revision:
+            raise util.CommandError(f"Migration {current_checkpoint} not found.")
+
+        op = Operations(context)
+        revision.module.upgrade(op)
+        context.stamp(script, current_checkpoint)
+        connection.commit()
+
+
+def get_current_revision() -> str | None:
+    """ Get the current revision of the database.
+
+    Returns:
+        str | None: The current revision of the database.
+    """
+    with db.engine.begin() as conn:
+        mc = MigrationContext.configure(conn)
+        return mc.get_current_revision()
+
 
 def migrate_to_head():
     """ Trying to update database to head revision.
@@ -26,14 +65,15 @@ def migrate_to_head():
     config.set_main_option('script_location', str(script_location_abc))
 
     script = ScriptDirectory.from_config(config)
-    with db.engine.begin() as conn:
-        mc = MigrationContext.configure(conn)
-        cur_revision = mc.get_current_revision()
+    cur_revision = get_current_revision()
+    if cur_revision is None:
+        apply_checkpoint_migration(script)
+        cur_revision = get_current_revision()
 
-        try:
-            script.revision_map.get_revision(cur_revision)
-        except ResolutionError:
-            raise Exception("Database version higher than application.")
+    try:
+        script.revision_map.get_revision(cur_revision)
+    except ResolutionError:
+        raise Exception("Database version higher than application.")
 
     head_rev = script.get_current_head()
     if cur_revision == head_rev:
@@ -47,7 +87,5 @@ def migrate_to_head():
 if __name__ == "__main__":
     # have to import this because
     # all env initialization happens here
-    from mindsdb.utilities.config import Config as MDBConfig
-    MDBConfig()
     db.init()
     migrate_to_head()

mindsdb/migrations/versions/2025-05-21_9f150e4f9a05_checkpoint_1.py

@@ -0,0 +1,360 @@
+"""checkpoint_1
+
+Revision ID: 9f150e4f9a05
+Revises: 53502b6d63bf
+Create Date: 2025-05-21 12:25:55.556388
+
+"""
+import datetime
+
+from alembic.operations import Operations
+import sqlalchemy as sa
+import mindsdb.interfaces.storage.db  # noqa
+
+
+# revision identifiers, used by Alembic.
+revision = '9f150e4f9a05'
+down_revision = '53502b6d63bf'
+branch_labels = None
+depends_on = None
+
+
+def upgrade(op: Operations = None):
+    # region skip migration if it is existing app, apply if it is new app
+    if op is None:
+        # 'op' is passed only from migrate.py when applying checkpoin migration
+        return
+    connection = op.get_bind()
+    inspector = sa.inspect(connection)
+    if 'alembic_version' in inspector.get_table_names():
+        # If version_num exists, then it is existing app
+        result = connection.execute(sa.text("SELECT version_num FROM alembic_version"))
+        current_version = result.scalar()
+        if current_version is not None:
+            return
+    # endregion
+
+    op.create_table(
+        'agents',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.Column('user_class', sa.Integer(), nullable=True),
+        sa.Column('name', sa.String(), nullable=False),
+        sa.Column('project_id', sa.Integer(), nullable=False),
+        sa.Column('model_name', sa.String(), nullable=True),
+        sa.Column('provider', sa.String(), nullable=True),
+        sa.Column('params', sa.JSON(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.Column('deleted_at', sa.DateTime(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'chat_bots_history',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('chat_bot_id', sa.Integer(), nullable=False),
+        sa.Column('type', sa.String(), nullable=True),
+        sa.Column('text', sa.String(), nullable=True),
+        sa.Column('user', sa.String(), nullable=True),
+        sa.Column('destination', sa.String(), nullable=True),
+        sa.Column('sent_at', sa.DateTime(), nullable=True),
+        sa.Column('error', sa.String(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'file',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('name', sa.String(), nullable=False),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.Column('source_file_path', sa.String(), nullable=False),
+        sa.Column('file_path', sa.String(), nullable=False),
+        sa.Column('row_count', sa.Integer(), nullable=False),
+        sa.Column('columns', mindsdb.interfaces.storage.db.Json(), nullable=False),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.Column('metadata', sa.JSON(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.PrimaryKeyConstraint('id'),
+        sa.UniqueConstraint('name', 'company_id', name='unique_file_name_company_id')
+    )
+    op.create_table(
+        'integration',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.Column('name', sa.String(), nullable=False),
+        sa.Column('engine', sa.String(), nullable=False),
+        sa.Column('data', mindsdb.interfaces.storage.db.Json(), nullable=True),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.PrimaryKeyConstraint('id'),
+        sa.UniqueConstraint('name', 'company_id', name='unique_integration_name_company_id')
+    )
+    op.create_table(
+        'jobs',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.Column('user_class', sa.Integer(), nullable=True),
+        sa.Column('active', sa.Boolean(), nullable=True),
+        sa.Column('name', sa.String(), nullable=False),
+        sa.Column('project_id', sa.Integer(), nullable=False),
+        sa.Column('query_str', sa.String(), nullable=False),
+        sa.Column('if_query_str', sa.String(), nullable=True),
+        sa.Column('start_at', sa.DateTime(), nullable=True),
+        sa.Column('end_at', sa.DateTime(), nullable=True),
+        sa.Column('next_run_at', sa.DateTime(), nullable=True),
+        sa.Column('schedule_str', sa.String(), nullable=True),
+        sa.Column('deleted_at', sa.DateTime(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'jobs_history',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.Column('job_id', sa.Integer(), nullable=True),
+        sa.Column('query_str', sa.String(), nullable=True),
+        sa.Column('start_at', sa.DateTime(), nullable=True),
+        sa.Column('end_at', sa.DateTime(), nullable=True),
+        sa.Column('error', sa.String(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.PrimaryKeyConstraint('id'),
+        sa.UniqueConstraint('job_id', 'start_at', name='uniq_job_history_job_id_start')
+    )
+    op.create_table(
+        'json_storage',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('resource_group', sa.String(), nullable=True),
+        sa.Column('resource_id', sa.Integer(), nullable=True),
+        sa.Column('name', sa.String(), nullable=True),
+        sa.Column('content', sa.JSON(), nullable=True),
+        sa.Column('encrypted_content', sa.LargeBinary(), nullable=True),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'llm_data',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('input', sa.String(), nullable=False),
+        sa.Column('output', sa.String(), nullable=False),
+        sa.Column('model_id', sa.Integer(), nullable=False),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'llm_log',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('company_id', sa.Integer(), nullable=False),
+        sa.Column('api_key', sa.String(), nullable=True),
+        sa.Column('model_id', sa.Integer(), nullable=True),
+        sa.Column('model_group', sa.String(), nullable=True),
+        sa.Column('input', sa.JSON(), nullable=True),
+        sa.Column('output', sa.JSON(), nullable=True),
+        sa.Column('start_time', sa.DateTime(), nullable=False),
+        sa.Column('end_time', sa.DateTime(), nullable=True),
+        sa.Column('cost', sa.Numeric(precision=5, scale=2), nullable=True),
+        sa.Column('prompt_tokens', sa.Integer(), nullable=True),
+        sa.Column('completion_tokens', sa.Integer(), nullable=True),
+        sa.Column('total_tokens', sa.Integer(), nullable=True),
+        sa.Column('success', sa.Boolean(), nullable=False),
+        sa.Column('exception', sa.String(), nullable=True),
+        sa.Column('traceback', sa.String(), nullable=True),
+        sa.Column('stream', sa.Boolean(), nullable=True, comment="Is this completion done in 'streaming' mode"),
+        sa.Column('metadata', sa.JSON(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'project',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('deleted_at', sa.DateTime(), nullable=True),
+        sa.Column('name', sa.String(), nullable=False),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.Column('metadata', sa.JSON(), nullable=True),
+        sa.PrimaryKeyConstraint('id'),
+        sa.UniqueConstraint('name', 'company_id', name='unique_project_name_company_id')
+    )
+    op.create_table(
+        'queries',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.Column('sql', sa.String(), nullable=False),
+        sa.Column('database', sa.String(), nullable=True),
+        sa.Column('started_at', sa.DateTime(), nullable=True),
+        sa.Column('finished_at', sa.DateTime(), nullable=True),
+        sa.Column('parameters', sa.JSON(), nullable=True),
+        sa.Column('context', sa.JSON(), nullable=True),
+        sa.Column('processed_rows', sa.Integer(), nullable=True),
+        sa.Column('error', sa.String(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'query_context',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.Column('query', sa.String(), nullable=False),
+        sa.Column('context_name', sa.String(), nullable=False),
+        sa.Column('values', sa.JSON(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'skills',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('name', sa.String(), nullable=False),
+        sa.Column('project_id', sa.Integer(), nullable=False),
+        sa.Column('type', sa.String(), nullable=False),
+        sa.Column('params', sa.JSON(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('deleted_at', sa.DateTime(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'tasks',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.Column('user_class', sa.Integer(), nullable=True),
+        sa.Column('object_type', sa.String(), nullable=False),
+        sa.Column('object_id', sa.Integer(), nullable=False),
+        sa.Column('last_error', sa.String(), nullable=True),
+        sa.Column('active', sa.Boolean(), nullable=True),
+        sa.Column('reload', sa.Boolean(), nullable=True),
+        sa.Column('run_by', sa.String(), nullable=True),
+        sa.Column('alive_time', sa.DateTime(timezone=True), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'triggers',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('name', sa.String(), nullable=False),
+        sa.Column('project_id', sa.Integer(), nullable=False),
+        sa.Column('database_id', sa.Integer(), nullable=False),
+        sa.Column('table_name', sa.String(), nullable=False),
+        sa.Column('query_str', sa.String(), nullable=False),
+        sa.Column('columns', sa.String(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'agent_skills',
+        sa.Column('agent_id', sa.Integer(), nullable=False),
+        sa.Column('skill_id', sa.Integer(), nullable=False),
+        sa.Column('parameters', sa.JSON(), nullable=True),
+        sa.ForeignKeyConstraint(['agent_id'], ['agents.id'], ),
+        sa.ForeignKeyConstraint(['skill_id'], ['skills.id'], ),
+        sa.PrimaryKeyConstraint('agent_id', 'skill_id')
+    )
+    op.create_table(
+        'chat_bots',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('name', sa.String(), nullable=False),
+        sa.Column('project_id', sa.Integer(), nullable=False),
+        sa.Column('agent_id', sa.Integer(), nullable=True),
+        sa.Column('model_name', sa.String(), nullable=True),
+        sa.Column('database_id', sa.Integer(), nullable=True),
+        sa.Column('params', sa.JSON(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.Column('webhook_token', sa.String(), nullable=True),
+        sa.ForeignKeyConstraint(['agent_id'], ['agents.id'], name='fk_agent_id'),
+        sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table(
+        'predictor',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.Column('deleted_at', sa.DateTime(), nullable=True),
+        sa.Column('name', sa.String(), nullable=True),
+        sa.Column('data', mindsdb.interfaces.storage.db.Json(), nullable=True),
+        sa.Column('to_predict', mindsdb.interfaces.storage.db.Array(), nullable=True),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.Column('mindsdb_version', sa.String(), nullable=True),
+        sa.Column('native_version', sa.String(), nullable=True),
+        sa.Column('integration_id', sa.Integer(), nullable=True),
+        sa.Column('data_integration_ref', mindsdb.interfaces.storage.db.Json(), nullable=True),
+        sa.Column('fetch_data_query', sa.String(), nullable=True),
+        sa.Column('learn_args', mindsdb.interfaces.storage.db.Json(), nullable=True),
+        sa.Column('update_status', sa.String(), nullable=True),
+        sa.Column('status', sa.String(), nullable=True),
+        sa.Column('active', sa.Boolean(), nullable=True),
+        sa.Column('training_data_columns_count', sa.Integer(), nullable=True),
+        sa.Column('training_data_rows_count', sa.Integer(), nullable=True),
+        sa.Column('training_start_at', sa.DateTime(), nullable=True),
+        sa.Column('training_stop_at', sa.DateTime(), nullable=True),
+        sa.Column('label', sa.String(), nullable=True),
+        sa.Column('version', sa.Integer(), nullable=True),
+        sa.Column('code', sa.String(), nullable=True),
+        sa.Column('lightwood_version', sa.String(), nullable=True),
+        sa.Column('dtype_dict', mindsdb.interfaces.storage.db.Json(), nullable=True),
+        sa.Column('project_id', sa.Integer(), nullable=False),
+        sa.Column('training_phase_current', sa.Integer(), nullable=True),
+        sa.Column('training_phase_total', sa.Integer(), nullable=True),
+        sa.Column('training_phase_name', sa.String(), nullable=True),
+        sa.Column('training_metadata', sa.JSON(), nullable=False),
+        sa.ForeignKeyConstraint(['integration_id'], ['integration.id'], name='fk_integration_id'),
+        sa.ForeignKeyConstraint(['project_id'], ['project.id'], name='fk_project_id'),
+        sa.PrimaryKeyConstraint('id')
+    )
+    with op.batch_alter_table('predictor', schema=None) as batch_op:
+        batch_op.create_index('predictor_index', ['company_id', 'name', 'version', 'active', 'deleted_at'], unique=True)
+
+    op.create_table(
+        'view',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('name', sa.String(), nullable=False),
+        sa.Column('company_id', sa.Integer(), nullable=True),
+        sa.Column('query', sa.String(), nullable=False),
+        sa.Column('project_id', sa.Integer(), nullable=False),
+        sa.ForeignKeyConstraint(['project_id'], ['project.id'], name='fk_project_id'),
+        sa.PrimaryKeyConstraint('id'),
+        sa.UniqueConstraint('name', 'company_id', name='unique_view_name_company_id')
+    )
+    op.create_table(
+        'knowledge_base',
+        sa.Column('id', sa.Integer(), nullable=False),
+        sa.Column('name', sa.String(), nullable=False),
+        sa.Column('project_id', sa.Integer(), nullable=False),
+        sa.Column('params', sa.JSON(), nullable=True),
+        sa.Column('vector_database_id', sa.Integer(), nullable=True),
+        sa.Column('vector_database_table', sa.String(), nullable=True),
+        sa.Column('embedding_model_id', sa.Integer(), nullable=True),
+        sa.Column('query_id', sa.Integer(), nullable=True),
+        sa.Column('created_at', sa.DateTime(), nullable=True),
+        sa.Column('updated_at', sa.DateTime(), nullable=True),
+        sa.ForeignKeyConstraint(['embedding_model_id'], ['predictor.id'], name='fk_knowledge_base_embedding_model_id'),
+        sa.ForeignKeyConstraint(['vector_database_id'], ['integration.id'], name='fk_knowledge_base_vector_database_id'),
+        sa.PrimaryKeyConstraint('id'),
+        sa.UniqueConstraint('name', 'project_id', name='unique_knowledge_base_name_project_id')
+    )
+
+    # Insert default project
+    op.bulk_insert(
+        sa.table(
+            'project',
+            sa.Column('name', sa.String()),
+            sa.Column('company_id', sa.Integer()),
+            sa.Column('metadata', sa.JSON()),
+            sa.Column('created_at', sa.DateTime())
+        ),
+        [{
+            'name': 'mindsdb',
+            'company_id': 0,
+            'metadata': {'is_default': True},
+            'created_at': datetime.datetime.now()
+        }]
+    )
+
+
+def downgrade():
+    # do nothging, since it is checkpoint migration
+    pass