update

backend/alembic/versions/8f43500ee275_add_index.py

@@ -18,13 +18,12 @@ depends_on = None
 def upgrade() -> None:
     # Create a basic index on the lowercase message column for direct text matching
     # Limit to 1500 characters to stay well under the 2856 byte limit of btree version 4
-    # op.execute(
-    #     """
-    #     CREATE INDEX idx_chat_message_message_lower
-    #     ON chat_message (LOWER(substring(message, 1, 1500)))
-    #     """
-    # )
-    pass
+    op.execute(
+        """
+        CREATE INDEX idx_chat_message_message_lower
+        ON chat_message (LOWER(substring(message, 1, 1500)))
+        """
+    )
 
 
 def downgrade() -> None:

backend/alembic/versions/9aadf32dfeb4_add_user_files.py

@@ -1,107 +0,0 @@
-"""add user files
-
-Revision ID: 9aadf32dfeb4
-Revises: 8f43500ee275
-Create Date: 2025-01-26 16:08:21.551022
-
-"""
-from alembic import op
-import sqlalchemy as sa
-import datetime
-
-
-# revision identifiers, used by Alembic.
-revision = "9aadf32dfeb4"
-down_revision = "8f43500ee275"
-branch_labels = None
-depends_on = None
-
-
-def upgrade() -> None:
-    # Create user_folder table without parent_id
-    op.create_table(
-        "user_folder",
-        sa.Column("id", sa.Integer(), primary_key=True, autoincrement=True),
-        sa.Column("user_id", sa.UUID(), sa.ForeignKey("user.id"), nullable=True),
-        sa.Column("name", sa.String(length=255), nullable=True),
-        sa.Column("description", sa.String(length=255), nullable=True),
-        sa.Column("display_priority", sa.Integer(), nullable=True, default=0),
-        sa.Column(
-            "created_at", sa.DateTime(timezone=True), server_default=sa.func.now()
-        ),
-    )
-
-    # Create user_file table with folder_id instead of parent_folder_id
-    op.create_table(
-        "user_file",
-        sa.Column("id", sa.Integer(), primary_key=True, autoincrement=True),
-        sa.Column("user_id", sa.UUID(), sa.ForeignKey("user.id"), nullable=True),
-        sa.Column(
-            "folder_id",
-            sa.Integer(),
-            sa.ForeignKey("user_folder.id"),
-            nullable=True,
-        ),
-        sa.Column("token_count", sa.Integer(), nullable=True),
-        sa.Column("file_type", sa.String(), nullable=True),
-        sa.Column("file_id", sa.String(length=255), nullable=False),
-        sa.Column("document_id", sa.String(length=255), nullable=False),
-        sa.Column("name", sa.String(length=255), nullable=False),
-        sa.Column(
-            "created_at",
-            sa.DateTime(),
-            default=datetime.datetime.utcnow,
-        ),
-        sa.Column(
-            "cc_pair_id",
-            sa.Integer(),
-            sa.ForeignKey("connector_credential_pair.id"),
-            nullable=True,
-            unique=True,
-        ),
-    )
-
-    # Create persona__user_file table
-    op.create_table(
-        "persona__user_file",
-        sa.Column(
-            "persona_id", sa.Integer(), sa.ForeignKey("persona.id"), primary_key=True
-        ),
-        sa.Column(
-            "user_file_id",
-            sa.Integer(),
-            sa.ForeignKey("user_file.id"),
-            primary_key=True,
-        ),
-    )
-
-    # Create persona__user_folder table
-    op.create_table(
-        "persona__user_folder",
-        sa.Column(
-            "persona_id", sa.Integer(), sa.ForeignKey("persona.id"), primary_key=True
-        ),
-        sa.Column(
-            "user_folder_id",
-            sa.Integer(),
-            sa.ForeignKey("user_folder.id"),
-            primary_key=True,
-        ),
-    )
-
-    op.add_column(
-        "connector_credential_pair",
-        sa.Column("is_user_file", sa.Boolean(), nullable=True),
-    )
-
-
-def downgrade() -> None:
-    # Drop the persona__user_folder table
-    op.drop_table("persona__user_folder")
-    # Drop the persona__user_file table
-    op.drop_table("persona__user_file")
-    # Drop the user_file table
-    op.drop_table("user_file")
-    # Drop the user_folder table
-    op.drop_table("user_folder")
-    op.drop_column("connector_credential_pair", "is_user_file")

backend/ee/onyx/background/celery/apps/primary.py

@@ -5,9 +5,11 @@ from onyx.background.celery.apps.primary import celery_app
 from onyx.background.task_utils import build_celery_task_wrapper
 from onyx.configs.app_configs import JOB_TIMEOUT
 from onyx.db.chat import delete_chat_sessions_older_than
-from onyx.db.engine import get_session_with_current_tenant
+from onyx.db.engine import get_session_with_tenant
 from onyx.server.settings.store import load_settings
 from onyx.utils.logger import setup_logger
+from shared_configs.configs import MULTI_TENANT
+from shared_configs.contextvars import CURRENT_TENANT_ID_CONTEXTVAR
 
 logger = setup_logger()
 
@@ -16,8 +18,10 @@ logger = setup_logger()
 
 @build_celery_task_wrapper(name_chat_ttl_task)
 @celery_app.task(soft_time_limit=JOB_TIMEOUT)
-def perform_ttl_management_task(retention_limit_days: int, *, tenant_id: str) -> None:
-    with get_session_with_current_tenant() as db_session:
+def perform_ttl_management_task(
+    retention_limit_days: int, *, tenant_id: str | None
+) -> None:
+    with get_session_with_tenant(tenant_id=tenant_id) as db_session:
         delete_chat_sessions_older_than(retention_limit_days, db_session)
 
 
@@ -31,19 +35,24 @@ def perform_ttl_management_task(retention_limit_days: int, *, tenant_id: str) ->
     ignore_result=True,
     soft_time_limit=JOB_TIMEOUT,
 )
-def check_ttl_management_task(*, tenant_id: str) -> None:
+def check_ttl_management_task(*, tenant_id: str | None) -> None:
     """Runs periodically to check if any ttl tasks should be run and adds them
     to the queue"""
+    token = None
+    if MULTI_TENANT and tenant_id is not None:
+        token = CURRENT_TENANT_ID_CONTEXTVAR.set(tenant_id)
 
     settings = load_settings()
     retention_limit_days = settings.maximum_chat_retention_days
-    with get_session_with_current_tenant() as db_session:
+    with get_session_with_tenant(tenant_id=tenant_id) as db_session:
         if should_perform_chat_ttl_check(retention_limit_days, db_session):
             perform_ttl_management_task.apply_async(
                 kwargs=dict(
                     retention_limit_days=retention_limit_days, tenant_id=tenant_id
                 ),
             )
+    if token is not None:
+        CURRENT_TENANT_ID_CONTEXTVAR.reset(token)
 
 
 @celery_app.task(
@@ -51,9 +60,9 @@ def check_ttl_management_task(*, tenant_id: str) -> None:
     ignore_result=True,
     soft_time_limit=JOB_TIMEOUT,
 )
-def autogenerate_usage_report_task(*, tenant_id: str) -> None:
+def autogenerate_usage_report_task(*, tenant_id: str | None) -> None:
     """This generates usage report under the /admin/generate-usage/report endpoint"""
-    with get_session_with_current_tenant() as db_session:
+    with get_session_with_tenant(tenant_id=tenant_id) as db_session:
         create_new_usage_report(
             db_session=db_session,
             user_id=None,

backend/ee/onyx/background/celery/tasks/vespa/tasks.py

@@ -18,7 +18,7 @@ logger = setup_logger()
 
 
 def monitor_usergroup_taskset(
-    tenant_id: str, key_bytes: bytes, r: Redis, db_session: Session
+    tenant_id: str | None, key_bytes: bytes, r: Redis, db_session: Session
 ) -> None:
     """This function is likely to move in the worker refactor happening next."""
     fence_key = key_bytes.decode("utf-8")

backend/ee/onyx/server/tenants/user_mapping.py

@@ -28,7 +28,7 @@ def get_tenant_id_for_email(email: str) -> str:
 
 
 def user_owns_a_tenant(email: str) -> bool:
-    with get_session_with_tenant(tenant_id=POSTGRES_DEFAULT_SCHEMA) as db_session:
+    with get_session_with_tenant(tenant_id=None) as db_session:
         result = (
             db_session.query(UserTenantMapping)
             .filter(UserTenantMapping.email == email)
@@ -38,7 +38,7 @@ def user_owns_a_tenant(email: str) -> bool:
 
 
 def add_users_to_tenant(emails: list[str], tenant_id: str) -> None:
-    with get_session_with_tenant(tenant_id=POSTGRES_DEFAULT_SCHEMA) as db_session:
+    with get_session_with_tenant(tenant_id=None) as db_session:
         try:
             for email in emails:
                 db_session.add(UserTenantMapping(email=email, tenant_id=tenant_id))
@@ -48,7 +48,7 @@ def add_users_to_tenant(emails: list[str], tenant_id: str) -> None:
 
 
 def remove_users_from_tenant(emails: list[str], tenant_id: str) -> None:
-    with get_session_with_tenant(tenant_id=POSTGRES_DEFAULT_SCHEMA) as db_session:
+    with get_session_with_tenant(tenant_id=None) as db_session:
         try:
             mappings_to_delete = (
                 db_session.query(UserTenantMapping)
@@ -71,7 +71,7 @@ def remove_users_from_tenant(emails: list[str], tenant_id: str) -> None:
 
 
 def remove_all_users_from_tenant(tenant_id: str) -> None:
-    with get_session_with_tenant(tenant_id=POSTGRES_DEFAULT_SCHEMA) as db_session:
+    with get_session_with_tenant(tenant_id=None) as db_session:
         db_session.query(UserTenantMapping).filter(
             UserTenantMapping.tenant_id == tenant_id
         ).delete()

backend/onyx/agents/agent_search/shared_graph_utils/utils.py

@@ -319,10 +319,8 @@ def dispatch_separated(
     sep: str = DISPATCH_SEP_CHAR,
 ) -> list[BaseMessage_Content]:
     num = 1
-    accumulated_tokens = ""
     streamed_tokens: list[BaseMessage_Content] = []
     for token in tokens:
-        accumulated_tokens += cast(str, token.content)
         content = cast(str, token.content)
         if sep in content:
             sub_question_parts = content.split(sep)

backend/onyx/auth/api_key.py

@@ -10,7 +10,6 @@ from pydantic import BaseModel
 
 from onyx.auth.schemas import UserRole
 from onyx.configs.app_configs import API_KEY_HASH_ROUNDS
-from shared_configs.configs import MULTI_TENANT
 
 
 _API_KEY_HEADER_NAME = "Authorization"
@@ -36,7 +35,8 @@ class ApiKeyDescriptor(BaseModel):
 
 
 def generate_api_key(tenant_id: str | None = None) -> str:
-    if not MULTI_TENANT or not tenant_id:
+    # For backwards compatibility, if no tenant_id, generate old style key
+    if not tenant_id:
         return _API_KEY_PREFIX + secrets.token_urlsafe(_API_KEY_LEN)
 
     encoded_tenant = quote(tenant_id)  # URL encode the tenant ID

backend/onyx/auth/email_utils.py

@@ -2,8 +2,6 @@ import smtplib
 from datetime import datetime
 from email.mime.multipart import MIMEMultipart
 from email.mime.text import MIMEText
-from email.utils import formatdate
-from email.utils import make_msgid
 
 from onyx.configs.app_configs import EMAIL_CONFIGURED
 from onyx.configs.app_configs import EMAIL_FROM
@@ -15,7 +13,6 @@ from onyx.configs.app_configs import WEB_DOMAIN
 from onyx.configs.constants import AuthType
 from onyx.configs.constants import TENANT_ID_COOKIE_NAME
 from onyx.db.models import User
-from shared_configs.configs import MULTI_TENANT
 
 HTML_EMAIL_TEMPLATE = """\
 <!DOCTYPE html>
@@ -153,9 +150,8 @@ def send_email(
     msg = MIMEMultipart("alternative")
     msg["Subject"] = subject
     msg["To"] = user_email
-    msg["From"] = mail_from
-    msg["Date"] = formatdate(localtime=True)
-    msg["Message-ID"] = make_msgid(domain="onyx.app")
+    if mail_from:
+        msg["From"] = mail_from
 
     part_text = MIMEText(text_body, "plain")
     part_html = MIMEText(html_body, "html")
@@ -177,7 +173,7 @@ def send_subscription_cancellation_email(user_email: str) -> None:
     subject = "Your Onyx Subscription Has Been Canceled"
     heading = "Subscription Canceled"
     message = (
-        "<p>We're sorry to see you go.</p>"
+        "<p>We’re sorry to see you go.</p>"
         "<p>Your subscription has been canceled and will end on your next billing date.</p>"
         "<p>If you change your mind, you can always come back!</p>"
     )
@@ -243,13 +239,13 @@ def send_user_email_invite(
 def send_forgot_password_email(
     user_email: str,
     token: str,
-    tenant_id: str,
     mail_from: str = EMAIL_FROM,
+    tenant_id: str | None = None,
 ) -> None:
     # Builds a forgot password email with or without fancy HTML
     subject = "Onyx Forgot Password"
     link = f"{WEB_DOMAIN}/auth/reset-password?token={token}"
-    if MULTI_TENANT:
+    if tenant_id:
         link += f"&{TENANT_ID_COOKIE_NAME}={tenant_id}"
     message = f"<p>Click the following link to reset your password:</p><p>{link}</p>"
     html_content = build_html_email("Reset Your Password", message)

backend/onyx/auth/users.py

@@ -214,7 +214,7 @@ def verify_email_is_invited(email: str) -> None:
     raise PermissionError("User not on allowed user whitelist")
 
 
-def verify_email_in_whitelist(email: str, tenant_id: str) -> None:
+def verify_email_in_whitelist(email: str, tenant_id: str | None = None) -> None:
     with get_session_with_tenant(tenant_id=tenant_id) as db_session:
         if not get_user_by_email(email, db_session):
             verify_email_is_invited(email)
@@ -553,7 +553,7 @@ class UserManager(UUIDIDMixin, BaseUserManager[User, uuid.UUID]):
             async_return_default_schema,
         )(email=user.email)
 
-        send_forgot_password_email(user.email, tenant_id=tenant_id, token=token)
+        send_forgot_password_email(user.email, token, tenant_id=tenant_id)
 
     async def on_after_request_verify(
         self, user: User, token: str, request: Optional[Request] = None

backend/onyx/background/celery/apps/app_base.py

@@ -2,7 +2,6 @@ import logging
 import multiprocessing
 import time
 from typing import Any
-from typing import cast
 
 import sentry_sdk
 from celery import Task
@@ -132,9 +131,9 @@ def on_task_postrun(
     # Get tenant_id directly from kwargs- each celery task has a tenant_id kwarg
     if not kwargs:
         logger.error(f"Task {task.name} (ID: {task_id}) is missing kwargs")
-        tenant_id = POSTGRES_DEFAULT_SCHEMA
+        tenant_id = None
     else:
-        tenant_id = cast(str, kwargs.get("tenant_id", POSTGRES_DEFAULT_SCHEMA))
+        tenant_id = kwargs.get("tenant_id")
 
     task_logger.debug(
         f"Task {task.name} (ID: {task_id}) completed with state: {state} "

backend/onyx/background/celery/celery_utils.py

@@ -34,7 +34,7 @@ def _get_deletion_status(
     connector_id: int,
     credential_id: int,
     db_session: Session,
-    tenant_id: str,
+    tenant_id: str | None = None,
 ) -> TaskQueueState | None:
     """We no longer store TaskQueueState in the DB for a deletion attempt.
     This function populates TaskQueueState by just checking redis.
@@ -67,7 +67,7 @@ def get_deletion_attempt_snapshot(
     connector_id: int,
     credential_id: int,
     db_session: Session,
-    tenant_id: str,
+    tenant_id: str | None = None,
 ) -> DeletionAttemptSnapshot | None:
     deletion_task = _get_deletion_status(
         connector_id, credential_id, db_session, tenant_id

backend/onyx/background/celery/tasks/connector_deletion/tasks.py

@@ -109,7 +109,9 @@ def revoke_tasks_blocking_deletion(
     trail=False,
     bind=True,
 )
-def check_for_connector_deletion_task(self: Task, *, tenant_id: str) -> bool | None:
+def check_for_connector_deletion_task(
+    self: Task, *, tenant_id: str | None
+) -> bool | None:
     r = get_redis_client()
     r_replica = get_redis_replica_client()
     r_celery: Redis = self.app.broker_connection().channel().client  # type: ignore
@@ -222,7 +224,7 @@ def try_generate_document_cc_pair_cleanup_tasks(
     cc_pair_id: int,
     db_session: Session,
     lock_beat: RedisLock,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> int | None:
     """Returns an int if syncing is needed. The int represents the number of sync tasks generated.
     Note that syncing can still be required even if the number of sync tasks generated is zero.
@@ -343,7 +345,7 @@ def try_generate_document_cc_pair_cleanup_tasks(
 
 
 def monitor_connector_deletion_taskset(
-    tenant_id: str, key_bytes: bytes, r: Redis
+    tenant_id: str | None, key_bytes: bytes, r: Redis
 ) -> None:
     fence_key = key_bytes.decode("utf-8")
     cc_pair_id_str = RedisConnector.get_id_from_fence_key(fence_key)
@@ -498,7 +500,7 @@ def monitor_connector_deletion_taskset(
 
 
 def validate_connector_deletion_fences(
-    tenant_id: str,
+    tenant_id: str | None,
     r: Redis,
     r_replica: Redis,
     r_celery: Redis,
@@ -538,7 +540,7 @@ def validate_connector_deletion_fences(
 
 
 def validate_connector_deletion_fence(
-    tenant_id: str,
+    tenant_id: str | None,
     key_bytes: bytes,
     queued_tasks: set[str],
     r: Redis,

backend/onyx/background/celery/tasks/doc_permission_syncing/tasks.py

@@ -221,7 +221,7 @@ def try_creating_permissions_sync_task(
     app: Celery,
     cc_pair_id: int,
     r: Redis,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> str | None:
     """Returns a randomized payload id on success.
     Returns None if no syncing is required."""
@@ -320,7 +320,7 @@ def try_creating_permissions_sync_task(
 def connector_permission_sync_generator_task(
     self: Task,
     cc_pair_id: int,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> None:
     """
     Permission sync task that handles document permission syncing for a given connector credential pair
@@ -410,6 +410,7 @@ def connector_permission_sync_generator_task(
                     cc_pair.connector.id,
                     cc_pair.credential.id,
                     db_session,
+                    tenant_id,
                     enforce_creation=False,
                 )
                 if not created:
@@ -509,7 +510,7 @@ def connector_permission_sync_generator_task(
 )
 def update_external_document_permissions_task(
     self: Task,
-    tenant_id: str,
+    tenant_id: str | None,
     serialized_doc_external_access: dict,
     source_string: str,
     connector_id: int,
@@ -584,7 +585,7 @@ def update_external_document_permissions_task(
 
 
 def validate_permission_sync_fences(
-    tenant_id: str,
+    tenant_id: str | None,
     r: Redis,
     r_replica: Redis,
     r_celery: Redis,
@@ -631,7 +632,7 @@ def validate_permission_sync_fences(
 
 
 def validate_permission_sync_fence(
-    tenant_id: str,
+    tenant_id: str | None,
     key_bytes: bytes,
     queued_tasks: set[str],
     reserved_tasks: set[str],
@@ -841,7 +842,7 @@ class PermissionSyncCallback(IndexingHeartbeatInterface):
 
 
 def monitor_ccpair_permissions_taskset(
-    tenant_id: str, key_bytes: bytes, r: Redis, db_session: Session
+    tenant_id: str | None, key_bytes: bytes, r: Redis, db_session: Session
 ) -> None:
     fence_key = key_bytes.decode("utf-8")
     cc_pair_id_str = RedisConnector.get_id_from_fence_key(fence_key)

backend/onyx/background/celery/tasks/external_group_syncing/tasks.py

@@ -123,7 +123,7 @@ def _is_external_group_sync_due(cc_pair: ConnectorCredentialPair) -> bool:
     soft_time_limit=JOB_TIMEOUT,
     bind=True,
 )
-def check_for_external_group_sync(self: Task, *, tenant_id: str) -> bool | None:
+def check_for_external_group_sync(self: Task, *, tenant_id: str | None) -> bool | None:
     # we need to use celery's redis client to access its redis data
     # (which lives on a different db number)
     r = get_redis_client()
@@ -220,7 +220,7 @@ def try_creating_external_group_sync_task(
     app: Celery,
     cc_pair_id: int,
     r: Redis,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> str | None:
     """Returns an int if syncing is needed. The int represents the number of sync tasks generated.
     Returns None if no syncing is required."""
@@ -306,7 +306,7 @@ def try_creating_external_group_sync_task(
 def connector_external_group_sync_generator_task(
     self: Task,
     cc_pair_id: int,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> None:
     """
     External group sync task for a given connector credential pair
@@ -392,6 +392,7 @@ def connector_external_group_sync_generator_task(
                     cc_pair.connector.id,
                     cc_pair.credential.id,
                     db_session,
+                    tenant_id,
                     enforce_creation=False,
                 )
                 if not created:
@@ -493,7 +494,7 @@ def connector_external_group_sync_generator_task(
 
 
 def validate_external_group_sync_fences(
-    tenant_id: str,
+    tenant_id: str | None,
     celery_app: Celery,
     r: Redis,
     r_replica: Redis,
@@ -525,7 +526,7 @@ def validate_external_group_sync_fences(
 
 
 def validate_external_group_sync_fence(
-    tenant_id: str,
+    tenant_id: str | None,
     key_bytes: bytes,
     reserved_tasks: set[str],
     r_celery: Redis,

backend/onyx/background/celery/tasks/indexing/tasks.py

@@ -182,7 +182,7 @@ class SimpleJobResult:
 
 
 class ConnectorIndexingContext(BaseModel):
-    tenant_id: str
+    tenant_id: str | None
     cc_pair_id: int
     search_settings_id: int
     index_attempt_id: int
@@ -210,7 +210,7 @@ class ConnectorIndexingLogBuilder:
 
 
 def monitor_ccpair_indexing_taskset(
-    tenant_id: str, key_bytes: bytes, r: Redis, db_session: Session
+    tenant_id: str | None, key_bytes: bytes, r: Redis, db_session: Session
 ) -> None:
     # if the fence doesn't exist, there's nothing to do
     fence_key = key_bytes.decode("utf-8")
@@ -358,7 +358,7 @@ def monitor_ccpair_indexing_taskset(
     soft_time_limit=300,
     bind=True,
 )
-def check_for_indexing(self: Task, *, tenant_id: str) -> int | None:
+def check_for_indexing(self: Task, *, tenant_id: str | None) -> int | None:
     """a lightweight task used to kick off indexing tasks.
     Occcasionally does some validation of existing state to clear up error conditions"""
 
@@ -598,7 +598,7 @@ def connector_indexing_task(
     cc_pair_id: int,
     search_settings_id: int,
     is_ee: bool,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> int | None:
     """Indexing task. For a cc pair, this task pulls all document IDs from the source
     and compares those IDs to locally stored documents and deletes all locally stored IDs missing
@@ -890,7 +890,7 @@ def connector_indexing_proxy_task(
     index_attempt_id: int,
     cc_pair_id: int,
     search_settings_id: int,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> None:
     """celery out of process task execution strategy is pool=prefork, but it uses fork,
     and forking is inherently unstable.
@@ -1170,7 +1170,7 @@ def connector_indexing_proxy_task(
     name=OnyxCeleryTask.CHECK_FOR_CHECKPOINT_CLEANUP,
     soft_time_limit=300,
 )
-def check_for_checkpoint_cleanup(*, tenant_id: str) -> None:
+def check_for_checkpoint_cleanup(*, tenant_id: str | None) -> None:
     """Clean up old checkpoints that are older than 7 days."""
     locked = False
     redis_client = get_redis_client(tenant_id=tenant_id)

backend/onyx/background/celery/tasks/indexing/utils.py

@@ -187,7 +187,7 @@ class IndexingCallback(IndexingCallbackBase):
 
 
 def validate_indexing_fence(
-    tenant_id: str,
+    tenant_id: str | None,
     key_bytes: bytes,
     reserved_tasks: set[str],
     r_celery: Redis,
@@ -311,7 +311,7 @@ def validate_indexing_fence(
 
 
 def validate_indexing_fences(
-    tenant_id: str,
+    tenant_id: str | None,
     r_replica: Redis,
     r_celery: Redis,
     lock_beat: RedisLock,
@@ -442,7 +442,7 @@ def try_creating_indexing_task(
     reindex: bool,
     db_session: Session,
     r: Redis,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> int | None:
     """Checks for any conditions that should block the indexing task from being
     created, then creates the task.

backend/onyx/background/celery/tasks/llm_model_update/tasks.py

@@ -59,7 +59,7 @@ def _process_model_list_response(model_list_json: Any) -> list[str]:
     trail=False,
     bind=True,
 )
-def check_for_llm_model_update(self: Task, *, tenant_id: str) -> bool | None:
+def check_for_llm_model_update(self: Task, *, tenant_id: str | None) -> bool | None:
     if not LLM_MODEL_UPDATE_API_URL:
         raise ValueError("LLM model update API URL not configured")
 

backend/onyx/background/celery/tasks/monitoring/tasks.py

@@ -91,7 +91,7 @@ class Metric(BaseModel):
         }
         task_logger.info(json.dumps(data))
 
-    def emit(self, tenant_id: str) -> None:
+    def emit(self, tenant_id: str | None) -> None:
         # Convert value to appropriate type based on the input value
         bool_value = None
         float_value = None
@@ -656,7 +656,7 @@ def build_job_id(
     queue=OnyxCeleryQueues.MONITORING,
     bind=True,
 )
-def monitor_background_processes(self: Task, *, tenant_id: str) -> None:
+def monitor_background_processes(self: Task, *, tenant_id: str | None) -> None:
     """Collect and emit metrics about background processes.
     This task runs periodically to gather metrics about:
     - Queue lengths for different Celery queues
@@ -864,7 +864,7 @@ def cloud_monitor_celery_queues(
 
 
 @shared_task(name=OnyxCeleryTask.MONITOR_CELERY_QUEUES, ignore_result=True, bind=True)
-def monitor_celery_queues(self: Task, *, tenant_id: str) -> None:
+def monitor_celery_queues(self: Task, *, tenant_id: str | None) -> None:
     return monitor_celery_queues_helper(self)
 
 

backend/onyx/background/celery/tasks/periodic/tasks.py

@@ -24,7 +24,7 @@ from onyx.db.engine import get_session_with_current_tenant
     bind=True,
     base=AbortableTask,
 )
-def kombu_message_cleanup_task(self: Any, tenant_id: str) -> int:
+def kombu_message_cleanup_task(self: Any, tenant_id: str | None) -> int:
     """Runs periodically to clean up the kombu_message table"""
 
     # we will select messages older than this amount to clean up

backend/onyx/background/celery/tasks/pruning/tasks.py

@@ -114,7 +114,7 @@ def _is_pruning_due(cc_pair: ConnectorCredentialPair) -> bool:
     soft_time_limit=JOB_TIMEOUT,
     bind=True,
 )
-def check_for_pruning(self: Task, *, tenant_id: str) -> bool | None:
+def check_for_pruning(self: Task, *, tenant_id: str | None) -> bool | None:
     r = get_redis_client()
     r_replica = get_redis_replica_client()
     r_celery: Redis = self.app.broker_connection().channel().client  # type: ignore
@@ -211,7 +211,7 @@ def try_creating_prune_generator_task(
     cc_pair: ConnectorCredentialPair,
     db_session: Session,
     r: Redis,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> str | None:
     """Checks for any conditions that should block the pruning generator task from being
     created, then creates the task.
@@ -333,7 +333,7 @@ def connector_pruning_generator_task(
     cc_pair_id: int,
     connector_id: int,
     credential_id: int,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> None:
     """connector pruning task. For a cc pair, this task pulls all document IDs from the source
     and compares those IDs to locally stored documents and deletes all locally stored IDs missing
@@ -521,7 +521,7 @@ def connector_pruning_generator_task(
 
 
 def monitor_ccpair_pruning_taskset(
-    tenant_id: str, key_bytes: bytes, r: Redis, db_session: Session
+    tenant_id: str | None, key_bytes: bytes, r: Redis, db_session: Session
 ) -> None:
     fence_key = key_bytes.decode("utf-8")
     cc_pair_id_str = RedisConnector.get_id_from_fence_key(fence_key)
@@ -567,7 +567,7 @@ def monitor_ccpair_pruning_taskset(
 
 
 def validate_pruning_fences(
-    tenant_id: str,
+    tenant_id: str | None,
     r: Redis,
     r_replica: Redis,
     r_celery: Redis,
@@ -615,7 +615,7 @@ def validate_pruning_fences(
 
 
 def validate_pruning_fence(
-    tenant_id: str,
+    tenant_id: str | None,
     key_bytes: bytes,
     reserved_tasks: set[str],
     queued_tasks: set[str],

backend/onyx/background/celery/tasks/shared/RetryDocumentIndex.py

@@ -32,7 +32,7 @@ class RetryDocumentIndex:
         self,
         doc_id: str,
         *,
-        tenant_id: str,
+        tenant_id: str | None,
         chunk_count: int | None,
     ) -> int:
         return self.index.delete_single(
@@ -50,7 +50,7 @@ class RetryDocumentIndex:
         self,
         doc_id: str,
         *,
-        tenant_id: str,
+        tenant_id: str | None,
         chunk_count: int | None,
         fields: VespaDocumentFields,
     ) -> int:

backend/onyx/background/celery/tasks/shared/tasks.py

@@ -76,7 +76,7 @@ def document_by_cc_pair_cleanup_task(
     document_id: str,
     connector_id: int,
     credential_id: int,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> bool:
     """A lightweight subtask used to clean up document to cc pair relationships.
     Created by connection deletion and connector pruning parent tasks."""
@@ -297,7 +297,7 @@ def cloud_beat_task_generator(
         return None
 
     last_lock_time = time.monotonic()
-    tenant_ids: list[str] = []
+    tenant_ids: list[str] | list[None] = []
 
     try:
         tenant_ids = get_all_tenant_ids()

backend/onyx/background/celery/tasks/vespa/tasks.py

@@ -76,7 +76,7 @@ logger = setup_logger()
     trail=False,
     bind=True,
 )
-def check_for_vespa_sync_task(self: Task, *, tenant_id: str) -> bool | None:
+def check_for_vespa_sync_task(self: Task, *, tenant_id: str | None) -> bool | None:
     """Runs periodically to check if any document needs syncing.
     Generates sets of tasks for Celery if syncing is needed."""
 
@@ -208,7 +208,7 @@ def try_generate_stale_document_sync_tasks(
     db_session: Session,
     r: Redis,
     lock_beat: RedisLock,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> int | None:
     # the fence is up, do nothing
 
@@ -284,7 +284,7 @@ def try_generate_document_set_sync_tasks(
     db_session: Session,
     r: Redis,
     lock_beat: RedisLock,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> int | None:
     lock_beat.reacquire()
 
@@ -361,7 +361,7 @@ def try_generate_user_group_sync_tasks(
     db_session: Session,
     r: Redis,
     lock_beat: RedisLock,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> int | None:
     lock_beat.reacquire()
 
@@ -448,7 +448,7 @@ def monitor_connector_taskset(r: Redis) -> None:
 
 
 def monitor_document_set_taskset(
-    tenant_id: str, key_bytes: bytes, r: Redis, db_session: Session
+    tenant_id: str | None, key_bytes: bytes, r: Redis, db_session: Session
 ) -> None:
     fence_key = key_bytes.decode("utf-8")
     document_set_id_str = RedisDocumentSet.get_id_from_fence_key(fence_key)
@@ -523,7 +523,9 @@ def monitor_document_set_taskset(
     time_limit=LIGHT_TIME_LIMIT,
     max_retries=3,
 )
-def vespa_metadata_sync_task(self: Task, document_id: str, *, tenant_id: str) -> bool:
+def vespa_metadata_sync_task(
+    self: Task, document_id: str, *, tenant_id: str | None
+) -> bool:
     start = time.monotonic()
 
     completion_status = OnyxCeleryTaskCompletionStatus.UNDEFINED

backend/onyx/background/indexing/run_indexing.py

@@ -55,7 +55,6 @@ from onyx.utils.logger import setup_logger
 from onyx.utils.logger import TaskAttemptSingleton
 from onyx.utils.telemetry import create_milestone_and_report
 from onyx.utils.variable_functionality import global_version
-from shared_configs.configs import MULTI_TENANT
 
 logger = setup_logger()
 
@@ -68,6 +67,7 @@ def _get_connector_runner(
     batch_size: int,
     start_time: datetime,
     end_time: datetime,
+    tenant_id: str | None,
     leave_connector_active: bool = LEAVE_CONNECTOR_ACTIVE_ON_INITIALIZATION_FAILURE,
 ) -> ConnectorRunner:
     """
@@ -86,6 +86,7 @@ def _get_connector_runner(
             input_type=task,
             connector_specific_config=attempt.connector_credential_pair.connector.connector_specific_config,
             credential=attempt.connector_credential_pair.credential,
+            tenant_id=tenant_id,
         )
 
         # validate the connector settings
@@ -240,7 +241,7 @@ def _check_failure_threshold(
 def _run_indexing(
     db_session: Session,
     index_attempt_id: int,
-    tenant_id: str,
+    tenant_id: str | None,
     callback: IndexingHeartbeatInterface | None = None,
 ) -> None:
     """
@@ -387,6 +388,7 @@ def _run_indexing(
                 batch_size=INDEX_BATCH_SIZE,
                 start_time=window_start,
                 end_time=window_end,
+                tenant_id=tenant_id,
             )
 
             # don't use a checkpoint if we're explicitly indexing from
@@ -679,7 +681,7 @@ def _run_indexing(
 
 def run_indexing_entrypoint(
     index_attempt_id: int,
-    tenant_id: str,
+    tenant_id: str | None,
     connector_credential_pair_id: int,
     is_ee: bool = False,
     callback: IndexingHeartbeatInterface | None = None,
@@ -699,7 +701,7 @@ def run_indexing_entrypoint(
         attempt = transition_attempt_to_in_progress(index_attempt_id, db_session)
 
         tenant_str = ""
-        if MULTI_TENANT:
+        if tenant_id is not None:
             tenant_str = f" for tenant {tenant_id}"
 
         connector_name = attempt.connector_credential_pair.connector.name

backend/onyx/chat/process_message.py

@@ -86,7 +86,6 @@ from onyx.document_index.factory import get_default_document_index
 from onyx.file_store.models import ChatFileType
 from onyx.file_store.models import FileDescriptor
 from onyx.file_store.utils import load_all_chat_files
-from onyx.file_store.utils import load_all_user_files
 from onyx.file_store.utils import save_files
 from onyx.llm.exceptions import GenAIDisabledException
 from onyx.llm.factory import get_llms_for_persona
@@ -263,11 +262,8 @@ def _get_force_search_settings(
     search_tool_available = any(isinstance(tool, SearchTool) for tool in tools)
 
     if not internet_search_available and not search_tool_available:
-        if new_msg_req.force_user_file_search:
-            return ForceUseTool(force_use=True, tool_name=SearchTool._NAME)
-        else:
-            # Does not matter much which tool is set here as force is false and neither tool is available
-            return ForceUseTool(force_use=False, tool_name=SearchTool._NAME)
+        # Does not matter much which tool is set here as force is false and neither tool is available
+        return ForceUseTool(force_use=False, tool_name=SearchTool._NAME)
 
     tool_name = SearchTool._NAME if search_tool_available else InternetSearchTool._NAME
     # Currently, the internet search tool does not support query override
@@ -283,7 +279,6 @@ def _get_force_search_settings(
 
     should_force_search = any(
         [
-            new_msg_req.force_user_file_search,
             new_msg_req.retrieval_options
             and new_msg_req.retrieval_options.run_search
             == OptionalSearchSetting.ALWAYS,
@@ -543,15 +538,6 @@ def stream_chat_message_objects(
         req_file_ids = [f["id"] for f in new_msg_req.file_descriptors]
         latest_query_files = [file for file in files if file.file_id in req_file_ids]
 
-        if not new_msg_req.force_user_file_search:
-            user_files = load_all_user_files(
-                new_msg_req.user_file_ids,
-                new_msg_req.user_folder_ids,
-                db_session,
-            )
-
-            latest_query_files += user_files
-
         if user_message:
             attach_files_to_chat_message(
                 chat_message=user_message,
@@ -695,7 +681,6 @@ def stream_chat_message_objects(
             user=user,
             llm=llm,
             fast_llm=fast_llm,
-            use_file_search=new_msg_req.force_user_file_search,
             search_tool_config=SearchToolConfig(
                 answer_style_config=answer_style_config,
                 document_pruning_config=document_pruning_config,

backend/onyx/configs/chat_configs.py

@@ -3,7 +3,7 @@ import os
 INPUT_PROMPT_YAML = "./onyx/seeding/input_prompts.yaml"
 PROMPTS_YAML = "./onyx/seeding/prompts.yaml"
 PERSONAS_YAML = "./onyx/seeding/personas.yaml"
-USER_FOLDERS_YAML = "./onyx/seeding/user_folders.yaml"
+
 NUM_RETURNED_HITS = 50
 # Used for LLM filtering and reranking
 # We want this to be approximately the number of results we want to show on the first page

backend/onyx/connectors/factory.py

@@ -5,6 +5,7 @@ from sqlalchemy.orm import Session
 
 from onyx.configs.app_configs import INTEGRATION_TESTS_MODE
 from onyx.configs.constants import DocumentSource
+from onyx.configs.constants import DocumentSourceRequiringTenantContext
 from onyx.connectors.airtable.airtable_connector import AirtableConnector
 from onyx.connectors.asana.connector import AsanaConnector
 from onyx.connectors.axero.connector import AxeroConnector
@@ -163,9 +164,13 @@ def instantiate_connector(
     input_type: InputType,
     connector_specific_config: dict[str, Any],
     credential: Credential,
+    tenant_id: str | None = None,
 ) -> BaseConnector:
     connector_class = identify_connector_class(source, input_type)
 
+    if source in DocumentSourceRequiringTenantContext:
+        connector_specific_config["tenant_id"] = tenant_id
+
     connector = connector_class(**connector_specific_config)
     new_credentials = connector.load_credentials(credential.credential_json)
 
@@ -179,6 +184,7 @@ def validate_ccpair_for_user(
     connector_id: int,
     credential_id: int,
     db_session: Session,
+    tenant_id: str | None,
     enforce_creation: bool = True,
 ) -> bool:
     if INTEGRATION_TESTS_MODE:
@@ -210,6 +216,7 @@ def validate_ccpair_for_user(
             input_type=connector.input_type,
             connector_specific_config=connector.connector_specific_config,
             credential=credential,
+            tenant_id=tenant_id,
         )
     except ConnectorValidationError as e:
         raise e

backend/onyx/connectors/file/connector.py

@@ -16,7 +16,7 @@ from onyx.connectors.interfaces import LoadConnector
 from onyx.connectors.models import BasicExpertInfo
 from onyx.connectors.models import Document
 from onyx.connectors.models import Section
-from onyx.db.engine import get_session_with_current_tenant
+from onyx.db.engine import get_session_with_tenant
 from onyx.file_processing.extract_file_text import detect_encoding
 from onyx.file_processing.extract_file_text import extract_file_text
 from onyx.file_processing.extract_file_text import get_file_ext
@@ -27,6 +27,8 @@ from onyx.file_processing.extract_file_text import read_pdf_file
 from onyx.file_processing.extract_file_text import read_text_file
 from onyx.file_store.file_store import get_default_file_store
 from onyx.utils.logger import setup_logger
+from shared_configs.configs import POSTGRES_DEFAULT_SCHEMA
+from shared_configs.contextvars import CURRENT_TENANT_ID_CONTEXTVAR
 
 logger = setup_logger()
 
@@ -163,10 +165,12 @@ class LocalFileConnector(LoadConnector):
     def __init__(
         self,
         file_locations: list[Path | str],
+        tenant_id: str = POSTGRES_DEFAULT_SCHEMA,
         batch_size: int = INDEX_BATCH_SIZE,
     ) -> None:
         self.file_locations = [Path(file_location) for file_location in file_locations]
         self.batch_size = batch_size
+        self.tenant_id = tenant_id
         self.pdf_pass: str | None = None
 
     def load_credentials(self, credentials: dict[str, Any]) -> dict[str, Any] | None:
@@ -175,8 +179,9 @@ class LocalFileConnector(LoadConnector):
 
     def load_from_state(self) -> GenerateDocumentsOutput:
         documents: list[Document] = []
+        token = CURRENT_TENANT_ID_CONTEXTVAR.set(self.tenant_id)
 
-        with get_session_with_current_tenant() as db_session:
+        with get_session_with_tenant(tenant_id=self.tenant_id) as db_session:
             for file_path in self.file_locations:
                 current_datetime = datetime.now(timezone.utc)
                 files = _read_files_and_metadata(
@@ -198,6 +203,8 @@ class LocalFileConnector(LoadConnector):
             if documents:
                 yield documents
 
+        CURRENT_TENANT_ID_CONTEXTVAR.reset(token)
+
 
 if __name__ == "__main__":
     connector = LocalFileConnector(file_locations=[os.environ["TEST_FILE"]])

backend/onyx/connectors/google_drive/doc_conversion.py

@@ -1,9 +1,7 @@
 import io
 from datetime import datetime
 from datetime import timezone
-from tempfile import NamedTemporaryFile
 
-import openpyxl  # type: ignore
 from googleapiclient.discovery import build  # type: ignore
 from googleapiclient.errors import HttpError  # type: ignore
 
@@ -45,15 +43,12 @@ def _extract_sections_basic(
 ) -> list[Section]:
     mime_type = file["mimeType"]
     link = file["webViewLink"]
-    supported_file_types = set(item.value for item in GDriveMimeType)
 
-    if mime_type not in supported_file_types:
+    if mime_type not in set(item.value for item in GDriveMimeType):
         # Unsupported file types can still have a title, finding this way is still useful
         return [Section(link=link, text=UNSUPPORTED_FILE_TYPE_CONTENT)]
 
     try:
-        # ---------------------------
-        # Google Sheets extraction
         if mime_type == GDriveMimeType.SPREADSHEET.value:
             try:
                 sheets_service = build(
@@ -114,53 +109,7 @@ def _extract_sections_basic(
                     f"Ran into exception '{e}' when pulling data from Google Sheet '{file['name']}'."
                     " Falling back to basic extraction."
                 )
-        # ---------------------------
-        # Microsoft Excel (.xlsx or .xls) extraction branch
-        elif mime_type in [
-            GDriveMimeType.SPREADSHEET_OPEN_FORMAT.value,
-            GDriveMimeType.SPREADSHEET_MS_EXCEL.value,
-        ]:
-            try:
-                response = service.files().get_media(fileId=file["id"]).execute()
 
-                with NamedTemporaryFile(suffix=".xlsx", delete=True) as tmp:
-                    tmp.write(response)
-                    tmp_path = tmp.name
-
-                    section_separator = "\n\n"
-                    workbook = openpyxl.load_workbook(tmp_path, read_only=True)
-
-                    # Work similarly to the xlsx_to_text function used for file connector
-                    # but returns Sections instead of a string
-                    sections = [
-                        Section(
-                            link=link,
-                            text=(
-                                f"Sheet: {sheet.title}\n\n"
-                                + section_separator.join(
-                                    ",".join(map(str, row))
-                                    for row in sheet.iter_rows(
-                                        min_row=1, values_only=True
-                                    )
-                                    if row
-                                )
-                            ),
-                        )
-                        for sheet in workbook.worksheets
-                    ]
-
-                return sections
-
-            except Exception as e:
-                logger.warning(
-                    f"Error extracting data from Excel file '{file['name']}': {e}"
-                )
-                return [
-                    Section(link=link, text="Error extracting data from Excel file")
-                ]
-
-        # ---------------------------
-        # Export for Google Docs, PPT, and fallback for spreadsheets
         if mime_type in [
             GDriveMimeType.DOC.value,
             GDriveMimeType.PPT.value,
@@ -179,8 +128,6 @@ def _extract_sections_basic(
             )
             return [Section(link=link, text=text)]
 
-        # ---------------------------
-        # Plain text and Markdown files
         elif mime_type in [
             GDriveMimeType.PLAIN_TEXT.value,
             GDriveMimeType.MARKDOWN.value,
@@ -194,8 +141,6 @@ def _extract_sections_basic(
                     .decode("utf-8"),
                 )
             ]
-        # ---------------------------
-        # Word, PowerPoint, PDF files
         if mime_type in [
             GDriveMimeType.WORD_DOC.value,
             GDriveMimeType.POWERPOINT.value,
@@ -225,11 +170,7 @@ def _extract_sections_basic(
                     Section(link=link, text=pptx_to_text(file=io.BytesIO(response)))
                 ]
 
-        # Catch-all case, should not happen since there should be specific handling
-        # for each of the supported file types
-        error_message = f"Unsupported file type: {mime_type}"
-        logger.error(error_message)
-        raise ValueError(error_message)
+        return [Section(link=link, text=UNSUPPORTED_FILE_TYPE_CONTENT)]
 
     except Exception:
         return [Section(link=link, text=UNSUPPORTED_FILE_TYPE_CONTENT)]

backend/onyx/connectors/google_drive/models.py

@@ -5,10 +5,6 @@ from typing import Any
 class GDriveMimeType(str, Enum):
     DOC = "application/vnd.google-apps.document"
     SPREADSHEET = "application/vnd.google-apps.spreadsheet"
-    SPREADSHEET_OPEN_FORMAT = (
-        "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
-    )
-    SPREADSHEET_MS_EXCEL = "application/vnd.ms-excel"
     PDF = "application/pdf"
     WORD_DOC = "application/vnd.openxmlformats-officedocument.wordprocessingml.document"
     PPT = "application/vnd.google-apps.presentation"

backend/onyx/context/search/models.py

@@ -98,7 +98,6 @@ class BaseFilters(BaseModel):
     document_set: list[str] | None = None
     time_cutoff: datetime | None = None
     tags: list[Tag] | None = None
-    user_file_ids: list[int] | None = None
 
 
 class IndexFilters(BaseFilters):

backend/onyx/context/search/preprocessing/preprocessing.py

@@ -160,16 +160,7 @@ def retrieval_preprocessing(
     user_acl_filters = (
         None if bypass_acl else build_access_filters_for_user(user, db_session)
     )
-    user_file_ids = preset_filters.user_file_ids
-    if persona and persona.user_files:
-        user_file_ids = user_file_ids + [
-            file.id
-            for file in persona.user_files
-            if file.id not in preset_filters.user_file_ids
-        ]
-
     final_filters = IndexFilters(
-        user_file_ids=user_file_ids,
         source_type=preset_filters.source_type or predicted_source_filters,
         document_set=preset_filters.document_set,
         time_cutoff=time_filter or predicted_time_cutoff,

backend/onyx/db/api_key.py

@@ -16,6 +16,7 @@ from onyx.configs.constants import UNNAMED_KEY_PLACEHOLDER
 from onyx.db.models import ApiKey
 from onyx.db.models import User
 from onyx.server.api_key.models import APIKeyArgs
+from shared_configs.configs import MULTI_TENANT
 from shared_configs.contextvars import get_current_tenant_id
 
 
@@ -72,7 +73,7 @@ def insert_api_key(
     # Get tenant_id from context var (will be default schema for single tenant)
     tenant_id = get_current_tenant_id()
 
-    api_key = generate_api_key(tenant_id)
+    api_key = generate_api_key(tenant_id if MULTI_TENANT else None)
     api_key_user_id = uuid.uuid4()
 
     display_name = api_key_args.name or UNNAMED_KEY_PLACEHOLDER

backend/onyx/db/connector_credential_pair.py

@@ -104,7 +104,6 @@ def get_connector_credential_pairs_for_user(
     get_editable: bool = True,
     ids: list[int] | None = None,
     eager_load_connector: bool = False,
-    include_user_files: bool = False,
     eager_load_credential: bool = False,
     eager_load_user: bool = False,
 ) -> list[ConnectorCredentialPair]:
@@ -127,9 +126,6 @@ def get_connector_credential_pairs_for_user(
     if ids:
         stmt = stmt.where(ConnectorCredentialPair.id.in_(ids))
 
-    if not include_user_files:
-        stmt = stmt.where(ConnectorCredentialPair.is_user_file != True)  # noqa: E712
-
     return list(db_session.scalars(stmt).unique().all())
 
 
@@ -157,16 +153,14 @@ def get_connector_credential_pairs_for_user_parallel(
 
 
 def get_connector_credential_pairs(
-    db_session: Session, ids: list[int] | None = None, include_user_files: bool = False
+    db_session: Session,
+    ids: list[int] | None = None,
 ) -> list[ConnectorCredentialPair]:
     stmt = select(ConnectorCredentialPair).distinct()
 
     if ids:
         stmt = stmt.where(ConnectorCredentialPair.id.in_(ids))
 
-    if not include_user_files:
-        stmt = stmt.where(ConnectorCredentialPair.is_user_file != True)  # noqa: E712
-
     return list(db_session.scalars(stmt).all())
 
 
@@ -452,7 +446,6 @@ def add_credential_to_connector(
     initial_status: ConnectorCredentialPairStatus = ConnectorCredentialPairStatus.ACTIVE,
     last_successful_index_time: datetime | None = None,
     seeding_flow: bool = False,
-    is_user_file: bool = False,
 ) -> StatusResponse:
     connector = fetch_connector_by_id(connector_id, db_session)
 
@@ -518,7 +511,6 @@ def add_credential_to_connector(
         access_type=access_type,
         auto_sync_options=auto_sync_options,
         last_successful_index_time=last_successful_index_time,
-        is_user_file=is_user_file,
     )
     db_session.add(association)
     db_session.flush()  # make sure the association has an id

backend/onyx/db/document.py

@@ -274,7 +274,7 @@ def get_document_counts_for_cc_pairs_parallel(
 def get_access_info_for_document(
     db_session: Session,
     document_id: str,
-) -> tuple[str, list[str | None], bool, list[int], list[int]] | None:
+) -> tuple[str, list[str | None], bool] | None:
     """Gets access info for a single document by calling the get_access_info_for_documents function
     and passing a list with a single document ID.
     Args:
@@ -294,7 +294,7 @@ def get_access_info_for_document(
 def get_access_info_for_documents(
     db_session: Session,
     document_ids: list[str],
-) -> Sequence[tuple[str, list[str | None], bool, list[int], list[int]]]:
+) -> Sequence[tuple[str, list[str | None], bool]]:
     """Gets back all relevant access info for the given documents. This includes
     the user_ids for cc pairs that the document is associated with + whether any
     of the associated cc pairs are intending to make the document globally public.

backend/onyx/db/document_set.py

@@ -605,6 +605,7 @@ def fetch_document_sets_for_document(
     result = fetch_document_sets_for_documents([document_id], db_session)
     if not result:
         return []
+
     return result[0][1]
 
 

backend/onyx/db/engine.py

@@ -258,11 +258,11 @@ class SqlEngine:
                 cls._engine = None
 
 
-def get_all_tenant_ids() -> list[str]:
+def get_all_tenant_ids() -> list[str] | list[None]:
     """Returning [None] means the only tenant is the 'public' or self hosted tenant."""
 
     if not MULTI_TENANT:
-        return [POSTGRES_DEFAULT_SCHEMA]
+        return [None]
 
     with get_session_with_shared_schema() as session:
         result = session.execute(
@@ -417,7 +417,7 @@ def get_session_with_shared_schema() -> Generator[Session, None, None]:
 
 
 @contextmanager
-def get_session_with_tenant(*, tenant_id: str) -> Generator[Session, None, None]:
+def get_session_with_tenant(*, tenant_id: str | None) -> Generator[Session, None, None]:
     """
     Generate a database session for a specific tenant.
     """

backend/onyx/db/models.py

@@ -205,11 +205,6 @@ class User(SQLAlchemyBaseUserTableUUID, Base):
         primaryjoin="User.id == foreign(ConnectorCredentialPair.creator_id)",
     )
 
-    folders: Mapped[list["UserFolder"]] = relationship(
-        "UserFolder", back_populates="user"
-    )
-    files: Mapped[list["UserFile"]] = relationship("UserFile", back_populates="user")
-
     @property
     def password_configured(self) -> bool:
         """
@@ -412,7 +407,6 @@ class ConnectorCredentialPair(Base):
     """
 
     __tablename__ = "connector_credential_pair"
-    is_user_file: Mapped[bool] = mapped_column(Boolean, default=False)
     # NOTE: this `id` column has to use `Sequence` instead of `autoincrement=True`
     # due to some SQLAlchemy quirks + this not being a primary key column
     id: Mapped[int] = mapped_column(
@@ -499,10 +493,6 @@ class ConnectorCredentialPair(Base):
         primaryjoin="foreign(ConnectorCredentialPair.creator_id) == remote(User.id)",
     )
 
-    user_file: Mapped["UserFile"] = relationship(
-        "UserFile", back_populates="cc_pair", uselist=False
-    )
-
     background_errors: Mapped[list["BackgroundError"]] = relationship(
         "BackgroundError", back_populates="cc_pair", cascade="all, delete-orphan"
     )
@@ -1723,17 +1713,6 @@ class Persona(Base):
         secondary="persona__user_group",
         viewonly=True,
     )
-    # Relationship to UserFile
-    user_files: Mapped[list["UserFile"]] = relationship(
-        "UserFile",
-        secondary="persona__user_file",
-        back_populates="assistants",
-    )
-    user_folders: Mapped[list["UserFolder"]] = relationship(
-        "UserFolder",
-        secondary="persona__user_folder",
-        back_populates="assistants",
-    )
     labels: Mapped[list["PersonaLabel"]] = relationship(
         "PersonaLabel",
         secondary=Persona__PersonaLabel.__table__,
@@ -1750,24 +1729,6 @@ class Persona(Base):
     )
 
 
-class Persona__UserFolder(Base):
-    __tablename__ = "persona__user_folder"
-
-    persona_id: Mapped[int] = mapped_column(ForeignKey("persona.id"), primary_key=True)
-    user_folder_id: Mapped[int] = mapped_column(
-        ForeignKey("user_folder.id"), primary_key=True
-    )
-
-
-class Persona__UserFile(Base):
-    __tablename__ = "persona__user_file"
-
-    persona_id: Mapped[int] = mapped_column(ForeignKey("persona.id"), primary_key=True)
-    user_file_id: Mapped[int] = mapped_column(
-        ForeignKey("user_file.id"), primary_key=True
-    )
-
-
 class PersonaLabel(Base):
     __tablename__ = "persona_label"
 
@@ -2289,68 +2250,6 @@ class InputPrompt__User(Base):
     disabled: Mapped[bool] = mapped_column(Boolean, nullable=False, default=False)
 
 
-class UserFolder(Base):
-    __tablename__ = "user_folder"
-
-    id: Mapped[int] = mapped_column(primary_key=True, autoincrement=True)
-    user_id: Mapped[int] = mapped_column(ForeignKey("user.id"), nullable=False)
-    name: Mapped[str] = mapped_column(nullable=False)
-    description: Mapped[str] = mapped_column(nullable=False)
-    created_at: Mapped[datetime.datetime] = mapped_column(
-        DateTime(timezone=True), server_default=func.now()
-    )
-
-    # Mapped[datetime.datetime] = mapped_column(
-    #     DateTime(timezone=True), server_default=func.now()
-    # )
-
-    user: Mapped["User"] = relationship(back_populates="folders")
-    files: Mapped[list["UserFile"]] = relationship(back_populates="folder")
-    assistants: Mapped[list["Persona"]] = relationship(
-        "Persona",
-        secondary=Persona__UserFolder.__table__,
-        back_populates="user_folders",
-    )
-
-
-class UserDocument(str, Enum):
-    CHAT = "chat"
-    RECENT = "recent"
-    FILE = "file"
-
-
-class UserFile(Base):
-    __tablename__ = "user_file"
-
-    id: Mapped[int] = mapped_column(primary_key=True, autoincrement=True)
-    user_id: Mapped[int | None] = mapped_column(ForeignKey("user.id"), nullable=False)
-    assistants: Mapped[list["Persona"]] = relationship(
-        "Persona",
-        secondary=Persona__UserFile.__table__,
-        back_populates="user_files",
-    )
-    folder_id: Mapped[int | None] = mapped_column(
-        ForeignKey("user_folder.id"), nullable=True
-    )
-
-    file_id: Mapped[str] = mapped_column(nullable=False)
-    document_id: Mapped[str] = mapped_column(nullable=False)
-    name: Mapped[str] = mapped_column(nullable=False)
-    created_at: Mapped[datetime.datetime] = mapped_column(
-        default=datetime.datetime.utcnow
-    )
-    user: Mapped["User"] = relationship(back_populates="files")
-    folder: Mapped["UserFolder"] = relationship(back_populates="files")
-    token_count: Mapped[int | None] = mapped_column(Integer, nullable=True)
-
-    cc_pair_id: Mapped[int | None] = mapped_column(
-        ForeignKey("connector_credential_pair.id"), nullable=True, unique=True
-    )
-    cc_pair: Mapped["ConnectorCredentialPair"] = relationship(
-        "ConnectorCredentialPair", back_populates="user_file"
-    )
-
-
 """
 Multi-tenancy related tables
 """

backend/onyx/db/persona.py

@@ -33,8 +33,6 @@ from onyx.db.models import StarterMessage
 from onyx.db.models import Tool
 from onyx.db.models import User
 from onyx.db.models import User__UserGroup
-from onyx.db.models import UserFile
-from onyx.db.models import UserFolder
 from onyx.db.models import UserGroup
 from onyx.db.notification import create_notification
 from onyx.server.features.persona.models import PersonaSharedNotificationData
@@ -239,8 +237,6 @@ def create_update_persona(
             llm_relevance_filter=create_persona_request.llm_relevance_filter,
             llm_filter_extraction=create_persona_request.llm_filter_extraction,
             is_default_persona=create_persona_request.is_default_persona,
-            user_file_ids=create_persona_request.user_file_ids,
-            user_folder_ids=create_persona_request.user_folder_ids,
         )
 
         versioned_make_persona_private = fetch_versioned_implementation(
@@ -335,8 +331,6 @@ def get_personas_for_user(
             selectinload(Persona.groups),
             selectinload(Persona.users),
             selectinload(Persona.labels),
-            selectinload(Persona.user_files),
-            selectinload(Persona.user_folders),
         )
 
     results = db_session.execute(stmt).scalars().all()
@@ -431,8 +425,6 @@ def upsert_persona(
     builtin_persona: bool = False,
     is_default_persona: bool = False,
     label_ids: list[int] | None = None,
-    user_file_ids: list[int] | None = None,
-    user_folder_ids: list[int] | None = None,
     chunks_above: int = CONTEXT_CHUNKS_ABOVE,
     chunks_below: int = CONTEXT_CHUNKS_BELOW,
 ) -> Persona:
@@ -458,7 +450,6 @@ def upsert_persona(
             user=user,
             get_editable=True,
         )
-
     # Fetch and attach tools by IDs
     tools = None
     if tool_ids is not None:
@@ -477,26 +468,6 @@ def upsert_persona(
         if not document_sets and document_set_ids:
             raise ValueError("document_sets not found")
 
-    # Fetch and attach user_files by IDs
-    user_files = None
-    if user_file_ids is not None:
-        user_files = (
-            db_session.query(UserFile).filter(UserFile.id.in_(user_file_ids)).all()
-        )
-        if not user_files and user_file_ids:
-            raise ValueError("user_files not found")
-
-    # Fetch and attach user_folders by IDs
-    user_folders = None
-    if user_folder_ids is not None:
-        user_folders = (
-            db_session.query(UserFolder)
-            .filter(UserFolder.id.in_(user_folder_ids))
-            .all()
-        )
-        if not user_folders and user_folder_ids:
-            raise ValueError("user_folders not found")
-
     # Fetch and attach prompts by IDs
     prompts = None
     if prompt_ids is not None:
@@ -561,14 +532,6 @@ def upsert_persona(
         if tools is not None:
             existing_persona.tools = tools or []
 
-        if user_file_ids is not None:
-            existing_persona.user_files.clear()
-            existing_persona.user_files = user_files or []
-
-        if user_folder_ids is not None:
-            existing_persona.user_folders.clear()
-            existing_persona.user_folders = user_folders or []
-
         # We should only update display priority if it is not already set
         if existing_persona.display_priority is None:
             existing_persona.display_priority = display_priority

backend/onyx/db/user_documents.py

@@ -1,175 +0,0 @@
-import datetime
-from typing import List
-
-from fastapi import UploadFile
-from sqlalchemy import and_
-from sqlalchemy.orm import Session
-
-from onyx.connectors.file.connector import _read_files_and_metadata
-from onyx.db.models import Persona
-from onyx.db.models import Persona__UserFile
-from onyx.db.models import User
-from onyx.db.models import UserFile
-from onyx.db.models import UserFolder
-from onyx.file_processing.extract_file_text import read_text_file
-from onyx.llm.factory import get_default_llms
-from onyx.natural_language_processing.utils import get_tokenizer
-from onyx.server.documents.connector import upload_files
-
-USER_FILE_CONSTANT = "USER_FILE_CONNECTOR"
-
-
-def create_user_files(
-    files: List[UploadFile],
-    folder_id: int | None,
-    user: User | None,
-    db_session: Session,
-) -> list[UserFile]:
-    upload_response = upload_files(files, db_session)
-    user_files = []
-
-    context_files = _read_files_and_metadata(
-        file_name=str(upload_response.file_paths[0]), db_session=db_session
-    )
-
-    content, _ = read_text_file(next(context_files)[1])
-    llm, _ = get_default_llms()
-
-    llm_tokenizer = get_tokenizer(
-        model_name=llm.config.model_name,
-        provider_type=llm.config.model_provider,
-    )
-    token_count = len(llm_tokenizer.encode(content))
-
-    for file_path, file in zip(upload_response.file_paths, files):
-        new_file = UserFile(
-            user_id=user.id if user else None,
-            folder_id=folder_id,
-            file_id=file_path,
-            document_id="USER_FILE_CONNECTOR__" + file_path,
-            name=file.filename,
-            token_count=token_count,
-        )
-        db_session.add(new_file)
-        user_files.append(new_file)
-    db_session.commit()
-    return user_files
-
-
-def get_user_files_from_folder(folder_id: int, db_session: Session) -> list[UserFile]:
-    return db_session.query(UserFile).filter(UserFile.folder_id == folder_id).all()
-
-
-def share_file_with_assistant(
-    file_id: int, assistant_id: int, db_session: Session
-) -> None:
-    file = db_session.query(UserFile).filter(UserFile.id == file_id).first()
-    assistant = db_session.query(Persona).filter(Persona.id == assistant_id).first()
-
-    if file and assistant:
-        file.assistants.append(assistant)
-        db_session.commit()
-
-
-def unshare_file_with_assistant(
-    file_id: int, assistant_id: int, db_session: Session
-) -> None:
-    db_session.query(Persona__UserFile).filter(
-        and_(
-            Persona__UserFile.user_file_id == file_id,
-            Persona__UserFile.persona_id == assistant_id,
-        )
-    ).delete()
-    db_session.commit()
-
-
-def share_folder_with_assistant(
-    folder_id: int, assistant_id: int, db_session: Session
-) -> None:
-    folder = db_session.query(UserFolder).filter(UserFolder.id == folder_id).first()
-    assistant = db_session.query(Persona).filter(Persona.id == assistant_id).first()
-
-    if folder and assistant:
-        for file in folder.files:
-            share_file_with_assistant(file.id, assistant_id, db_session)
-
-
-def unshare_folder_with_assistant(
-    folder_id: int, assistant_id: int, db_session: Session
-) -> None:
-    folder = db_session.query(UserFolder).filter(UserFolder.id == folder_id).first()
-
-    if folder:
-        for file in folder.files:
-            unshare_file_with_assistant(file.id, assistant_id, db_session)
-
-
-def fetch_user_files_for_documents(
-    document_ids: list[str],
-    db_session: Session,
-) -> dict[str, None | int]:
-    # Query UserFile objects for the given document_ids
-    user_files = (
-        db_session.query(UserFile).filter(UserFile.document_id.in_(document_ids)).all()
-    )
-
-    # Create a dictionary mapping document_ids to UserFile objects
-    result = {doc_id: None for doc_id in document_ids}
-    for user_file in user_files:
-        result[user_file.document_id] = user_file.id
-
-    return result
-
-
-def upsert_user_folder(
-    db_session: Session,
-    id: int | None = None,
-    user_id: int | None = None,
-    name: str | None = None,
-    description: str | None = None,
-    created_at: datetime.datetime | None = None,
-    user: User | None = None,
-    files: list[UserFile] | None = None,
-    assistants: list[Persona] | None = None,
-) -> UserFolder:
-    if id is not None:
-        user_folder = db_session.query(UserFolder).filter_by(id=id).first()
-    else:
-        user_folder = (
-            db_session.query(UserFolder).filter_by(name=name, user_id=user_id).first()
-        )
-
-    if user_folder:
-        if user_id is not None:
-            user_folder.user_id = user_id
-        if name is not None:
-            user_folder.name = name
-        if description is not None:
-            user_folder.description = description
-        if created_at is not None:
-            user_folder.created_at = created_at
-        if user is not None:
-            user_folder.user = user
-        if files is not None:
-            user_folder.files = files
-        if assistants is not None:
-            user_folder.assistants = assistants
-    else:
-        user_folder = UserFolder(
-            id=id,
-            user_id=user_id,
-            name=name,
-            description=description,
-            created_at=created_at or datetime.datetime.utcnow(),
-            user=user,
-            files=files or [],
-            assistants=assistants or [],
-        )
-        db_session.add(user_folder)
-
-    db_session.flush()
-    return user_folder
-
-
-def get_user_folder_by_name(db_session: Session, name: str) -> UserFolder | None:
-    return db_session.query(UserFolder).filter(UserFolder.name == name).first()

backend/onyx/document_index/document_index_utils.py

@@ -81,7 +81,7 @@ def translate_boost_count_to_multiplier(boost: int) -> float:
 # Vespa's Document API.
 def get_document_chunk_ids(
     enriched_document_info_list: list[EnrichedDocumentIndexingInfo],
-    tenant_id: str,
+    tenant_id: str | None,
     large_chunks_enabled: bool,
 ) -> list[UUID]:
     doc_chunk_ids = []
@@ -139,7 +139,7 @@ def get_uuid_from_chunk_info(
     *,
     document_id: str,
     chunk_id: int,
-    tenant_id: str,
+    tenant_id: str | None,
     large_chunk_id: int | None = None,
 ) -> UUID:
     """NOTE: be VERY carefuly about changing this function. If changed without a migration,
@@ -154,7 +154,7 @@ def get_uuid_from_chunk_info(
         "large_" + str(large_chunk_id) if large_chunk_id is not None else str(chunk_id)
     )
     unique_identifier_string = "_".join([doc_str, chunk_index])
-    if MULTI_TENANT:
+    if tenant_id and MULTI_TENANT:
         unique_identifier_string += "_" + tenant_id
 
     uuid_value = uuid.uuid5(uuid.NAMESPACE_X500, unique_identifier_string)

backend/onyx/document_index/interfaces.py

@@ -43,7 +43,7 @@ class IndexBatchParams:
 
     doc_id_to_previous_chunk_cnt: dict[str, int | None]
     doc_id_to_new_chunk_cnt: dict[str, int]
-    tenant_id: str
+    tenant_id: str | None
     large_chunks_enabled: bool
 
 
@@ -222,7 +222,7 @@ class Deletable(abc.ABC):
         self,
         doc_id: str,
         *,
-        tenant_id: str,
+        tenant_id: str | None,
         chunk_count: int | None,
     ) -> int:
         """
@@ -249,7 +249,7 @@ class Updatable(abc.ABC):
         self,
         doc_id: str,
         *,
-        tenant_id: str,
+        tenant_id: str | None,
         chunk_count: int | None,
         fields: VespaDocumentFields,
     ) -> int:
@@ -270,7 +270,9 @@ class Updatable(abc.ABC):
         raise NotImplementedError
 
     @abc.abstractmethod
-    def update(self, update_requests: list[UpdateRequest], *, tenant_id: str) -> None:
+    def update(
+        self, update_requests: list[UpdateRequest], *, tenant_id: str | None
+    ) -> None:
         """
         Updates some set of chunks. The document and fields to update are specified in the update
         requests. Each update request in the list applies its changes to a list of document ids.

backend/onyx/document_index/vespa/app_config/schemas/danswer_chunk.sd

@@ -112,16 +112,6 @@ schema DANSWER_CHUNK_NAME {
             rank: filter
             attribute: fast-search
         }
-        field user_file type int {
-            indexing: summary | attribute
-            rank: filter
-            attribute: fast-search
-        }
-        field user_folders type weightedset<int> {
-            indexing: summary | attribute
-            rank: filter
-            attribute: fast-search
-        }
     }
 
     # If using different tokenization settings, the fieldset has to be removed, and the field must

backend/onyx/document_index/vespa/index.py

@@ -468,7 +468,9 @@ class VespaIndex(DocumentIndex):
                         failure_msg = f"Failed to update document: {future_to_document_id[future]}"
                         raise requests.HTTPError(failure_msg) from e
 
-    def update(self, update_requests: list[UpdateRequest], *, tenant_id: str) -> None:
+    def update(
+        self, update_requests: list[UpdateRequest], *, tenant_id: str | None
+    ) -> None:
         logger.debug(f"Updating {len(update_requests)} documents in Vespa")
 
         # Handle Vespa character limitations
@@ -616,7 +618,7 @@ class VespaIndex(DocumentIndex):
         doc_id: str,
         *,
         chunk_count: int | None,
-        tenant_id: str,
+        tenant_id: str | None,
         fields: VespaDocumentFields,
     ) -> int:
         """Note: if the document id does not exist, the update will be a no-op and the
@@ -645,8 +647,6 @@ class VespaIndex(DocumentIndex):
                     tenant_id=tenant_id,
                     large_chunks_enabled=large_chunks_enabled,
                 )
-                logger.error("CHECKing chunks")
-                logger.error(doc_chunk_ids)
 
                 doc_chunk_count += len(doc_chunk_ids)
 
@@ -661,7 +661,7 @@ class VespaIndex(DocumentIndex):
         self,
         doc_id: str,
         *,
-        tenant_id: str,
+        tenant_id: str | None,
         chunk_count: int | None,
     ) -> int:
         total_chunks_deleted = 0
@@ -693,7 +693,6 @@ class VespaIndex(DocumentIndex):
                     tenant_id=tenant_id,
                     large_chunks_enabled=large_chunks_enabled,
                 )
-
                 for doc_chunk_ids_batch in batch_generator(
                     chunks_to_delete, BATCH_SIZE
                 ):

backend/onyx/document_index/vespa/indexing_utils.py

@@ -47,7 +47,6 @@ from onyx.document_index.vespa_constants import SOURCE_TYPE
 from onyx.document_index.vespa_constants import TENANT_ID
 from onyx.document_index.vespa_constants import TITLE
 from onyx.document_index.vespa_constants import TITLE_EMBEDDING
-from onyx.document_index.vespa_constants import USER_FILE
 from onyx.indexing.models import DocMetadataAwareIndexChunk
 from onyx.utils.logger import setup_logger
 
@@ -199,8 +198,6 @@ def _index_vespa_chunk(
         # which only calls VespaIndex.update
         ACCESS_CONTROL_LIST: {acl_entry: 1 for acl_entry in chunk.access.to_acl()},
         DOCUMENT_SETS: {document_set: 1 for document_set in chunk.document_sets},
-        USER_FILE: chunk.user_file if chunk.user_file is not None else None,
-        # USER_FOLDERS: {user_folder: 1 for user_folder in chunk.user_folders},
         BOOST: chunk.boost,
     }
 

backend/onyx/document_index/vespa/shared_utils/vespa_request_builders.py

@@ -5,6 +5,7 @@ from datetime import timezone
 from onyx.configs.constants import INDEX_SEPARATOR
 from onyx.context.search.models import IndexFilters
 from onyx.document_index.interfaces import VespaChunkRequest
+from onyx.document_index.vespa_constants import ACCESS_CONTROL_LIST
 from onyx.document_index.vespa_constants import CHUNK_ID
 from onyx.document_index.vespa_constants import DOC_UPDATED_AT
 from onyx.document_index.vespa_constants import DOCUMENT_ID
@@ -13,7 +14,6 @@ from onyx.document_index.vespa_constants import HIDDEN
 from onyx.document_index.vespa_constants import METADATA_LIST
 from onyx.document_index.vespa_constants import SOURCE_TYPE
 from onyx.document_index.vespa_constants import TENANT_ID
-from onyx.document_index.vespa_constants import USER_FILE
 from onyx.utils.logger import setup_logger
 from shared_configs.configs import MULTI_TENANT
 
@@ -27,26 +27,14 @@ def build_vespa_filters(
     remove_trailing_and: bool = False,  # Set to True when using as a complete Vespa query
 ) -> str:
     def _build_or_filters(key: str, vals: list[str] | None) -> str:
-        """For string-based 'contains' filters, e.g. WSET fields or array<string> fields."""
-        if not key or not vals:
-            return ""
-        eq_elems = [f'{key} contains "{val}"' for val in vals if val]
-        if not eq_elems:
-            return ""
-        or_clause = " or ".join(eq_elems)
-        return f"({or_clause}) and "
-
-    def _build_int_or_filters(key: str, vals: list[int] | None) -> str:
-        """
-        For an integer field filter.
-        If vals is not None, we want *only* docs whose key matches one of vals.
-        """
-        # If `vals` is None => skip the filter entirely
-        if vals is None or not vals:
+        if vals is None:
             return ""
 
-        # Otherwise build the OR filter
-        eq_elems = [f"{key} = {val}" for val in vals]
+        valid_vals = [val for val in vals if val]
+        if not key or not valid_vals:
+            return ""
+
+        eq_elems = [f'{key} contains "{elem}"' for elem in valid_vals]
         or_clause = " or ".join(eq_elems)
         result = f"({or_clause}) and "
 
@@ -54,55 +42,53 @@ def build_vespa_filters(
 
     def _build_time_filter(
         cutoff: datetime | None,
+        # Slightly over 3 Months, approximately 1 fiscal quarter
         untimed_doc_cutoff: timedelta = timedelta(days=92),
     ) -> str:
         if not cutoff:
             return ""
+
+        # For Documents that don't have an updated at, filter them out for queries asking for
+        # very recent documents (3 months) default. Documents that don't have an updated at
+        # time are assigned 3 months for time decay value
         include_untimed = datetime.now(timezone.utc) - untimed_doc_cutoff > cutoff
         cutoff_secs = int(cutoff.timestamp())
 
         if include_untimed:
+            # Documents without updated_at are assigned -1 as their date
             return f"!({DOC_UPDATED_AT} < {cutoff_secs}) and "
+
         return f"({DOC_UPDATED_AT} >= {cutoff_secs}) and "
 
-    # Start building the filter string
     filter_str = f"!({HIDDEN}=true) and " if not include_hidden else ""
 
-    # If running in multi-tenant mode
+    # If running in multi-tenant mode, we may want to filter by tenant_id
     if filters.tenant_id and MULTI_TENANT:
         filter_str += f'({TENANT_ID} contains "{filters.tenant_id}") and '
 
-    # ACL filters
-    # if filters.access_control_list is not None:
-    #     filter_str += _build_or_filters(ACCESS_CONTROL_LIST, filters.access_control_list)
+    # CAREFUL touching this one, currently there is no second ACL double-check post retrieval
+    if filters.access_control_list is not None:
+        filter_str += _build_or_filters(
+            ACCESS_CONTROL_LIST, filters.access_control_list
+        )
 
-    # Source type filters
     source_strs = (
         [s.value for s in filters.source_type] if filters.source_type else None
     )
     filter_str += _build_or_filters(SOURCE_TYPE, source_strs)
 
-    # Tag filters
     tag_attributes = None
-    if filters.tags:
-        # build e.g. "tag_key|tag_value"
-        tag_attributes = [
-            f"{tag.tag_key}{INDEX_SEPARATOR}{tag.tag_value}" for tag in filters.tags
-        ]
+    tags = filters.tags
+    if tags:
+        tag_attributes = [tag.tag_key + INDEX_SEPARATOR + tag.tag_value for tag in tags]
     filter_str += _build_or_filters(METADATA_LIST, tag_attributes)
 
-    # Document sets
     filter_str += _build_or_filters(DOCUMENT_SETS, filters.document_set)
 
-    # New: user_file_ids as integer filters
-    filter_str += _build_int_or_filters(USER_FILE, filters.user_file_ids)
-
-    # Time filter
     filter_str += _build_time_filter(filters.time_cutoff)
 
-    # Trim trailing " and "
     if remove_trailing_and and filter_str.endswith(" and "):
-        filter_str = filter_str[:-5]
+        filter_str = filter_str[:-5]  # We remove the trailing " and "
 
     return filter_str
 

backend/onyx/document_index/vespa_constants.py

@@ -66,8 +66,6 @@ EMBEDDINGS = "embeddings"
 TITLE_EMBEDDING = "title_embedding"
 ACCESS_CONTROL_LIST = "access_control_list"
 DOCUMENT_SETS = "document_sets"
-USER_FILE = "user_file"
-USER_FOLDERS = "user_folders"
 LARGE_CHUNK_REFERENCE_IDS = "large_chunk_reference_ids"
 METADATA = "metadata"
 METADATA_LIST = "metadata_list"

backend/onyx/file_processing/unstructured.py

@@ -37,7 +37,6 @@ def delete_unstructured_api_key() -> None:
 def _sdk_partition_request(
     file: IO[Any], file_name: str, **kwargs: Any
 ) -> operations.PartitionRequest:
-    file.seek(0, 0)
     try:
         request = operations.PartitionRequest(
             partition_parameters=shared.PartitionParameters(

backend/onyx/file_store/utils.py

@@ -10,10 +10,7 @@ from sqlalchemy.orm import Session
 from onyx.configs.constants import FileOrigin
 from onyx.db.engine import get_session_with_current_tenant
 from onyx.db.models import ChatMessage
-from onyx.db.models import UserFile
-from onyx.db.models import UserFolder
 from onyx.file_store.file_store import get_default_file_store
-from onyx.file_store.models import ChatFileType
 from onyx.file_store.models import FileDescriptor
 from onyx.file_store.models import InMemoryChatFile
 from onyx.utils.b64 import get_image_type
@@ -56,53 +53,6 @@ def load_all_chat_files(
     return files
 
 
-
-
-
-def load_user_folder(folder_id: int, db_session: Session) -> list[InMemoryChatFile]:
-    user_files = (
-        db_session.query(UserFile).filter(UserFile.folder_id == folder_id).all()
-    )
-    return [load_user_file(file.id, db_session) for file in user_files]
-
-
-def load_user_file(file_id: int, db_session: Session) -> InMemoryChatFile:
-    user_file = db_session.query(UserFile).filter(UserFile.id == file_id).first()
-    if not user_file:
-        raise ValueError(f"User file with id {file_id} not found")
-
-    file_io = get_default_file_store(db_session).read_file(
-        user_file.document_id, mode="b"
-    )
-    return InMemoryChatFile(
-        file_id=str(user_file.id),
-        content=file_io.read(),
-        file_type=ChatFileType.PLAIN_TEXT,
-        filename=user_file.name,
-    )
-
-
-def load_all_user_files(
-    user_file_ids: list[int],
-    user_folder_ids: list[int],
-    db_session: Session,
-) -> list[InMemoryChatFile]:
-    return cast(
-        list[InMemoryChatFile],
-        run_functions_tuples_in_parallel(
-            [(load_user_file, (file_id, db_session)) for file_id in user_file_ids]
-        )
-        + [
-            file
-            for folder_id in user_folder_ids
-            for file in load_user_folder(folder_id, db_session)
-        ],
-    )
-
-
-
-
-
 def save_file_from_url(url: str) -> str:
     """NOTE: using multiple sessions here, since this is often called
     using multithreading. In practice, sharing a session has resulted in
@@ -178,39 +128,3 @@ def save_files(urls: list[str], base64_files: list[str]) -> list[str]:
     ]
 
     return run_functions_tuples_in_parallel(funcs)
-
-
-def load_all_persona_files_for_chat(
-    persona_id: int, db_session: Session
-) -> tuple[list[InMemoryChatFile], list[int]]:
-    from onyx.db.models import Persona
-    from sqlalchemy.orm import joinedload
-
-    persona = (
-        db_session.query(Persona)
-        .filter(Persona.id == persona_id)
-        .options(
-            joinedload(Persona.user_files),
-            joinedload(Persona.user_folders).joinedload(UserFolder.files),
-        )
-        .one()
-    )
-
-    persona_file_calls = [
-        (load_user_file, (user_file.id, db_session)) for user_file in persona.user_files
-    ]
-    persona_loaded_files = run_functions_tuples_in_parallel(persona_file_calls)
-
-    persona_folder_files = []
-    persona_folder_file_ids = []
-    for user_folder in persona.user_folders:
-        folder_files = load_user_folder(user_folder.id, db_session)
-        persona_folder_files.extend(folder_files)
-        persona_folder_file_ids.extend([file.id for file in user_folder.files])
-
-    persona_files = list(persona_loaded_files) + persona_folder_files
-    persona_file_ids = [
-        file.id for file in persona.user_files
-    ] + persona_folder_file_ids
-
-    return persona_files, persona_file_ids

backend/onyx/indexing/indexing_pipeline.py

@@ -31,7 +31,6 @@ from onyx.db.models import Document as DBDocument
 from onyx.db.search_settings import get_current_search_settings
 from onyx.db.tag import create_or_add_document_tag
 from onyx.db.tag import create_or_add_document_tag_list
-from onyx.db.user_documents import fetch_user_files_for_documents
 from onyx.document_index.document_index_utils import (
     get_multipass_config,
 )
@@ -159,8 +158,8 @@ def index_doc_batch_with_handler(
     document_batch: list[Document],
     index_attempt_metadata: IndexAttemptMetadata,
     db_session: Session,
-    tenant_id: str,
     ignore_time_skip: bool = False,
+    tenant_id: str | None = None,
 ) -> IndexingPipelineResult:
     try:
         index_pipeline_result = index_doc_batch(
@@ -318,8 +317,8 @@ def index_doc_batch(
     document_index: DocumentIndex,
     index_attempt_metadata: IndexAttemptMetadata,
     db_session: Session,
-    tenant_id: str,
     ignore_time_skip: bool = False,
+    tenant_id: str | None = None,
     filter_fnc: Callable[[list[Document]], list[Document]] = filter_documents,
 ) -> IndexingPipelineResult:
     """Takes different pieces of the indexing pipeline and applies it to a batch of documents
@@ -403,10 +402,6 @@ def index_doc_batch(
             )
         }
 
-        doc_id_to_user_file_id: dict[str, int | None] = fetch_user_files_for_documents(
-            document_ids=updatable_ids, db_session=db_session
-        )
-
         doc_id_to_previous_chunk_cnt: dict[str, int | None] = {
             document_id: chunk_count
             for document_id, chunk_count in fetch_chunk_counts_for_documents(
@@ -438,7 +433,6 @@ def index_doc_batch(
                 document_sets=set(
                     doc_id_to_document_set.get(chunk.source_document.id, [])
                 ),
-                user_file=doc_id_to_user_file_id.get(chunk.source_document.id, None),
                 boost=(
                     ctx.id_to_db_doc_map[chunk.source_document.id].boost
                     if chunk.source_document.id in ctx.id_to_db_doc_map
@@ -531,9 +525,9 @@ def build_indexing_pipeline(
     embedder: IndexingEmbedder,
     document_index: DocumentIndex,
     db_session: Session,
-    tenant_id: str,
     chunker: Chunker | None = None,
     ignore_time_skip: bool = False,
+    tenant_id: str | None = None,
     callback: IndexingHeartbeatInterface | None = None,
 ) -> IndexingPipelineProtocol:
     """Builds a pipeline which takes in a list (batch) of docs and indexes them."""

backend/onyx/indexing/models.py

@@ -84,11 +84,9 @@ class DocMetadataAwareIndexChunk(IndexChunk):
            negative -> ranked lower.
     """
 
-    tenant_id: str
+    tenant_id: str | None = None
     access: "DocumentAccess"
     document_sets: set[str]
-    user_file: int | None
-    # user_folders: list[int]
     boost: int
 
     @classmethod
@@ -97,18 +95,14 @@ class DocMetadataAwareIndexChunk(IndexChunk):
         index_chunk: IndexChunk,
         access: "DocumentAccess",
         document_sets: set[str],
-        user_file: int | None,
-        # user_folder: list[int],
         boost: int,
-        tenant_id: str,
+        tenant_id: str | None,
     ) -> "DocMetadataAwareIndexChunk":
         index_chunk_data = index_chunk.model_dump()
         return cls(
             **index_chunk_data,
             access=access,
             document_sets=document_sets,
-            user_file=user_file,
-            # user_folders=user_folders,
             boost=boost,
             tenant_id=tenant_id,
         )

backend/onyx/main.py

@@ -97,7 +97,6 @@ from onyx.server.settings.api import basic_router as settings_router
 from onyx.server.token_rate_limits.api import (
     router as token_rate_limit_settings_router,
 )
-from onyx.server.user_documents.api import router as user_documents_router
 from onyx.server.utils import BasicAuthenticationError
 from onyx.setup import setup_multitenant_onyx
 from onyx.setup import setup_onyx
@@ -220,7 +219,7 @@ async def lifespan(app: FastAPI) -> AsyncGenerator[None, None]:
 
         # If we are multi-tenant, we need to only set up initial public tables
         with Session(engine) as db_session:
-            setup_onyx(db_session, POSTGRES_DEFAULT_SCHEMA)
+            setup_onyx(db_session, None)
     else:
         setup_multitenant_onyx()
 
@@ -296,7 +295,6 @@ def get_application() -> FastAPI:
     include_router_with_global_prefix_prepended(application, input_prompt_router)
     include_router_with_global_prefix_prepended(application, admin_input_prompt_router)
     include_router_with_global_prefix_prepended(application, cc_pair_router)
-    include_router_with_global_prefix_prepended(application, user_documents_router)
     include_router_with_global_prefix_prepended(application, folder_router)
     include_router_with_global_prefix_prepended(application, document_set_router)
     include_router_with_global_prefix_prepended(application, search_settings_router)

backend/onyx/onyxbot/slack/blocks.py

@@ -410,7 +410,7 @@ def _build_qa_response_blocks(
 
 
 def _build_continue_in_web_ui_block(
-    tenant_id: str,
+    tenant_id: str | None,
     message_id: int | None,
 ) -> Block:
     if message_id is None:
@@ -482,7 +482,7 @@ def build_follow_up_resolved_blocks(
 
 def build_slack_response_blocks(
     answer: ChatOnyxBotResponse,
-    tenant_id: str,
+    tenant_id: str | None,
     message_info: SlackMessageInfo,
     channel_conf: ChannelConfig | None,
     use_citations: bool,

backend/onyx/onyxbot/slack/handlers/handle_buttons.py

@@ -151,7 +151,7 @@ def handle_slack_feedback(
     user_id_to_post_confirmation: str,
     channel_id_to_post_confirmation: str,
     thread_ts_to_post_confirmation: str,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> None:
     message_id, doc_id, doc_rank = decompose_action_id(feedback_id)
 

backend/onyx/onyxbot/slack/handlers/handle_message.py

@@ -109,7 +109,7 @@ def handle_message(
     slack_channel_config: SlackChannelConfig,
     client: WebClient,
     feedback_reminder_id: str | None,
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> bool:
     """Potentially respond to the user message depending on filters and if an answer was generated
 

backend/onyx/onyxbot/slack/handlers/handle_regular_answer.py

@@ -72,7 +72,7 @@ def handle_regular_answer(
     channel: str,
     logger: OnyxLoggingAdapter,
     feedback_reminder_id: str | None,
-    tenant_id: str,
+    tenant_id: str | None,
     num_retries: int = DANSWER_BOT_NUM_RETRIES,
     thread_context_percent: float = MAX_THREAD_CONTEXT_PERCENTAGE,
     should_respond_with_error_msgs: bool = DANSWER_BOT_DISPLAY_ERROR_MSGS,

backend/onyx/onyxbot/slack/listener.py

@@ -123,13 +123,13 @@ _OFFICIAL_SLACKBOT_USER_ID = "USLACKBOT"
 class SlackbotHandler:
     def __init__(self) -> None:
         logger.info("Initializing SlackbotHandler")
-        self.tenant_ids: Set[str] = set()
+        self.tenant_ids: Set[str | None] = set()
         # The keys for these dictionaries are tuples of (tenant_id, slack_bot_id)
-        self.socket_clients: Dict[tuple[str, int], TenantSocketModeClient] = {}
-        self.slack_bot_tokens: Dict[tuple[str, int], SlackBotTokens] = {}
+        self.socket_clients: Dict[tuple[str | None, int], TenantSocketModeClient] = {}
+        self.slack_bot_tokens: Dict[tuple[str | None, int], SlackBotTokens] = {}
 
         # Store Redis lock objects here so we can release them properly
-        self.redis_locks: Dict[str, Lock] = {}
+        self.redis_locks: Dict[str | None, Lock] = {}
 
         self.running = True
         self.pod_id = self.get_pod_id()
@@ -193,7 +193,7 @@ class SlackbotHandler:
             self._shutdown_event.wait(timeout=TENANT_HEARTBEAT_INTERVAL)
 
     def _manage_clients_per_tenant(
-        self, db_session: Session, tenant_id: str, bot: SlackBot
+        self, db_session: Session, tenant_id: str | None, bot: SlackBot
     ) -> None:
         """
         - If the tokens are missing or empty, close the socket client and remove them.
@@ -385,7 +385,7 @@ class SlackbotHandler:
             finally:
                 CURRENT_TENANT_ID_CONTEXTVAR.reset(token)
 
-    def _remove_tenant(self, tenant_id: str) -> None:
+    def _remove_tenant(self, tenant_id: str | None) -> None:
         """
         Helper to remove a tenant from `self.tenant_ids` and close any socket clients.
         (Lock release now happens in `acquire_tenants()`, not here.)
@@ -415,7 +415,7 @@ class SlackbotHandler:
             )
 
     def start_socket_client(
-        self, slack_bot_id: int, tenant_id: str, slack_bot_tokens: SlackBotTokens
+        self, slack_bot_id: int, tenant_id: str | None, slack_bot_tokens: SlackBotTokens
     ) -> None:
         socket_client: TenantSocketModeClient = _get_socket_client(
             slack_bot_tokens, tenant_id, slack_bot_id
@@ -912,7 +912,7 @@ def create_process_slack_event() -> (
 
 
 def _get_socket_client(
-    slack_bot_tokens: SlackBotTokens, tenant_id: str, slack_bot_id: int
+    slack_bot_tokens: SlackBotTokens, tenant_id: str | None, slack_bot_id: int
 ) -> TenantSocketModeClient:
     # For more info on how to set this up, checkout the docs:
     # https://docs.onyx.app/slack_bot_setup

backend/onyx/onyxbot/slack/utils.py

@@ -570,7 +570,7 @@ def read_slack_thread(
 
 
 def slack_usage_report(
-    action: str, sender_id: str | None, client: WebClient, tenant_id: str
+    action: str, sender_id: str | None, client: WebClient, tenant_id: str | None
 ) -> None:
     if DISABLE_TELEMETRY:
         return
@@ -663,7 +663,9 @@ def get_feedback_visibility() -> FeedbackVisibility:
 
 
 class TenantSocketModeClient(SocketModeClient):
-    def __init__(self, tenant_id: str, slack_bot_id: int, *args: Any, **kwargs: Any):
+    def __init__(
+        self, tenant_id: str | None, slack_bot_id: int, *args: Any, **kwargs: Any
+    ):
         super().__init__(*args, **kwargs)
         self.tenant_id = tenant_id
         self.slack_bot_id = slack_bot_id

backend/onyx/redis/redis_connector.py

@@ -16,10 +16,10 @@ class RedisConnector:
     """Composes several classes to simplify interacting with a connector and its
     associated background tasks / associated redis interactions."""
 
-    def __init__(self, tenant_id: str, id: int) -> None:
+    def __init__(self, tenant_id: str | None, id: int) -> None:
         """id: a connector credential pair id"""
 
-        self.tenant_id: str = tenant_id
+        self.tenant_id: str | None = tenant_id
         self.id: int = id
         self.redis: redis.Redis = get_redis_client(tenant_id=tenant_id)
 

backend/onyx/redis/redis_connector_credential_pair.py

@@ -31,7 +31,7 @@ class RedisConnectorCredentialPair(RedisObjectHelper):
     PREFIX = "connectorsync"
     TASKSET_PREFIX = PREFIX + "_taskset"
 
-    def __init__(self, tenant_id: str, id: int) -> None:
+    def __init__(self, tenant_id: str | None, id: int) -> None:
         super().__init__(tenant_id, str(id))
 
         # documents that should be skipped
@@ -60,7 +60,7 @@ class RedisConnectorCredentialPair(RedisObjectHelper):
         db_session: Session,
         redis_client: Redis,
         lock: RedisLock,
-        tenant_id: str,
+        tenant_id: str | None,
     ) -> tuple[int, int] | None:
         """We can limit the number of tasks generated here, which is useful to prevent
         one tenant from overwhelming the sync queue.

backend/onyx/redis/redis_connector_delete.py

@@ -39,8 +39,8 @@ class RedisConnectorDelete:
     ACTIVE_PREFIX = PREFIX + "_active"
     ACTIVE_TTL = 3600
 
-    def __init__(self, tenant_id: str, id: int, redis: redis.Redis) -> None:
-        self.tenant_id: str = tenant_id
+    def __init__(self, tenant_id: str | None, id: int, redis: redis.Redis) -> None:
+        self.tenant_id: str | None = tenant_id
         self.id = id
         self.redis = redis
 

backend/onyx/redis/redis_connector_doc_perm_sync.py

@@ -52,8 +52,8 @@ class RedisConnectorPermissionSync:
     ACTIVE_PREFIX = PREFIX + "_active"
     ACTIVE_TTL = CELERY_PERMISSIONS_SYNC_LOCK_TIMEOUT * 2
 
-    def __init__(self, tenant_id: str, id: int, redis: redis.Redis) -> None:
-        self.tenant_id: str = tenant_id
+    def __init__(self, tenant_id: str | None, id: int, redis: redis.Redis) -> None:
+        self.tenant_id: str | None = tenant_id
         self.id = id
         self.redis = redis
 

backend/onyx/redis/redis_connector_ext_group_sync.py

@@ -44,8 +44,8 @@ class RedisConnectorExternalGroupSync:
     ACTIVE_PREFIX = PREFIX + "_active"
     ACTIVE_TTL = 3600
 
-    def __init__(self, tenant_id: str, id: int, redis: redis.Redis) -> None:
-        self.tenant_id: str = tenant_id
+    def __init__(self, tenant_id: str | None, id: int, redis: redis.Redis) -> None:
+        self.tenant_id: str | None = tenant_id
         self.id = id
         self.redis = redis
 

backend/onyx/redis/redis_connector_index.py

@@ -52,12 +52,12 @@ class RedisConnectorIndex:
 
     def __init__(
         self,
-        tenant_id: str,
+        tenant_id: str | None,
         id: int,
         search_settings_id: int,
         redis: redis.Redis,
     ) -> None:
-        self.tenant_id: str = tenant_id
+        self.tenant_id: str | None = tenant_id
         self.id = id
         self.search_settings_id = search_settings_id
         self.redis = redis

backend/onyx/redis/redis_connector_prune.py

@@ -52,8 +52,8 @@ class RedisConnectorPrune:
     ACTIVE_PREFIX = PREFIX + "_active"
     ACTIVE_TTL = CELERY_PRUNING_LOCK_TIMEOUT * 2
 
-    def __init__(self, tenant_id: str, id: int, redis: redis.Redis) -> None:
-        self.tenant_id: str = tenant_id
+    def __init__(self, tenant_id: str | None, id: int, redis: redis.Redis) -> None:
+        self.tenant_id: str | None = tenant_id
         self.id = id
         self.redis = redis
 

backend/onyx/redis/redis_connector_stop.py

@@ -13,8 +13,8 @@ class RedisConnectorStop:
     TIMEOUT_PREFIX = f"{PREFIX}_timeout"
     TIMEOUT_TTL = 300
 
-    def __init__(self, tenant_id: str, id: int, redis: redis.Redis) -> None:
-        self.tenant_id: str = tenant_id
+    def __init__(self, tenant_id: str | None, id: int, redis: redis.Redis) -> None:
+        self.tenant_id: str | None = tenant_id
         self.id: int = id
         self.redis = redis
 

backend/onyx/redis/redis_document_set.py

@@ -23,7 +23,7 @@ class RedisDocumentSet(RedisObjectHelper):
     FENCE_PREFIX = PREFIX + "_fence"
     TASKSET_PREFIX = PREFIX + "_taskset"
 
-    def __init__(self, tenant_id: str, id: int) -> None:
+    def __init__(self, tenant_id: str | None, id: int) -> None:
         super().__init__(tenant_id, str(id))
 
     @property
@@ -58,7 +58,7 @@ class RedisDocumentSet(RedisObjectHelper):
         db_session: Session,
         redis_client: Redis,
         lock: RedisLock,
-        tenant_id: str,
+        tenant_id: str | None,
     ) -> tuple[int, int] | None:
         """Max tasks is ignored for now until we can build the logic to mark the
         document set up to date over multiple batches.

backend/onyx/redis/redis_object_helper.py

@@ -14,8 +14,8 @@ class RedisObjectHelper(ABC):
     FENCE_PREFIX = PREFIX + "_fence"
     TASKSET_PREFIX = PREFIX + "_taskset"
 
-    def __init__(self, tenant_id: str, id: str):
-        self._tenant_id: str = tenant_id
+    def __init__(self, tenant_id: str | None, id: str):
+        self._tenant_id: str | None = tenant_id
         self._id: str = id
         self.redis = get_redis_client(tenant_id=tenant_id)
 
@@ -87,7 +87,7 @@ class RedisObjectHelper(ABC):
         db_session: Session,
         redis_client: Redis,
         lock: RedisLock,
-        tenant_id: str,
+        tenant_id: str | None,
     ) -> tuple[int, int] | None:
         """First element should be the number of actual tasks generated, second should
         be the number of docs that were candidates to be synced for the cc pair.

backend/onyx/redis/redis_usergroup.py

@@ -24,7 +24,7 @@ class RedisUserGroup(RedisObjectHelper):
     FENCE_PREFIX = PREFIX + "_fence"
     TASKSET_PREFIX = PREFIX + "_taskset"
 
-    def __init__(self, tenant_id: str, id: int) -> None:
+    def __init__(self, tenant_id: str | None, id: int) -> None:
         super().__init__(tenant_id, str(id))
 
     @property
@@ -59,7 +59,7 @@ class RedisUserGroup(RedisObjectHelper):
         db_session: Session,
         redis_client: Redis,
         lock: RedisLock,
-        tenant_id: str,
+        tenant_id: str | None,
     ) -> tuple[int, int] | None:
         """Max tasks is ignored for now until we can build the logic to mark the
         user group up to date over multiple batches.

backend/onyx/seeding/load_docs.py

@@ -37,15 +37,13 @@ from onyx.key_value_store.interface import KvKeyNotFoundError
 from onyx.server.documents.models import ConnectorBase
 from onyx.utils.logger import setup_logger
 from onyx.utils.variable_functionality import fetch_versioned_implementation
-from shared_configs.configs import MULTI_TENANT
-from shared_configs.configs import POSTGRES_DEFAULT_SCHEMA
 
 logger = setup_logger()
 
 
 def _create_indexable_chunks(
     preprocessed_docs: list[dict],
-    tenant_id: str,
+    tenant_id: str | None,
 ) -> tuple[list[Document], list[DocMetadataAwareIndexChunk]]:
     ids_to_documents = {}
     chunks = []
@@ -88,10 +86,9 @@ def _create_indexable_chunks(
                 mini_chunk_embeddings=[],
             ),
             title_embedding=preprocessed_doc["title_embedding"],
-            tenant_id=tenant_id if MULTI_TENANT else POSTGRES_DEFAULT_SCHEMA,
+            tenant_id=tenant_id,
             access=default_public_access,
             document_sets=set(),
-            user_file=None,
             boost=DEFAULT_BOOST,
             large_chunk_id=None,
         )
@@ -114,7 +111,7 @@ def load_processed_docs(cohere_enabled: bool) -> list[dict]:
 
 
 def seed_initial_documents(
-    db_session: Session, tenant_id: str, cohere_enabled: bool = False
+    db_session: Session, tenant_id: str | None, cohere_enabled: bool = False
 ) -> None:
     """
     Seed initial documents so users don't have an empty index to start

backend/onyx/seeding/load_yamls.py

@@ -5,7 +5,6 @@ from onyx.configs.chat_configs import INPUT_PROMPT_YAML
 from onyx.configs.chat_configs import MAX_CHUNKS_FED_TO_CHAT
 from onyx.configs.chat_configs import PERSONAS_YAML
 from onyx.configs.chat_configs import PROMPTS_YAML
-from onyx.configs.chat_configs import USER_FOLDERS_YAML
 from onyx.context.search.enums import RecencyBiasSetting
 from onyx.db.document_set import get_or_create_document_set_by_name
 from onyx.db.input_prompt import insert_input_prompt_if_not_exists
@@ -16,30 +15,6 @@ from onyx.db.models import Tool as ToolDBModel
 from onyx.db.persona import upsert_persona
 from onyx.db.prompts import get_prompt_by_name
 from onyx.db.prompts import upsert_prompt
-from onyx.db.user_documents import upsert_user_folder
-
-
-def load_user_folders_from_yaml(
-    db_session: Session,
-    user_folders_yaml: str = USER_FOLDERS_YAML,
-) -> None:
-    with open(user_folders_yaml, "r") as file:
-        data = yaml.safe_load(file)
-
-    all_user_folders = data.get("user_folders", [])
-    for user_folder in all_user_folders:
-        upsert_user_folder(
-            db_session=db_session,
-            id=user_folder.get("id"),
-            user_id=user_folder.get("user_id"),
-            name=user_folder.get("name"),
-            description=user_folder.get("description"),
-            created_at=user_folder.get("created_at"),
-            user=user_folder.get("user"),
-            files=user_folder.get("files"),
-            assistants=user_folder.get("assistants"),
-        )
-    db_session.flush()
 
 
 def load_prompts_from_yaml(
@@ -204,4 +179,3 @@ def load_chat_yamls(
     load_prompts_from_yaml(db_session, prompt_yaml)
     load_personas_from_yaml(db_session, personas_yaml)
     load_input_prompts_from_yaml(db_session, input_prompts_yaml)
-    load_user_folders_from_yaml(db_session)

backend/onyx/seeding/user_folders.yaml

@@ -1,6 +0,0 @@
-user_folders:
-  - id: -1
-    name: "Recent Documents"
-    description: "Documents uploaded by the user"
-    files: []
-    assistants: []

backend/onyx/server/documents/cc_pair.py

@@ -620,7 +620,7 @@ def associate_credential_to_connector(
     )
 
     try:
-        validate_ccpair_for_user(connector_id, credential_id, db_session)
+        validate_ccpair_for_user(connector_id, credential_id, db_session, tenant_id)
 
         response = add_credential_to_connector(
             db_session=db_session,

backend/onyx/server/documents/connector.py

@@ -389,7 +389,12 @@ def check_drive_tokens(
     return AuthStatus(authenticated=True)
 
 
-def upload_files(files: list[UploadFile], db_session: Session) -> FileUploadResponse:
+@router.post("/admin/connector/file/upload")
+def upload_files(
+    files: list[UploadFile],
+    _: User = Depends(current_curator_or_admin_user),
+    db_session: Session = Depends(get_session),
+) -> FileUploadResponse:
     for file in files:
         if not file.filename:
             raise HTTPException(status_code=400, detail="File name cannot be empty")
@@ -450,15 +455,6 @@ def upload_files(files: list[UploadFile], db_session: Session) -> FileUploadResp
     return FileUploadResponse(file_paths=deduped_file_paths)
 
 
-@router.post("/admin/connector/file/upload")
-def upload_files_api(
-    files: list[UploadFile],
-    _: User = Depends(current_curator_or_admin_user),
-    db_session: Session = Depends(get_session),
-) -> FileUploadResponse:
-    return upload_files(files, db_session)
-
-
 @router.get("/admin/connector")
 def get_connectors_by_credential(
     _: User = Depends(current_curator_or_admin_user),
@@ -906,6 +902,7 @@ def create_connector_with_mock_credential(
             connector_id=connector_id,
             credential_id=credential_id,
             db_session=db_session,
+            tenant_id=tenant_id,
         )
         response = add_credential_to_connector(
             db_session=db_session,
@@ -1046,16 +1043,55 @@ def connector_run_once(
             status_code=400,
             detail="Connector has no valid credentials, cannot create index attempts.",
         )
-    try:
-        num_triggers = trigger_indexing_for_cc_pair(
-            credential_ids,
-            connector_id,
-            run_info.from_beginning,
-            tenant_id,
-            db_session,
+
+    # Prevents index attempts for cc pairs that already have an index attempt currently running
+    skipped_credentials = [
+        credential_id
+        for credential_id in credential_ids
+        if get_index_attempts_for_cc_pair(
+            cc_pair_identifier=ConnectorCredentialPairIdentifier(
+                connector_id=run_info.connector_id,
+                credential_id=credential_id,
+            ),
+            only_current=True,
+            db_session=db_session,
+            disinclude_finished=True,
         )
-    except ValueError as e:
-        raise HTTPException(status_code=400, detail=str(e))
+    ]
+
+    connector_credential_pairs = [
+        get_connector_credential_pair(
+            db_session=db_session,
+            connector_id=connector_id,
+            credential_id=credential_id,
+        )
+        for credential_id in credential_ids
+        if credential_id not in skipped_credentials
+    ]
+
+    num_triggers = 0
+    for cc_pair in connector_credential_pairs:
+        if cc_pair is not None:
+            indexing_mode = IndexingMode.UPDATE
+            if run_info.from_beginning:
+                indexing_mode = IndexingMode.REINDEX
+
+            mark_ccpair_with_indexing_trigger(cc_pair.id, indexing_mode, db_session)
+            num_triggers += 1
+
+            logger.info(
+                f"connector_run_once - marking cc_pair with indexing trigger: "
+                f"connector={run_info.connector_id} "
+                f"cc_pair={cc_pair.id} "
+                f"indexing_trigger={indexing_mode}"
+            )
+
+    # run the beat task to pick up the triggers immediately
+    primary_app.send_task(
+        OnyxCeleryTask.CHECK_FOR_INDEXING,
+        priority=OnyxCeleryPriority.HIGH,
+        kwargs={"tenant_id": tenant_id},
+    )
 
     logger.info("connector_run_once - running check_for_indexing")
 
@@ -1229,82 +1265,3 @@ def get_basic_connector_indexing_status(
         for cc_pair in cc_pairs
         if cc_pair.connector.source != DocumentSource.INGESTION_API
     ]
-
-
-def trigger_indexing_for_cc_pair(
-    specified_credential_ids: list[int],
-    connector_id: int,
-    from_beginning: bool,
-    tenant_id: str,
-    db_session: Session,
-) -> int:
-    try:
-        possible_credential_ids = get_connector_credential_ids(connector_id, db_session)
-    except ValueError as e:
-        raise ValueError(f"Connector by id {connector_id} does not exist: {str(e)}")
-
-    if not specified_credential_ids:
-        credential_ids = possible_credential_ids
-    else:
-        if set(specified_credential_ids).issubset(set(possible_credential_ids)):
-            credential_ids = specified_credential_ids
-        else:
-            raise ValueError(
-                "Not all specified credentials are associated with connector"
-            )
-
-    if not credential_ids:
-        raise ValueError(
-            "Connector has no valid credentials, cannot create index attempts."
-        )
-
-    # Prevents index attempts for cc pairs that already have an index attempt currently running
-    skipped_credentials = [
-        credential_id
-        for credential_id in credential_ids
-        if get_index_attempts_for_cc_pair(
-            cc_pair_identifier=ConnectorCredentialPairIdentifier(
-                connector_id=connector_id,
-                credential_id=credential_id,
-            ),
-            only_current=True,
-            db_session=db_session,
-            disinclude_finished=True,
-        )
-    ]
-
-    connector_credential_pairs = [
-        get_connector_credential_pair(
-            db_session=db_session,
-            connector_id=connector_id,
-            credential_id=credential_id,
-        )
-        for credential_id in credential_ids
-        if credential_id not in skipped_credentials
-    ]
-
-    num_triggers = 0
-    for cc_pair in connector_credential_pairs:
-        if cc_pair is not None:
-            indexing_mode = IndexingMode.UPDATE
-            if from_beginning:
-                indexing_mode = IndexingMode.REINDEX
-
-            mark_ccpair_with_indexing_trigger(cc_pair.id, indexing_mode, db_session)
-            num_triggers += 1
-
-            logger.info(
-                f"connector_run_once - marking cc_pair with indexing trigger: "
-                f"connector={connector_id} "
-                f"cc_pair={cc_pair.id} "
-                f"indexing_trigger={indexing_mode}"
-            )
-
-    # run the beat task to pick up the triggers immediately
-    primary_app.send_task(
-        OnyxCeleryTask.CHECK_FOR_INDEXING,
-        priority=OnyxCeleryPriority.HIGH,
-        kwargs={"tenant_id": tenant_id},
-    )
-
-    return num_triggers

backend/onyx/server/documents/credential.py

@@ -18,6 +18,7 @@ from onyx.db.credentials import fetch_credentials_by_source_for_user
 from onyx.db.credentials import fetch_credentials_for_user
 from onyx.db.credentials import swap_credentials_connector
 from onyx.db.credentials import update_credential
+from onyx.db.engine import get_current_tenant_id
 from onyx.db.engine import get_session
 from onyx.db.models import DocumentSource
 from onyx.db.models import User
@@ -99,11 +100,13 @@ def swap_credentials_for_connector(
     credential_swap_req: CredentialSwapRequest,
     user: User | None = Depends(current_user),
     db_session: Session = Depends(get_session),
+    tenant_id: str | None = Depends(get_current_tenant_id),
 ) -> StatusResponse:
     validate_ccpair_for_user(
         credential_swap_req.connector_id,
         credential_swap_req.new_credential_id,
         db_session,
+        tenant_id,
     )
 
     connector_credential_pair = swap_credentials_connector(

backend/onyx/server/documents/models.py

@@ -122,7 +122,6 @@ class CredentialBase(BaseModel):
     name: str | None = None
     curator_public: bool = False
     groups: list[int] = Field(default_factory=list)
-    is_user_file: bool = False
 
 
 class CredentialSnapshot(CredentialBase):
@@ -393,7 +392,7 @@ class FileUploadResponse(BaseModel):
 
 
 class ObjectCreationIdResponse(BaseModel):
-    id: int
+    id: int | str
     credential: CredentialSnapshot | None = None
 
 

backend/onyx/server/features/folder/api.py

@@ -18,9 +18,9 @@ from onyx.db.models import User
 from onyx.server.features.folder.models import DeleteFolderOptions
 from onyx.server.features.folder.models import FolderChatSessionRequest
 from onyx.server.features.folder.models import FolderCreationRequest
+from onyx.server.features.folder.models import FolderResponse
 from onyx.server.features.folder.models import FolderUpdateRequest
 from onyx.server.features.folder.models import GetUserFoldersResponse
-from onyx.server.features.folder.models import UserFolderSnapshot
 from onyx.server.models import DisplayPriorityRequest
 from onyx.server.query_and_chat.models import ChatSessionDetails
 
@@ -39,7 +39,7 @@ def get_folders(
     folders.sort()
     return GetUserFoldersResponse(
         folders=[
-            UserFolderSnapshot(
+            FolderResponse(
                 folder_id=folder.id,
                 folder_name=folder.name,
                 display_priority=folder.display_priority,

backend/onyx/server/features/folder/models.py

@@ -5,7 +5,7 @@ from pydantic import BaseModel
 from onyx.server.query_and_chat.models import ChatSessionDetails
 
 
-class UserFolderSnapshot(BaseModel):
+class FolderResponse(BaseModel):
     folder_id: int
     folder_name: str | None
     display_priority: int
@@ -13,7 +13,7 @@ class UserFolderSnapshot(BaseModel):
 
 
 class GetUserFoldersResponse(BaseModel):
-    folders: list[UserFolderSnapshot]
+    folders: list[FolderResponse]
 
 
 class FolderCreationRequest(BaseModel):

backend/onyx/server/features/persona/api.py

@@ -26,7 +26,6 @@ from onyx.db.persona import create_assistant_label
 from onyx.db.persona import create_update_persona
 from onyx.db.persona import delete_persona_label
 from onyx.db.persona import get_assistant_labels
-from shared_configs.contextvars import get_current_tenant_id
 from onyx.db.persona import get_persona_by_id
 from onyx.db.persona import get_personas_for_user
 from onyx.db.persona import mark_persona_as_deleted
@@ -56,9 +55,11 @@ from onyx.server.models import DisplayPriorityRequest
 from onyx.tools.utils import is_image_generation_available
 from onyx.utils.logger import setup_logger
 from onyx.utils.telemetry import create_milestone_and_report
+from shared_configs.contextvars import get_current_tenant_id
 
 logger = setup_logger()
 
+
 admin_router = APIRouter(prefix="/admin/persona")
 basic_router = APIRouter(prefix="/persona")
 
@@ -209,7 +210,6 @@ def create_persona(
         and len(persona_upsert_request.prompt_ids) > 0
         else None
     )
-
     prompt = upsert_prompt(
         db_session=db_session,
         user=user,

backend/onyx/server/features/persona/models.py

@@ -85,8 +85,6 @@ class PersonaUpsertRequest(BaseModel):
     label_ids: list[int] | None = None
     is_default_persona: bool = False
     display_priority: int | None = None
-    user_file_ids: list[int] | None = None
-    user_folder_ids: list[int] | None = None
 
 
 class PersonaSnapshot(BaseModel):

backend/onyx/server/manage/llm/models.py

@@ -4,7 +4,6 @@ from pydantic import BaseModel
 from pydantic import Field
 
 from onyx.llm.llm_provider_options import fetch_models_for_provider
-from onyx.llm.utils import get_max_input_tokens
 
 
 if TYPE_CHECKING:
@@ -36,36 +35,22 @@ class LLMProviderDescriptor(BaseModel):
     fast_default_model_name: str | None
     is_default_provider: bool | None
     display_model_names: list[str] | None
-    model_token_limits: dict[str, int] | None = None
 
     @classmethod
     def from_model(
         cls, llm_provider_model: "LLMProviderModel"
     ) -> "LLMProviderDescriptor":
-        model_names = (
-            llm_provider_model.model_names
-            or fetch_models_for_provider(llm_provider_model.provider)
-            or [llm_provider_model.default_model_name]
-        )
-
-        model_token_rate = (
-            {
-                model_name: get_max_input_tokens(
-                    model_name, llm_provider_model.provider
-                )
-                for model_name in model_names
-            }
-            if model_names is not None
-            else None
-        )
         return cls(
             name=llm_provider_model.name,
             provider=llm_provider_model.provider,
             default_model_name=llm_provider_model.default_model_name,
             fast_default_model_name=llm_provider_model.fast_default_model_name,
             is_default_provider=llm_provider_model.is_default_provider,
-            model_names=model_names,
-            model_token_limits=model_token_rate,
+            model_names=(
+                llm_provider_model.model_names
+                or fetch_models_for_provider(llm_provider_model.provider)
+                or [llm_provider_model.default_model_name]
+            ),
             display_model_names=llm_provider_model.display_model_names,
         )
 
@@ -95,7 +80,6 @@ class FullLLMProvider(LLMProvider):
     id: int
     is_default_provider: bool | None = None
     model_names: list[str]
-    model_token_limits: dict[str, int] | None = None
 
     @classmethod
     def from_model(cls, llm_provider_model: "LLMProviderModel") -> "FullLLMProvider":
@@ -116,14 +100,6 @@ class FullLLMProvider(LLMProvider):
                 or fetch_models_for_provider(llm_provider_model.provider)
                 or [llm_provider_model.default_model_name]
             ),
-            model_token_limits={
-                model_name: get_max_input_tokens(
-                    model_name, llm_provider_model.provider
-                )
-                for model_name in llm_provider_model.model_names
-            }
-            if llm_provider_model.model_names is not None
-            else None,
             is_public=llm_provider_model.is_public,
             groups=[group.id for group in llm_provider_model.groups],
             deployment_name=llm_provider_model.deployment_name,

backend/onyx/server/query_and_chat/chat_backend.py

@@ -3,7 +3,6 @@ import datetime
 import io
 import json
 import os
-import time
 import uuid
 from collections.abc import Callable
 from collections.abc import Generator
@@ -30,12 +29,10 @@ from onyx.chat.prompt_builder.citations_prompt import (
     compute_max_document_tokens_for_persona,
 )
 from onyx.configs.app_configs import WEB_DOMAIN
-from onyx.configs.constants import DocumentSource
 from onyx.configs.constants import FileOrigin
 from onyx.configs.constants import MessageType
 from onyx.configs.constants import MilestoneRecordType
 from onyx.configs.model_configs import LITELLM_PASS_THROUGH_HEADERS
-from onyx.connectors.models import InputType
 from onyx.db.chat import add_chats_to_session_from_slack_thread
 from onyx.db.chat import create_chat_session
 from onyx.db.chat import create_new_chat_message
@@ -50,18 +47,13 @@ from onyx.db.chat import get_or_create_root_message
 from onyx.db.chat import set_as_latest_chat_message
 from onyx.db.chat import translate_db_message_to_chat_message_detail
 from onyx.db.chat import update_chat_session
-from onyx.db.connector import create_connector
-from onyx.db.connector_credential_pair import add_credential_to_connector
-from onyx.db.credentials import create_credential
 from onyx.db.chat_search import search_chat_sessions
 from onyx.db.engine import get_session
 from onyx.db.engine import get_session_with_tenant
-from onyx.db.enums import AccessType
 from onyx.db.feedback import create_chat_message_feedback
 from onyx.db.feedback import create_doc_retrieval_feedback
 from onyx.db.models import User
 from onyx.db.persona import get_persona_by_id
-from onyx.db.user_documents import create_user_files
 from onyx.file_processing.extract_file_text import docx_to_txt_filename
 from onyx.file_processing.extract_file_text import extract_file_text
 from onyx.file_store.file_store import get_default_file_store
@@ -74,8 +66,6 @@ from onyx.natural_language_processing.utils import get_tokenizer
 from onyx.secondary_llm_flows.chat_session_naming import (
     get_renamed_conversation_name,
 )
-from onyx.server.documents.models import ConnectorBase
-from onyx.server.documents.models import CredentialBase
 from onyx.server.query_and_chat.models import ChatFeedbackRequest
 from onyx.server.query_and_chat.models import ChatMessageIdentifier
 from onyx.server.query_and_chat.models import ChatRenameRequest
@@ -101,7 +91,6 @@ from onyx.utils.logger import setup_logger
 from onyx.utils.telemetry import create_milestone_and_report
 from shared_configs.contextvars import get_current_tenant_id
 
-RECENT_DOCS_FOLDER_ID = -1
 
 logger = setup_logger()
 
@@ -658,7 +647,7 @@ def seed_chat_from_slack(
 def upload_files_for_chat(
     files: list[UploadFile],
     db_session: Session = Depends(get_session),
-    user: User | None = Depends(current_user),
+    _: User | None = Depends(current_user),
 ) -> dict[str, list[FileDescriptor]]:
     image_content_types = {"image/jpeg", "image/png", "image/webp"}
     csv_content_types = {"text/csv"}
@@ -696,11 +685,17 @@ def upload_files_for_chat(
             if file.content_type in image_content_types:
                 error_detail = "Unsupported image file type. Supported image types include .jpg, .jpeg, .png, .webp."
             elif file.content_type in text_content_types:
-                error_detail = "Unsupported text file type."
+                error_detail = "Unsupported text file type. Supported text types include .txt, .csv, .md, .mdx, .conf, "
+                ".log, .tsv."
             elif file.content_type in csv_content_types:
-                error_detail = "Unsupported CSV file type."
+                error_detail = (
+                    "Unsupported CSV file type. Supported CSV types include .csv."
+                )
             else:
-                error_detail = "Unsupported document file type."
+                error_detail = (
+                    "Unsupported document file type. Supported document types include .pdf, .docx, .pptx, .xlsx, "
+                    ".json, .xml, .yml, .yaml, .eml, .epub."
+                )
             raise HTTPException(status_code=400, detail=error_detail)
 
         if (
@@ -748,12 +743,11 @@ def upload_files_for_chat(
             file_type=new_content_type or file_type.value,
         )
 
-        # 4) If the file is a doc, extract text and store that separately
+        # if the file is a doc, extract text and store that so we don't need
+        # to re-extract it every time we send a message
         if file_type == ChatFileType.DOC:
-            # Re-wrap bytes in a fresh BytesIO so we start at position 0
-            extracted_text_io = io.BytesIO(file_content)
             extracted_text = extract_file_text(
-                file=extracted_text_io,  # use the bytes we already read
+                file=file_content_io,  # use the bytes we already read
                 file_name=file.filename or "",
             )
             text_file_id = str(uuid.uuid4())
@@ -765,57 +759,13 @@ def upload_files_for_chat(
                 file_origin=FileOrigin.CHAT_UPLOAD,
                 file_type="text/plain",
             )
-            # Return the text file as the "main" file descriptor for doc types
+            # for DOC type, just return this for the FileDescriptor
+            # as we would always use this as the ID to attach to the
+            # message
             file_info.append((text_file_id, file.filename, ChatFileType.PLAIN_TEXT))
         else:
             file_info.append((file_id, file.filename, file_type))
 
-        # 5) Create a user file for each uploaded file
-        user_files = create_user_files([file], RECENT_DOCS_FOLDER_ID, user, db_session)
-        for user_file in user_files:
-            # 6) Create connector
-            connector_base = ConnectorBase(
-                name=f"UserFile-{int(time.time())}",
-                source=DocumentSource.FILE,
-                input_type=InputType.LOAD_STATE,
-                connector_specific_config={
-                    "file_locations": [user_file.file_id],
-                },
-                refresh_freq=None,
-                prune_freq=None,
-                indexing_start=None,
-            )
-            connector = create_connector(
-                db_session=db_session,
-                connector_data=connector_base,
-            )
-
-            # 7) Create credential
-            credential_info = CredentialBase(
-                credential_json={},
-                admin_public=True,
-                source=DocumentSource.FILE,
-                curator_public=True,
-                groups=[],
-                name=f"UserFileCredential-{int(time.time())}",
-                is_user_file=True,
-            )
-            credential = create_credential(credential_info, user, db_session)
-
-            # 8) Create connector credential pair
-            cc_pair = add_credential_to_connector(
-                db_session=db_session,
-                user=user,
-                connector_id=connector.id,
-                credential_id=credential.id,
-                cc_pair_name=f"UserFileCCPair-{int(time.time())}",
-                access_type=AccessType.PRIVATE,
-                auto_sync_options=None,
-                groups=[],
-            )
-            user_file.cc_pair_id = cc_pair.data
-            db_session.commit()
-
     return {
         "files": [
             {"id": file_id, "type": file_type, "name": file_name}

backend/onyx/server/query_and_chat/models.py

@@ -92,8 +92,6 @@ class CreateChatMessageRequest(ChunkContext):
     message: str
     # Files that we should attach to this message
     file_descriptors: list[FileDescriptor]
-    user_file_ids: list[int] = []
-    user_folder_ids: list[int] = []
 
     # If no prompt provided, uses the largest prompt of the chat session
     # but really this should be explicitly specified, only in the simplified APIs is this inferred
@@ -120,7 +118,7 @@ class CreateChatMessageRequest(ChunkContext):
     # this does persist in the chat thread details
     temperature_override: float | None = None
 
-    # allow user to specify an alternate assistant
+    # allow user to specify an alternate assistnat
     alternate_assistant_id: int | None = None
 
     # This takes the priority over the prompt_override
@@ -137,8 +135,6 @@ class CreateChatMessageRequest(ChunkContext):
     # https://platform.openai.com/docs/guides/structured-outputs/introduction
     structured_response_format: dict | None = None
 
-    force_user_file_search: bool = False
-
     # If true, ignores most of the search options and uses pro search instead.
     # TODO: decide how many of the above options we want to pass through to pro search
     use_agentic_search: bool = False

backend/onyx/server/settings/models.py

@@ -5,7 +5,6 @@ from pydantic import BaseModel
 
 from onyx.configs.constants import NotificationType
 from onyx.db.models import Notification as NotificationDBModel
-from shared_configs.configs import POSTGRES_DEFAULT_SCHEMA
 
 
 class PageType(str, Enum):
@@ -55,4 +54,4 @@ class Settings(BaseModel):
 class UserSettings(Settings):
     notifications: list[Notification]
     needs_reindexing: bool
-    tenant_id: str = POSTGRES_DEFAULT_SCHEMA
+    tenant_id: str | None = None

backend/onyx/server/user_documents/api.py

@@ -1,443 +0,0 @@
-import io
-import time
-from typing import List
-
-import requests
-import sqlalchemy.exc
-from bs4 import BeautifulSoup
-from fastapi import APIRouter
-from fastapi import Depends
-from fastapi import File
-from fastapi import Form
-from fastapi import HTTPException
-from fastapi import UploadFile
-from pydantic import BaseModel
-from sqlalchemy.orm import Session
-
-from onyx.auth.users import current_user
-from onyx.configs.constants import DocumentSource
-from onyx.connectors.models import InputType
-from onyx.db.connector import create_connector
-from onyx.db.connector_credential_pair import add_credential_to_connector
-from onyx.db.credentials import create_credential
-from onyx.db.engine import get_session
-from onyx.db.enums import AccessType
-from onyx.db.models import User
-from onyx.db.models import UserFile
-from onyx.db.models import UserFolder
-from onyx.db.user_documents import create_user_files
-from onyx.db.user_documents import share_file_with_assistant
-from onyx.db.user_documents import share_folder_with_assistant
-from onyx.db.user_documents import unshare_file_with_assistant
-from onyx.db.user_documents import unshare_folder_with_assistant
-from onyx.file_processing.html_utils import web_html_cleanup
-from onyx.server.documents.models import ConnectorBase
-from onyx.server.documents.models import CredentialBase
-from onyx.server.documents.models import FileUploadResponse
-from onyx.server.user_documents.models import MessageResponse
-from onyx.server.user_documents.models import UserFileSnapshot
-from onyx.server.user_documents.models import UserFolderSnapshot
-
-router = APIRouter()
-
-
-class FolderCreationRequest(BaseModel):
-    name: str
-    description: str
-
-
-@router.post("/user/folder")
-def create_folder(
-    request: FolderCreationRequest,
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> UserFolderSnapshot:
-    try:
-        new_folder = UserFolder(
-            user_id=user.id if user else None,
-            name=request.name,
-            description=request.description,
-        )
-        db_session.add(new_folder)
-        db_session.commit()
-        return UserFolderSnapshot.from_model(new_folder)
-    except sqlalchemy.exc.DataError as e:
-        if "StringDataRightTruncation" in str(e):
-            raise HTTPException(
-                status_code=400,
-                detail="Folder name or description is too long. Please use a shorter name or description.",
-            )
-        raise
-
-
-@router.get(
-    "/user/folder",
-)
-def get_folders(
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> list[UserFolderSnapshot]:
-    user_id = user.id if user else None
-    folders = db_session.query(UserFolder).filter(UserFolder.user_id == user_id).all()
-    return [UserFolderSnapshot.from_model(folder) for folder in folders]
-
-
-@router.get("/user/folder/{folder_id}")
-def get_folder(
-    folder_id: int,
-    user: User | None = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> UserFolderSnapshot:
-    user_id = user.id if user else None
-    folder = (
-        db_session.query(UserFolder)
-        .filter(UserFolder.id == folder_id, UserFolder.user_id == user_id)
-        .first()
-    )
-    if not folder:
-        raise HTTPException(status_code=404, detail="Folder not found")
-
-    return UserFolderSnapshot.from_model(folder)
-
-
-RECENT_DOCS_FOLDER_ID = -1
-
-
-@router.post("/user/file/upload")
-def upload_user_files(
-    files: List[UploadFile] = File(...),
-    folder_id: int | None = Form(None),
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> FileUploadResponse:
-    if folder_id == 0:
-        folder_id = None
-
-    user_files = create_user_files(files, folder_id, user, db_session)
-    for user_file in user_files:
-        connector_base = ConnectorBase(
-            name=f"UserFile-{user_file.file_id}-{int(time.time())}",
-            source=DocumentSource.FILE,
-            input_type=InputType.LOAD_STATE,
-            connector_specific_config={
-                "file_locations": [user_file.file_id],
-            },
-            refresh_freq=None,
-            prune_freq=None,
-            indexing_start=None,
-        )
-
-        connector = create_connector(
-            db_session=db_session,
-            connector_data=connector_base,
-        )
-
-        credential_info = CredentialBase(
-            credential_json={},
-            admin_public=True,
-            source=DocumentSource.FILE,
-            curator_public=True,
-            groups=[],
-            name=f"UserFileCredential-{user_file.file_id}-{int(time.time())}",
-            is_user_file=True,
-        )
-        credential = create_credential(credential_info, user, db_session)
-
-        cc_pair = add_credential_to_connector(
-            db_session=db_session,
-            user=user,
-            connector_id=connector.id,
-            credential_id=credential.id,
-            cc_pair_name=f"UserFileCCPair-{user_file.file_id}-{int(time.time())}",
-            access_type=AccessType.PRIVATE,
-            auto_sync_options=None,
-            groups=[],
-            is_user_file=True,
-        )
-        user_file.cc_pair_id = cc_pair.data
-        print("A")
-        db_session.commit()
-
-    db_session.commit()
-    # TODO: functional document indexing
-    # trigger_document_indexing(db_session, user.id)
-    return FileUploadResponse(
-        file_paths=[user_file.file_id for user_file in user_files],
-    )
-
-
-@router.put("/user/folder/{folder_id}")
-def update_folder(
-    folder_id: int,
-    name: str,
-    user: User | None = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> UserFolderSnapshot:
-    user_id = user.id if user else None
-    folder = (
-        db_session.query(UserFolder)
-        .filter(UserFolder.id == folder_id, UserFolder.user_id == user_id)
-        .first()
-    )
-    if not folder:
-        raise HTTPException(status_code=404, detail="Folder not found")
-    folder.name = name
-
-    db_session.commit()
-
-    return UserFolderSnapshot.from_model(folder)
-
-
-@router.delete("/user/folder/{folder_id}")
-def delete_folder(
-    folder_id: int,
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> MessageResponse:
-    user_id = user.id if user else None
-    folder = (
-        db_session.query(UserFolder)
-        .filter(UserFolder.id == folder_id, UserFolder.user_id == user_id)
-        .first()
-    )
-    if not folder:
-        raise HTTPException(status_code=404, detail="Folder not found")
-    db_session.delete(folder)
-    db_session.commit()
-    return MessageResponse(message="Folder deleted successfully")
-
-
-@router.delete("/user/file/{file_id}")
-def delete_file(
-    file_id: int,
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> MessageResponse:
-    user_id = user.id if user else None
-    file = (
-        db_session.query(UserFile)
-        .filter(UserFile.id == file_id, UserFile.user_id == user_id)
-        .first()
-    )
-    if not file:
-        raise HTTPException(status_code=404, detail="File not found")
-    db_session.delete(file)
-    db_session.commit()
-    return MessageResponse(message="File deleted successfully")
-
-
-class FileMoveRequest(BaseModel):
-    new_folder_id: int | None
-
-
-@router.put("/user/file/{file_id}/move")
-def move_file(
-    file_id: int,
-    request: FileMoveRequest,
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> UserFileSnapshot:
-    user_id = user.id if user else None
-    file = (
-        db_session.query(UserFile)
-        .filter(UserFile.id == file_id, UserFile.user_id == user_id)
-        .first()
-    )
-    if not file:
-        raise HTTPException(status_code=404, detail="File not found")
-    file.folder_id = request.new_folder_id
-    db_session.commit()
-    return UserFileSnapshot.from_model(file)
-
-
-@router.get("/user/file-system")
-def get_file_system(
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> list[UserFolderSnapshot]:
-    user_id = user.id if user else None
-    folders = db_session.query(UserFolder).filter(UserFolder.user_id == user_id).all()
-    return [UserFolderSnapshot.from_model(folder) for folder in folders]
-
-
-@router.put("/user/file/{file_id}/rename")
-def rename_file(
-    file_id: int,
-    name: str,
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> UserFileSnapshot:
-    user_id = user.id if user else None
-    file = (
-        db_session.query(UserFile)
-        .filter(UserFile.id == file_id, UserFile.user_id == user_id)
-        .first()
-    )
-    if not file:
-        raise HTTPException(status_code=404, detail="File not found")
-    file.name = name
-    db_session.commit()
-    return UserFileSnapshot.from_model(file)
-
-
-class ShareRequest(BaseModel):
-    assistant_id: int
-
-
-@router.post("/user/file/{file_id}/share")
-def share_file(
-    file_id: int,
-    request: ShareRequest,
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> MessageResponse:
-    user_id = user.id if user else None
-    file = (
-        db_session.query(UserFile)
-        .filter(UserFile.id == file_id, UserFile.user_id == user_id)
-        .first()
-    )
-    if not file:
-        raise HTTPException(status_code=404, detail="File not found")
-
-    share_file_with_assistant(file_id, request.assistant_id, db_session)
-    return MessageResponse(message="File shared successfully with the assistant")
-
-
-@router.post("/user/file/{file_id}/unshare")
-def unshare_file(
-    file_id: int,
-    request: ShareRequest,
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> MessageResponse:
-    user_id = user.id if user else None
-    file = (
-        db_session.query(UserFile)
-        .filter(UserFile.id == file_id, UserFile.user_id == user_id)
-        .first()
-    )
-    if not file:
-        raise HTTPException(status_code=404, detail="File not found")
-
-    unshare_file_with_assistant(file_id, request.assistant_id, db_session)
-    return MessageResponse(message="File unshared successfully from the assistant")
-
-
-@router.post("/user/folder/{folder_id}/share")
-def share_folder(
-    folder_id: int,
-    request: ShareRequest,
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> MessageResponse:
-    user_id = user.id if user else None
-    folder = (
-        db_session.query(UserFolder)
-        .filter(UserFolder.id == folder_id, UserFolder.user_id == user_id)
-        .first()
-    )
-    if not folder:
-        raise HTTPException(status_code=404, detail="Folder not found")
-
-    share_folder_with_assistant(folder_id, request.assistant_id, db_session)
-    return MessageResponse(
-        message="Folder and its files shared successfully with the assistant"
-    )
-
-
-@router.post("/user/folder/{folder_id}/unshare")
-def unshare_folder(
-    folder_id: int,
-    request: ShareRequest,
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> MessageResponse:
-    user_id = user.id if user else None
-    folder = (
-        db_session.query(UserFolder)
-        .filter(UserFolder.id == folder_id, UserFolder.user_id == user_id)
-        .first()
-    )
-    if not folder:
-        raise HTTPException(status_code=404, detail="Folder not found")
-
-    unshare_folder_with_assistant(folder_id, request.assistant_id, db_session)
-    return MessageResponse(
-        message="Folder and its files unshared successfully from the assistant"
-    )
-
-
-class CreateFileFromLinkRequest(BaseModel):
-    url: str
-    folder_id: int | None
-
-
-@router.post("/user/file/create-from-link")
-def create_file_from_link(
-    request: CreateFileFromLinkRequest,
-    user: User = Depends(current_user),
-    db_session: Session = Depends(get_session),
-) -> FileUploadResponse:
-    try:
-        response = requests.get(request.url)
-        response.raise_for_status()
-        content = response.text
-        soup = BeautifulSoup(content, "html.parser")
-        parsed_html = web_html_cleanup(soup, mintlify_cleanup_enabled=False)
-
-        file_name = f"{parsed_html.title or 'Untitled'}.txt"
-        file_content = parsed_html.cleaned_text.encode()
-
-        file = UploadFile(filename=file_name, file=io.BytesIO(file_content))
-        user_files = create_user_files([file], request.folder_id, user, db_session)
-
-        # Create connector and credential (same as in upload_user_files)
-        for user_file in user_files:
-            connector_base = ConnectorBase(
-                name=f"UserFile-{user_file.file_id}-{int(time.time())}",
-                source=DocumentSource.FILE,
-                input_type=InputType.LOAD_STATE,
-                connector_specific_config={
-                    "file_locations": [user_file.file_id],
-                },
-                refresh_freq=None,
-                prune_freq=None,
-                indexing_start=None,
-            )
-
-            connector = create_connector(
-                db_session=db_session,
-                connector_data=connector_base,
-            )
-
-            credential_info = CredentialBase(
-                credential_json={},
-                admin_public=True,
-                source=DocumentSource.FILE,
-                curator_public=True,
-                groups=[],
-                name=f"UserFileCredential-{user_file.file_id}-{int(time.time())}",
-            )
-            credential = create_credential(credential_info, user, db_session)
-
-            cc_pair = add_credential_to_connector(
-                db_session=db_session,
-                user=user,
-                connector_id=connector.id,
-                credential_id=credential.id,
-                cc_pair_name=f"UserFileCCPair-{int(time.time())}",
-                access_type=AccessType.PRIVATE,
-                auto_sync_options=None,
-                groups=[],
-                is_user_file=True,
-            )
-            user_file.cc_pair_id = cc_pair.data
-            db_session.commit()
-
-        db_session.commit()
-        return FileUploadResponse(
-            file_paths=[user_file.file_id for user_file in user_files]
-        )
-    except requests.RequestException as e:
-        raise HTTPException(status_code=400, detail=f"Failed to fetch URL: {str(e)}")

backend/onyx/server/user_documents/models.py

@@ -1,70 +0,0 @@
-from datetime import datetime
-from typing import List
-
-from pydantic import BaseModel
-
-from onyx.db.models import UserFile
-from onyx.db.models import UserFolder
-
-
-class UserFileSnapshot(BaseModel):
-    id: int
-    name: str
-    document_id: str
-    folder_id: int | None = None
-    user_id: int | None
-    file_id: str
-    created_at: datetime
-    assistant_ids: List[int] = []  # List of assistant IDs
-    token_count: int | None
-    indexed: bool
-
-    @classmethod
-    def from_model(cls, model: UserFile) -> "UserFileSnapshot":
-        return cls(
-            id=model.id,
-            name=model.name,
-            folder_id=model.folder_id,
-            document_id=model.document_id,
-            user_id=model.user_id,
-            file_id=model.file_id,
-            created_at=model.created_at,
-            assistant_ids=[assistant.id for assistant in model.assistants],
-            token_count=model.token_count,
-            indexed=model.cc_pair.last_successful_index_time is not None
-            if model.cc_pair
-            else False,
-        )
-
-
-class UserFolderSnapshot(BaseModel):
-    id: int
-    name: str
-    description: str
-    files: List[UserFileSnapshot]
-    created_at: datetime
-    user_id: int | None
-    assistant_ids: List[int] = []  # List of assistant IDs
-    token_count: int | None
-
-    @classmethod
-    def from_model(cls, model: UserFolder) -> "UserFolderSnapshot":
-        return cls(
-            id=model.id,
-            name=model.name,
-            description=model.description,
-            files=[UserFileSnapshot.from_model(file) for file in model.files],
-            created_at=model.created_at,
-            user_id=model.user_id,
-            assistant_ids=[assistant.id for assistant in model.assistants],
-            token_count=sum(file.token_count or 0 for file in model.files) or None,
-        )
-
-
-class MessageResponse(BaseModel):
-    message: str
-
-
-class FileSystemResponse(BaseModel):
-    folders: list[UserFolderSnapshot]
-    files: list[UserFileSnapshot]

backend/onyx/setup.py

@@ -65,7 +65,7 @@ logger = setup_logger()
 
 
 def setup_onyx(
-    db_session: Session, tenant_id: str, cohere_enabled: bool = False
+    db_session: Session, tenant_id: str | None, cohere_enabled: bool = False
 ) -> None:
     """
     Setup Onyx for a particular tenant. In the Single Tenant case, it will set it up for the default schema

backend/onyx/tools/tool_constructor.py

@@ -138,7 +138,6 @@ def construct_tools(
     user: User | None,
     llm: LLM,
     fast_llm: LLM,
-    use_file_search: bool,
     search_tool_config: SearchToolConfig | None = None,
     internet_search_tool_config: InternetSearchToolConfig | None = None,
     image_generation_tool_config: ImageGenerationToolConfig | None = None,
@@ -252,33 +251,6 @@ def construct_tools(
     for tool_list in tool_dict.values():
         tools.extend(tool_list)
 
-    if use_file_search:
-        search_tool_config = SearchToolConfig()
-
-        search_tool = SearchTool(
-            db_session=db_session,
-            user=user,
-            persona=persona,
-            retrieval_options=search_tool_config.retrieval_options,
-            prompt_config=prompt_config,
-            llm=llm,
-            fast_llm=fast_llm,
-            pruning_config=search_tool_config.document_pruning_config,
-            answer_style_config=search_tool_config.answer_style_config,
-            selected_sections=search_tool_config.selected_sections,
-            chunks_above=search_tool_config.chunks_above,
-            chunks_below=search_tool_config.chunks_below,
-            full_doc=search_tool_config.full_doc,
-            evaluation_type=(
-                LLMEvaluationType.BASIC
-                if persona.llm_relevance_filter
-                else LLMEvaluationType.SKIP
-            ),
-            rerank_settings=search_tool_config.rerank_settings,
-            bypass_acl=search_tool_config.bypass_acl,
-        )
-        tool_dict[1] = [search_tool]
-
     # factor in tool definition size when pruning
     if search_tool_config:
         search_tool_config.document_pruning_config.tool_num_tokens = (

backend/onyx/tools/tool_implementations/custom/custom_tool.py

@@ -64,7 +64,7 @@ logger = setup_logger()
 CUSTOM_TOOL_RESPONSE_ID = "custom_tool_response"
 
 
-class CustomToolUserFileSnapshot(BaseModel):
+class CustomToolFileResponse(BaseModel):
     file_ids: List[str]  # References to saved images or CSVs
 
 
@@ -131,7 +131,7 @@ class CustomTool(BaseTool):
         response = cast(CustomToolCallSummary, args[0].response)
 
         if response.response_type == "image" or response.response_type == "csv":
-            image_response = cast(CustomToolUserFileSnapshot, response.tool_result)
+            image_response = cast(CustomToolFileResponse, response.tool_result)
             return json.dumps({"file_ids": image_response.file_ids})
 
         # For JSON or other responses, return as-is
@@ -267,14 +267,14 @@ class CustomTool(BaseTool):
             file_ids = self._save_and_get_file_references(
                 response.content, content_type
             )
-            tool_result = CustomToolUserFileSnapshot(file_ids=file_ids)
+            tool_result = CustomToolFileResponse(file_ids=file_ids)
             response_type = "csv"
 
         elif "image/" in content_type:
             file_ids = self._save_and_get_file_references(
                 response.content, content_type
             )
-            tool_result = CustomToolUserFileSnapshot(file_ids=file_ids)
+            tool_result = CustomToolFileResponse(file_ids=file_ids)
             response_type = "image"
 
         else:
@@ -358,7 +358,7 @@ class CustomTool(BaseTool):
 
     def final_result(self, *args: ToolResponse) -> JSON_ro:
         response = cast(CustomToolCallSummary, args[0].response)
-        if isinstance(response.tool_result, CustomToolUserFileSnapshot):
+        if isinstance(response.tool_result, CustomToolFileResponse):
             return response.tool_result.model_dump()
         return response.tool_result
 

backend/scripts/debugging/onyx_vespa.py

@@ -260,7 +260,7 @@ def get_documents_for_tenant_connector(
 def search_for_document(
     index_name: str,
     document_id: str | None = None,
-    tenant_id: str = POSTGRES_DEFAULT_SCHEMA,
+    tenant_id: str | None = None,
     max_hits: int | None = 10,
 ) -> List[Dict[str, Any]]:
     yql_query = f"select * from sources {index_name}"
@@ -444,15 +444,12 @@ def get_document_acls(
         response = vespa_client.get(document_url)
         if response.status_code == 200:
             fields = response.json().get("fields", {})
-
             document_id = fields.get("document_id") or fields.get(
                 "documentid", "Unknown"
             )
             acls = fields.get("access_control_list", {})
             title = fields.get("title", "")
             source_type = fields.get("source_type", "")
-            doc_sets = fields.get("document_sets", [])
-            user_file = fields.get("user_file", None)
             source_links_raw = fields.get("source_links", "{}")
             try:
                 source_links = json.loads(source_links_raw)
@@ -465,8 +462,6 @@ def get_document_acls(
             print(f"Source Links: {source_links}")
             print(f"Title: {title}")
             print(f"Source Type: {source_type}")
-            print(f"Document Sets: {doc_sets}")
-            print(f"User File: {user_file}")
             if MULTI_TENANT:
                 print(f"Tenant ID: {fields.get('tenant_id', 'N/A')}")
             print("-" * 80)
@@ -512,9 +507,9 @@ def get_number_of_chunks_we_think_exist(
 
 class VespaDebugging:
     # Class for managing Vespa debugging actions.
-    def __init__(self, tenant_id: str = POSTGRES_DEFAULT_SCHEMA):
+    def __init__(self, tenant_id: str | None = None):
         CURRENT_TENANT_ID_CONTEXTVAR.set(tenant_id)
-        self.tenant_id = tenant_id
+        self.tenant_id = POSTGRES_DEFAULT_SCHEMA if not tenant_id else tenant_id
         self.index_name = get_index_name(self.tenant_id)
 
     def sample_document_counts(self) -> None:
@@ -608,7 +603,7 @@ class VespaDebugging:
         delete_documents_for_tenant(self.index_name, self.tenant_id, count=count)
 
     def search_for_document(
-        self, document_id: str | None = None, tenant_id: str = POSTGRES_DEFAULT_SCHEMA
+        self, document_id: str | None = None, tenant_id: str | None = None
     ) -> List[Dict[str, Any]]:
         return search_for_document(self.index_name, document_id, tenant_id)
 

backend/scripts/force_delete_connector_by_id.py

@@ -8,7 +8,6 @@ from sqlalchemy.orm import Session
 from onyx.db.document import delete_documents_complete__no_commit
 from onyx.db.enums import ConnectorCredentialPairStatus
 from onyx.db.search_settings import get_active_search_settings
-from shared_configs.configs import POSTGRES_DEFAULT_SCHEMA
 
 # Modify sys.path
 current_dir = os.path.dirname(os.path.abspath(__file__))
@@ -75,7 +74,7 @@ def _unsafe_deletion(
         for document in documents:
             document_index.delete_single(
                 doc_id=document.id,
-                tenant_id=POSTGRES_DEFAULT_SCHEMA,
+                tenant_id=None,
                 chunk_count=document.chunk_count,
             )
 

backend/scripts/orphan_doc_cleanup_script.py

@@ -6,7 +6,6 @@ from sqlalchemy import text
 from sqlalchemy.orm import Session
 
 from onyx.document_index.document_index_utils import get_multipass_config
-from shared_configs.configs import POSTGRES_DEFAULT_SCHEMA
 
 # makes it so `PYTHONPATH=.` is not required when running this script
 parent_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
@@ -97,9 +96,7 @@ def main() -> None:
                     try:
                         print(f"Deleting document {doc_id} in Vespa")
                         chunks_deleted = vespa_index.delete_single(
-                            doc_id,
-                            tenant_id=POSTGRES_DEFAULT_SCHEMA,
-                            chunk_count=document.chunk_count,
+                            doc_id, tenant_id=None, chunk_count=document.chunk_count
                         )
                         if chunks_deleted > 0:
                             print(

backend/shared_configs/contextvars.py

@@ -18,7 +18,5 @@ CURRENT_TENANT_ID_CONTEXTVAR: contextvars.ContextVar[
 def get_current_tenant_id() -> str:
     tenant_id = CURRENT_TENANT_ID_CONTEXTVAR.get()
     if tenant_id is None:
-        if not MULTI_TENANT:
-            return POSTGRES_DEFAULT_SCHEMA
         raise RuntimeError("Tenant ID is not set. This should never happen.")
     return tenant_id

backend/tests/daily/connectors/confluence/test_confluence_basic.py

@@ -87,7 +87,7 @@ def test_confluence_connector_basic(
     assert len(txt_doc.sections) == 1
     assert txt_doc.sections[0].text == "small"
     assert txt_doc.primary_owners
-    assert txt_doc.primary_owners[0].email == "chris@onyx.app"
+    assert txt_doc.primary_owners[0].email == "chris@danswer.ai"
     assert (
         txt_doc.sections[0].link
         == "https://danswerai.atlassian.net/wiki/pages/viewpageattachments.action?pageId=52494430&preview=%2F52494430%2F52527123%2Fsmall-file.txt"

deployment/docker_compose/docker-compose.prod.yml

@@ -3,7 +3,7 @@ services:
     image: onyxdotapp/onyx-backend:${IMAGE_TAG:-latest}
     build:
       context: ../../backend
-      is dockerfile: Dockerfile
+      dockerfile: Dockerfile
     command: >
       /bin/sh -c "
       alembic upgrade head &&

tsconfig.json

@@ -1,20 +0,0 @@
-{
-  "compilerOptions": {
-    "target": "es5",
-    "lib": ["dom", "dom.iterable", "esnext"],
-    "allowJs": true,
-    "skipLibCheck": true,
-    "strict": true,
-    "forceConsistentCasingInFileNames": true,
-    "noEmit": true,
-    "esModuleInterop": true,
-    "module": "esnext",
-    "moduleResolution": "node",
-    "resolveJsonModule": true,
-    "isolatedModules": true,
-    "jsx": "preserve",
-    "incremental": true
-  },
-  "include": ["next-env.d.ts", "**/*.ts", "**/*.tsx"],
-  "exclude": ["node_modules"]
-} 

web/package.json

@@ -20,13 +20,11 @@
     "@phosphor-icons/react": "^2.0.8",
     "@radix-ui/react-accordion": "^1.2.2",
     "@radix-ui/react-checkbox": "^1.1.2",
-    "@radix-ui/react-context-menu": "^2.2.5",
     "@radix-ui/react-collapsible": "^1.1.2",
     "@radix-ui/react-dialog": "^1.1.6",
     "@radix-ui/react-dropdown-menu": "^2.1.6",
     "@radix-ui/react-label": "^2.1.1",
     "@radix-ui/react-popover": "^1.1.6",
-    "@radix-ui/react-progress": "^1.1.1",
     "@radix-ui/react-radio-group": "^1.2.2",
     "@radix-ui/react-scroll-area": "^1.2.2",
     "@radix-ui/react-select": "^2.1.6",
@@ -75,8 +73,6 @@
     "recharts": "^2.13.1",
     "rehype-katex": "^7.0.1",
     "rehype-prism-plus": "^2.0.0",
-    "rehype-sanitize": "^6.0.0",
-    "rehype-stringify": "^10.0.1",
     "remark-gfm": "^4.0.0",
     "remark-math": "^6.0.0",
     "semver": "^7.5.4",

web/src/app/admin/assistants/AssistantEditor.tsx

@@ -64,10 +64,10 @@ import { debounce } from "lodash";
 import { FullLLMProvider } from "../configuration/llm/interfaces";
 import StarterMessagesList from "./StarterMessageList";
 
-import { SwitchField } from "@/components/ui/switch";
+import { Switch, SwitchField } from "@/components/ui/switch";
 import { generateIdenticon } from "@/components/assistants/AssistantIcon";
 import { BackButton } from "@/components/BackButton";
-import { Checkbox } from "@/components/ui/checkbox";
+import { Checkbox, CheckboxField } from "@/components/ui/checkbox";
 import { AdvancedOptionsToggle } from "@/components/AdvancedOptionsToggle";
 import { MinimalUserSnapshot } from "@/lib/types";
 import { useUserGroups } from "@/lib/hooks";
@@ -76,26 +76,12 @@ import {
   Option as DropdownOption,
 } from "@/components/Dropdown";
 import { SourceChip } from "@/app/chat/input/ChatInputBar";
-import {
-  TagIcon,
-  UserIcon,
-  FileIcon,
-  FolderIcon,
-  InfoIcon,
-} from "lucide-react";
+import { TagIcon, UserIcon, XIcon, InfoIcon } from "lucide-react";
 import { LLMSelector } from "@/components/llm/LLMSelector";
 import useSWR from "swr";
 import { errorHandlingFetcher } from "@/lib/fetcher";
 import { ConfirmEntityModal } from "@/components/modals/ConfirmEntityModal";
-
-import { FilePickerModal } from "@/app/chat/my-documents/components/FilePicker";
-import { useDocumentsContext } from "@/app/chat/my-documents/DocumentsContext";
-import {
-  FileResponse,
-  FolderResponse,
-} from "@/app/chat/my-documents/DocumentsContext";
-import { RadioGroup } from "@/components/ui/radio-group";
-import { RadioGroupItemField } from "@/components/ui/RadioGroupItemField";
+import Title from "@/components/ui/title";
 import { SEARCH_TOOL_ID } from "@/app/chat/tools/constants";
 
 function findSearchTool(tools: ToolSnapshot[]) {
@@ -161,7 +147,6 @@ export function AssistantEditor({
     "#6FFFFF",
   ];
 
-  const [filePickerModalOpen, setFilePickerModalOpen] = useState(false);
   const [showAdvancedOptions, setShowAdvancedOptions] = useState(false);
 
   // state to persist across formik reformatting
@@ -236,16 +221,6 @@ export function AssistantEditor({
     enabledToolsMap[tool.id] = personaCurrentToolIds.includes(tool.id);
   });
 
-  const {
-    selectedFiles,
-    selectedFolders,
-    addSelectedFile,
-    removeSelectedFile,
-    addSelectedFolder,
-    removeSelectedFolder,
-    clearSelectedItems,
-  } = useDocumentsContext();
-
   const [showVisibilityWarning, setShowVisibilityWarning] = useState(false);
 
   const initialValues = {
@@ -284,9 +259,6 @@ export function AssistantEditor({
         (u) => u.id !== existingPersona.owner?.id
       ) ?? [],
     selectedGroups: existingPersona?.groups ?? [],
-    user_file_ids: existingPersona?.user_file_ids ?? [],
-    user_folder_ids: existingPersona?.user_folder_ids ?? [],
-    knowledge_source: "user_files",
     is_default_persona: existingPersona?.is_default_persona ?? false,
   };
 
@@ -396,24 +368,6 @@ export function AssistantEditor({
           <BackButton />
         </div>
       )}
-      {filePickerModalOpen && (
-        <FilePickerModal
-          selectedFiles={selectedFiles}
-          selectedFolders={selectedFolders}
-          addSelectedFile={addSelectedFile}
-          removeSelectedFile={removeSelectedFile}
-          addSelectedFolder={addSelectedFolder}
-          isOpen={filePickerModalOpen}
-          onClose={() => {
-            setFilePickerModalOpen(false);
-          }}
-          onSave={() => {
-            setFilePickerModalOpen(false);
-          }}
-          title="Add Documents to your Assistant"
-          buttonContent="Add to Assistant"
-        />
-      )}
 
       {labelToDelete && (
         <ConfirmEntityModal
@@ -480,7 +434,6 @@ export function AssistantEditor({
             label_ids: Yup.array().of(Yup.number()),
             selectedUsers: Yup.array().of(Yup.object()),
             selectedGroups: Yup.array().of(Yup.number()),
-            knowledge_source: Yup.string().required(),
             is_default_persona: Yup.boolean().required(),
           })
           .test(
@@ -569,12 +522,9 @@ export function AssistantEditor({
               ? new Date(values.search_start_date)
               : null,
             num_chunks: numChunks,
-            user_file_ids: selectedFiles.map((file) => file.id),
-            user_folder_ids: selectedFolders.map((folder) => folder.id),
           };
 
           let personaResponse;
-
           if (isUpdate) {
             personaResponse = await updatePersona(
               existingPersona.id,
@@ -896,168 +846,77 @@ export function AssistantEditor({
                     values.enabled_tools_map[searchTool.id] &&
                     !(user?.role != "admin" && documentSets.length === 0) && (
                       <CollapsibleSection>
-                        <div>
-                          <Label>Knowledge Source</Label>
-                          <RadioGroup
-                            className="flex flex-col gap-y-4 mt-2"
-                            value={values.knowledge_source}
-                            onValueChange={(value: string) => {
-                              setFieldValue("knowledge_source", value);
-                            }}
-                          >
-                            <RadioGroupItemField
-                              value="user_files"
-                              id="user_files"
-                              label="User Files"
-                              sublabel="Select specific user files and folders for this Assistant to use"
-                            />
-                            <RadioGroupItemField
-                              value="team_knowledge"
-                              id="team_knowledge"
-                              label="Team Knowledge"
-                              sublabel="Use team-wide document sets for this Assistant"
-                            />
-                          </RadioGroup>
-
-                          {values.knowledge_source === "user_files" &&
-                            !existingPersona?.is_default_persona &&
-                            !admin && (
-                              <div className="mt-4">
-                                <div className="flex justify-start gap-x-2 items-center">
-                                  <Label>User Files</Label>
-                                  <span
-                                    className="cursor-pointer text-xs text-primary hover:underline"
-                                    onClick={() => setFilePickerModalOpen(true)}
-                                  >
-                                    Attach Files and Folders
-                                  </span>
-                                </div>
-
+                        <div className="mt-2">
+                          {ccPairs.length > 0 && (
+                            <>
+                              <Label small>Document Sets</Label>
+                              <div>
                                 <SubLabel>
-                                  Select which of your user files and folders
-                                  this Assistant should use to inform its
-                                  responses. If none are specified, the
-                                  Assistant will not have access to any
-                                  user-specific documents.
+                                  <>
+                                    Select which{" "}
+                                    {!user || user.role === "admin" ? (
+                                      <Link
+                                        href="/admin/documents/sets"
+                                        className="font-semibold underline hover:underline text-text"
+                                        target="_blank"
+                                      >
+                                        Document Sets
+                                      </Link>
+                                    ) : (
+                                      "Document Sets"
+                                    )}{" "}
+                                    this Assistant should use to inform its
+                                    responses. If none are specified, the
+                                    Assistant will reference all available
+                                    documents.
+                                  </>
                                 </SubLabel>
-
-                                <div className="mt-2 mb-4">
-                                  <h4 className="text-xs font-normal mb-2">
-                                    Selected Files and Folders
-                                  </h4>
-                                  <div className="flex flex-wrap gap-2">
-                                    {selectedFiles.map((file: FileResponse) => (
-                                      <SourceChip
-                                        key={file.id}
-                                        onRemove={() => {
-                                          removeSelectedFile(file);
-                                          setFieldValue(
-                                            "selectedFiles",
-                                            values.selectedFiles.filter(
-                                              (f: FileResponse) =>
-                                                f.id !== file.id
-                                            )
-                                          );
-                                        }}
-                                        title={file.name}
-                                        icon={<FileIcon size={12} />}
-                                      />
-                                    ))}
-                                    {selectedFolders.map(
-                                      (folder: FolderResponse) => (
-                                        <SourceChip
-                                          key={folder.id}
-                                          onRemove={() => {
-                                            removeSelectedFolder(folder);
-                                            setFieldValue(
-                                              "selectedFolders",
-                                              values.selectedFolders.filter(
-                                                (f: FolderResponse) =>
-                                                  f.id !== folder.id
-                                              )
-                                            );
-                                          }}
-                                          title={folder.name}
-                                          icon={<FolderIcon size={12} />}
-                                        />
-                                      )
-                                    )}
-                                  </div>
-                                </div>
                               </div>
-                            )}
 
-                          {values.knowledge_source === "team_knowledge" &&
-                            ccPairs.length > 0 && (
-                              <div className="mt-4">
-                                <Label>Team Knowledge</Label>
-                                <div>
-                                  <SubLabel>
-                                    <>
-                                      Select which{" "}
-                                      {!user || user.role === "admin" ? (
-                                        <Link
-                                          href="/admin/documents/sets"
-                                          className="font-semibold underline hover:underline text-text"
-                                          target="_blank"
-                                        >
-                                          Team Document Sets
-                                        </Link>
-                                      ) : (
-                                        "Team Document Sets"
-                                      )}{" "}
-                                      this Assistant should use to inform its
-                                      responses. If none are specified, the
-                                      Assistant will reference all available
-                                      documents.
-                                    </>
-                                  </SubLabel>
-                                </div>
-
-                                {documentSets.length > 0 ? (
-                                  <FieldArray
-                                    name="document_set_ids"
-                                    render={(arrayHelpers: ArrayHelpers) => (
-                                      <div>
-                                        <div className="mb-3 mt-2 flex gap-2 flex-wrap text-sm">
-                                          {documentSets.map((documentSet) => (
-                                            <DocumentSetSelectable
-                                              key={documentSet.id}
-                                              documentSet={documentSet}
-                                              isSelected={values.document_set_ids.includes(
-                                                documentSet.id
-                                              )}
-                                              onSelect={() => {
-                                                const index =
-                                                  values.document_set_ids.indexOf(
-                                                    documentSet.id
-                                                  );
-                                                if (index !== -1) {
-                                                  arrayHelpers.remove(index);
-                                                } else {
-                                                  arrayHelpers.push(
-                                                    documentSet.id
-                                                  );
-                                                }
-                                              }}
-                                            />
-                                          ))}
-                                        </div>
+                              {documentSets.length > 0 ? (
+                                <FieldArray
+                                  name="document_set_ids"
+                                  render={(arrayHelpers: ArrayHelpers) => (
+                                    <div>
+                                      <div className="mb-3 mt-2 flex gap-2 flex-wrap text-sm">
+                                        {documentSets.map((documentSet) => (
+                                          <DocumentSetSelectable
+                                            key={documentSet.id}
+                                            documentSet={documentSet}
+                                            isSelected={values.document_set_ids.includes(
+                                              documentSet.id
+                                            )}
+                                            onSelect={() => {
+                                              const index =
+                                                values.document_set_ids.indexOf(
+                                                  documentSet.id
+                                                );
+                                              if (index !== -1) {
+                                                arrayHelpers.remove(index);
+                                              } else {
+                                                arrayHelpers.push(
+                                                  documentSet.id
+                                                );
+                                              }
+                                            }}
+                                          />
+                                        ))}
                                       </div>
-                                    )}
-                                  />
-                                ) : (
-                                  <p className="text-sm">
-                                    <Link
-                                      href="/admin/documents/sets/new"
-                                      className="text-primary hover:underline"
-                                    >
-                                      + Create Document Set
-                                    </Link>
-                                  </p>
-                                )}
-                              </div>
-                            )}
+                                    </div>
+                                  )}
+                                />
+                              ) : (
+                                <p className="text-sm">
+                                  <Link
+                                    href="/admin/documents/sets/new"
+                                    className="text-primary hover:underline"
+                                  >
+                                    + Create Document Set
+                                  </Link>
+                                </p>
+                              )}
+                            </>
+                          )}
                         </div>
                       </CollapsibleSection>
                     )}

web/src/app/admin/assistants/assistantFileUtils.ts

@@ -1,106 +0,0 @@
-import {
-  FileResponse,
-  FolderResponse,
-} from "@/app/chat/my-documents/DocumentsContext";
-
-export interface AssistantFileChanges {
-  filesToShare: number[];
-  filesToUnshare: number[];
-  foldersToShare: number[];
-  foldersToUnshare: number[];
-}
-
-export function calculateFileChanges(
-  existingFileIds: number[],
-  existingFolderIds: number[],
-  selectedFiles: FileResponse[],
-  selectedFolders: FolderResponse[]
-): AssistantFileChanges {
-  const selectedFileIds = selectedFiles.map((file) => file.id);
-  const selectedFolderIds = selectedFolders.map((folder) => folder.id);
-
-  return {
-    filesToShare: selectedFileIds.filter((id) => !existingFileIds.includes(id)),
-    filesToUnshare: existingFileIds.filter(
-      (id) => !selectedFileIds.includes(id)
-    ),
-    foldersToShare: selectedFolderIds.filter(
-      (id) => !existingFolderIds.includes(id)
-    ),
-    foldersToUnshare: existingFolderIds.filter(
-      (id) => !selectedFolderIds.includes(id)
-    ),
-  };
-}
-
-export async function shareFiles(
-  assistantId: number,
-  fileIds: number[]
-): Promise<void> {
-  for (const fileId of fileIds) {
-    await fetch(`/api/user/file/${fileId}/share`, {
-      method: "POST",
-      headers: {
-        "Content-Type": "application/json",
-      },
-      body: JSON.stringify({ assistant_id: assistantId }),
-    });
-  }
-}
-
-export async function unshareFiles(
-  assistantId: number,
-  fileIds: number[]
-): Promise<void> {
-  for (const fileId of fileIds) {
-    await fetch(`/api/user/file/${fileId}/unshare`, {
-      method: "POST",
-      headers: {
-        "Content-Type": "application/json",
-      },
-      body: JSON.stringify({ assistant_id: assistantId }),
-    });
-  }
-}
-
-export async function shareFolders(
-  assistantId: number,
-  folderIds: number[]
-): Promise<void> {
-  for (const folderId of folderIds) {
-    await fetch(`/api/user/folder/${folderId}/share`, {
-      method: "POST",
-      headers: {
-        "Content-Type": "application/json",
-      },
-      body: JSON.stringify({ assistant_id: assistantId }),
-    });
-  }
-}
-
-export async function unshareFolders(
-  assistantId: number,
-  folderIds: number[]
-): Promise<void> {
-  for (const folderId of folderIds) {
-    await fetch(`/api/user/folder/${folderId}/unshare`, {
-      method: "POST",
-      headers: {
-        "Content-Type": "application/json",
-      },
-      body: JSON.stringify({ assistant_id: assistantId }),
-    });
-  }
-}
-
-export async function updateAssistantFiles(
-  assistantId: number,
-  changes: AssistantFileChanges
-): Promise<void> {
-  await Promise.all([
-    shareFiles(assistantId, changes.filesToShare),
-    unshareFiles(assistantId, changes.filesToUnshare),
-    shareFolders(assistantId, changes.foldersToShare),
-    unshareFolders(assistantId, changes.foldersToUnshare),
-  ]);
-}