type fixes

Author: Subash-Mohan

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

@@ -28,9 +28,6 @@
 from onyx.db.connector_credential_pair import (
     delete_connector_credential_pair__no_commit,
 )
-from onyx.db.connector_credential_pair import (
-    delete_userfiles_for_cc_pair__no_commit,
-)
 from onyx.db.connector_credential_pair import get_connector_credential_pair_from_id
 from onyx.db.connector_credential_pair import get_connector_credential_pairs
 from onyx.db.document import (
@@ -484,12 +481,6 @@ def monitor_connector_deletion_taskset(
             # related to the deleted DocumentByConnectorCredentialPair during commit
             db_session.expire(cc_pair)
 
-            # delete all userfiles for the cc_pair
-            delete_userfiles_for_cc_pair__no_commit(
-                db_session=db_session,
-                cc_pair_id=cc_pair_id,
-            )
-
             # finally, delete the cc-pair
             delete_connector_credential_pair__no_commit(
                 db_session=db_session,

backend/onyx/db/chat.py

@@ -33,7 +33,6 @@
 from onyx.auth.schemas import UserRole
 from onyx.chat.models import DocumentRelevance
 from onyx.configs.chat_configs import HARD_DELETE_CHATS
-from onyx.configs.constants import DocumentSource
 from onyx.configs.constants import MessageType
 from onyx.context.search.models import InferenceSection
 from onyx.context.search.models import RetrievalDocs
@@ -53,12 +52,10 @@
 from onyx.db.models import SearchDoc as DBSearchDoc
 from onyx.db.models import ToolCall
 from onyx.db.models import User
-from onyx.db.models import UserFile
 from onyx.db.persona import get_best_persona_id_for_user
 from onyx.db.tools import get_tool_by_id
 from onyx.file_store.file_store import get_default_file_store
 from onyx.file_store.models import FileDescriptor
-from onyx.file_store.models import InMemoryChatFile
 from onyx.llm.override_models import LLMOverride
 from onyx.llm.override_models import PromptOverride
 from onyx.server.query_and_chat.models import ChatMessageDetail
@@ -1193,90 +1190,6 @@ def get_db_search_doc_by_document_id(
     return search_doc
 
 
-def create_search_doc_from_user_file(
-    db_user_file: UserFile, associated_chat_file: InMemoryChatFile, db_session: Session
-) -> SearchDoc:
-    """Create a SearchDoc in the database from a UserFile and return it.
-    This ensures proper ID generation by SQLAlchemy and prevents duplicate key errors.
-    """
-    blurb = ""
-    if associated_chat_file and associated_chat_file.content:
-        try:
-            # Try to decode as UTF-8, but handle errors gracefully
-            content_sample = associated_chat_file.content[:100]
-            # Remove null bytes which can cause SQL errors
-            content_sample = content_sample.replace(b"\x00", b"")
-
-            # NOTE(rkuo): this used to be "replace" instead of strict, but
-            # that would bypass the binary handling below
-            blurb = content_sample.decode("utf-8", errors="strict")
-        except Exception:
-            # If decoding fails completely, provide a generic description
-            blurb = f"[Binary file: {db_user_file.name}]"
-
-    db_search_doc = SearchDoc(
-        document_id=db_user_file.document_id,
-        chunk_ind=0,  # Default to 0 for user files
-        semantic_id=db_user_file.name,
-        link=db_user_file.link_url,
-        blurb=blurb,
-        source_type=DocumentSource.FILE,  # Assuming internal source for user files
-        boost=0,  # Default boost
-        hidden=False,  # Default visibility
-        doc_metadata={},  # Empty metadata
-        score=0.0,  # Default score of 0.0 instead of None
-        is_relevant=None,  # No relevance initially
-        relevance_explanation=None,  # No explanation initially
-        match_highlights=[],  # No highlights initially
-        updated_at=db_user_file.created_at,  # Use created_at as updated_at
-        primary_owners=[],  # Empty list instead of None
-        secondary_owners=[],  # Empty list instead of None
-        is_internet=False,  # Not from internet
-    )
-
-    db_session.add(db_search_doc)
-    db_session.flush()  # Get the ID but don't commit yet
-
-    return db_search_doc
-
-
-def translate_db_user_file_to_search_doc(
-    db_user_file: UserFile, associated_chat_file: InMemoryChatFile
-) -> SearchDoc:
-    blurb = ""
-    if associated_chat_file and associated_chat_file.content:
-        try:
-            # Try to decode as UTF-8, but handle errors gracefully
-            content_sample = associated_chat_file.content[:100]
-            # Remove null bytes which can cause SQL errors
-            content_sample = content_sample.replace(b"\x00", b"")
-            blurb = content_sample.decode("utf-8", errors="replace")
-        except Exception:
-            # If decoding fails completely, provide a generic description
-            blurb = f"[Binary file: {db_user_file.name}]"
-
-    return SearchDoc(
-        # Don't set ID - let SQLAlchemy auto-generate it
-        document_id=db_user_file.document_id,
-        chunk_ind=0,  # Default to 0 for user files
-        semantic_id=db_user_file.name,
-        link=db_user_file.link_url,
-        blurb=blurb,
-        source_type=DocumentSource.FILE,  # Assuming internal source for user files
-        boost=0,  # Default boost
-        hidden=False,  # Default visibility
-        doc_metadata={},  # Empty metadata
-        score=0.0,  # Default score of 0.0 instead of None
-        is_relevant=None,  # No relevance initially
-        relevance_explanation=None,  # No explanation initially
-        match_highlights=[],  # No highlights initially
-        updated_at=db_user_file.created_at,  # Use created_at as updated_at
-        primary_owners=[],  # Empty list instead of None
-        secondary_owners=[],  # Empty list instead of None
-        is_internet=False,  # Not from internet
-    )
-
-
 def translate_db_search_doc_to_server_search_doc(
     db_search_doc: SearchDoc,
     remove_doc_content: bool = False,

backend/onyx/db/connector_credential_pair.py

@@ -34,7 +34,6 @@
 from onyx.db.models import SearchSettings
 from onyx.db.models import User
 from onyx.db.models import User__UserGroup
-from onyx.db.models import UserFile
 from onyx.db.models import UserGroup__ConnectorCredentialPair
 from onyx.db.models import UserRole
 from onyx.server.models import StatusResponse
@@ -805,31 +804,3 @@ def find_latest_index_attempt(
     )
 
     db_session.commit()
-
-
-def get_connector_credential_pairs_with_user_files(
-    db_session: Session,
-) -> list[ConnectorCredentialPair]:
-    """
-    Get all connector credential pairs that have associated user files.
-
-    Args:
-        db_session: Database session
-
-    Returns:
-        List of ConnectorCredentialPair objects that have user files
-    """
-    return (
-        db_session.query(ConnectorCredentialPair)
-        .join(UserFile, UserFile.cc_pair_id == ConnectorCredentialPair.id)
-        .distinct()
-        .all()
-    )
-
-
-def delete_userfiles_for_cc_pair__no_commit(
-    db_session: Session,
-    cc_pair_id: int,
-) -> None:
-    stmt = delete(UserFile).where(UserFile.cc_pair_id == cc_pair_id)
-    db_session.execute(stmt)

backend/onyx/db/user_file.py

@@ -6,7 +6,7 @@
 
 
 def fetch_chunk_counts_for_user_files(
-    user_file_ids: list[int],
+    user_file_ids: list[str],
     db_session: Session,
 ) -> list[tuple[str, int]]:
     """
@@ -56,6 +56,6 @@ def fetch_user_project_ids_for_user_files(
     stmt = select(UserFile).where(UserFile.id.in_(user_file_ids))
     results = db_session.execute(stmt).scalars().all()
     return {
-        user_file.id: [project.id for project in user_file.projects]
+        str(user_file.id): [project.id for project in user_file.projects]
         for user_file in results
     }

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

@@ -170,8 +170,11 @@ def _build_time_filter(
     # Document sets
     filter_str += _build_or_filters(DOCUMENT_SETS, filters.document_set)
 
-    # New: user_file_ids as integer filters
-    filter_str += _build_or_filters(DOCUMENT_ID, filters.user_file_ids)
+    # Convert UUIDs to strings for user_file_ids
+    user_file_ids_str = (
+        [str(uuid) for uuid in filters.user_file_ids] if filters.user_file_ids else None
+    )
+    filter_str += _build_or_filters(DOCUMENT_ID, user_file_ids_str)
 
     # Time filter
     filter_str += _build_time_filter(filters.time_cutoff)

backend/onyx/file_store/utils.py

@@ -22,12 +22,12 @@
 logger = setup_logger()
 
 
-def user_file_id_to_plaintext_file_name(user_file_id: int) -> str:
+def user_file_id_to_plaintext_file_name(user_file_id: UUID) -> str:
     """Generate a consistent file name for storing plaintext content of a user file."""
     return f"plaintext_{user_file_id}"
 
 
-def store_user_file_plaintext(user_file_id: int, plaintext_content: str) -> bool:
+def store_user_file_plaintext(user_file_id: UUID, plaintext_content: str) -> bool:
     """
     Store plaintext content for a user file in the file store.
 

backend/onyx/indexing/adapters/user_file_indexing_adapter.py

@@ -145,7 +145,7 @@ def build_metadata_aware_chunks(
             llm_tokenizer = None
 
         user_file_id_to_raw_text: dict[str, str] = {}
-        user_file_id_to_token_count: dict[int, int | None] = {}
+        user_file_id_to_token_count: dict[str, int | None] = {}
         for user_file_id in updatable_ids:
             user_file_chunks = [
                 chunk
@@ -156,14 +156,14 @@ def build_metadata_aware_chunks(
                 combined_content = " ".join(
                     [chunk.content for chunk in user_file_chunks]
                 )
-                user_file_id_to_raw_text[user_file_id] = combined_content
+                user_file_id_to_raw_text[str(user_file_id)] = combined_content
                 token_count = (
                     len(llm_tokenizer.encode(combined_content)) if llm_tokenizer else 0
                 )
-                user_file_id_to_token_count[user_file_id] = token_count
+                user_file_id_to_token_count[str(user_file_id)] = token_count
             else:
-                user_file_id_to_raw_text[user_file_id] = None
-                user_file_id_to_token_count[user_file_id] = None
+                user_file_id_to_raw_text[str(user_file_id)] = ""
+                user_file_id_to_token_count[str(user_file_id)] = None
 
         access_aware_chunks = [
             DocMetadataAwareIndexChunk.from_index_chunk(