From 636a0ba37f703b8c9acc5b39e1e0941a3aa60d3e Mon Sep 17 00:00:00 2001 From: Bowen Liang Date: Thu, 10 Apr 2025 17:12:48 +0800 Subject: [PATCH] chore: skip document segments fetching with non-existed dataset of DatasetDocument in add_document_to_index_task task (#17784) --- api/tasks/add_document_to_index_task.py | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/api/tasks/add_document_to_index_task.py b/api/tasks/add_document_to_index_task.py index 0b7d2ad31f..be88881efc 100644 --- a/api/tasks/add_document_to_index_task.py +++ b/api/tasks/add_document_to_index_task.py @@ -37,6 +37,10 @@ def add_document_to_index_task(dataset_document_id: str): indexing_cache_key = "document_{}_indexing".format(dataset_document.id) try: + dataset = dataset_document.dataset + if not dataset: + raise Exception(f"Document {dataset_document.id} dataset {dataset_document.dataset_id} doesn't exist.") + segments = ( db.session.query(DocumentSegment) .filter( @@ -77,11 +81,6 @@ def add_document_to_index_task(dataset_document_id: str): document.children = child_documents documents.append(document) - dataset = dataset_document.dataset - - if not dataset: - raise Exception("Document has no dataset") - index_type = dataset.doc_form index_processor = IndexProcessorFactory(index_type).init_index_processor() index_processor.load(dataset, documents)