update document and segment word count (#10449)

This commit is contained in:
Jyong 2024-11-08 17:32:27 +08:00 committed by GitHub
commit 4f1a56f0f0
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 41 additions and 4 deletions

View file

@ -57,7 +57,7 @@ def batch_create_segment_to_index_task(
model_type=ModelType.TEXT_EMBEDDING,
model=dataset.embedding_model,
)
word_count_change = 0
for segment in content:
content = segment["content"]
doc_id = str(uuid.uuid4())
@ -86,8 +86,13 @@ def batch_create_segment_to_index_task(
)
if dataset_document.doc_form == "qa_model":
segment_document.answer = segment["answer"]
segment_document.word_count += len(segment["answer"])
word_count_change += segment_document.word_count
db.session.add(segment_document)
document_segments.append(segment_document)
# update document word count
dataset_document.word_count += word_count_change
db.session.add(dataset_document)
# add index to db
indexing_runner = IndexingRunner()
indexing_runner.batch_add_segments(document_segments, dataset)