hz/ebiz-dify-ai
1
0
Fork 0
mirror of http://112.124.100.131/huang.ze/ebiz-dify-ai.git synced 2025-12-10 11:26:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix/create document by api with metadata (#16307)

Browse Source 
Co-authored-by: zxhlyh <jasonapring2015@outlook.com>
This commit is contained in:
Jyong
2025-03-20 14:33:32 +08:00
committed by GitHub
parent c1f3d968bf
commit 2c9af712a2
7 changed files with 75 additions and 527 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
api/services/dataset_service.py
View File

@@ -46,7 +46,6 @@ from models.source import DataSourceOauthBinding
from services.entities.knowledge_entities.knowledge_entities import (
ChildChunkUpdateArgs,
KnowledgeConfig,
MetaDataConfig,
RerankingModel,
RetrievalModel,
SegmentUpdateArgs,
@@ -999,9 +998,6 @@ class DocumentService:
document.data_source_info = json.dumps(data_source_info)
document.batch = batch
document.indexing_status = "waiting"
if knowledge_config.metadata:
document.doc_type = knowledge_config.metadata.doc_type
document.metadata = knowledge_config.metadata.doc_metadata
db.session.add(document)
documents.append(document)
duplicate_document_ids.append(document.id)
@@ -1018,7 +1014,6 @@ class DocumentService:
account,
file_name,
batch,
knowledge_config.metadata,
)
db.session.add(document)
db.session.flush()
@@ -1076,7 +1071,6 @@ class DocumentService:
account,
truncated_page_name,
batch,
knowledge_config.metadata,
)
db.session.add(document)
db.session.flush()
@@ -1117,7 +1111,6 @@ class DocumentService:
account,
document_name,
batch,
knowledge_config.metadata,
)
db.session.add(document)
db.session.flush()
@@ -1155,7 +1148,6 @@ class DocumentService:
account: Account,
name: str,
batch: str,
metadata: Optional[MetaDataConfig] = None,
):
document = Document(
tenant_id=dataset.tenant_id,
@@ -1180,9 +1172,6 @@ class DocumentService:
BuiltInField.last_update_date: datetime.datetime.now(datetime.UTC).strftime("%Y-%m-%d %H:%M:%S"),
BuiltInField.source: data_source_type,
}
if metadata is not None:
doc_metadata.update(metadata.doc_metadata)
document.doc_type = metadata.doc_type
if doc_metadata:
document.doc_metadata = doc_metadata
return document
@@ -1297,10 +1286,6 @@ class DocumentService:
# update document name
if document_data.name:
document.name = document_data.name
# update doc_type and doc_metadata if provided
if document_data.metadata is not None:
document.doc_metadata = document_data.metadata.doc_metadata
document.doc_type = document_data.metadata.doc_type
# update document to be waiting
document.indexing_status = "waiting"
document.completed_at = None