AI/dify
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/langgenius/dify.git synced 2025-08-12 21:09:04 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix multiple retrieval in knowledge node (#8942)

Browse Source
This commit is contained in:
Jyong 2024-09-30 23:07:04 +08:00 committed by GitHub
parent 7850a28ec8
commit 625e4c4c72
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
api/core/workflow/nodes/knowledge_retrieval/knowledge_retrieval_node.py
View File

@ -229,7 +229,9 @@ class KnowledgeRetrievalNode(BaseNode):
source["content"] = segment.get_sign_content() source["content"] = segment.get_sign_content()
retrieval_resource_list.append(source) retrieval_resource_list.append(source)
if retrieval_resource_list: if retrieval_resource_list:
retrieval_resource_list = sorted(retrieval_resource_list, key=lambda x: x.get("score"), reverse=True) retrieval_resource_list = sorted(
retrieval_resource_list, key=lambda x: x.get("metadata").get("score"), reverse=True
)
position = 1 position = 1
for item in retrieval_resource_list: for item in retrieval_resource_list:
item["metadata"]["position"] = position item["metadata"]["position"] = position

4
api/services/website_service.py
View File

@ -166,6 +166,8 @@ class WebsiteService:
@classmethod @classmethod
def get_crawl_url_data(cls, job_id: str, provider: str, url: str, tenant_id: str) -> dict | None: def get_crawl_url_data(cls, job_id: str, provider: str, url: str, tenant_id: str) -> dict | None:
credentials = ApiKeyAuthService.get_auth_credentials(tenant_id, "website", provider) credentials = ApiKeyAuthService.get_auth_credentials(tenant_id, "website", provider)
# decrypt api_key
api_key = encrypter.decrypt_token(tenant_id=tenant_id, token=credentials.get("config").get("api_key"))
if provider == "firecrawl": if provider == "firecrawl":
file_key = "website_files/" + job_id + ".txt" file_key = "website_files/" + job_id + ".txt"
if storage.exists(file_key): if storage.exists(file_key):
@ -173,8 +175,6 @@ class WebsiteService:
if data: if data:
data = json.loads(data.decode("utf-8")) data = json.loads(data.decode("utf-8"))
else: else:
# decrypt api_key
api_key = encrypter.decrypt_token(tenant_id=tenant_id, token=credentials.get("config").get("api_key"))
firecrawl_app = FirecrawlApp(api_key=api_key, base_url=credentials.get("config").get("base_url", None)) firecrawl_app = FirecrawlApp(api_key=api_key, base_url=credentials.get("config").get("base_url", None))
result = firecrawl_app.check_crawl_status(job_id) result = firecrawl_app.check_crawl_status(job_id)
if result.get("status") != "completed": if result.get("status") != "completed":