Merge pull request #190 from xinsenyan/main

优化逻辑与去除冗余代码
This commit is contained in:
zstar 2025-07-07 21:50:47 +08:00 committed by GitHub
commit 0a61254d65
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 12 additions and 11 deletions

View File

@ -1,3 +1,4 @@
import logging import logging
logger = logging.getLogger(__name__) logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)

View File

@ -484,11 +484,8 @@ def perform_parse(doc_id, doc_info, file_info, embedding_config, kb_info):
# 4. 更新文本块的图像信息 # 4. 更新文本块的图像信息
if image_info_list and chunk_ids_list: if image_info_list and chunk_ids_list:
conn = None
cursor = None
try: try:
conn = get_db_connection()
cursor = conn.cursor()
# 为每个文本块找到最近的图片 # 为每个文本块找到最近的图片
for i, chunk_id in enumerate(chunk_ids_list): for i, chunk_id in enumerate(chunk_ids_list):
@ -516,11 +513,7 @@ def perform_parse(doc_id, doc_info, file_info, embedding_config, kb_info):
except Exception as e: except Exception as e:
logger.error(f"[Parser-ERROR] 更新文本块图片关联失败: {e}") logger.error(f"[Parser-ERROR] 更新文本块图片关联失败: {e}")
raise Exception(f"[Parser-ERROR] 更新文本块图片关联失败: {e}") raise Exception(f"[Parser-ERROR] 更新文本块图片关联失败: {e}")
finally:
if cursor:
cursor.close()
if conn:
conn.close()
# 5. 更新最终状态 # 5. 更新最终状态
process_duration = time.time() - start_time process_duration = time.time() - start_time

View File

@ -711,12 +711,19 @@ class KnowledgebaseService:
cursor = conn.cursor(dictionary=True) cursor = conn.cursor(dictionary=True)
# 先检查文档是否存在 # 先检查文档是否存在
# check_query = """
# SELECT
# d.kb_id,
# kb.created_by AS tenant_id -- 获取 tenant_id (knowledgebase的创建者)
# FROM document d
# JOIN knowledgebase kb ON d.kb_id = kb.id -- JOIN knowledgebase 表
# WHERE d.id = %s
# """
check_query = """ check_query = """
SELECT SELECT
d.kb_id, d.kb_id,
kb.created_by AS tenant_id -- 获取 tenant_id (knowledgebase的创建者) d.created_by AS tenant_id
FROM document d FROM document d
JOIN knowledgebase kb ON d.kb_id = kb.id -- JOIN knowledgebase
WHERE d.id = %s WHERE d.id = %s
""" """
cursor.execute(check_query, (doc_id,)) cursor.execute(check_query, (doc_id,))