glowz/ragflow_api_test
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

更新文档处理逻辑，添加从文件读取文本块并上传至知识库的功能

Browse Source
This commit is contained in:
glowz
2025-07-05 11:36:29 +08:00
parent 587305f070
commit bc2aac4eea
1 changed files with 30 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
chunk.py
View File

@@ -1,28 +1,48 @@
from ragflow_sdk import RAGFlow from ragflow_sdk import RAGFlow
api_key = "ragflow-I5ZDNjMWNhNTdlMjExZjBiOTEwMzI0ZT" #api_key = "ragflow-I5ZDNjMWNhNTdlMjExZjBiOTEwMzI0ZT"
base_url = "http://192.168.107.165:8099" #base_url = "http://192.168.107.165:8099"
base_url = "http://localhost"
api_key = "ragflow-MyMjM2ODE2NThlMTExZjBiMzJlNzY5Mj"
rag_object = RAGFlow(api_key=api_key, base_url=base_url) rag_object = RAGFlow(api_key=api_key, base_url=base_url)
#dataset = rag_object.create_dataset(name="kb_1") #dataset = rag_object.create_dataset(name="kb_1")
datasets = rag_object.list_datasets()
#dataset = rag_object.list_datasets(name="kb_1")
dataset = rag_object.list_datasets(name="制度")
dataset = rag_object.list_datasets(name="kb_1")
dataset = dataset[0] dataset = dataset[0]
# filename1 = "ragflow.txt" # filename1 = "ragflow.txt"
# blob = open(filename1 , "rb").read() # blob = open(filename1 , "rb").read()
# dataset.upload_documents([{"display_name":filename1,"blob":blob}]) # dataset.upload_documents([{"display_name":filename1,"blob":blob}])
for doc in dataset.list_documents( page=0, page_size=12): # for doc in dataset.list_documents( page=0, page_size=12):
print(doc) # print(doc)
print("=========================================") # print("=========================================")
doc = dataset.list_documents(name= 'ragflow.txt') doc = dataset.list_documents(name= '科技创新管理办法（试行）.pdf')
doc = doc[0] doc = doc[0]
doc.update({"parser_config": {"chunk_token_count": 256}}) # doc.update({"parser_config": {"chunk_token_count": 256}})
chunk = doc.add_chunk(content="xxxxxxx") file_path ="G:\\11\\ragflow_api_test\\1.txt"
print(doc) with open(file_path, 'r', encoding='utf-8') as file:
file_content = file.read()
for num,txt_chunk in enumerate(file_content.split('\n\n')):
print(f"处理文本块: {txt_chunk[:30]}...") # 打印前30个字符以示例
if txt_chunk.strip(): # 确保不是空行
chunk = doc.add_chunk(content=txt_chunk)
print(f"第{num+1} Chunk添加成功! ID: {chunk.id}")
# content = '''
# 第二章 部门职责>第六条 【财务管理部】
# （一） 配合投标相关费用的办理工作；
# （二） 负责提供投标所需的相关财务资料。'''
# chunk = doc.add_chunk(content=content)
# print(f"Chunk添加成功! ID: {chunk.id}")
#dataset.upload_documents([{"display_name": "1.txt", "blob": open('1.txt',"rb").read()}]) #dataset.upload_documents([{"display_name": "1.txt", "blob": open('1.txt',"rb").read()}])
# 查询所有知识库 # 查询所有知识库