mirror of
https://github.com/infiniflow/ragflow.git
synced 2025-12-08 20:42:30 +08:00
dab92ac1e8
### What problem does this PR solve? Refactor Chunk API #2846 ### Type of change - [x] Refactoring --------- Co-authored-by: liuhua <10215101452@stu.ecun.edu.cn> Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
59 lines
1.9 KiB
Python
59 lines
1.9 KiB
Python
import time
|
|
|
|
from PIL.ImageFile import raise_oserror
|
|
|
|
from .base import Base
|
|
from .chunk import Chunk
|
|
from typing import List
|
|
|
|
|
|
class Document(Base):
|
|
def __init__(self, rag, res_dict):
|
|
self.id = ""
|
|
self.name = ""
|
|
self.thumbnail = None
|
|
self.knowledgebase_id = None
|
|
self.parser_method = ""
|
|
self.parser_config = {"pages": [[1, 1000000]]}
|
|
self.source_type = "local"
|
|
self.type = ""
|
|
self.created_by = ""
|
|
self.size = 0
|
|
self.token_count = 0
|
|
self.chunk_count = 0
|
|
self.progress = 0.0
|
|
self.progress_msg = ""
|
|
self.process_begin_at = None
|
|
self.process_duration = 0.0
|
|
self.run = "0"
|
|
self.status = "1"
|
|
for k in list(res_dict.keys()):
|
|
if k not in self.__dict__:
|
|
res_dict.pop(k)
|
|
super().__init__(rag, res_dict)
|
|
|
|
def list_chunks(self,offset=0, limit=30, keywords="", id:str=None):
|
|
data={"document_id": self.id,"keywords": keywords,"offset":offset,"limit":limit,"id":id}
|
|
res = self.get(f'/dataset/{self.knowledgebase_id}/document/{self.id}/chunk', data)
|
|
res = res.json()
|
|
if res.get("code") == 0:
|
|
chunks=[]
|
|
for data in res["data"].get("chunks"):
|
|
chunk = Chunk(self.rag,data)
|
|
chunks.append(chunk)
|
|
return chunks
|
|
raise Exception(res.get("message"))
|
|
|
|
|
|
def add_chunk(self, content: str):
|
|
res = self.post(f'/dataset/{self.knowledgebase_id}/document/{self.id}/chunk', {"content":content})
|
|
res = res.json()
|
|
if res.get("code") == 0:
|
|
return Chunk(self.rag,res["data"].get("chunk"))
|
|
raise Exception(res.get("message"))
|
|
|
|
def delete_chunks(self,ids:List[str]):
|
|
res = self.rm(f"dataset/{self.knowledgebase_id}/document/{self.id}/chunk",{"ids":ids})
|
|
res = res.json()
|
|
if res.get("code")!=0:
|
|
raise Exception(res.get("message")) |