fix parameter error (#1925)

### What problem does this PR solve?


### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
Kevin Hu 2024-08-13 11:42:38 +08:00 committed by GitHub
parent da2d8b8267
commit 77f0fb03e3
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

View File

@ -15,7 +15,7 @@ from rag.nlp import find_codec,num_tokens_from_string
import re
class RAGFlowTxtParser:
def __call__(self, fnm, binary=None, chunk_token_num=128):
def __call__(self, fnm, binary=None, chunk_token_num=128, delimiter="\n!?;。;!?"):
txt = ""
if binary:
encoding = find_codec(binary)
@ -27,7 +27,7 @@ class RAGFlowTxtParser:
if not l:
break
txt += l
return self.parser_txt(txt, chunk_token_num)
return self.parser_txt(txt, chunk_token_num, delimiter)
@classmethod
def parser_txt(cls, txt, chunk_token_num=128, delimiter="\n!?;。;!?"):