8 lines
365 B
Python
8 lines
365 B
Python
from langchain_text_splitter import RecursiveCharacterTextSplitter
|
|
from langchain.schema import Document
|
|
|
|
def get_chunks(content: str, title: str):
|
|
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
|
|
texts = splitter.split_text(content)
|
|
return [Document(page_content=t, metadata={"title": title, "content": t}) for t in texts]
|