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