Add function get_similar_content_by_query_with_vectorstore.

[dzerkes]

Fixes a bug when calling GPTResearcher with report_source='langchain_vectorstore'.

code to test:

import asyncio 
from gpt_researcher import GPTResearcher
import os 
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_openai import OpenAIEmbeddings 
from langchain_community.vectorstores import FAISS 
from langchain_core.documents import Document 
from gpt_researcher.document import DocumentLoader, LangChainDocumentLoader 

os.environ['OPENAI_API_KEY']="add your key"
os.environ['TAVILY_API_KEY']="add your key"

async def main():
    document_data = await DocumentLoader("./my-docs").load()
    langchain_docs = [Document(page_content=item["raw_content"], metadata={"source":item["url"]}) 
                      for item in document_data]
    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1400, chunk_overlap=200)
    splitted_docs = text_splitter.split_documents(langchain_docs)
    vector_store = FAISS.from_documents(splitted_docs, OpenAIEmbeddings())
    
    researcher = GPTResearcher(
        query="Write me a summary about formalin or formaldehyde", 
        report_source="langchain_vectorstore",
        vector_store=vector_store
    )
    
    context = await researcher.conduct_research()

if __name__ == "__main__":
    asyncio.run(main())

Without the addition of the function get_similar_content_by_query_with_vectorstore, it would throw an error.

P.S.: Before some weeks, the problem was that while this function existed it used double underscore, hence when it was called, it was not called correctly. But now, I see that this function disappeared completely.

[assafelovic]

Thanks wonder how this was missing. Appreciate it!