MCPcopy
hub / github.com/langchain-ai/langchain / _ingest

Function _ingest

templates/rag-mongo/rag_mongo/chain.py:65–80  ·  view source on GitHub ↗
(url: str)

Source from the content-addressed store, hash-verified

63
64
65def _ingest(url: str) -> dict:
66 loader = PyPDFLoader(url)
67 data = loader.load()
68
69 # Split docs
70 text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=0)
71 docs = text_splitter.split_documents(data)
72
73 # Insert the documents in MongoDB Atlas Vector Search
74 _ = MongoDBAtlasVectorSearch.from_documents(
75 documents=docs,
76 embedding=OpenAIEmbeddings(disallowed_special=()),
77 collection=MONGODB_COLLECTION,
78 index_name=ATLAS_VECTOR_SEARCH_INDEX_NAME,
79 )
80 return {}
81
82
83ingest = RunnableLambda(_ingest)

Callers

nothing calls this directly

Calls 6

PyPDFLoaderClass · 0.90
OpenAIEmbeddingsClass · 0.90
loadMethod · 0.45
split_documentsMethod · 0.45
from_documentsMethod · 0.45

Tested by

no test coverage detected