How to Load Documents with LangChain Document Loaders

What this does

LangChain document loaders provide a unified interface for ingesting data from dozens of sources into Document objects that the rest of a RAG pipeline consumes. Each loader returns Document instances wrapping page content and optional metadata.

Steps

1. Install extras for your specific loader.

   pip install langchain-community pypdf beautifulsoup4
   

2. Import the loader and load documents.

   from langchain_community.document_loaders import PyPDFLoader
   from langchain_core.documents import Document
   
   loader = PyPDFLoader("/data/report.pdf")
   pages = loader.load()
   for page in pages:
       print(f"Page {page.metadata.get('page')}: {len(page.page_content)} chars")
   

3. Load from a text file or directory.

   from langchain_community.document_loaders import TextLoader, DirectoryLoader
   
   txt_loader = TextLoader("/data/notes.txt")
   txt_docs = txt_loader.load()
   
   dir_loader = DirectoryLoader("/data/docs", glob="**/*.txt", loader_cls=TextLoader)
   dir_docs = dir_loader.load()
   

4. Load from a URL with BSHTMLLoader.

   from langchain_community.document_loaders import BSHTMLLoader
   
   html_loader = BSHTMLLoader("https://example.com/page.html")
   html_docs = html_loader.load()
   

Verification

python -c "
from langchain_community.document_loaders import TextLoader
loader = TextLoader('/etc/hostname')
docs = loader.load()
print(f'Loaded {len(docs)} document(s), {len(docs[0].page_content)} chars')
"
# Expected: Loaded 1 document(s), <N> chars

Common failures

• ImportError for PyPDFLoader. Missing langchain-community or pypdf. Install with pip install langchain-community pypdf.
• FileNotFoundError. Verify the path exists with ls -la /data/report.pdf.
• Empty page_content on all pages. Encrypted PDF or image-only scan. Run pdftotext to confirm text extraction works first.
• AttributeError on metadata. Metadata field missing for some pages. Use .get("page", 0) with a fallback default.
• Version mismatch - The installed package or runtime differs from the command shown; check the version first and rerun the smallest verification command.
• Local environment drift - Another service, virtual environment, model, or path is being used; print the active binary path and configuration before changing the guide steps.

Operator checkpoint

Before treating this as solved, write down the local runtime, model or package version, hardware/backend if relevant, and the verification output. This keeps the guide useful as a Will-It-Run style decision instead of a one-off command transcript.

Related guides

• batch-ingest-documents-vector-database
• extract-metadata-documents-filtering