Loading PDFs with PyPDFLoader in LangChain

Introduction

PyPDFLoader helps you read PDF files easily so you can use their text in your programs.

You want to extract text from a PDF document for analysis.

You need to load PDF content to feed into a language model.

You want to automate reading multiple PDF files in a project.

You are building a search tool that indexes PDF documents.

You want to convert PDF text into other formats or summaries.

Syntax

LangChain

from langchain.document_loaders import PyPDFLoader

loader = PyPDFLoader("path/to/file.pdf")
documents = loader.load()

Replace "path/to/file.pdf" with your actual PDF file path.

The load() method reads the PDF and returns a list of documents with text.

Examples

Loads a PDF named "example.pdf" from the current folder.

LangChain

from langchain.document_loaders import PyPDFLoader

loader = PyPDFLoader("example.pdf")
documents = loader.load()

Loads a PDF from an absolute path on your computer.

LangChain

from langchain.document_loaders import PyPDFLoader

loader = PyPDFLoader("/home/user/docs/report.pdf")
documents = loader.load()

Loads an empty PDF file and prints how many document chunks were found (likely 0).

LangChain

from langchain.document_loaders import PyPDFLoader

loader = PyPDFLoader("empty.pdf")
documents = loader.load()
print(len(documents))

Loads a single-page PDF and prints the text content of the first page.

LangChain

from langchain.document_loaders import PyPDFLoader

loader = PyPDFLoader("single_page.pdf")
documents = loader.load()
print(documents[0].page_content)

Sample Program

This program loads a PDF named "sample.pdf" from the current folder. It prints how many pages it found and then prints the text from each page.

LangChain

from langchain.document_loaders import PyPDFLoader

# Create loader for the PDF file
loader = PyPDFLoader("sample.pdf")

# Load the documents (pages) from the PDF
documents = loader.load()

# Print how many pages were loaded
print(f"Number of pages loaded: {len(documents)}")

# Print the text content of each page
for index, document in enumerate(documents, start=1):
    print(f"--- Page {index} content ---")
    print(document.page_content)
    print()

Important Notes

Loading PDFs with PyPDFLoader reads the file page by page, returning a list of document objects.

Time complexity depends on PDF size; larger PDFs take longer to load.

Common mistake: forgetting to provide the correct file path causes errors.

Use PyPDFLoader when you want to work with PDF text directly; for other file types, use their specific loaders.

Summary

PyPDFLoader makes it easy to read PDF files and get their text.

It returns a list of documents, each representing a page.

Always check your file path and handle empty or single-page PDFs carefully.