pypdf

6.9.2 · active · verified Sun Mar 29

pypdf is a free and open-source pure-Python library designed for PDF file manipulation. It can split, merge, crop, and transform PDF files, as well as add custom data, viewing options, and passwords. It also supports retrieving text and metadata from PDFs. Currently at version 6.9.2, the library is actively maintained with a rapid release cadence, often seeing multiple updates per month to address bugs, security, and performance improvements.

Warnings

breaking The library was renamed from PyPDF2 to pypdf. This involved a package name change and significant class and method renames (e.g., `PdfFileReader` to `PdfReader`, `PdfFileWriter` to `PdfWriter`, `PdfFileMerger` to `PdfMerger`, and methods like `getNumPages()` to `len(reader.pages)`).
Fix: Update your import statements from `from PyPDF2 import ...` to `from pypdf import ...` and adapt to the new class and method names following PEP8 conventions (e.g., `reader.getNumPages()` becomes `len(reader.pages)`). Consult the migration guide for a full list of changes.
deprecated Support for abbreviations in `decode_stream_data` was deprecated.
Fix: Avoid using abbreviations when calling `decode_stream_data`. Review the official documentation for the recommended usage.
gotcha Calling `PageObject.replace_contents()` for pages not assigned to a `PdfWriter` is deprecated. This can lead to unexpected behavior and will be removed in pypdf 7.0.0.
Fix: Ensure that `PageObject.replace_contents()` is only called on `PageObject` instances that are part of a `PdfWriter` object. The documentation advises against using it directly on pages from a `PdfReader`.
gotcha Older versions of pypdf might experience significant performance degradation (O(n²) complexity) when dealing with frequent `NameObject` read/write operations, especially with complex or large PDF files.
Fix: Upgrade to pypdf 6.9.0 or later, which includes performance improvements for `NameObject` handling.
gotcha Processing untrusted or malformed PDF files, especially with older versions of pypdf, can lead to security vulnerabilities such as infinite loops, excessive resource consumption, or crashes. Numerous security fixes address issues like circular references and stream length limits.
Fix: Keep `pypdf` updated to the latest version to benefit from security patches. Consider using `strict=True` when initializing `PdfReader` to raise exceptions for non-standard compliant PDFs, allowing for explicit error handling.

Install

pip install pypdf Default installation
pip install pypdf[crypto] For AES encryption/decryption support

Imports

PdfReader
```
from pypdf import PdfReader
```
PdfWriter
```
from pypdf import PdfWriter
```
PdfMerger
```
from pypdf import PdfMerger
```
The library was renamed from PyPDF2 to pypdf, and class names were standardized. `PdfFileMerger` is now `PdfMerger`.

Quickstart

This quickstart demonstrates how to create two simple PDF files and then merge them into a single output PDF using PdfReader and PdfWriter.

from pypdf import PdfReader, PdfWriter

# Create dummy PDF files for the example
with open("document1.pdf", "wb") as f:
    writer = PdfWriter()
    writer.add_blank_page(width=72, height=72)
    writer.add_page(writer.add_blank_page(width=72, height=72))
    writer.write(f)

with open("document2.pdf", "wb") as f:
    writer = PdfWriter()
    writer.add_blank_page(width=72, height=72)
    writer.add_page(writer.add_blank_page(width=72, height=72))
    writer.write(f)

# Merge multiple PDF files into one
writer = PdfWriter()

# Add pages from document1.pdf
reader1 = PdfReader("document1.pdf")
for page in reader1.pages:
    writer.add_page(page)

# Add pages from document2.pdf
reader2 = PdfReader("document2.pdf")
for page in reader2.pages:
    writer.add_page(page)

# Write the merged PDF to a new file
with open("merged_document.pdf", "wb") as output_pdf:
    writer.write(output_pdf)

print("PDFs merged successfully into merged_document.pdf")

view raw JSON →