pdfplumber: Extract Text, Tables, and Objects from PDFs

Overview

pdfplumber parses PDFs into detailed, positioned objects such as characters, lines, rectangles, curves, and images. Built on pdfminer.six, it adds higher-level text and table extraction tools for documents where layout and object coordinates matter.

It is aimed at Python developers working with machine-generated PDFs. For scanned documents, OCR is needed elsewhere; pdfplumber does not provide OCR or PDF editing and generation.

Key Features

• Inspect page objects and their properties, including character text, font, size, and coordinates.
• Extract page text, words, and searchable text matches with bounding boxes.
• Find and extract tables using graphical lines, text alignment, or explicitly supplied separators.
• Crop and filter pages to focus extraction on specific regions or objects.
• Render pages and overlay detected objects or table boundaries for visual debugging.
• Use a command-line interface to export object data as CSV or JSON, or page text as plain text.

Use Cases

• Data engineers can turn machine-generated reports into structured text or table data for downstream processing.
• Analysts can extract tables from PDFs and tune detection when documents have unusual layouts or weakly defined cell borders.
• Developers investigating extraction errors can inspect object coordinates and visualize page structure.
• Teams processing forms can use its PDF parsing wrappers to access form-field values, though there is no dedicated form API.

Project Facts

• Language: Python
• License: MIT
• Stars: 10.8k
• Forks: 928
• Topics: pdf, pdf-parsing, table-extraction
• Archived: No

Getting Started

Install with pip:

pip install pdfplumber

For usage examples and configuration details, see the README.

Alternatives

• pdfminer.six: pdfminer.six focuses on parsing PDF text and layout details, while pdfplumber adds a higher-level interface for inspecting objects and extracting tables.
• pypdf: pypdf combines text extraction with PDF editing tasks such as merging and splitting, while pdfplumber focuses on layout-aware inspection and extraction.
• Marker: Marker converts PDFs and other documents into Markdown or structured formats, including scanned files, while pdfplumber focuses on inspecting machine-generated PDFs.

Considerations

• Works best with machine-generated PDFs. Scanned PDFs and OCR-derived documents are not its strength, and OCR is not included.
• Parsing is built on pdfminer.six and is slower than PyMuPDF according to the project’s comparison; choose another tool if speed is the priority.
• Table extraction is configurable, but results depend on document layout and may require cropping or tuning settings.
• Text layout extraction is marked experimental, as are some search and text-line features.
• The library exposes image positions and metadata but does not reconstruct image content. It also does not generate or modify PDFs.