PDFDancer Changelog

Last updated: December 2, 2025

Follow along with the latest improvements to PDFDancer. We ship updates frequently to make editing, inspecting, and automating PDF workflows feel first-class for developers.

Week of November 30 - December 6, 2025

Core Platform

Redaction

Redact sensitive content from PDFs including text, images, paths, and form fields
Elements are permanently removed and replaced with placeholder boxes
Redact by element ID for precise control over what gets removed

Text Reflow

Replace text and automatically reflow to fit the original bounds
Options for horizontal expansion and vertical scaling
Works on both paragraphs and individual text lines

Java SDK (v0.2.1)

New Features

Redact content from PDFs: textLine.redact(), image.redact(), path.redact(), formField.redact()
Select form fields by name: selectFormFieldByName("fieldName")

Improvements

Default HTTP timeout increased to 60 seconds for large documents

Python SDK (v0.3.1)

New Features

Redact content from PDFs with redact() method on text lines, images, paths, and form fields

TypeScript SDK (v2.0.1)

New Features

Redact content from PDFs with redact() method on text lines, images, paths, and form fields

What's PDFDancer? PDFDancer is a PDF manipulation platform with SDKs for Java, Python, and TypeScript, plus a Model Context Protocol (MCP) server for AI assistants. This week introduces redaction support across all SDKs and text reflow for automatic text fitting.

Week of November 23-29, 2025

Core Platform

Find and Replace

Find and replace text across PDFs with two replacement modes:
- Visual replacement: overlays new text while preserving original structure
- True replacement: removes original text completely
Supports case-sensitive and case-insensitive searches
Preserves font styling and positioning

PDF Processing

Page orientation now preserved when loading and creating pages

Java SDK (v0.2.0)

Breaking Changes

Page numbering is now 1-based instead of 0-based
pageIndex renamed to pageNumber throughout the API

Python SDK (v0.3.0)

Breaking Changes

Page numbering is now 1-based instead of 0-based
page_index renamed to page_number throughout the API
Removed line_spacing option from text line editing

TypeScript SDK (v2.0.0)

Breaking Changes

Page numbering is now 1-based instead of 0-based
pageIndex renamed to pageNumber throughout the API

New Features

Open PDFs directly from file paths: PDFDancer.open("document.pdf")
Support for File and ArrayBuffer types in open() method
Environment variable loading with dotenv support

What's PDFDancer? PDFDancer is a PDF manipulation platform with SDKs for Java, Python, and TypeScript, plus a Model Context Protocol (MCP) server for AI assistants. This week brings a major API version update with 1-based page numbering across all SDKs and a new find-and-replace feature.

Week of November 16-22, 2025

Core Platform

Text Manipulation

More accurate text positioning and alignment in processed PDFs
Better paragraph detection for multi-column documents
Edit text lines that contain multiple styled text elements

Form Fields

Form field validation ensures you're setting the right value types

Java SDK (v0.1.8)

New Features

Select text lines using pattern matching: selectTextLinesMatching("pattern") and selectTextLineMatching("pattern")
Build text line edits with fluent API for clearer code
Delete pages with page().delete()
Better Gradle 9.x compatibility

API Changes

name() method renamed to getName() for consistency

Documentation

OpenAPI specification published
API coverage report shows what's available

Python SDK (v0.2.28)

New Features

Edit text in existing PDFs by modifying text lines
Change fonts and colors on text lines

Bug Fixes

Text and font properties now preserved correctly during edits
Clearer error messages when operations aren't supported

Documentation

Complete PyPI package information for easier discovery

TypeScript SDK (v1.0.22)

New Features

Create vector paths programmatically with the new PathBuilder API
Visual debugging helpers for development

Bug Fixes

Better handling of page references in complex documents
Improved Node.js compatibility

Documentation

Visual debugging examples
OpenAPI specification updated

MCP Server (v0.1.2)

Improvements

Verified compatibility with Node.js 18-25
Full Windows and macOS support
Works in PowerShell and WSL environments

What's PDFDancer? PDFDancer is a PDF manipulation platform with SDKs for Java, Python, and TypeScript, plus a Model Context Protocol (MCP) server for AI assistants. This week brought text editing improvements and the new PathBuilder API for creating vector graphics.

Week of November 9-15, 2025

Core Platform

Text Processing

More accurate text positioning and sizing
Better text merging for cleaner output

Vector Graphics

Support for shading fills and gradients
Enhanced annotation handling
Better line width rendering
Support for CMYK colors

OCR Detection

Automatically detect if a PDF contains OCR'd text
Page-level OCR detection

PDF Features

Support for link annotations
Better handling of specialized fonts

Font Management

API endpoints for managing fonts

Java SDK (v0.1.4)

New Features

Select single elements at coordinates: selectTextLineAt(), selectPathAt(), selectElementAt()
Automatic retry for failed requests with configurable backoff
Graceful handling of rate limits with Retry-After support
Enhanced form field type validation

Documentation

API coverage analysis

Python SDK (v0.2.24)

New Features

Automatic retry for network errors
Graceful rate limit handling with Retry-After support
Select single elements: selectTextLineAt(), selectPathAt(), selectElementAt()

Configuration

Control retry behavior with PDFDANCER_RETRY_BACKOFF_FACTOR environment variable

TypeScript SDK (v1.0.17)

New Features

Automatic retry for failed requests with Retry-After support
Select single elements: selectTextLineAt(), selectPathAt(), selectElementAt()
ImageBuilder.apply() method for easier image operations

Improvements

Better Node.js environment handling

MCP Server (v0.1.2)

Major Changes

Focused on PDFDancer SDK documentation only
Simplified search with clear syntax guide

Publishing

Published to npm as @pdfdancer/mcp-server

Documentation

Installation guides for Claude Desktop, Cline, Zed, and Sourcegraph Cody
Demo video showing Claude Code integration

What's PDFDancer? PDFDancer is a PDF manipulation platform with SDKs for Java, Python, and TypeScript, plus a Model Context Protocol (MCP) server for AI assistants. This week focused on reliability with automatic retries and rate limiting.

Week of November 2-8, 2025

Core Platform

PDF Metadata

All generated PDFs now include metadata
Tenant ID tracking in documents

Documentation

Coordinate system visualization tool
Element selection demonstration tool

Java SDK (v0.1.2)

New Features

Configurable retry mechanism for network calls
Session management and PDF processing

Configuration

Default endpoint: https://api.pdfdancer.com
Configurable base URLs

Publishing

Published to Maven Central

Python SDK (v0.2.22)

New Features

Automatic retry for network failures
Select single elements at coordinates

Configuration

PDFDANCER_RETRY_BACKOFF_FACTOR environment variable

Bug Fixes

Fixed attribute delegation in TextLineObject

TypeScript SDK (v1.0.16)

New Features

Full TypeScript support with type safety
Works in browsers and Node.js
String file path support
Select single elements at coordinates

Configuration

Default endpoint: https://api.pdfdancer.com
Configurable base URLs

MCP Server (v0.1.0)

Initial Release

Access PDFDancer SDK documentation through MCP protocol
Search documentation and retrieve code samples
Compatible with AI assistants

What's PDFDancer? PDFDancer is a PDF manipulation platform with SDKs for Java, Python, and TypeScript, plus a Model Context Protocol (MCP) server for AI assistants.