trafilatura Releases

A minor bug fix release for the chunkr-services component, identical to v2.2.1, focusing on model replacement for better hardware performance.

Jul 31, 2025

A major version update focusing on stability and simplicity by removing unused components from the web interface.

Jul 30, 2025

A minor version update for the chunkr-services component focusing on stability and simplicity by removing unused components.

Jul 30, 2025

A major release introducing natural language search, dynamic sorting, streaming conversations, and extensive performance improvements, bug fixes, and API enhancements.

Jun 30, 2025

Major version release with significant breaking changes including Python 3.6/3.7 deprecation, refactored bare_extraction() to return Document class instances, removal of deprecated GUI, and extensive type hinting additions.

Dec 3, 2024

Maintenance release adding SOCKS proxy support and various extraction fixes for tables, images, and spider functionality.

Sep 10, 2024

Navigation and bugfix release focusing on crawler improvements, memory error fixes, and code maintenance with type annotations.

Aug 20, 2024

Breaking release enforcing fixed output formats and improving extraction accuracy with better justext fallback and XPath rules.

Jul 30, 2024

Major metadata handling change with metadata now skipped by default, plus HTML output format addition and extraction performance improvements.

Jun 27, 2024

This release focuses on stability and code organization. It introduces stricter error handling for deprecated arguments, reorganizes the deduplication module, and adds a port of readability.js's `is_probably_readerable` function. Various extraction and CLI bugs were also fixed.

May 30, 2024

This release emphasizes output flexibility and performance. It adds explicit Markdown output support, improves extraction recall and speed, and introduces file creation date extraction. It also deprecates the `process_record()` function.

May 2, 2024

Apr 23, 2024

A maintenance release focused on dependency stability and extraction accuracy. It pins the LXML dependency to prevent breakage and fixes formatting issues related to element generation and space handling.

Apr 3, 2024

A performance and precision update that introduces better extraction accuracy, CSV output improvements, and LXML optimizations. It also changes the project license to Apache 2.0.

Mar 20, 2024

This release focuses on LXML compatibility and download utilities. It adds support for LXML v5+, introduces an advanced fetch function, and improves the `html2txt()` function.

Jan 25, 2024

A maintenance release fixing MacOS setup issues and improving navigation robustness. It introduces a redirect limit configuration and fixes feed detection.

Jan 8, 2024

Focuses on metadata precision and URL normalization. It updates the htmldate library for better date extraction and improves URL handling using the courlan library.

Nov 29, 2023

Improves parsing robustness and CLI usability. It allows more lenient HTML parsing, adds a `--probe` option to check content, and adds support for Python 3.12.

Sep 6, 2023

Bug fixes for metadata extraction and navigation. It simplifies JSON parsing, fixes author handling, and improves the robustness of the spider and link processing.

Jun 15, 2023

An early 1.6 series release introducing content hashing, sitemap improvements, and safety checks. It adds a new `is_live_test()` function and improves download/sitemap efficiency.

May 11, 2023