v2.1.0

adbar/trafilaturav2.1.0Jun 7, 2026by adbar

AI Summary

This release focuses on performance improvements using XSLT extensions and updates to the lxml dependency. It includes a large volume of bug fixes regarding table extraction, image handling, and code formatting, alongside general code modernization and test improvements.

Key Highlights

Significant performance boost using XSLT extensions for XPath operations
Updated lxml dependency with minimal code changes
Enhanced robustness with improved error handling and deprecation warnings
More accurate table and image extraction logic
Codebase modernization including formatting, CI updates, and test coverage

New Features

Added `extract_with_metadata` method
Added support for extracting images from meta name tags
Added support for preserving images when links are included
Added support for preserving links in tables when `--links` is enabled

Full Release Notes

Major changes:
- Dependencies updated, lxml in particular (with minimal changes in the code)
- Faster XPath performance using XSLT extensions by @Honesty-of-the-Cavernous-Tissue (#793)
- More deprecation warnings
- More robust code

Fixes:
- Fix node pruning raising an AttributeError in certain cases by @PLPeeters (#761)
- Refine img src url and fix table extraction bugs by @unsleepy22 (#762)
- Fix eager doctype regex matching by @UVMvmfee (#764)
- Add extract_with_metadata method and ut by @unsleepy22 (#765)
- More robust table extraction by @unsleepy22 (#767)
- Detect code in pre-tag using indicators and fix code formatting issues by @steineggerroland (#776)
- Fix tokenizer not stripping "Ideographic Full Stop" by @reinoldus (#783)
- Improve spacing in item, cell and code blocks by @unsleepy22 (#772)
- Changed code to handle potential empty string by @seanswyi (#790)
- Optimize PUNCT_TBL creation by @deedy5 (#791)
- Replace `None`s in code blocks by @crackcomm in (#797)
- Propagate no_fallback setting to fast argument by @unsleepy22 (#800)
- Fix bug #807 by @arcombe012 (#808)
- Added support to meta name image by @felipehertzer (#809)
- Preserve images when links are included by @rmarchei (#841)
- Clean fencedframe element during extraction by @eyupcanakman (#847)
- Add all Justext languages known to the detector with @cikay (#810)
- Safer lxml element init by @adbar (#856)
- Preserve links in tables when --links is enabled by @rmarchei (#858)

Maintenance:
- Update docs by @adbar (#773)
- Update README: fix /dependents links by @vbarbaresi (#785)
- Update CI pipeline by @adbar (#801)
- Update setup, fix code and tests by @adbar (#853)
- Review consistency of code, tests and docs by @adbar (#855)
- Add tests to prevent metadata bugs by @adbar (#857)
- Tests: check consistency by @adbar (#859)
- Modernize code base by @adbar (#860)
- Format the code with ruff by @adbar (#861)