What’s new in babel 26.7

2026-05-03

Cyrillic LICR to Unicode mapping

Some languages may require extending the built-in declarations in LaTeX for the TU encoding, because a package may rely solely on them (like \cyrzh instead of ж).

The file cyrl2uni.def provides the necessary definitions to map LaTeX internal character representation (LICR) commands to their corresponding Unicode points when using Unicode-aware engines like XeTeX or LuaTeX. The mapping is direct from the LICR macros to the corresponding Unicode characters.

It’s still a beta release and currently only Cyrillic characters are defined. Further scripts may be added in the future.

The definition files are standalone and don’t require babel. You can input them directly.

Package option licr

Related to the previous feature.

Babel loads cyrl2uni.def with the Cyrillic locales. You can deactivate these extra declarations (in case you have devised your own, for example) with licr=unextended. They are ignored with lazy loading, so if you need them you have to declare explicitly the language in the preamble (or to input directly cyrl2uni.def).