The lenient release
Mostly fixes, although there is a notable change to markup.
Previously anything in square brackets was considered a tag, which resulted in markup errors with Python list literals. i.e. [1,2,3] would be interpreted as a tag. The Content markup parser has been made more lenient in these cases, and treats them as literal text.
[3.4.0] - 2025-06-14
Fixed
- Fixed issues with initial flicker in
TextArearendering https://github.com/Textualize/textual/issues/5841vcomm - Fixed issue with workers that have large parameter lists breaking dev tools #5850
- Fixed post_message failing on 3.8 #5848
- Fixed log not working from threads #5863
Added
- Added experimental opt-in support for https://github.com/willmcgugan/textual-speedups
Changed
- Content markup is now more lenient; if a 'tag' doesn't contain a valid style it will be included verbatim. #5851