langchain-text-splitters
A core text-processing component in the LangChain ecosystem responsible for splitting long documents into text chunks suitable for embedding models and large language model context windows. It supports recursive character splitting, code-language-aware splitting, and JSON structured data processing, serving as a foundational tool in RAG pipelines.
Timeline (last 90 days)
langchain-text-splitters released version 1.1.3.
langchain-text-splitters is responsible for splitting long documents into text chunks suitable for embedding models and large language model context windows.
Version 1.1.3 removes invalid and duplicate separators in Kotlin, Rust, and Haskell (#37039).
Version 1.1.3 removes incorrect delimiter keywords in C# and Elixir (#37037).
Version 1.1.3 restores the lazy imports mechanism for heavy optional dependencies (#35469), which was accidentally reverted in version 1.1.2.
Version 1.1.3 replaces mypy with ty for type checking (#38658) and unifies the type checking configuration across the monorepo.
The highest proportion of changes in this release are dependency version upgrades (chore(deps)), involving dozens of PRs.
All Facts (6)
langchain-text-splitters is responsible for splitting long documents into text chunks suitable for embedding models and large language model context windows.
50%UnverifiedVersion 1.1.3 removes invalid and duplicate separators in Kotlin, Rust, and Haskell (#37039).
50%UnverifiedVersion 1.1.3 removes incorrect delimiter keywords in C# and Elixir (#37037).
50%UnverifiedVersion 1.1.3 restores the lazy imports mechanism for heavy optional dependencies (#35469), which was accidentally reverted in version 1.1.2.
50%UnverifiedVersion 1.1.3 replaces mypy with ty for type checking (#38658) and unifies the type checking configuration across the monorepo.
50%UnverifiedThe highest proportion of changes in this release are dependency version upgrades (chore(deps)), involving dozens of PRs.
50%