text-splitter 0.13.1

Split text into semantic chunks, up to a desired chunk size. Supports calculating length by characters and tokens, and is callable from Rust and Python.
Documentation
[workspace]
members = ["bindings/*"]

[workspace.package]
version = "0.13.1"
authors = ["Ben Brandt <benjamin.j.brandt@gmail.com>"]
edition = "2021"
description = "Split text into semantic chunks, up to a desired chunk size. Supports calculating length by characters and tokens, and is callable from Rust and Python."
repository = "https://github.com/benbrandt/text-splitter"
license = "MIT"
keywords = ["text", "split", "tokenizer", "nlp", "ai"]
categories = ["text-processing"]


[workspace.lints.rust]
future_incompatible = { level = "warn", priority = -1 }
missing_debug_implementations = "warn"
missing_docs = "warn"
nonstandard_style = { level = "warn", priority = -1 }
rust_2018_compatibility = { level = "warn", priority = -1 }
rust_2018_idioms = { level = "warn", priority = -1 }
rust_2021_compatibility = { level = "warn", priority = -1 }
rust_2024_compatibility = { level = "warn", priority = -1 }
unused = { level = "warn", priority = -1 }

[workspace.lints.clippy]
cargo = "warn"
pedantic = "warn"

[package]
name = "text-splitter"
version.workspace = true
authors.workspace = true
edition.workspace = true
description.workspace = true
repository.workspace = true
license.workspace = true
keywords.workspace = true
categories.workspace = true
exclude = [
    ".github/**",
    ".vscode/**",
    "/bindings/**",
    "/benches/output.txt",
    # Rely on large test files
    "/tests/snapshots/**",
    "/tests/text_splitter_snapshots.rs",
    "/tests/inputs/**",
    "/tests/tokenizers/**",
    "*.yml",
    "*.yaml",
]
rust-version = "1.75.0"

[package.metadata.docs.rs]
all-features = true
rustdoc-args = ["--cfg", "docsrs"]

# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html

[dependencies]
ahash = "0.8.11"
auto_enums = "0.8.5"
either = "1.11.0"
itertools = "0.12.1"
once_cell = "1.19.0"
pulldown-cmark = { version = "0.10.3", default-features = false, optional = true }
regex = "1.10.4"
rust_tokenizers = { version = "8.1.1", optional = true }
strum = { version = "0.26.2", features = ["derive"] }
thiserror = "1.0.60"
tiktoken-rs = { version = "0.5.8", optional = true }
tokenizers = { version = "0.19.1", default_features = false, features = [
    "onig",
], optional = true }
unicode-segmentation = "1.11.0"

[dev-dependencies]
cached-path = { version = "0.6.1", default-features = false, features = [
    "rustls-tls",
] }
dirs = "5.0.1"
divan = "0.1.14"
fake = "2.9.2"
insta = { version = "1.38.0", features = ["glob", "yaml"] }
more-asserts = "0.3.1"
rayon = "1.10.0"
tokenizers = { version = "0.19.1", default-features = false, features = [
    "onig",
    "http",
] }

[[bench]]
name = "chunk_size"
harness = false

[features]
markdown = ["dep:pulldown-cmark"]
rust-tokenizers = ["dep:rust_tokenizers"]
tiktoken-rs = ["dep:tiktoken-rs"]
tokenizers = ["dep:tokenizers"]

[lints]
workspace = true

# Tokenizers and indirect deps can cause slow runtime
[profile.dev.package."*"]
opt-level = 1

[profile.dev.package.insta]
opt-level = 3

[profile.dev.package.similar]
opt-level = 3