[package]
authors = ["Anthony MOI <m.anthony.moi@gmail.com>", "Nicolas Patry <patry.nicolas@protonmail.com>"]
edition = "2018"
name = "tokenizers"
version = "0.21.0"
homepage = "https://github.com/huggingface/tokenizers"
repository = "https://github.com/huggingface/tokenizers"
documentation = "https://docs.rs/tokenizers/"
license = "Apache-2.0"
keywords = ["tokenizer", "NLP", "huggingface", "BPE", "WordPiece"]
readme = "./README.md"
description = """
Provides an implementation of today's most used tokenizers,
with a focus on performances and versatility.
"""
exclude = [ "rust-toolchain", "target/*", "Cargo.lock", "benches/*.txt", "benches/*.json", "data/*" ]

[lib]
name = "tokenizers"
path = "src/lib.rs"
bench = false

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

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

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

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

[[bench]]
name = "llama3"
required-features = ["http"]
harness = false

[dependencies]
lazy_static = "1.4"
rand = "0.8"
onig = { version = "6.4", default-features = false, optional = true }
regex = "1.10"
regex-syntax = "0.8"
rayon = "1.10"
rayon-cond = "0.3"
serde = { version = "1.0", features = [ "derive" ] }
serde_json = "1.0"
unicode-normalization-alignments = "0.1"
unicode_categories = "0.1"
unicode-segmentation = "1.11"
indicatif = {version = "0.17", optional = true}
itertools = "0.12"
log = "0.4"
derive_builder = "0.20"
spm_precompiled = "0.1.3"
hf-hub = { version = "0.3.2", optional = true }
aho-corasick = "1.1"
paste = "1.0.14"
macro_rules_attribute = "0.2.0"
thiserror = "1.0.49"
fancy-regex = { version = "0.13", optional = true}
getrandom = { version = "0.2.10" }
esaxx-rs = { version = "0.1.10", default-features = false, features=[]}
monostate = "0.1.12"

[features]
default = ["progressbar", "onig", "esaxx_fast"]
esaxx_fast = ["esaxx-rs/cpp"]
progressbar = ["indicatif"]
http = ["hf-hub"]
unstable_wasm = ["fancy-regex", "getrandom/js"]

[dev-dependencies]
criterion = "0.5"
tempfile = "3.10"
assert_approx_eq = "1.1"
tracing = "0.1"
tracing-subscriber = "0.3.18"

[profile.release]
lto = "fat"

[[example]]
name = "encode_batch"
required-features = ["http"]