[package]
name = "milli"
edition = "2021"
publish = false

version.workspace = true
authors.workspace = true
description.workspace = true
homepage.workspace = true
readme.workspace = true
# edition.workspace = true
license.workspace = true

[dependencies]
big_s.workspace = true
bimap = { workspace = true, features = ["serde"] }
bincode.workspace = true
bstr.workspace = true
bytemuck = { workspace = true, features = ["extern_crate_alloc"] }
byteorder.workspace = true
charabia = { workspace = true, default-features = false }
concat-arrays.workspace = true
convert_case.workspace = true
crossbeam-channel.workspace = true
deserr.workspace = true
either = { workspace = true, features = ["serde"] }
flatten-serde-json.path = "../flatten-serde-json"
fst.workspace = true
fxhash.workspace = true
geoutils.workspace = true
grenad = { workspace = true, default-features = false, features = [
    "rayon",
    "tempfile",
] }
heed = { workspace = true, default-features = false, features = [
    "serde-json",
    "serde-bincode",
] }
indexmap = { workspace = true, features = ["serde"] }
json-depth-checker.path = "../json-depth-checker"
levenshtein_automata = { workspace = true, features = ["fst_automaton"] }
memchr.workspace = true
memmap2.workspace = true
obkv.workspace = true
once_cell.workspace = true
ordered-float.workspace = true
rayon.workspace = true
roaring = { workspace = true, features = ["serde"] }
rstar = { workspace = true, features = ["serde"] }
serde = { workspace = true, features = ["derive"] }
serde_json = { workspace = true, features = ["preserve_order", "raw_value"] }
slice-group-by.workspace = true
smallstr = { workspace = true, features = ["serde"] }
smallvec.workspace = true
smartstring.workspace = true
tempfile.workspace = true
thiserror.workspace = true
time = { workspace = true, features = [
    "serde-well-known",
    "formatting",
    "parsing",
    "macros",
] }
uuid = { workspace = true, features = ["v4"] }

filter-parser.path = "../filter-parser"

# documents words self-join
itertools.workspace = true

csv.workspace = true
candle-core = { workspace = true }
candle-transformers = { workspace = true }
candle-nn = { workspace = true }
tokenizers = { workspace = true, default-features = false, features = [
    "onig",
] }
hf-hub = { workspace = true, default-features = false, features = [
    "online",
] }
tiktoken-rs.workspace = true
liquid.workspace = true
rhai = { workspace = true, features = [
    "serde",
    "no_module",
    "no_custom_syntax",
    "no_time",
    "sync",
] }
arroy.workspace = true
rand.workspace = true
tracing.workspace = true
ureq = { workspace = true, features = ["json"] }
url.workspace = true
hashbrown.workspace = true
bumpalo.workspace = true
bumparaw-collections.workspace = true
thread_local.workspace = true
rustc-hash.workspace = true
enum-iterator.workspace = true
bbqueue.workspace = true
flume.workspace = true
utoipa = { workspace = true, features = [
    "non_strict_integers",
    "preserve_order",
    "uuid",
    "time",
    "openapi_extensions",
] }
lru.workspace = true

[dev-dependencies]
mimalloc.workspace = true
insta.workspace = true
maplit.workspace = true
md5.workspace = true
meili-snap.path = "../meili-snap"
rand = { workspace = true, features = ["small_rng"] }

[features]
all-tokenizations = ["charabia/default"]

# Use POSIX semaphores instead of SysV semaphores in LMDB
# For more information on this feature, see heed's Cargo.toml
lmdb-posix-sem = ["heed/posix-sem"]

# allow chinese specialized tokenization
chinese = ["charabia/chinese"]
chinese-pinyin = ["chinese", "charabia/chinese-normalization-pinyin"]

# allow hebrew specialized tokenization
hebrew = ["charabia/hebrew"]

# allow japanese specialized tokenization
japanese = ["charabia/japanese"]
japanese-transliteration = ["charabia/japanese-transliteration"]

# allow korean specialized tokenization
korean = ["charabia/korean"]

# allow thai specialized tokenization
thai = ["charabia/thai"]

# allow greek specialized tokenization
greek = ["charabia/greek"]

# allow khmer specialized tokenization
khmer = ["charabia/khmer"]

# allow vietnamese specialized tokenization
vietnamese = ["charabia/vietnamese"]

# allow german specialized tokenization
german = ["charabia/german-segmentation"]

# force swedish character recomposition
swedish-recomposition = ["charabia/swedish-recomposition"]

# allow turkish specialized tokenization
turkish = ["charabia/turkish"]

# allow CUDA support, see <https://github.com/meilisearch/meilisearch/issues/4306>
cuda = ["candle-core/cuda"]