[package]
name = "craft-text"
version.workspace = true
edition.workspace = true
license.workspace = true
repository.workspace = true
rust-version.workspace = true
description = "Text processing below the Craft composer: hyphenation with a bundled US English dictionary and patterns, other languages through hypher"

[lints]
workspace = true

[dependencies]
hypher = { version = "0.1.8", default-features = false, features = ["alloc"], optional = true }
miniz_oxide = { version = "0.8", optional = true }
unicode-general-category = { version = "1.1", optional = true }
unicode-linebreak = { version = "0.1.5", optional = true }
unicode-segmentation = { version = "1.13", optional = true }
icu_segmenter = { version = "2.3", default-features = false, features = ["compiled_data"], optional = true }
unicode-bidi = { version = "0.3.18", optional = true }
unicode-bidi-mirroring = { version = "0.4.0", optional = true }

[dev-dependencies]
unicode-segmentation = "1.13"

[[example]]
name = "hyphgen"
required-features = ["generator"]

[features]
default = ["hyphen", "bidi", "segment", "linebreak"]
# Hyphenation points; English (bundled dictionary and patterns) needs no further feature.
hyphen = ["dep:hypher", "dep:miniz_oxide", "dep:unicode-general-category"]
# Bidirectional text: embedding levels, line reordering and mirroring (UAX #9).
bidi = ["dep:unicode-bidi", "dep:unicode-bidi-mirroring"]
# Line breaking on a box/glue/penalty model (Knuth–Plass, greedy, balanced); no dependencies.
linebreak = []
# The dictionary writer (`hyphen::dict`) and pattern trainer (`hyphen::patgen`) used by
# examples/hyphgen.rs.
generator = ["hyphen"]
# Line-break opportunities (UAX #14 with kinsoku) and grapheme-cluster emergency breaks; Thai, Lao,
# Khmer and Burmese words need their language feature.
segment = ["dep:unicode-linebreak", "dep:unicode-segmentation"]
# With `thai`, `lao`, `khmer` or `burmese`: find their words with ICU4X's LSTM models instead of its
# dictionaries. Much smaller (for web builds), slower to run, and its words differ slightly.
lstm = ["segment", "icu_segmenter?/lstm"]
# Each feature below adds the data for one more language: for `hyphen`, hypher's patterns (see
# NOTICE for their licenses; Slovak is `slovak-gpl`, outside `all-languages`); for `segment`, ICU4X's word data for Thai, Lao, Khmer and Burmese
# (one shared data set, Unicode License v3).
all-languages = [
    "afrikaans",
    "albanian",
    "assamese",
    "belarusian",
    "bengali",
    "bulgarian",
    "burmese",
    "catalan",
    "croatian",
    "czech",
    "danish",
    "dutch",
    "estonian",
    "finnish",
    "french",
    "galician",
    "georgian",
    "german",
    "greek",
    "gujarati",
    "hindi",
    "hungarian",
    "icelandic",
    "italian",
    "kannada",
    "khmer",
    "kurmanji",
    "lao",
    "latin",
    "lithuanian",
    "malayalam",
    "marathi",
    "mongolian",
    "norwegian",
    "oriya",
    "panjabi",
    "polish",
    "portuguese",
    "russian",
    "sanskrit",
    "serbian",
    "slovenian",
    "spanish",
    "swedish",
    "tamil",
    "telugu",
    "thai",
    "turkish",
    "turkmen",
    "ukrainian",
]
afrikaans = ["hyphen", "hypher/afrikaans"]
albanian = ["hyphen", "hypher/albanian"]
assamese = ["hyphen", "hypher/assamese"]
belarusian = ["hyphen", "hypher/belarusian"]
bengali = ["hyphen", "hypher/bengali"]
bulgarian = ["hyphen", "hypher/bulgarian"]
catalan = ["hyphen", "hypher/catalan"]
croatian = ["hyphen", "hypher/croatian"]
czech = ["hyphen", "hypher/czech"]
danish = ["hyphen", "hypher/danish"]
dutch = ["hyphen", "hypher/dutch"]
estonian = ["hyphen", "hypher/estonian"]
finnish = ["hyphen", "hypher/finnish"]
french = ["hyphen", "hypher/french"]
galician = ["hyphen", "hypher/galician"]
georgian = ["hyphen", "hypher/georgian"]
german = ["hyphen", "hypher/german"]
greek = ["hyphen", "hypher/greek"]
gujarati = ["hyphen", "hypher/gujarati"]
hindi = ["hyphen", "hypher/hindi"]
hungarian = ["hyphen", "hypher/hungarian"]
icelandic = ["hyphen", "hypher/icelandic"]
italian = ["hyphen", "hypher/italian"]
kannada = ["hyphen", "hypher/kannada"]
kurmanji = ["hyphen", "hypher/kurmanji"]
latin = ["hyphen", "hypher/latin"]
lithuanian = ["hyphen", "hypher/lithuanian"]
malayalam = ["hyphen", "hypher/malayalam"]
marathi = ["hyphen", "hypher/marathi"]
mongolian = ["hyphen", "hypher/mongolian"]
norwegian = ["hyphen", "hypher/norwegian"]
oriya = ["hyphen", "hypher/oriya"]
panjabi = ["hyphen", "hypher/panjabi"]
polish = ["hyphen", "hypher/polish"]
portuguese = ["hyphen", "hypher/portuguese"]
russian = ["hyphen", "hypher/russian"]
sanskrit = ["hyphen", "hypher/sanskrit"]
serbian = ["hyphen", "hypher/serbian"]
# Slovak patterns are under the GPL, version 2 or later (hyph-sk.tex), unlike every other language
# here, so `all-languages` leaves them out. A build that enables this ships GPL-licensed data. Any
# other dependency that enables `hypher/slovak` (hypher's defaults do) embeds them too, through
# feature unification.
slovak-gpl = ["hyphen", "hypher/slovak"]
slovenian = ["hyphen", "hypher/slovenian"]
spanish = ["hyphen", "hypher/spanish"]
swedish = ["hyphen", "hypher/swedish"]
tamil = ["hyphen", "hypher/tamil"]
telugu = ["hyphen", "hypher/telugu"]
turkish = ["hyphen", "hypher/turkish"]
turkmen = ["hyphen", "hypher/turkmen"]
ukrainian = ["hyphen", "hypher/ukrainian"]
burmese = ["segment", "dep:icu_segmenter"]
khmer = ["segment", "dep:icu_segmenter"]
lao = ["segment", "dep:icu_segmenter"]
thai = ["segment", "dep:icu_segmenter"]