Refactor properties to separate crate

.github/workflows/build-test.yml

@@ -460,9 +460,9 @@ jobs:
       matrix:
         component:
           - components/locid
-          - components/uniset
           - components/plurals
           - components/datetime
+          - utils/uniset
           - utils/fixed_decimal
 
 

Cargo.toml

@@ -14,7 +14,7 @@ members = [
     "components/locid",
     "components/locid/macros",
     "components/plurals",
-    "components/uniset",
+    "components/properties",
     "experimental/bies",
     "experimental/formatted_string_builder",
     "experimental/list_formatter",
@@ -37,6 +37,7 @@ members = [
     "utils/fixed_decimal",
     "utils/litemap",
     "utils/pattern",
+    "utils/uniset",
     "utils/writeable",
     "utils/yoke",
     "utils/yoke/derive",

README.md

@@ -87,7 +87,7 @@ ICU4X will provide an ECMA-402-compatible API surface in the target client-side
 The [performance benchmarks](docs/process/benchmarking.md) are all run on Ubuntu, and are broken out by component.
 
 * [locid](https://unicode-org.github.io/icu4x-docs/benchmarks/perf/components/locid)
-* [uniset](https://unicode-org.github.io/icu4x-docs/benchmarks/perf/components/uniset)
+* [uniset](https://unicode-org.github.io/icu4x-docs/benchmarks/perf/utils/uniset)
 * [fixed_decimal](https://unicode-org.github.io/icu4x-docs/benchmarks/perf/utils/fixed_decimal)
 * [plurals](https://unicode-org.github.io/icu4x-docs/benchmarks/perf/components/plurals)
 * [datetime](https://unicode-org.github.io/icu4x-docs/benchmarks/perf/components/datetime)

components/icu/Cargo.toml

@@ -66,9 +66,9 @@ version = "0.3"
 path = "../plurals"
 default-features = false
 
-[dependencies.icu_uniset]
+[dependencies.icu_properties]
 version = "0.3"
-path = "../uniset"
+path = "../../components/properties"
 default-features = false
 
 [dependencies.fixed_decimal]
@@ -79,10 +79,11 @@ default-features = false
 [dev-dependencies]
 icu_provider = { version = "0.3", path = "../../provider/core" }
 icu_testdata = { version = "0.3", path = "../../provider/testdata" }
+icu_uniset = { version = "0.3", path = "../../utils/uniset" }
 writeable = { version = "0.2", path = "../../utils/writeable" }
 
 [features]
-std = ["icu_datetime/std", "icu_locid/std", "icu_plurals/std", "icu_uniset/std", "fixed_decimal/std"]
+std = ["icu_datetime/std", "icu_locid/std", "icu_plurals/std", "icu_properties/std", "fixed_decimal/std"]
 default = ["provider_serde"]
 serde = [
     "icu_locid/serde"
@@ -92,5 +93,5 @@ provider_serde = [
     "icu_decimal/provider_serde",
     "icu_locale_canonicalizer/provider_serde",
     "icu_plurals/provider_serde",
-    "icu_uniset/provider_serde",
+    "icu_properties/provider_serde",
 ]

components/icu/examples/tui.rs

@@ -10,8 +10,8 @@
 use icu::datetime::DateTimeFormatOptions;
 use icu::locid::{macros::langid, Locale};
 use icu::plurals::{PluralCategory, PluralRuleType, PluralRules};
-use icu::uniset::UnicodeSetBuilder;
 use icu_datetime::{mock::zoned_datetime::MockZonedDateTime, ZonedDateTimeFormat};
+use icu_uniset::UnicodeSetBuilder;
 use std::env;
 
 fn print<T: AsRef<str>>(_input: T) {

components/icu/src/lib.rs

@@ -367,50 +367,18 @@ pub mod plurals {
     pub use icu_plurals::*;
 }
 
-pub mod uniset {
-    //! Unicode Set operations
+pub mod properties {
+    //! `icu_properties` is a utility crate of the [`ICU4X`] project.
     //!
-    //! This API provides necessary functionality for highly efficient querying of sets of Unicode characters.
+    //! This component provides definitions of [Unicode Properties] and APIs for
+    //! retrieving property data in an appropriate data structure.
     //!
-    //! It is an implementation of the existing [ICU4C UnicodeSet API](https://unicode-org.github.io/icu-docs/apidoc/released/icu4c/classicu_1_1UnicodeSet.html).
+    //! Currently, only binary property APIs are supported, with APIs that return
+    //! a [`UnicodeSet`]. See the [`sets`] module for more details.
     //!
-    //! # Architecture
-    //! ICU4X `UnicodeSet` is split up into independent levels, with [`UnicodeSet`] representing the membership/query API,
-    //! and [`UnicodeSetBuilder`] representing the builder API. A [Properties API](http://userguide.icu-project.org/strings/properties)
-    //! is in future works.
-    //!
-    //! # Examples:
-    //!
-    //! ## Creating a `UnicodeSet`
-    //!
-    //! UnicodeSets are created from either serialized UnicodeSets,
-    //! represented by [inversion lists](http://userguide.icu-project.org/strings/properties),
-    //! the [`UnicodeSetBuilder`], or from the TBA Properties API.
-    //!
-    //! ```
-    //! use icu::uniset::{UnicodeSet, UnicodeSetBuilder};
-    //!
-    //! let mut builder = UnicodeSetBuilder::new();
-    //! builder.add_range(&('A'..'Z'));
-    //! let set: UnicodeSet = builder.build();
-    //!
-    //! assert!(set.contains('A'));
-    //! ```
-    //!
-    //! ## Querying a `UnicodeSet`
-    //!
-    //! Currently, you can check if a character/range of characters exists in the UnicodeSet, or iterate through the characters.
-    //!
-    //! ```
-    //! use icu::uniset::{UnicodeSet, UnicodeSetBuilder};
-    //!
-    //! let mut builder = UnicodeSetBuilder::new();
-    //! builder.add_range(&('A'..'Z'));
-    //! let set: UnicodeSet = builder.build();
-    //!
-    //! assert!(set.contains('A'));
-    //! assert!(set.contains_range(&('A'..='C')));
-    //! assert_eq!(set.iter_chars().next(), Some('A'));
-    //! ```
-    pub use icu_uniset::*;
+    //! [`ICU4X`]: ../icu/index.html
+    //! [Unicode Properties]: https://unicode-org.github.io/icu/userguide/strings/properties.html
+    //! [`UnicodeSet`]: ../../icu_uniset/struct.UnicodeSet.html
+    //! [`sets`]: sets
+    pub use icu_properties::*;
 }

components/properties/Cargo.toml

@@ -0,0 +1,51 @@
+# This file is part of ICU4X. For terms of use, please see the file
+# called LICENSE at the top level of the ICU4X source tree
+# (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ).
+
+[package]
+name = "icu_properties"
+description = "Definitions for Unicode properties"
+version = "0.3.0"
+authors = ["The ICU4X Project Developers"]
+edition = "2018"
+readme = "README.md"
+repository = "https://github.com/unicode-org/icu4x"
+license-file = "LICENSE"
+categories = ["internationalization"]
+# Keep this in sync with other crates unless there are exceptions
+include = [
+    "src/**/*",
+    "examples/**/*",
+    "benches/**/*",
+    "tests/**/*",
+    "Cargo.toml",
+    "LICENSE",
+    "README.md"
+]
+
+[package.metadata.cargo-all-features]
+skip_optional_dependencies = true
+# Bench feature gets tested separately and is only relevant for CI
+denylist = ["bench"]
+
+[package.metadata.docs.rs]
+all-features = true
+
+[dependencies]
+icu_provider = { version = "0.3", path = "../../provider/core", features = ["macros"] }
+icu_uniset = { version = "0.3", path = "../../utils/uniset", features = ["serde"] }
+num_enum = { version = "0.5.4", default-features = false }
+serde = { version = "1.0", default-features = false, features = ["derive", "alloc"], optional = true }
+zerovec = { version = "0.3", path = "../../utils/zerovec", features = ["serde"] }
+
+[dev-dependencies]
+icu = { path = "../../components/icu", default-features = false }
+
+[lib]
+bench = false  # This option is required for Benchmark CI
+path = "src/lib.rs"
+
+[features]
+std = ["icu_provider/std"]
+default = ["provider_serde"]
+provider_serde = ["serde"]

components/properties/README.md

@@ -0,0 +1,18 @@
+# icu_properties [![crates.io](http://meritbadge.herokuapp.com/icu_properties)](https://crates.io/crates/icu_properties)
+
+`icu_properties` is a utility crate of the [`ICU4X`] project.
+
+This component provides definitions of [Unicode Properties] and APIs for
+retrieving property data in an appropriate data structure.
+
+Currently, only binary property APIs are supported, with APIs that return
+a [`UnicodeSet`]. See the [`sets`] module for more details.
+
+[`ICU4X`]: ../icu/index.html
+[Unicode Properties]: https://unicode-org.github.io/icu/userguide/strings/properties.html
+[`UnicodeSet`]: icu_uniset::UnicodeSet
+[`sets`]: crate::sets
+
+## More Information
+
+For more information on development, authorship, contributing etc. please visit [`ICU4X home page`](https://github.com/unicode-org/icu4x).

components/properties/src/lib.rs

@@ -0,0 +1,25 @@
+// This file is part of ICU4X. For terms of use, please see the file
+// called LICENSE at the top level of the ICU4X source tree
+// (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ).
+
+//! `icu_properties` is a utility crate of the [`ICU4X`] project.
+//!
+//! This component provides definitions of [Unicode Properties] and APIs for
+//! retrieving property data in an appropriate data structure.
+//!
+//! Currently, only binary property APIs are supported, with APIs that return
+//! a [`UnicodeSet`]. See the [`sets`] module for more details.
+//!
+//! [`ICU4X`]: ../icu/index.html
+//! [Unicode Properties]: https://unicode-org.github.io/icu/userguide/strings/properties.html
+//! [`UnicodeSet`]: icu_uniset::UnicodeSet
+//! [`sets`]: crate::sets
+
+#![no_std]
+
+mod props;
+pub mod provider;
+pub mod sets;
+mod ule;
+
+pub use props::*;

components/uniset/src/enum_props.rs → components/properties/src/props.rs

@@ -8,7 +8,7 @@ use num_enum::{TryFromPrimitive, UnsafeFromPrimitive};
 
 /// Selection constants for Unicode properties.
 /// These constants are used to select one of the Unicode properties.
-/// See UProperty in ICU4C.
+/// See `UProperty` in ICU4C.
 #[derive(Clone, PartialEq, Debug)]
 #[non_exhaustive]
 pub enum EnumeratedProperty {
@@ -21,8 +21,8 @@ pub enum EnumeratedProperty {
 }
 
 /// Enumerated Unicode general category types.
-/// GeneralSubcategory only supports specific subcategories (eg UppercaseLetter).
-/// It does not support grouped categories (eg Letter). For grouped categories, use GeneralCategory.
+/// GeneralSubcategory only supports specific subcategories (eg `UppercaseLetter`).
+/// It does not support grouped categories (eg `Letter`). For grouped categories, use [`GeneralCategory`].
 #[derive(Copy, Clone, PartialEq, Debug, TryFromPrimitive, UnsafeFromPrimitive)]
 #[repr(u8)]
 pub enum GeneralSubcategory {
@@ -99,8 +99,8 @@ pub enum GeneralSubcategory {
 /// The discriminants correspond to the U_GC_XX_MASK constants in ICU4C.
 /// Unlike GeneralSubcategory, this supports groups of general categories: for example, `Letter`
 /// is the union of `UppercaseLetter`, `LowercaseLetter`, etc...
-/// See https://www.unicode.org/reports/tr44/ .
-/// See UCharCategory and U_GET_GC_MASK in ICU4C.
+/// See <https://www.unicode.org/reports/tr44/> .
+/// See `UCharCategory` and `U_GET_GC_MASK` in ICU4C.
 #[derive(Copy, Clone, PartialEq, Debug, Eq)]
 #[repr(transparent)]
 pub struct GeneralCategory(pub(crate) u32);
@@ -231,7 +231,7 @@ impl From<GeneralSubcategory> for GeneralCategory {
 /// a particular subset of scripts will be in more than one Script_Extensions set.
 /// For example, DEVANAGARI DIGIT NINE has Script=Devanagari, but is also in the
 /// Script_Extensions set for Dogra, Kaithi, and Mahajani.
-/// For more information, see UAX #24: http://www.unicode.org/reports/tr24/.
+/// For more information, see UAX #24: <http://www.unicode.org/reports/tr24/>.
 /// See UScriptCode in ICU4C.
 #[derive(Copy, Clone, Debug, Eq, PartialEq)]
 #[repr(transparent)]

components/uniset/src/provider.rs → components/properties/src/provider.rs

@@ -6,9 +6,9 @@
 //!
 //! Read more about data providers: [`icu_provider`]
 
-use crate::builder::UnicodeSetBuilder;
-use crate::uniset::UnicodeSet;
 use icu_provider::yoke::{self, *};
+use icu_uniset::UnicodeSet;
+use icu_uniset::UnicodeSetBuilder;
 
 //
 // resource key structs - the structs used directly by users of data provider