Move icu_uniset into collections component as codepointinvlist module

.github/workflows/build-test.yml

@@ -630,7 +630,7 @@ jobs:
           - components/locid
           - components/plurals
           - components/datetime
-          - utils/uniset
+          - components/collections
           - utils/fixed_decimal
 
 
@@ -752,7 +752,7 @@ jobs:
               icu_locid/filter_langids
               icu_plurals/unread_emails
               icu_plurals/elevator_floors
-              icu_uniset/unicode_bmp_blocks_selector
+              icu_collections/unicode_bmp_blocks_selector
               fixed_decimal/permyriad
               writeable/writeable_message
               litemap/language_names_lite_map

Cargo.toml

@@ -46,7 +46,6 @@ members = [
     "utils/pattern",
     "utils/tinystr",
     "utils/tzif",
-    "utils/uniset",
     "utils/writeable",
     "utils/yoke",
     "utils/yoke/derive",

README.md

@@ -84,7 +84,7 @@ 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/utils/uniset)
+* [collections](https://unicode-org.github.io/icu4x-docs/benchmarks/perf/components/collections)
 * [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/collator/Cargo.toml

@@ -40,7 +40,6 @@ icu_provider = { version = "0.6", path = "../../provider/core", features = ["mac
 icu_locid = { version = "0.6", path = "../../components/locid" }
 icu_normalizer = { version = "0.6", path = "../../components/normalizer" }
 icu_properties = { version = "0.6", path = "../../components/properties" }
-icu_uniset = { version = "0.5", path = "../../utils/uniset" }
 serde = { version = "1.0", default-features = false, features = ["derive", "alloc"], optional = true }
 zerovec = { version = "0.7", path = "../../utils/zerovec", features = ["serde"] }
 utf8_iter = "1.0"
@@ -59,5 +58,5 @@ bench = false  # This option is required for Benchmark CI
 
 [features]
 default = []
-serde = ["dep:serde", "zerovec/serde", "icu_char16trie/serde", "icu_properties/serde", "icu_normalizer/serde", "icu_uniset/serde", "icu_collections/serde"]
-datagen = ["serde", "databake", "zerovec/databake", "icu_char16trie/databake", "icu_properties/databake", "icu_normalizer/databake", "icu_uniset/databake", "icu_collections/databake"]
+serde = ["dep:serde", "zerovec/serde", "icu_char16trie/serde", "icu_properties/serde", "icu_normalizer/serde", "icu_collections/serde"]
+datagen = ["serde", "databake", "zerovec/databake", "icu_char16trie/databake", "icu_properties/databake", "icu_normalizer/databake", "icu_collections/databake"]

components/collections/Cargo.toml

@@ -33,7 +33,6 @@ all-features = true
 
 [dependencies]
 displaydoc = { version = "0.2.3", default-features = false }
-icu_uniset = { version = "0.5.0", path = "../../utils/uniset" }
 yoke = { version = "0.6.0", path = "../../utils/yoke", features = ["derive"] }
 zerofrom = { version = "0.1.0", path = "../../utils/zerofrom", features = ["derive"] }
 zerovec = { version = "0.7", path = "../../utils/zerovec", features = ["yoke"] }
@@ -45,7 +44,9 @@ databake = { version = "0.1.0", path = "../../utils/databake", features = ["deri
 postcard = { version = "1.0.0-alpha.4", features = ["alloc"] }
 toml = "0.5"
 serde = { version = "1.0", features = ["derive"] }
+serde_json = "1.0"
 criterion = "0.3.4"
+icu_benchmark_macros = { version = "0.6", path = "../../tools/benchmark/macros" }
 iai = "0.1.1"
 icu = { path = "../icu", default-features = false }
 
@@ -55,15 +56,26 @@ path = "src/lib.rs"
 
 [features]
 default = []
-serde = ["dep:serde", "zerovec/serde", "icu_uniset/serde"]
+serde = ["dep:serde", "zerovec/serde"]
 bench = []
 databake = ["dep:databake", "zerovec/databake"]
 
-# [[bench]]
-# name = "codepointtrie"
-# harness = false
-# 
-# [[bench]]
-# name = "iai_cpt"
-# harness = false
-# required-features = ["bench"]
+[[bench]]
+name = "codepointtrie"
+harness = false
+path = "src/codepointtrie/benches/codepointtrie.rs"
+
+[[bench]]
+name = "iai_cpt"
+harness = false
+required-features = ["bench"]
+path = "src/codepointtrie/benches/iai_cpt.rs"
+
+[[bench]]
+name = "inv_list"
+harness = false
+path = "src/codepointinvlist/benches/inv_list.rs"
+
+[[example]]
+name = "unicode_bmp_blocks_selector"
+path = "src/codepointinvlist/examples/unicode_bmp_blocks_selector.rs"

components/collections/README.md

@@ -5,10 +5,13 @@ Collections of API for use in ICU.
 This module is published as its own crate ([`icu_collections`](https://docs.rs/icu_collections/latest/icu_collections/))
 and as part of the [`icu`](https://docs.rs/icu/latest/icu/) crate. See the latter for more details on the ICU4X project.
 
-ICU4X [`CodePointTrie`](crate::codepointtrie::CodePointTrie) is designed to provide a read-only view of CodePointTrie data that is exported
+ICU4X [`CodePointTrie`](crate::codepointtrie::CodePointTrie) provides a read-only view of CodePointTrie data that is exported
 from ICU4C. Detailed information about the design of the data structure can be found in the documentation
 for the [`CodePointTrie`](crate::codepointtrie::CodePointTrie) struct.
 
+ICU4X [`CodePointInversionList`](`crate::codepointinvlist::CodePointInversionList`) provides necessary functionality for highly efficient querying of sets of Unicode characters.
+It is an implementation of the existing [ICU4C UnicodeSet API](https://unicode-org.github.io/icu-docs/apidoc/released/icu4c/classicu_1_1UnicodeSet.html).
+
 
 ## More Information
 

utils/uniset/README.md → components/collections/src/codepointinvlist/README.md

@@ -19,7 +19,7 @@ represented by [inversion lists](http://userguide.icu-project.org/strings/proper
 the [`CodePointInversionListBuilder`], or from the Properties API.
 
 ```rust
-use icu_uniset::{CodePointInversionList, CodePointInversionListBuilder};
+use icu_collections::codepointinvlist::{CodePointInversionList, CodePointInversionListBuilder};
 
 let mut builder = CodePointInversionListBuilder::new();
 builder.add_range(&('A'..'Z'));
@@ -33,7 +33,7 @@ assert!(set.contains('A'));
 Currently, you can check if a character/range of characters exists in the [`CodePointInversionList`], or iterate through the characters.
 
 ```rust
-use icu_uniset::{CodePointInversionList, CodePointInversionListBuilder};
+use icu_collections::codepointinvlist::{CodePointInversionList, CodePointInversionListBuilder};
 
 let mut builder = CodePointInversionListBuilder::new();
 builder.add_range(&('A'..'Z'));

utils/uniset/benches/inv_list.rs → components/collections/src/codepointinvlist/benches/inv_list.rs

@@ -3,7 +3,7 @@
 // (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ).
 
 use criterion::{criterion_group, criterion_main, Criterion};
-use icu_uniset::CodePointInversionList;
+use icu_collections::codepointinvlist::CodePointInversionList;
 use std::char;
 
 fn uniset_bench(c: &mut Criterion) {

utils/uniset/src/builder.rs → components/collections/src/codepointinvlist/builder.rs

@@ -6,7 +6,7 @@ use alloc::vec;
 use alloc::vec::Vec;
 use core::{char, cmp::Ordering, ops::RangeBounds};
 
-use crate::{uniset::CodePointInversionList, utils::deconstruct_range};
+use crate::codepointinvlist::{utils::deconstruct_range, CodePointInversionList};
 use zerovec::{ule::AsULE, ZeroVec};
 
 /// A builder for [`CodePointInversionList`].
@@ -90,7 +90,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::CodePointInversionListBuilder;
+    /// use icu_collections::codepointinvlist::CodePointInversionListBuilder;
     /// let mut builder = CodePointInversionListBuilder::new();
     /// builder.add_char('a');
     /// let check = builder.build();
@@ -111,7 +111,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::CodePointInversionListBuilder;
+    /// use icu_collections::codepointinvlist::CodePointInversionListBuilder;
     /// let mut builder = CodePointInversionListBuilder::new();
     /// builder.add_u32(0x41);
     /// let check = builder.build();
@@ -129,7 +129,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::CodePointInversionListBuilder;
+    /// use icu_collections::codepointinvlist::CodePointInversionListBuilder;
     /// let mut builder = CodePointInversionListBuilder::new();
     /// builder.add_range(&('A'..='Z'));
     /// let check = builder.build();
@@ -145,7 +145,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::CodePointInversionListBuilder;
+    /// use icu_collections::codepointinvlist::CodePointInversionListBuilder;
     /// let mut builder = CodePointInversionListBuilder::new();
     /// builder.add_range_u32(&(0xd800..=0xdfff));
     /// let check = builder.build();
@@ -163,7 +163,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::{CodePointInversionList, CodePointInversionListBuilder};
+    /// use icu_collections::codepointinvlist::{CodePointInversionList, CodePointInversionListBuilder};
     /// let mut builder = CodePointInversionListBuilder::new();
     /// let set = CodePointInversionList::from_inversion_list_slice(&[0x41, 0x4C]).unwrap();
     /// builder.add_set(&set);
@@ -210,7 +210,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::CodePointInversionListBuilder;
+    /// use icu_collections::codepointinvlist::CodePointInversionListBuilder;
     /// let mut builder = CodePointInversionListBuilder::new();
     /// builder.add_range(&('A'..='Z'));
     /// builder.remove_char('A');
@@ -226,7 +226,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::CodePointInversionListBuilder;
+    /// use icu_collections::codepointinvlist::CodePointInversionListBuilder;
     /// let mut builder = CodePointInversionListBuilder::new();
     /// builder.add_range(&('A'..='Z'));
     /// builder.remove_range(&('A'..='C'));
@@ -242,7 +242,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::{CodePointInversionList, CodePointInversionListBuilder};
+    /// use icu_collections::codepointinvlist::{CodePointInversionList, CodePointInversionListBuilder};
     /// let mut builder = CodePointInversionListBuilder::new();
     /// let set = CodePointInversionList::from_inversion_list_slice(&[0x41, 0x46]).unwrap();
     /// builder.add_range(&('A'..='Z'));
@@ -271,7 +271,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::CodePointInversionListBuilder;
+    /// use icu_collections::codepointinvlist::CodePointInversionListBuilder;
     /// let mut builder = CodePointInversionListBuilder::new();
     /// builder.add_range(&('A'..='Z'));
     /// builder.retain_char('A');
@@ -291,7 +291,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::CodePointInversionListBuilder;
+    /// use icu_collections::codepointinvlist::CodePointInversionListBuilder;
     /// let mut builder = CodePointInversionListBuilder::new();
     /// builder.add_range(&('A'..='Z'));
     /// builder.retain_range(&('A'..='B'));
@@ -312,7 +312,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::{CodePointInversionList, CodePointInversionListBuilder};
+    /// use icu_collections::codepointinvlist::{CodePointInversionList, CodePointInversionListBuilder};
     /// let mut builder = CodePointInversionListBuilder::new();
     /// let set = CodePointInversionList::from_inversion_list_slice(&[65, 70]).unwrap();
     /// builder.add_range(&('A'..='Z'));
@@ -379,7 +379,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::{CodePointInversionList, CodePointInversionListBuilder};
+    /// use icu_collections::codepointinvlist::{CodePointInversionList, CodePointInversionListBuilder};
     /// let mut builder = CodePointInversionListBuilder::new();
     /// let set =
     ///     CodePointInversionList::from_inversion_list_slice(&[0x0, 0x41, 0x46, (std::char::MAX as u32) + 1])
@@ -413,7 +413,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::CodePointInversionListBuilder;
+    /// use icu_collections::codepointinvlist::CodePointInversionListBuilder;
     /// let mut builder = CodePointInversionListBuilder::new();
     /// builder.add_range(&('A'..='D'));
     /// builder.complement_char('A');
@@ -434,7 +434,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::CodePointInversionListBuilder;
+    /// use icu_collections::codepointinvlist::CodePointInversionListBuilder;
     /// let mut builder = CodePointInversionListBuilder::new();
     /// builder.add_range(&('A'..='D'));
     /// builder.complement_range(&('C'..='F'));
@@ -454,7 +454,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::{CodePointInversionList, CodePointInversionListBuilder};
+    /// use icu_collections::codepointinvlist::{CodePointInversionList, CodePointInversionListBuilder};
     /// let mut builder = CodePointInversionListBuilder::new();
     /// let set = CodePointInversionList::from_inversion_list_slice(&[0x41, 0x46, 0x4B, 0x5A]).unwrap();
     /// builder.add_range(&('C'..='N')); // 67 - 78
@@ -473,7 +473,7 @@ impl CodePointInversionListBuilder {
     /// # Examples
     ///
     /// ```
-    /// use icu_uniset::{CodePointInversionList, CodePointInversionListBuilder};
+    /// use icu_collections::codepointinvlist::{CodePointInversionList, CodePointInversionListBuilder};
     /// let mut builder = CodePointInversionListBuilder::new();
     /// let check = builder.build();
     /// assert!(check.is_empty());