unicode-intervals 0.3.0__tar.gz → 0.3.1__tar.gz

{unicode_intervals-0.3.0 → unicode_intervals-0.3.1}/Cargo.lock

@@ -991,7 +991,7 @@ checksum = "e6e4313cd5fcd3dad5cafa179702e2b244f760991f45397d14d4ebf38247da75"
 
 [[package]]
 name = "unicode-intervals"
-version = "0.3.0"
+version = "0.3.1"
 dependencies = [
  "codspeed-criterion-compat",
  "test-case",
@@ -999,7 +999,7 @@ dependencies = [
 
 [[package]]
 name = "unicode-intervals-py"
-version = "0.3.0"
+version = "0.3.1"
 dependencies = [
  "built",
  "pyo3",

{unicode_intervals-0.3.0 → unicode_intervals-0.3.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: unicode_intervals
-Version: 0.3.0
+Version: 0.3.1
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
@@ -26,7 +26,6 @@ Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
 # unicode-intervals
 
 Search Unicode code point intervals by category, codepoint range, and characters.
-Fast Rust core (via PyO3), output-compatible with Hypothesis's `charmap`.
 
 ## Installation
 
@@ -60,7 +59,7 @@ assert unicode_intervals.query(
     categories=["Lu"], min_codepoint=65, max_codepoint=90, exclude_characters="AEIOU"
 ) == ((66, 68), (70, 72), (74, 78), (80, 84), (86, 90))
 
-# Exclude whole categories (an extra over charmap): drop digits from U+0030..U+003A
+# Exclude whole categories: drop digits from U+0030..U+003A
 assert unicode_intervals.query(
     exclude_categories=["Nd"], min_codepoint=48, max_codepoint=58
 ) == ((58, 58),)

{unicode_intervals-0.3.0/crates/unicode-intervals-py → unicode_intervals-0.3.1}/README.md

@@ -1,7 +1,6 @@
 # unicode-intervals
 
 Search Unicode code point intervals by category, codepoint range, and characters.
-Fast Rust core (via PyO3), output-compatible with Hypothesis's `charmap`.
 
 ## Installation
 
@@ -35,7 +34,7 @@ assert unicode_intervals.query(
     categories=["Lu"], min_codepoint=65, max_codepoint=90, exclude_characters="AEIOU"
 ) == ((66, 68), (70, 72), (74, 78), (80, 84), (86, 90))
 
-# Exclude whole categories (an extra over charmap): drop digits from U+0030..U+003A
+# Exclude whole categories: drop digits from U+0030..U+003A
 assert unicode_intervals.query(
     exclude_categories=["Nd"], min_codepoint=48, max_codepoint=58
 ) == ((58, 58),)

{unicode_intervals-0.3.0 → unicode_intervals-0.3.1}/crates/unicode-intervals/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "unicode-intervals"
-version = "0.3.0"
+version = "0.3.1"
 authors = ["Dmitry Dygalo <dmitry@dygalo.dev>"]
 edition = "2021"
 documentation = "https://github.com/Stranger6667/unicode-intervals"

{unicode_intervals-0.3.0 → unicode_intervals-0.3.1}/crates/unicode-intervals/benches/bench.rs

@@ -1,12 +1,10 @@
-use criterion::{black_box, criterion_group, criterion_main, Criterion};
+use criterion::{black_box, criterion_group, criterion_main, BatchSize, Criterion};
 use unicode_intervals::{internals, UnicodeCategory, UnicodeCategorySet, UnicodeVersion};
 
 fn version(c: &mut Criterion) {
     let version = black_box(UnicodeVersion::V15_0_0);
     c.bench_function("version - normalized_categories", |b| {
-        b.iter(|| {
-            let _ = version.normalized_categories();
-        })
+        b.iter(|| version.normalized_categories())
     });
 }
 
@@ -32,27 +30,42 @@ fn intervals(c: &mut Criterion) {
     c.bench_function("intervals - subtract sparse", |b| {
         b.iter(|| internals::intervals::subtract(large.to_vec(), &few))
     });
+    // Concatenated, sorted-per-category slices that interleave across the codepoint space:
+    // the input shape `query` feeds to `merge` for a multi-category set.
+    let multi = internals::query::intervals_for_set(
+        UnicodeVersion::V15_0_0,
+        UnicodeCategory::Lu
+            | UnicodeCategory::Ll
+            | UnicodeCategory::Lo
+            | UnicodeCategory::Nd
+            | UnicodeCategory::Po
+            | UnicodeCategory::Sm,
+    )
+    .into_owned();
+    c.bench_function("intervals - merge multi-category", |b| {
+        b.iter_batched(
+            || multi.clone(),
+            |mut intervals| internals::intervals::merge(&mut intervals),
+            BatchSize::SmallInput,
+        )
+    });
 }
 
 fn categories(c: &mut Criterion) {
     let all_categories = black_box(UnicodeCategorySet::all());
     c.bench_function("categories - set - display - all", |b| {
-        b.iter(|| {
-            let _ = all_categories.to_string();
-        })
+        b.iter(|| all_categories.to_string())
     });
     let few_categories = UnicodeCategory::Lm | UnicodeCategory::Sk | UnicodeCategory::Zl;
     c.bench_function("categories - set - display - few", |b| {
-        b.iter(|| {
-            let _ = few_categories.to_string();
-        })
+        b.iter(|| few_categories.to_string())
     });
     c.bench_function("categories - merge", |b| {
         b.iter(|| {
-            let _ = internals::categories::merge(
+            internals::categories::merge(
                 Some(all_categories),
                 black_box(UnicodeCategory::Lu.into()),
-            );
+            )
         })
     });
 }
@@ -82,60 +95,98 @@ fn query(c: &mut Criterion) {
             )
         })
     });
+    // Full range + multiple categories: hits the concat + sort/merge path over the whole set.
+    c.bench_function("query - multiple categories - full range", |b| {
+        b.iter(|| {
+            version
+                .query()
+                .include_categories(black_box(
+                    UnicodeCategory::Lu
+                        | UnicodeCategory::Ll
+                        | UnicodeCategory::Lo
+                        | UnicodeCategory::Nd
+                        | UnicodeCategory::Po
+                        | UnicodeCategory::Sm,
+                ))
+                .intervals()
+        })
+    });
     let exclude_categories = black_box(UnicodeCategory::Lu);
     let min_codepoint = black_box(0);
     let max_codepoint = black_box(128);
     c.bench_function("query - top level - only codepoints", |b| {
         b.iter(|| {
-            let _ = version
+            version
                 .query()
                 .exclude_categories(exclude_categories)
                 .min_codepoint(min_codepoint)
                 .max_codepoint(max_codepoint)
-                .intervals();
+                .intervals()
         })
     });
     c.bench_function("query - top level - exclude chars", |b| {
         b.iter(|| {
-            let _ = version
+            version
                 .query()
                 .exclude_categories(exclude_categories)
                 .exclude_characters(black_box("A@т"))
                 .min_codepoint(min_codepoint)
                 .max_codepoint(max_codepoint)
-                .intervals();
+                .intervals()
         })
     });
     c.bench_function("query - top level - include and exclude chars", |b| {
         b.iter(|| {
-            let _ = version
+            version
                 .query()
                 .exclude_categories(exclude_categories)
                 .include_characters(black_box("0123456789"))
                 .exclude_characters(black_box("QWERTYUIOP"))
                 .min_codepoint(min_codepoint)
                 .max_codepoint(max_codepoint)
-                .intervals();
+                .intervals()
         })
     });
     c.bench_function("query - top level - include only", |b| {
         b.iter(|| {
-            let _ = version
+            version
                 .query()
                 .include_categories(UnicodeCategory::Ll)
                 .include_characters("ABC")
                 .min_codepoint(0)
                 .max_codepoint(50)
-                .intervals();
+                .intervals()
         })
     });
     c.bench_function("query - single category - codepoint range", |b| {
         b.iter(|| {
-            let _ = version
+            version
                 .query()
                 .include_categories(black_box(UnicodeCategory::Lo))
                 .max_codepoint(black_box(0x1_0000))
-                .intervals();
+                .intervals()
+        })
+    });
+    // High `min_codepoint`: most of the category's intervals sit below the range and are
+    // skipped by the binary search in `extend_clamped`.
+    c.bench_function("query - single category - high range", |b| {
+        b.iter(|| {
+            version
+                .query()
+                .include_categories(black_box(UnicodeCategory::Lo))
+                .min_codepoint(black_box(0x2_0000))
+                .max_codepoint(black_box(0x2_4000))
+                .intervals()
+        })
+    });
+    c.bench_function("query - multiple categories - high range", |b| {
+        b.iter(|| {
+            version
+                .query()
+                .include_categories(black_box(UnicodeCategory::Lu | UnicodeCategory::Ll))
+                .min_codepoint(black_box(0x1_0000))
+                .max_codepoint(black_box(0x1_0500))
+                .intervals()
         })
     });
     let interval_set = UnicodeVersion::V15_0_0
@@ -144,9 +195,7 @@ fn query(c: &mut Criterion) {
         .interval_set()
         .expect("Invalid query input");
     c.bench_function("query - interval set - codepoint_at", |b| {
-        b.iter(|| {
-            let _ = interval_set.codepoint_at(27);
-        })
+        b.iter(|| interval_set.codepoint_at(27))
     });
 }