autocorrect-rb 2.1.2-x86_64-darwin → 2.2.0-x86_64-darwin

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ac2670b585d58cdc13556aeec1d38e7bcfa244f5ff6aae457e762c58a508d57e
-  data.tar.gz: 42714a112f82c5be94237b3798467baca0542e5ca653b9efcd1d3c05f9e6eec6
+  metadata.gz: a0ec060a8fe207e43dc3885963bd225b5ffdc5ef4b624e3bea0758f53c7ee579
+  data.tar.gz: 2c635f8ce87b8f31d366c087a4d27221b8d88d02416fac13a16a67d7013245aa
 SHA512:
-  metadata.gz: e5c7e13b2765d28ac5bbfeb7fcd3f27baa9e42ae20ba4f629c9f03e73759839ba92e8357123c5f53970f71a7c9beea166b4eee5ef50bc86fb4ec8deabe5bbf49
-  data.tar.gz: 9d803fe8ae109576f400b8c0082c7472ee0f6361835edee8f45e2b50a9fa3d5f972ed010b75ea58b770e1c5b601e0d4392d4c1141b6028f519a64d4928fd3bc2
+  metadata.gz: c41437d796c6c25eadb86e2ae971a3d209cf7ce51f4ba98133aa320acd6985cbb087b16a7f658d2b478e8a3e9a882575351a6b69eb7baf518395f00d1f049159
+  data.tar.gz: 6f8539d02b45841faa316287ffdfde30f3a48b068cea11b1f224bd6602a3523e480946c3ccdaa60ec3be59eb0213e0a610c07ede9649574f98a97a22080cdffd

data/README.md

@@ -18,7 +18,7 @@ $ bundle add autocorrect-rb
 ## Usage
 
 ```rb
-require('autocorrect-rb');
+require('autocorrect-rb')
 
 out = AutoCorrect.format('Hello你好.')
 puts out
@@ -28,7 +28,7 @@ out = AutoCorrect.format_for("title = 'Hello你好。'", 'rb')
 puts out
 # title = 'Hello 你好。'
 
-result = AutoCorrect.lint_lor("title = 'Hello你好。'", 'rb')
+result = AutoCorrect.lint_for("title = 'Hello你好。'", 'rb')
 puts result
 # {
 #   filepath: 'rb',
@@ -38,3 +38,57 @@ puts result
 #   error: ''
 # }
 ```
+
+## Benchmarks
+
+🎊 Rust version is 3x faster than the Ruby (pure) version.
+
+> NOTE: In this case Rust version has more complex rules.
+
+### Rust version
+
+```bash
+Warming up --------------------------------------
+     format 50 chars    11.348k i/100ms
+    format 100 chars     6.033k i/100ms
+    format 400 chars     1.772k i/100ms
+         format_html   545.000  i/100ms
+Calculating -------------------------------------
+     format 50 chars    111.904k (± 3.1%) i/s -    567.400k in   5.075674s
+    format 100 chars     58.684k (± 2.1%) i/s -    295.617k in   5.039837s
+    format 400 chars     17.266k (± 2.9%) i/s -     86.828k in   5.033234s
+         format_html      5.448k (± 1.5%) i/s -     27.250k in   5.002853s
+```
+
+### Pure [Ruby version](https://rubygems.org/gems/auto-correct/versions/1.0.0) result:
+
+```bash
+Warming up --------------------------------------
+     format 50 chars     3.167k i/100ms
+    format 100 chars     1.588k i/100ms
+    format 400 chars   496.000  i/100ms
+         format_html   166.000  i/100ms
+Calculating -------------------------------------
+     format 50 chars     31.589k (± 2.5%) i/s -    158.350k in   5.016131s
+    format 100 chars     16.122k (± 1.2%) i/s -     80.988k in   5.024082s
+    format 400 chars      4.946k (± 2.6%) i/s -     24.800k in   5.017711s
+         format_html      1.659k (± 1.7%) i/s -      8.300k in   5.003164s
+```
+
+## Know issues
+
+Bundler install error:
+
+```
+Could not find gem 'autocorrect-rb' with platform 'ruby' in rubygems repository https://rubygems.org/ or installed locally.
+```
+
+To fix this you can run:
+
+```bash
+$ bundle lock --remove-platform ruby
+```
+
+Because of autocorrect-rb not release the gem for `platform: ruby`, but your `Gemfile.lock` specialed that. This command will remove `ruby` platform from your `Gemfile.lock`
+
+

data/Rakefile

@@ -7,6 +7,7 @@ rescue LoadError
 end
 
 require "bundler/gem_tasks"
+require "rubygems/package_task"
 require "rake/testtask"
 require "rake/extensiontask"
 require "bundler"
@@ -19,22 +20,57 @@ CROSS_PLATFORMS = %w[
   x86_64-linux
 ]
 
-spec = Bundler.load_gemspec("autocorrect-rb.gemspec")
-
 Rake::TestTask.new(:test) do |t|
   t.libs << "test"
   t.libs << "lib"
   t.test_files = FileList["test/**/*_test.rb"]
 end
 
-Rake::ExtensionTask.new("autocorrect") do |ext|
-  ext.gem_spec = spec
+spec = Bundler.load_gemspec("autocorrect-rb.gemspec")
+
+# Make a autocorrect-rb-0.0.0.gem
+# Disable release gem for platform: ruby, because autocorrect rust source build is complex.
+# Gem::PackageTask.new(spec).define
+
+# Make a native gem, eg. autocorrect-rb-0.0.0-x86_64-linux.gem
+Rake::ExtensionTask.new("autocorrect", spec) do |ext|
+  # Compile output autocorrect.so into lib/autocorrect/autocorrect.so
   ext.lib_dir = "lib/autocorrect"
   ext.source_pattern = "*.{rs,toml}"
   ext.cross_compile = true
   ext.cross_platform = CROSS_PLATFORMS
 end
 
-task build: :compile
+task :bench do
+  require "benchmark/ips"
+  require "./lib/autocorrect-rb"
+
+  html = open("./test/fixtures/example.txt").read
+
+  Benchmark.ips do |x|
+    x.report("format 50 chars") do |n|
+      n.times do
+        AutoCorrect.format("【野村：重申吉利汽车(00175)“买入”评级 上调目标价至17.9港元】智通财经APP获悉，野村发布报告称")
+      end
+    end
+    x.report("format 100 chars") do |n|
+      n.times do
+        AutoCorrect.format("【野村：重申吉利汽车(00175)“买入”评级 上调目标价至17.9港元】智通财经APP获悉，野村发布报告称，【野村：重申吉利汽车(00175)“买入”评级 上调目标价至17.9港元】智通财经APP获悉，野村发布报告称")
+      end
+    end
+    x.report("format 400 chars") do |n|
+      n.times do
+        AutoCorrect.format("【野村：重申吉利汽车(00175)“买入”评级 上调目标价至17.9港元】智通财经APP获悉，野村发布报告称，上调吉利汽车(00175)目标价12.58%，由15.9港元升至17.9港元，并维持吉汽为行业首选股，重申对其“买入”评级，坚信吉汽长远可成为行业赢家。 该行称，随着公司销量持续复苏及产品组合改善，预计今年销量可达148万辆，同比升9%，较公司原定目标销量141万辆为高。 该行又称称，上调公司今明两年每股盈利预测各13%及升毛利率0.1个百分点，以反映销量较预期高2%及产品组合改善，主要是由领克品牌带动。公司自去年8月开始已持续投资领克品牌及进行市场推广，带动领克销量环比有所改变，预期今明两年领克将占整体销量的11%及14%。 该行表示，由于低端国产车品牌在欠缺新车款及科技下，行业整合度将提升。另外，公司从去年第二季到12月为止，一直都积极推动经销商去库存，这将有利公司今年利润率复苏。")
+      end
+    end
+
+    x.report("format_html") do |n|
+      n.times do
+        AutoCorrect.format_for(html, "text.html")
+      end
+    end
+  end
+end
+
 task test: :compile
-task default: %i[compile test]
+task default: %i[test]

data/ext/autocorrect/Cargo.toml

@@ -0,0 +1,13 @@
+[package]
+edition = "2021"
+name = "autocorrect-rb"
+version = "2.1.1"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+[lib]
+crate-type = ["cdylib"]
+name = "autocorrect"
+
+[dependencies]
+autocorrect = {path = "../../../autocorrect", version = ">=1.0.0"}
+magnus = "0.3"

data/ext/autocorrect/extconf.rb

@@ -0,0 +1,4 @@
+require "mkmf"
+require "rb_sys/mkmf"
+
+create_rust_makefile("autocorrect/autocorrect")

data/ext/autocorrect/src/lib.rs

@@ -0,0 +1,125 @@
+use magnus::{define_class, function, Error, Object};
+
+#[derive(Debug, Clone)]
+pub struct LineResult {
+    line: usize,
+    col: usize,
+    new: String,
+    old: String,
+    severity: usize,
+}
+
+impl LineResult {
+    pub fn line(&self) -> usize {
+        self.line
+    }
+
+    pub fn col(&self) -> usize {
+        self.col
+    }
+
+    pub fn get_new(&self) -> String {
+        self.new.clone()
+    }
+
+    pub fn old(&self) -> String {
+        self.old.clone()
+    }
+
+    pub fn severity(&self) -> usize {
+        self.severity
+    }
+
+    pub fn inspect(&self) -> String {
+        format!("{:?}", self)
+    }
+
+    pub fn to_hash(&self) -> Result<magnus::RHash, Error> {
+        let hash = magnus::RHash::new();
+        hash.aset("line", self.line())?;
+        hash.aset("col", self.col())?;
+        hash.aset("new", self.get_new())?;
+        hash.aset("old", self.old())?;
+        hash.aset("severity", self.severity())?;
+        Ok(hash)
+    }
+}
+
+#[derive(Debug, Clone)]
+pub struct LintResult {
+    pub filepath: String,
+    pub lines: Vec<LineResult>,
+    pub error: String,
+}
+
+impl LintResult {
+    pub fn filepath(&self) -> String {
+        self.filepath.clone()
+    }
+
+    pub fn lines(&self) -> Vec<LineResult> {
+        self.lines.clone()
+    }
+
+    pub fn error(&self) -> String {
+        self.error.clone()
+    }
+
+    pub fn inspect(&self) -> String {
+        format!("{:?}", self)
+    }
+
+    pub fn to_hash(&self) -> Result<magnus::RHash, Error> {
+        let hash = magnus::RHash::new();
+        hash.aset("filepath", self.filepath())?;
+        hash.aset(
+            "lines",
+            self.lines()
+                .iter()
+                .map(|l| l.to_hash().unwrap())
+                .collect::<Vec<magnus::RHash>>(),
+        )?;
+        hash.aset("error", self.error())?;
+        Ok(hash)
+    }
+}
+
+pub fn format(input: String) -> String {
+    autocorrect::format(&input)
+}
+
+pub fn format_for(input: String, filename_or_ext: String) -> String {
+    autocorrect::format_for(&input, &filename_or_ext).out
+}
+
+pub fn lint_for(input: String, filename_or_ext: String) -> magnus::RHash {
+    let result = autocorrect::lint_for(&input, &filename_or_ext);
+
+    LintResult {
+        filepath: filename_or_ext,
+        lines: result
+            .lines
+            .iter()
+            .map(|l| LineResult {
+                line: l.line,
+                col: l.col,
+                new: l.new.clone(),
+                old: l.old.clone(),
+                severity: l.severity.clone() as usize,
+            })
+            .collect::<_>(),
+        error: result.error,
+    }
+    .to_hash()
+    .unwrap()
+}
+
+#[magnus::init(name = "autocorrect")]
+fn init() -> Result<(), Error> {
+    let class = define_class("AutoCorrect", Default::default())?;
+    class.define_singleton_method("format", function!(format, 1))?;
+    class.define_singleton_method("format_for", function!(format_for, 2))?;
+    class.define_singleton_method("lint_for", function!(lint_for, 2))?;
+
+    Ok(())
+}

data/lib/autocorrect-rb.rb

@@ -3,10 +3,5 @@ begin
   ruby_version = /(\d+\.\d+)/.match(::RUBY_VERSION)
   require_relative "autocorrect/#{ruby_version}/autocorrect"
 rescue LoadError
-  # fall back to the extension compiled upon installation.
-  # use "require" instead of "require_relative" because non-native gems will place C extension files
-  # in Gem::BasicSpecification#extension_dir after compilation (during normal installation), which
-  # is in $LOAD_PATH but not necessarily relative to this file
-  # (see https://github.com/sparklemotion/nokogiri/issues/2300 for more)
   require "autocorrect/autocorrect"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: autocorrect-rb
 version: !ruby/object:Gem::Version
-  version: 2.1.2
+  version: 2.2.0
 platform: x86_64-darwin
 authors:
 - Jason Lee
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2022-10-12 00:00:00.000000000 Z
+date: 2022-10-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rb_sys
@@ -34,6 +34,9 @@ extra_rdoc_files: []
 files:
 - README.md
 - Rakefile
+- ext/autocorrect/Cargo.toml
+- ext/autocorrect/extconf.rb
+- ext/autocorrect/src/lib.rs
 - lib/autocorrect-rb.rb
 - lib/autocorrect/2.7/autocorrect.bundle
 - lib/autocorrect/3.0/autocorrect.bundle