whatlang 0.1.8 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 69c49c403761528be76daea679fbf38b5c7fbf2f508a819d9bef3e5b6bf14b48
-  data.tar.gz: 748275ab31ce880de74a32c92ebd680f3b05af4303cf29e3985cfb774076d0e9
+  metadata.gz: e5b245173c946709df258efa77b2e253d062ecd79087cde657919f488d3f08b8
+  data.tar.gz: ec4cd29f02a29bfd44c66b31ee8cd5e2aec9b5be6db126897dbf7214ccb7ad73
 SHA512:
-  metadata.gz: 8c7acef11c2721c3ea347f4c64d199114c22ebd3258cef88acbb8cd81633d9ec1a0843f8fa3543c667b1c416ffa22631a71fe0e0ddbb5b7803a7bc0d581f729d
-  data.tar.gz: 381ad3386f4bae91ed497e1a49f5bf60df9346457f775f0fe7f6df335cb05d2b5ca5123a30f4e2fcae518c81d2a0e9da12aa1c69841b3a192b3349909d96e618
+  metadata.gz: e13e54ad577c1949673739ab8a02dab3405dd19f7ac3602d47232cf5f746576be8c0d23de36efc02ac1837e8bbaeb41a21fff9a7eecf6170bf0a6b782fbb246f
+  data.tar.gz: 89832e91d3b3be94a019416f95627c5fe8e46981a0066a85a0b3a758bae333a72572dba8a31197ef3665c0ec8fe61a5b601792beb18367c306ad17f07c6ac37c

data/.gitignore

@@ -7,5 +7,6 @@
 /spec/reports/
 /tmp/
 Gemfile.lock
-/lib/whatlang-rb/whatlang_rb.bundle
-/whatlang-rb/
+/lib/whatlang.bundle
+/whatlang.bundle
+/ext/target/

data/CHANGELOG.md

@@ -1,3 +1,16 @@
+0.2.0
+=====
+
+* Implement all lib code in Rust
+* Add rb-sys/stable-api-compiled-fallback for some environment
+* Implement `Whatlang::Lang.all`
+
+0.1.9
+=====
+
+* Fix install problem
+* Update Whatlang version to 0.18.0, which includes Welsh support
+
 0.1.8
 =====
 

data/Gemfile

@@ -4,5 +4,3 @@ source "https://rubygems.org"
 
 # Specify your gem's dependencies in whatlang.gemspec
 gemspec
-
-gem "fiddle", "~> 1.1"

data/README.md

@@ -64,6 +64,20 @@ Usage
     Whatlang.detect(text, allowlist: list).lang.code # => "ita"
     Whatlang.detect(text, denylist: list).lang.code # => "epo"
 
+If you need available languages, use `Whatlang::Lang.all`:
+
+    puts Whatlang::Lang.all.collect {|lang| "#{lang.code}: #{lang.eng_name}"}
+    # epo: Esperanto
+    # eng: English
+    # rus: Russian
+    # cmn: Mandarin
+    # spa: Spanish
+    # por: Portuguese
+    #   :
+    #   :
+
+Use `Whatlang::Lang#code` for allow and deny list.
+
 If you're interested in only language information, use `detect_lang` function:
 
     lang = Whatlang.detect_lang(text)

data/Rakefile

@@ -1,23 +1,40 @@
-# frozen_string_literal: true
-
 require "rake/clean"
 require "rubygems/ext"
 require "rubygems/tasks"
 require "rake/testtask"
 require "yard"
+require "shellwords"
 
 task default: :test
 
-Gem::Tasks.new
-YARD::Rake::YardocTask.new
+GEMSPEC = Gem::Specification.load("whatlang.gemspec")
+MANIFEST = GEMSPEC.extensions.first
+
+CARGO_LOCK = "ext/Cargo.lock"
+file CARGO_LOCK => MANIFEST do |t|
+  pkgid = `cargo pkgid --manifest-path=#{t.source.shellescape}`.chomp
+  system "cargo", "update", "--manifest-path", t.source, pkgid, exception: true
+end
 
-EXTENSION = "lib/whatlang-rb/whatlang_rb.#{RbConfig::CONFIG["DLEXT"]}"
-file EXTENSION do
+SO_NAME = "#{GEMSPEC.name}.#{RbConfig::CONFIG["DLEXT"]}"
+SO_PATH = File.join("lib", SO_NAME)
+SRC = FileList["ext/src/**/*.rs"]
+file SO_PATH => [CARGO_LOCK] + SRC do
   results = Rake.verbose == true ? $stdout : []
-  Gem::Ext::CargoBuilder.new.build "ext/whatlang-rb/Cargo.toml", ".", results, [], "lib", File.expand_path("ext/whatlang-rb")
+  Gem::Ext::CargoBuilder.new.build MANIFEST, ".", results, [], "lib", File.expand_path("ext")
 end
-CLEAN.include "whatlang-rb"
-CLOBBER.include EXTENSION
+CLEAN.include SO_NAME
+CLOBBER.include SO_PATH
+
+Gem::Tasks.new
+task build: CARGO_LOCK
+CLOBBER.include("pkg/#{GEMSPEC.file_name}")
 
 Rake::TestTask.new
-task test: EXTENSION
+task test: SO_PATH
+
+YARD::Rake::YardocTask.new
+desc "Generate Ruby and Rust documentation"
+task doc: :yard do
+  system "cargo", "doc", "--manifest-path", MANIFEST, exception: true
+end

data/ext/{whatlang-rb/Cargo.lock → Cargo.lock}

@@ -2,18 +2,6 @@
 # It is not intended for manual editing.
 version = 4
 
-[[package]]
-name = "ahash"
-version = "0.8.12"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5a15f179cd60c4584b8a8c596927aadc462e27f2ca70c04e0071964a73ba7a75"
-dependencies = [
- "cfg-if",
- "once_cell",
- "version_check",
- "zerocopy",
-]
-
 [[package]]
 name = "aho-corasick"
 version = "1.1.3"
@@ -87,6 +75,18 @@ version = "1.15.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "48c757948c5ede0e46177b7add2e67155f70e33c07fea8284df6576da70b3719"
 
+[[package]]
+name = "equivalent"
+version = "1.0.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "877a4ace8713b0bcf2a4e7eec82529c029f1d0619886d18145fea96c3ffe5c0f"
+
+[[package]]
+name = "foldhash"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d9c4f5dac5e15c24eb999c26181a6ca40b39fe946cbe4c263c7209467bc83af2"
+
 [[package]]
 name = "glob"
 version = "0.3.3"
@@ -95,12 +95,13 @@ checksum = "0cc23270f6e1808e30a928bdc84dea0b9b4136a8bc82338574f23baf47bbd280"
 
 [[package]]
 name = "hashbrown"
-version = "0.14.5"
+version = "0.15.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e5274423e17b7c9fc20b6e7e208532f9b19825d82dfd615708b70edd83df41f1"
+checksum = "9229cfe53dfd69f0609a49f65461bd93001ea1ef889cd5529dd176593f5338a1"
 dependencies = [
- "ahash",
  "allocator-api2",
+ "equivalent",
+ "foldhash",
 ]
 
 [[package]]
@@ -185,12 +186,6 @@ dependencies = [
  "minimal-lexical",
 ]
 
-[[package]]
-name = "once_cell"
-version = "1.21.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "42f5e15c9953c5e4ccceeb2e7382a716482c34515315f7b03532b8b4e8393d2d"
-
 [[package]]
 name = "proc-macro2"
 version = "1.0.101"
@@ -309,28 +304,22 @@ version = "1.0.20"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "462eeb75aeb73aea900253ce739c8e18a67423fadf006037cd3ff27e82748a06"
 
-[[package]]
-name = "version_check"
-version = "0.9.5"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0b928f33d975fc6ad9f86c8f283853ad26bdd5b10b7f1542aa2fa15e2289105a"
-
 [[package]]
 name = "whatlang"
-version = "0.16.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "471d1c1645d361eb782a1650b1786a8fb58dd625e681a04c09f5ff7c8764a7b0"
+version = "0.2.0"
 dependencies = [
- "hashbrown",
- "once_cell",
+ "magnus",
+ "rb-sys",
+ "whatlang 0.18.0",
 ]
 
 [[package]]
-name = "whatlang-rb"
-version = "0.1.7"
+name = "whatlang"
+version = "0.18.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f5e8f38b596e2a359b755342473520a99421e43658548c79489ee221b728c107"
 dependencies = [
- "magnus",
- "whatlang",
+ "hashbrown",
 ]
 
 [[package]]
@@ -338,23 +327,3 @@ name = "windows-link"
 version = "0.2.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f0805222e57f7521d6a62e36fa9163bc891acd422f971defe97d64e70d0a4fe5"
-
-[[package]]
-name = "zerocopy"
-version = "0.8.27"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0894878a5fa3edfd6da3f88c4805f4c8558e2b996227a3d864f47fe11e38282c"
-dependencies = [
- "zerocopy-derive",
-]
-
-[[package]]
-name = "zerocopy-derive"
-version = "0.8.27"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "88d2b8d9c68ad2b9e4340d7832716a4d21a22a1154777ad56ea55c51a9cf3831"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn",
-]

data/ext/Cargo.toml

@@ -0,0 +1,14 @@
+[package]
+name = "whatlang"
+version = "0.2.0"
+authors = ["Kitaiti Makoto <KitaitiMakoto@gmail.com>"]
+edition = "2024"
+publish = false
+
+[dependencies]
+magnus = "0.8"
+whatlang-rs = { package = "whatlang", version = "0.18.0", registry = "crates-io" }
+rb-sys = { version = "*", features = ["stable-api-compiled-fallback"] }
+
+[lib]
+crate-type = ["cdylib"]

data/ext/{whatlang-rb/src → src}/lib.rs

@@ -1,14 +1,19 @@
 use magnus::{
-    Error, Ruby, Value, function, method,
+    Error, RArray, Ruby, Value, function, method,
     prelude::*,
     scan_args::{get_kwargs, scan_args},
 };
-use whatlang as wl;
+use whatlang_rs as wl;
 
 #[magnus::wrap(class = "Whatlang::Lang")]
 struct Lang(wl::Lang);
 
 impl Lang {
+    fn all(ruby: &Ruby) -> RArray {
+        let langs = wl::Lang::all();
+        ruby.ary_from_iter(langs.iter().map(|&lang| Lang(lang)))
+    }
+
     fn code(&self) -> &str {
         self.0.code()
     }
@@ -75,17 +80,21 @@ fn detect_without_options(text: String) -> Option<Info> {
 }
 
 fn detect_with_allowlist(text: String, allowlist: Vec<String>) -> Option<Info> {
-    wl::Detector::with_allowlist(allowlist.iter().filter_map(wl::Lang::from_code).collect())
+    wl::Detector::with_allowlist(lang_list(allowlist))
         .detect(&text)
         .map(Info)
 }
 
 fn detect_with_denylist(text: String, denylist: Vec<String>) -> Option<Info> {
-    wl::Detector::with_denylist(denylist.iter().filter_map(wl::Lang::from_code).collect())
+    wl::Detector::with_denylist(lang_list(denylist))
         .detect(&text)
         .map(Info)
 }
 
+fn lang_list(list: Vec<String>) -> Vec<wl::Lang> {
+    list.iter().filter_map(wl::Lang::from_code).collect()
+}
+
 fn detect_lang(text: String) -> Option<Lang> {
     wl::detect_lang(&text).map(Lang)
 }
@@ -108,6 +117,7 @@ fn init(ruby: &Ruby) -> Result<(), Error> {
     module.define_singleton_method("detect_script", function!(detect_script, 1))?;
 
     let lang = module.define_class("Lang", ruby.class_object())?;
+    lang.define_singleton_method("all", function!(Lang::all, 0))?;
     lang.define_method("code", method!(Lang::code, 0))?;
     lang.define_method("name", method!(Lang::name, 0))?;
     lang.define_method("eng_name", method!(Lang::eng_name, 0))?;

data/test/test_package.rb

@@ -1,4 +1,5 @@
 require "test/unit"
+require "test/unit/notify"
 require 'tempfile'
 require 'tmpdir'
 require 'shellwords'
@@ -32,7 +33,7 @@ class TestPackage < Test::Unit::TestCase
     private
 
     def assert_installed(dir, version)
-      assert_path_exist File.join(dir, "gems/whatlang-#{version}/lib/whatlang-rb", "whatlang_rb.#{RbConfig::CONFIG["DLEXT"]}")
+      assert_path_exist File.join(dir, "gems/whatlang-#{version}/lib", "whatlang.#{RbConfig::CONFIG["DLEXT"]}")
     end
   end
 end

data/test/test_whatlang.rb

@@ -1,6 +1,6 @@
-# frozen_string_literal: true
-
-require_relative "helper"
+require "test/unit"
+require "test/unit/notify"
+require "whatlang"
 
 class TestWhatlang < Test::Unit::TestCase
   def test_detect
@@ -62,4 +62,10 @@ class TestWhatlang < Test::Unit::TestCase
   def test_numeric
     assert_nil Whatlang.detect(1234)
   end
+
+  def test_lang_all
+    all_langs = Whatlang::Lang.all
+    assert_equal 70, all_langs.length
+    assert_all(all_langs) {|lang| lang.instance_of? Whatlang::Lang}
+  end
 end

data/whatlang.gemspec

@@ -2,7 +2,7 @@ require "json"
 
 Gem::Specification.new do |spec|
   spec.name          = "whatlang"
-  spec.version       = JSON.parse(`cargo metadata --no-deps --format-version=1 --manifest-path=ext/whatlang-rb/Cargo.toml`)["packages"][0]["version"]
+  spec.version       = JSON.parse(`cargo metadata --no-deps --format-version=1 --manifest-path=ext/Cargo.toml`)["packages"][0]["version"]
   spec.license       = "Ruby"
   spec.authors       = ["Kitaiti Makoto"]
   spec.email         = ["KitaitiMakoto@gmail.com"]
@@ -18,10 +18,12 @@ Gem::Specification.new do |spec|
   spec.files = Dir.chdir(File.expand_path(__dir__)) do
     `git ls-files -z`.split("\x0")
   end
-  spec.extensions = ["ext/whatlang-rb/Cargo.toml"]
+  spec.extensions = ["ext/Cargo.toml"]
 
   spec.add_development_dependency "test-unit"
+  spec.add_development_dependency "test-unit-notify"
   spec.add_development_dependency "rake"
   spec.add_development_dependency "yard"
   spec.add_development_dependency "rubygems-tasks"
+  spec.add_development_dependency "terminal-notifier"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: whatlang
 version: !ruby/object:Gem::Version
-  version: 0.1.8
+  version: 0.2.0
 platform: ruby
 authors:
 - Kitaiti Makoto
@@ -23,6 +23,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: test-unit-notify
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
@@ -65,12 +79,26 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: terminal-notifier
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 description: Ruby bindings for Whatlang, a natural language detection for Rust.
 email:
 - KitaitiMakoto@gmail.com
 executables: []
 extensions:
-- ext/whatlang-rb/Cargo.toml
+- ext/Cargo.toml
 extra_rdoc_files: []
 files:
 - ".gitignore"
@@ -82,11 +110,9 @@ files:
 - Gemfile
 - README.md
 - Rakefile
-- ext/whatlang-rb/Cargo.lock
-- ext/whatlang-rb/Cargo.toml
-- ext/whatlang-rb/src/lib.rs
-- lib/whatlang.rb
-- test/helper.rb
+- ext/Cargo.lock
+- ext/Cargo.toml
+- ext/src/lib.rs
 - test/test_package.rb
 - test/test_whatlang.rb
 - whatlang.gemspec

data/ext/whatlang-rb/Cargo.toml

@@ -1,12 +0,0 @@
-[package]
-name = "whatlang-rb"
-version = "0.1.8"
-authors = ["Kitaiti Makoto <KitaitiMakoto@gmail.com>"]
-edition = "2024"
-
-[dependencies]
-magnus = "0.8"
-whatlang = "0.16.4"
-
-[lib]
-crate-type = ["cdylib"]

data/lib/whatlang.rb

@@ -1 +0,0 @@
-require "whatlang-rb/whatlang_rb"

data/test/helper.rb

@@ -1,2 +0,0 @@
-require "test/unit"
-require "whatlang"