ruby : extend API (#2551)

* Handle objs in Ruby code

* Add task to make Makefile

* Share commont constance in test suites

* Add model-related APIs

* Add Whisper::Model class

* Add tests for Whisper::Model

* Add missing LDFLAG -lstdc++

* Add tests for Whisper.log_set

* Add Whisper.set_log

* Define log level

* Add document on logging

* Add license section to README

* Add document on Whisper::Model

* Fix examples in README

* Add test for Model with GC

* Make dependency on Makefile more accurate

* Fix bug about Whisper::Model and GC

bindings/ruby/tests/helper.rb

@@ -0,0 +1,7 @@
+require "test/unit"
+require "whisper"
+
+class TestBase < Test::Unit::TestCase
+  MODEL = File.join(__dir__, "..", "..", "..", "models", "ggml-base.en.bin")
+  AUDIO = File.join(__dir__, "..", "..", "..", "samples", "jfk.wav")
+end

bindings/ruby/tests/test_model.rb

@@ -0,0 +1,44 @@
+require_relative "helper"
+
+class TestModel < TestBase
+  def test_model
+    whisper = Whisper::Context.new(MODEL)
+    assert_instance_of Whisper::Model, whisper.model
+  end
+
+  def test_attributes
+    whisper = Whisper::Context.new(MODEL)
+    model = whisper.model
+
+    assert_equal 51864, model.n_vocab
+    assert_equal 1500, model.n_audio_ctx
+    assert_equal 512, model.n_audio_state
+    assert_equal 8, model.n_audio_head
+    assert_equal 6, model.n_audio_layer
+    assert_equal 448, model.n_text_ctx
+    assert_equal 512, model.n_text_state
+    assert_equal 8, model.n_text_head
+    assert_equal 6, model.n_text_layer
+    assert_equal 80, model.n_mels
+    assert_equal 1, model.ftype
+    assert_equal "base", model.type
+  end
+
+  def test_gc
+    model = Whisper::Context.new(MODEL).model
+    GC.start
+
+    assert_equal 51864, model.n_vocab
+    assert_equal 1500, model.n_audio_ctx
+    assert_equal 512, model.n_audio_state
+    assert_equal 8, model.n_audio_head
+    assert_equal 6, model.n_audio_layer
+    assert_equal 448, model.n_text_ctx
+    assert_equal 512, model.n_text_state
+    assert_equal 8, model.n_text_head
+    assert_equal 6, model.n_text_layer
+    assert_equal 80, model.n_mels
+    assert_equal 1, model.ftype
+    assert_equal "base", model.type
+  end
+end

bindings/ruby/tests/test_package.rb

@@ -1,9 +1,9 @@
-require 'test/unit'
+require_relative "helper"
 require 'tempfile'
 require 'tmpdir'
 require 'shellwords'
 
-class TestPackage < Test::Unit::TestCase
+class TestPackage < TestBase
   def test_build
     Tempfile.create do |file|
       assert system("gem", "build", "whispercpp.gemspec", "--output", file.to_path.shellescape, exception: true)

bindings/ruby/tests/test_params.rb

@@ -1,7 +1,6 @@
-require 'test/unit'
-require 'whisper'
+require_relative "helper"
 
-class TestParams < Test::Unit::TestCase
+class TestParams < TestBase
   def setup
     @params  = Whisper::Params.new
   end

bindings/ruby/tests/test_segment.rb

@@ -1,18 +1,14 @@
-require "test/unit"
-require "whisper"
-
-class TestSegment < Test::Unit::TestCase
-  TOPDIR = File.expand_path(File.join(File.dirname(__FILE__), '..'))
+require_relative "helper"
 
+class TestSegment < TestBase
   class << self
     attr_reader :whisper
 
     def startup
-      @whisper = Whisper::Context.new(File.join(TOPDIR, '..', '..', 'models', 'ggml-base.en.bin'))
+      @whisper = Whisper::Context.new(TestBase::MODEL)
       params = Whisper::Params.new
       params.print_timestamps = false
-      jfk = File.join(TOPDIR, '..', '..', 'samples', 'jfk.wav')
-      @whisper.transcribe(jfk, params)
+      @whisper.transcribe(TestBase::AUDIO, params)
     end
   end
 
@@ -60,7 +56,7 @@ class TestSegment < Test::Unit::TestCase
       end
       index += 1
     end
-    whisper.transcribe(File.join(TOPDIR, '..', '..', 'samples', 'jfk.wav'), params)
+    whisper.transcribe(AUDIO, params)
     assert_equal 0, seg.start_time
     assert_match /ask not what your country can do for you, ask what you can do for your country/, seg.text
   end
@@ -76,7 +72,7 @@ class TestSegment < Test::Unit::TestCase
       assert_same seg, segment
       return
     end
-    whisper.transcribe(File.join(TOPDIR, '..', '..', 'samples', 'jfk.wav'), params)
+    whisper.transcribe(AUDIO, params)
   end
 
   private

bindings/ruby/tests/test_whisper.rb

@@ -1,20 +1,20 @@
-require 'whisper'
-require 'test/unit'
+require_relative "helper"
+require "stringio"
 
-class TestWhisper < Test::Unit::TestCase
-  TOPDIR = File.expand_path(File.join(File.dirname(__FILE__), '..'))
+# Exists to detect memory-related bug
+Whisper.log_set ->(level, buffer, user_data) {}, nil
 
+class TestWhisper < TestBase
   def setup
     @params  = Whisper::Params.new
   end
 
   def test_whisper
-    @whisper = Whisper::Context.new(File.join(TOPDIR, '..', '..', 'models', 'ggml-base.en.bin'))
+    @whisper = Whisper::Context.new(MODEL)
     params  = Whisper::Params.new
     params.print_timestamps = false
 
-    jfk = File.join(TOPDIR, '..', '..', 'samples', 'jfk.wav')
-    @whisper.transcribe(jfk, params) {|text|
+    @whisper.transcribe(AUDIO, params) {|text|
       assert_match /ask not what your country can do for you, ask what you can do for your country/, text
     }
   end
@@ -24,11 +24,10 @@ class TestWhisper < Test::Unit::TestCase
       attr_reader :whisper
 
       def startup
-        @whisper = Whisper::Context.new(File.join(TOPDIR, '..', '..', 'models', 'ggml-base.en.bin'))
+        @whisper = Whisper::Context.new(TestBase::MODEL)
         params = Whisper::Params.new
         params.print_timestamps = false
-        jfk = File.join(TOPDIR, '..', '..', 'samples', 'jfk.wav')
-        @whisper.transcribe(jfk, params)
+        @whisper.transcribe(TestBase::AUDIO, params)
       end
     end
 
@@ -96,4 +95,33 @@ class TestWhisper < Test::Unit::TestCase
       Whisper.lang_str_full(Whisper.lang_max_id + 1)
     end
   end
+
+  def test_log_set
+    user_data = Object.new
+    logs = []
+    log_callback = ->(level, buffer, udata) {
+      logs << [level, buffer, udata]
+    }
+    Whisper.log_set log_callback, user_data
+    Whisper::Context.new(MODEL)
+
+    assert logs.length > 30
+    logs.each do |log|
+      assert_equal Whisper::LOG_LEVEL_INFO, log[0]
+      assert_same user_data, log[2]
+    end
+  end
+
+  def test_log_suppress
+    stderr = $stderr
+    Whisper.log_set ->(level, buffer, user_data) {
+      # do nothing
+    }, nil
+    dev = StringIO.new("")
+    $stderr = dev
+    Whisper::Context.new(MODEL)
+    assert_empty dev.string
+  ensure
+    $stderr = stderr
+  end
 end