ruby : extend API (ggml-org#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

Author: KitaitiMakoto

bindings/ruby/README.md

@@ -107,5 +107,63 @@ whisper.transcribe("path/to/audio.wav", params)
 
 ```
 
+You can see model information:
+
+```ruby
+whisper = Whisper::Context.new("path/to/model.bin")
+model = whisper.model
+
+model.n_vocab # => 51864
+model.n_audio_ctx # => 1500
+model.n_audio_state # => 512
+model.n_audio_head # => 8
+model.n_audio_layer # => 6
+model.n_text_ctx # => 448
+model.n_text_state # => 512
+model.n_text_head # => 8
+model.n_text_layer # => 6
+model.n_mels # => 80
+model.ftype # => 1
+model.type # => "base"
+
+```
+
+You can set log callback:
+
+```ruby
+prefix = "[MyApp] "
+log_callback = ->(level, buffer, user_data) {
+  case level
+  when Whisper::LOG_LEVEL_NONE
+    puts "#{user_data}none: #{buffer}"
+  when Whisper::LOG_LEVEL_INFO
+    puts "#{user_data}info: #{buffer}"
+  when Whisper::LOG_LEVEL_WARN
+    puts "#{user_data}warn: #{buffer}"
+  when Whisper::LOG_LEVEL_ERROR
+    puts "#{user_data}error: #{buffer}"
+  when Whisper::LOG_LEVEL_DEBUG
+    puts "#{user_data}debug: #{buffer}"
+  when Whisper::LOG_LEVEL_CONT
+    puts "#{user_data}same to previous: #{buffer}"
+  end
+}
+Whisper.log_set log_callback, prefix
+```
+
+Using this feature, you are also able to suppress log:
+
+```ruby
+Whisper.log_set ->(level, buffer, user_data) {
+  # do nothing
+}, nil
+Whisper::Context.new(MODEL)
+```
+
+License
+-------
+
+The same to [whisper.cpp][].
+
 [whisper.cpp]: https://github.com/ggerganov/whisper.cpp
 [models]: https://github.com/ggerganov/whisper.cpp/tree/master/models

bindings/ruby/Rakefile

@@ -23,30 +23,39 @@ CLEAN.include FileList[
                 "ext/depend"
               ]
 
-task build: SOURCES + FileList[
-                        "ext/extconf.rb",
-                        "ext/ruby_whisper.h",
-                        "ext/ruby_whisper.cpp",
-                        "whispercpp.gemspec",
-                      ]
+task build: FileList[
+       "ext/Makefile",
+       "ext/ruby_whisper.h",
+       "ext/ruby_whisper.cpp",
+       "whispercpp.gemspec",
+     ]
 
 directory "pkg"
 CLOBBER.include "pkg"
 
 TEST_MODEL = "../../models/ggml-base.en.bin"
 LIB_NAME = "whisper".ext(RbConfig::CONFIG["DLEXT"])
+SO_FILE = File.join("ext", LIB_NAME)
 LIB_FILE = File.join("lib", LIB_NAME)
 
-directory "lib"
-task LIB_FILE => SOURCES + ["lib"] do |t|
+file "ext/Makefile" => ["ext/extconf.rb", "ext/ruby_whisper.h", "ext/ruby_whisper.cpp"] + SOURCES do |t|
+  Dir.chdir "ext" do
+    ruby "extconf.rb"
+  end
+end
+
+file SO_FILE => "ext/Makefile" do |t|
   Dir.chdir "ext" do
-    sh "ruby extconf.rb"
     sh "make"
   end
-  mv "ext/#{LIB_NAME}", t.name
 end
 CLEAN.include LIB_FILE
 
+directory "lib"
+file LIB_FILE => [SO_FILE, "lib"] do |t|
+  copy t.source, t.name
+end
+
 Rake::TestTask.new do |t|
   t.test_files = FileList["tests/test_*.rb"]
 end

bindings/ruby/ext/extconf.rb

@@ -2,6 +2,9 @@
 
 # need to use c++ compiler flags
 $CXXFLAGS << ' -std=c++11'
+
+$LDFLAGS << ' -lstdc++'
+
 # Set to true when building binary gems
 if enable_config('static-stdlib', false)
   $LDFLAGS << ' -static-libgcc -static-libstdc++'
@@ -12,34 +15,6 @@
   $CXXFLAGS << ' -march=native -mtune=native'
 end
 
-def with_disabling_unsupported_files
-  disabled_files = []
-
-  unless $GGML_METAL
-    disabled_files << 'ggml-metal.h' << 'ggml-metal.m'
-  end
-
-  unless $GGML_METAL_EMBED_LIBRARY
-    disabled_files << 'ggml-metal.metal'
-  end
-
-  unless $OBJ_ALL&.include? 'ggml-blas.o'
-    disabled_files << 'ggml-blas.h' << 'ggml-blas.cpp'
-  end
-
-  disabled_files.filter! {|file| File.exist? file}
-
-  disabled_files.each do |file|
-    File.rename file, "#{file}.disabled"
-  end
-
-  yield
-
-  disabled_files.each do |file|
-    File.rename "#{file}.disabled", file
-  end
-end
-
 if ENV['WHISPER_METAL']
   $GGML_METAL ||= true
   $DEPRECATE_WARNING ||= true
@@ -66,10 +41,10 @@ def with_disabling_unsupported_files
 $MK_NVCCFLAGS = '-std=c++11'
 $MK_LDFLAGS = ''
 
-$OBJ_GGML = ''
-$OBJ_WHISPER = ''
-$OBJ_COMMON = ''
-$OBJ_SDL = ''
+$OBJ_GGML = []
+$OBJ_WHISPER = []
+$OBJ_COMMON = []
+$OBJ_SDL = []
 
 $MK_CPPFLAGS << ' -D_XOPEN_SOURCE=600'
 
@@ -152,50 +127,51 @@ def with_disabling_unsupported_files
     $MK_CPPFLAGS << ' -DACCELERATE_NEW_LAPACK'
     $MK_CPPFLAGS << ' -DACCELERATE_LAPACK_ILP64'
     $MK_LDFLAGS  << ' -framework Accelerate'
-    $OBJ_GGML    << ' ggml-blas.o'
+    $OBJ_GGML    << 'ggml-blas.o'
   end
 end
 
 if ENV['GGML_OPENBLAS']
   $MK_CPPFLAGS << " -DGGML_USE_BLAS #{`pkg-config --cflags-only-I openblas`.chomp}"
   $MK_CFLAGS   << " #{`pkg-config --cflags-only-other openblas)`.chomp}"
   $MK_LDFLAGS  << " #{`pkg-config --libs openblas`}"
-  $OBJ_GGML    << ' ggml-blas.o'
+  $OBJ_GGML    << 'ggml-blas.o'
 end
 
 if ENV['GGML_OPENBLAS64']
   $MK_CPPFLAGS << " -DGGML_USE_BLAS #{`pkg-config --cflags-only-I openblas64`.chomp}"
   $MK_CFLAGS   << " #{`pkg-config --cflags-only-other openblas64)`.chomp}"
   $MK_LDFLAGS  << " #{`pkg-config --libs openblas64`}"
-  $OBJ_GGML    << ' ggml-blas.o'
+  $OBJ_GGML    << 'ggml-blas.o'
 end
 
 if $GGML_METAL
   $MK_CPPFLAGS << ' -DGGML_USE_METAL'
   $MK_LDFLAGS  << ' -framework Foundation -framework Metal -framework MetalKit'
-  $OBJ_GGML    << ' ggml-metal.o'
+  $OBJ_GGML    << 'ggml-metal.o'
 
   if ENV['GGML_METAL_NDEBUG']
     $MK_CPPFLAGS << ' -DGGML_METAL_NDEBUG'
   end
 
   if $GGML_METAL_EMBED_LIBRARY
     $MK_CPPFLAGS << ' -DGGML_METAL_EMBED_LIBRARY'
-    $OBJ_GGML    << ' ggml-metal-embed.o'
+    $OBJ_GGML    << 'ggml-metal-embed.o'
   end
 end
 
 $OBJ_GGML <<
-  ' ggml.o' <<
-  ' ggml-alloc.o' <<
-  ' ggml-backend.o' <<
-  ' ggml-quants.o' <<
-  ' ggml-aarch64.o'
+  'ggml.o' <<
+  'ggml-alloc.o' <<
+  'ggml-backend.o' <<
+  'ggml-quants.o' <<
+  'ggml-aarch64.o'
 
 $OBJ_WHISPER <<
-  ' whisper.o'
+  'whisper.o'
 
-$OBJ_ALL = "#{$OBJ_GGML} #{$OBJ_WHISPER} #{$OBJ_COMMON} #{$OBJ_SDL}"
+$objs = $OBJ_GGML + $OBJ_WHISPER + $OBJ_COMMON + $OBJ_SDL
+$objs << "ruby_whisper.o"
 
 $CPPFLAGS  = "#{$MK_CPPFLAGS} #{$CPPFLAGS}"
 $CFLAGS    = "#{$CPPFLAGS} #{$MK_CFLAGS} #{$GF_CFLAGS} #{$CFLAGS}"
@@ -204,26 +180,13 @@ def with_disabling_unsupported_files
 $NVCCFLAGS = "#{$MK_NVCCFLAGS} #{$NVCCFLAGS}"
 $LDFLAGS   = "#{$MK_LDFLAGS} #{$LDFLAGS}"
 
-if $GGML_METAL_EMBED_LIBRARY
-  File.write 'depend', "$(OBJS): $(OBJS) ggml-metal-embed.o\n"
-end
-
-with_disabling_unsupported_files do
-
-  create_makefile('whisper')
-
-end
+create_makefile('whisper')
 
 File.open 'Makefile', 'a' do |file|
   file.puts 'include get-flags.mk'
 
   if $GGML_METAL
     if $GGML_METAL_EMBED_LIBRARY
-      # mkmf determines object files to compile dependent on existing *.{c,cpp,m} files
-      # but ggml-metal-embed.c doesn't exist on creating Makefile.
-      file.puts "objs := $(OBJS)"
-      file.puts "OBJS = $(objs) 'ggml-metal-embed.o'"
-
       file.puts 'include metal-embed.mk'
     end
   end