Add CLI executable with model registry support

.rubocop.yml

@@ -29,42 +29,48 @@ Metrics/BlockLength:
   Exclude:
     - 'test/**/*'
     - 'lib/classifier/extensions/vector.rb'
+    - 'lib/classifier/cli.rb'
 
-# Allow longer methods in complex algorithms (SVD, etc.)
+# Allow longer methods in complex algorithms (SVD, etc.) and CLI
 Metrics/MethodLength:
   Max: 25
   Exclude:
     - 'test/**/*'
     - 'lib/classifier/extensions/vector.rb'
     - 'lib/classifier/lsi/content_node.rb'
+    - 'lib/classifier/cli.rb'
 
-# Allow higher complexity for mathematical algorithms
+# Allow higher complexity for mathematical algorithms and CLI
 Metrics/AbcSize:
   Max: 30
   Exclude:
     - 'test/**/*'
     - 'lib/classifier/extensions/vector.rb'
     - 'lib/classifier/lsi.rb'
     - 'lib/classifier/lsi/content_node.rb'
+    - 'lib/classifier/cli.rb'
 
 Metrics/CyclomaticComplexity:
   Max: 10
   Exclude:
     - 'lib/classifier/extensions/vector.rb'
     - 'lib/classifier/lsi/content_node.rb'
+    - 'lib/classifier/cli.rb'
 
 Metrics/PerceivedComplexity:
   Max: 10
   Exclude:
     - 'lib/classifier/extensions/vector.rb'
     - 'lib/classifier/lsi/content_node.rb'
+    - 'lib/classifier/cli.rb'
 
-# Class length limits - algorithms and tests can be longer
+# Class length limits - algorithms, tests and CLI can be longer
 Metrics/ClassLength:
   Max: 250
   Exclude:
     - 'test/**/*'
     - 'lib/classifier/lsi.rb'
+    - 'lib/classifier/cli.rb'
 
 # SV_decomp is a standard algorithm name
 Naming/MethodName:

Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    classifier (2.2.0)
+    classifier (2.3.0)
       fast-stemmer (~> 1.0)
       matrix
       mutex_m (~> 0.2)
@@ -23,12 +23,17 @@ GEM
       securerandom (>= 0.3)
       tzinfo (~> 2.0, >= 2.0.5)
       uri (>= 0.13.1)
+    addressable (2.8.8)
+      public_suffix (>= 2.0.2, < 8.0)
     ast (2.4.3)
     base64 (0.3.0)
     bigdecimal (4.0.1)
     cgi (0.5.1)
     concurrent-ruby (1.3.6)
     connection_pool (3.0.2)
+    crack (1.0.1)
+      bigdecimal
+      rexml
     csv (3.3.5)
     date (3.5.1)
     docile (1.4.1)
@@ -39,6 +44,7 @@ GEM
     ffi (1.17.2-arm64-darwin)
     ffi (1.17.2-x86_64-linux-gnu)
     fileutils (1.8.0)
+    hashdiff (1.2.1)
     i18n (1.14.8)
       concurrent-ruby (~> 1.0)
     json (2.18.0)
@@ -61,6 +67,7 @@ GEM
     psych (5.3.1)
       date
       stringio
+    public_suffix (7.0.0)
     racc (1.8.1)
     rainbow (3.1.1)
     rake (13.3.1)
@@ -80,6 +87,7 @@ GEM
       psych (>= 4.0.0)
       tsort
     regexp_parser (2.11.3)
+    rexml (3.4.4)
     rubocop (1.82.1)
       json (~> 2.3)
       language_server-protocol (~> 3.17.0.2)
@@ -134,6 +142,10 @@ GEM
       unicode-emoji (~> 4.1)
     unicode-emoji (4.2.0)
     uri (1.1.1)
+    webmock (3.26.1)
+      addressable (>= 2.8.0)
+      crack (>= 0.3.2)
+      hashdiff (>= 0.4.0, < 2.0.0)
 
 PLATFORMS
   arm64-darwin-22
@@ -156,6 +168,7 @@ DEPENDENCIES
   rubocop-minitest
   simplecov
   steep
+  webmock
 
 BUNDLED WITH
   4.0.3

Steepfile

@@ -5,6 +5,12 @@ target :lib do
 
   check 'lib'
 
+  # Stdlib dependencies for CLI
+  library 'fileutils'
+  library 'uri'
+  library 'net-http'
+  library 'json'
+
   # Strict mode: report methods without type annotations
   configure_code_diagnostics(D::Ruby.strict)
 

classifier.gemspec

@@ -1,6 +1,8 @@
+require_relative 'lib/classifier/version'
+
 Gem::Specification.new do |s|
   s.name        = 'classifier'
-  s.version     = '2.2.0'
+  s.version     = Classifier::VERSION
   s.summary     = 'Text classification with Bayesian, LSI, Logistic Regression, kNN, and TF-IDF vectorization.'
   s.description = 'A Ruby library for text classification featuring Naive Bayes, LSI (Latent Semantic Indexing), ' \
                   'Logistic Regression, and k-Nearest Neighbors classifiers. Includes TF-IDF vectorization, ' \
@@ -16,7 +18,9 @@ Gem::Specification.new do |s|
     'changelog_uri' => 'https://github.com/cardmagic/classifier/releases'
   }
   s.required_ruby_version = '>= 3.1'
-  s.files = Dir['{lib,sig}/**/*.{rb,rbs}', 'ext/**/*.{c,h,rb}', 'bin/*', 'LICENSE', '*.md', 'test/*']
+  s.files = Dir['{lib,sig,exe}/**/*.{rb,rbs}', 'ext/**/*.{c,h,rb}', 'exe/*', 'bin/*', 'LICENSE', '*.md', 'test/*']
+  s.bindir = 'exe'
+  s.executables = ['classifier']
   s.extensions = ['ext/classifier/extconf.rb']
   s.license = 'LGPL'
 
@@ -28,4 +32,5 @@ Gem::Specification.new do |s|
   s.add_development_dependency 'rbs-inline'
   s.add_development_dependency 'rdoc'
   s.add_development_dependency 'rake-compiler'
+  s.add_development_dependency 'webmock'
 end

exe/classifier

@@ -0,0 +1,9 @@
+#!/usr/bin/env ruby
+require 'classifier/cli'
+
+result = Classifier::CLI.new(ARGV).run
+
+warn result[:error] unless result[:error].empty?
+puts result[:output] unless result[:output].empty?
+
+exit result[:exit_code]

lib/classifier.rb

@@ -25,6 +25,7 @@
 # License::   LGPL
 
 require 'rubygems'
+require 'classifier/version'
 require 'classifier/errors'
 require 'classifier/storage'
 require 'classifier/streaming'