better_translate 0.3.1 → 0.4.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: a2ff034a6db3498226ddff2111cf66e94a371f65741f82986b6ec8324daaf2dc
-  data.tar.gz: 991a78781adfc647d789729f385bca6e7688a6269bf102e0bff8b40b7bf8aca7
+  metadata.gz: cf5e2fea9a6f6702124591e888a922072386b7dc3eca153008828448eae45357
+  data.tar.gz: 5c11456be0bda2b86ea385235a769d405af58233c096d306f710d451eecf6f2a
 SHA512:
-  metadata.gz: 68f06a1006c8fc3e43184485d8f71ec9b080649fa68e2f8c5563d55b20c356b501a20a308f0c00f967b5387ac8d3f62abf11d837ca25b325a646658bf8126146
-  data.tar.gz: 719cb1a9bb0217934ca0ecacd455bcc6c81afab081fbc43d24b8add5a50eb33b7947ab715a37938460d162fdcf88e9e629b8cc18d08e1f47ecec085101c35e8c
+  metadata.gz: 8ca97b7c5796b961e80de8a3247aecaf7e781965dfed89197977be2fc59326f8800d6a4b82ed91fbff592c82958b23088e09493d50c42f82458a697628a2a607
+  data.tar.gz: 6404f6b92fa55708db6127ade139d6f3f12170ef5c6617553ccbe383be3cb0971d31b2f956b9ac1bf0f8b4bc65bc46f702f55be13cc49cfc11541af319c72811

data/CHANGELOG.md

@@ -5,6 +5,32 @@ All notable changes to BetterTranslate will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.4.1] - 2025-03-11
+
+### Fixed
+- Migliorati i test RSpec per garantire maggiore affidabilità:
+  - Corretti gli stub per i provider di traduzione
+  - Migliorata la gestione delle richieste HTTP nei test
+  - Ottimizzati i file YAML temporanei per i test di similarità
+- Risolti problemi di compatibilità con WebMock
+
+### Changed
+- Sostituito l'approccio di stubbing specifico con pattern più flessibili
+- Migliorata la struttura dei test per il SimilarityAnalyzer
+
+## [0.4.0] - 2025-03-11
+
+### Added
+- New Translation Similarity Analyzer:
+  - Identifies similar translations across language files
+  - Generates detailed JSON reports and human-readable summaries
+  - Uses Levenshtein distance for similarity calculation
+  - Configurable similarity threshold
+- New Rails generator: `rails generate better_translate:analyze`
+  - Analyzes all YAML files in the locales directory
+  - Provides immediate feedback in the console
+  - Generates comprehensive similarity reports
+
 ## [0.3.1] - 2025-03-11
 
 ### Added

data/README.md

@@ -14,6 +14,7 @@ BetterTranslate simplifies the translation of YAML files in your Ruby/Rails appl
 - 📊 Real-time progress tracking
 - 🧪 Comprehensive test coverage
 - ⚡️ LRU caching for performance
+- 🔍 Translation similarity analysis
 
 ## Why BetterTranslate? 🤔
 
@@ -30,6 +31,10 @@ BetterTranslate simplifies the translation of YAML files in your Ruby/Rails appl
   - Comprehensive test suite
   - LRU caching for performance
   - Progress tracking
+- 🔍 **Translation Analysis**:
+  - Similarity detection
+  - Detailed reports
+  - Optimization suggestions
 - 🔧 **Flexible Helpers**:
   - Single text translation
   - Bulk text translation
@@ -146,6 +151,25 @@ The gem includes generators to simplify tasks:
 
   The `better_translate:translate` generator will trigger the translation process (via `BetterTranslate.magic`) and display progress in the terminal.
 
+- **Analyze Translations:**
+
+  ```bash
+  rails generate better_translate:analyze
+  ```
+
+  The `better_translate:analyze` generator will:
+  - Scan all YAML files in your locales directory
+  - Find similar translations using Levenshtein distance
+  - Generate two reports:
+    - `translation_similarities.json`: Detailed JSON report
+    - `translation_similarities_summary.txt`: Human-readable summary
+  
+  This helps you:
+  - Identify potentially redundant translations
+  - Maintain consistency across your translations
+  - Optimize your translation files
+  - Reduce translation costs
+
 ### Translation Helpers
 
 BetterTranslate provides helper methods to simplify translation tasks.

data/lib/better_translate/providers/chatgpt_provider.rb

@@ -13,7 +13,7 @@ module BetterTranslate
         body = {
           model: "gpt-3.5-turbo",
           messages: [
-            { role: "system", content: "You are a professional translator. Translate the following text exactly from #{BetterTranslate.configuration.source_language} to #{target_lang_name} without adding comments, explanations, or alternatives. Provide only the direct translation:" },
+            { role: "system", content: "You are a professional translator. Translate the following text exactly from #{BetterTranslate.configuration.source_language} to #{target_lang_name}. Provide ONLY the direct translation without any explanations, alternatives, or additional text. Do not include the original text. Do not use markdown formatting. Do not add any prefixes or notes. Just return the plain translated text." },
             { role: "user", content: "#{text}" }
           ],
           temperature: 0.3
@@ -30,8 +30,10 @@ module BetterTranslate
           translated_text = json.dig("choices", 0, "message", "content")
           translated_text ? translated_text.strip : text
         else
-          raise "Error during translation with ChatGPT: #{response.body}"
+          raise "Errore HTTP #{response.code}: #{response.body}"
         end
+      rescue StandardError => e
+        raise "Errore durante la traduzione con ChatGPT: #{e.message}"
       end
     end
   end

data/lib/better_translate/providers/gemini_provider.rb

@@ -1,33 +1,55 @@
+# frozen_string_literal: true
+
+require "net/http"
+require "json"
+require "uri"
+
 module BetterTranslate
   module Providers
     class GeminiProvider < BaseProvider
-      # Esempio di implementazione per Google Gemini.
-      # Nota: L'endpoint e i parametri sono ipotetici e vanno sostituiti con quelli reali secondo la documentazione ufficiale.
+      GEMINI_API_URL = "https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:generateContent"
+
       def translate_text(text, target_lang_code, target_lang_name)
-        uri = URI("https://gemini.googleapis.com/v1/translate")
-        headers = {
-          "Content-Type"  => "application/json",
-          "Authorization" => "Bearer #{@api_key}"
-        }
+        url = "#{GEMINI_API_URL}?key=#{@api_key}"
+        uri = URI(url)
 
+        headers = { "Content-Type" => "application/json" }
         body = {
-          input_text: text,
-          target_language: target_lang_code,
-          model: "gemini" # oppure altri parametri richiesti dall'API
+          contents: [{
+            parts: [{
+              text: "Translate the following text to #{target_lang_name}. Provide ONLY the direct translation without any explanations, alternatives, or additional text. Do not include the original text. Do not use markdown formatting. Do not add any prefixes or notes. Just return the plain translated text:\n\n#{text}"
+            }]
+          }]
         }
 
         http = Net::HTTP.new(uri.host, uri.port)
         http.use_ssl = true
-        request = Net::HTTP::Post.new(uri.path, headers)
+        request = Net::HTTP::Post.new(uri.path + "?" + uri.query, headers)
         request.body = body.to_json
 
-        response = http.request(request)
-        if response.is_a?(Net::HTTPSuccess)
-          json = JSON.parse(response.body)
-          translated_text = json["translatedText"]
-          translated_text ? translated_text.strip : text
-        else
-          raise "Errore durante la traduzione con Gemini: #{response.body}"
+        begin
+          response = http.request(request)
+          
+          if response.is_a?(Net::HTTPSuccess)
+            json = JSON.parse(response.body)
+            
+            if json["candidates"]&.any? && json["candidates"][0]["content"]["parts"]&.any?
+              translated_text = json["candidates"][0]["content"]["parts"][0]["text"]
+              
+              # Pulizia minima del testo, dato che il prompt è già specifico
+              cleaned_text = translated_text.strip
+                .gsub(/[\*\`\n"]/, '') # Rimuovi markdown, newline e virgolette
+                .gsub(/\s+/, ' ') # Riduci spazi multipli a uno singolo
+              
+              cleaned_text.empty? ? text : cleaned_text
+            else
+              raise "Risposta Gemini non valida: #{json}"
+            end
+          else
+            raise "Errore HTTP #{response.code}: #{response.body}"
+          end
+        rescue => e
+          raise "Errore durante la traduzione con Gemini: #{e.message}"
         end
       end
     end

data/lib/better_translate/service.rb

@@ -60,7 +60,7 @@ module BetterTranslate
                              when :chatgpt
                                Providers::ChatgptProvider.new(BetterTranslate.configuration.openai_key)
                              when :gemini
-                               Providers::GeminiProvider.new(BetterTranslate.configuration.google_gemini_key)
+                               Providers::GeminiProvider.new(BetterTranslate.configuration.gemini_key)
                              else
                                raise "Provider non supportato: #{@provider_name}"
                              end

data/lib/better_translate/similarity_analyzer.rb

@@ -0,0 +1,151 @@
+# frozen_string_literal: true
+
+require "yaml"
+require "set"
+require "json"
+require "time"
+
+module BetterTranslate
+  class SimilarityAnalyzer
+    SIMILARITY_THRESHOLD = 0.75 # Abbassiamo la soglia per trovare più similarità
+    REPORT_FILE = "translation_similarities.json"
+
+    def initialize(yaml_files)
+      @yaml_files = yaml_files
+      @similarities = {}
+    end
+
+    def analyze
+      translations_by_language = load_translations
+      find_similarities(translations_by_language)
+      generate_report
+    end
+
+    private
+
+    def load_translations
+      translations = {}
+      puts "Loading YAML files..."
+      
+      @yaml_files.each do |file|
+        lang_code = File.basename(file, ".yml")
+        translations[lang_code] = YAML.load_file(file)
+        puts "  - Loaded #{lang_code}.yml"
+      end
+
+      translations
+    end
+
+    def find_similarities(translations_by_language)
+      translations_by_language.each do |lang, translations|
+        puts "\nAnalyzing #{lang} translations..."
+        flattened = flatten_translations(translations)
+        keys = flattened.keys
+        similar_found = 0
+        
+        keys.each_with_index do |key1, i|
+          value1 = flattened[key1]
+          
+          # Confronta solo con le chiavi successive per evitare duplicati
+          keys[(i + 1)..-1].each do |key2|
+            value2 = flattened[key2]
+            
+            similarity = calculate_similarity(value1.to_s, value2.to_s)
+            if similarity >= SIMILARITY_THRESHOLD
+              record_similarity(lang, key1, key2, value1, value2, similarity)
+              similar_found += 1
+            end
+          end
+        end
+        
+        puts "  Found #{similar_found} similar translations"
+      end
+    end
+
+    def flatten_translations(hash, prefix = "", result = {})
+      hash.each do |key, value|
+        current_key = prefix.empty? ? key.to_s : "#{prefix}.#{key}"
+        
+        if value.is_a?(Hash)
+          flatten_translations(value, current_key, result)
+        else
+          result[current_key] = value
+        end
+      end
+      
+      result
+    end
+
+    def calculate_similarity(str1, str2)
+      # Implementazione della distanza di Levenshtein normalizzata
+      matrix = Array.new(str1.length + 1) { Array.new(str2.length + 1) }
+      
+      (0..str1.length).each { |i| matrix[i][0] = i }
+      (0..str2.length).each { |j| matrix[0][j] = j }
+      
+      (1..str1.length).each do |i|
+        (1..str2.length).each do |j|
+          cost = str1[i-1] == str2[j-1] ? 0 : 1
+          matrix[i][j] = [
+            matrix[i-1][j] + 1,
+            matrix[i][j-1] + 1,
+            matrix[i-1][j-1] + cost
+          ].min
+        end
+      end
+      
+      max_length = [str1.length, str2.length].max
+      1 - (matrix[str1.length][str2.length].to_f / max_length)
+    end
+
+    def record_similarity(lang, key1, key2, value1, value2, similarity)
+      @similarities[lang] ||= []
+      @similarities[lang] << {
+        key1: key1,
+        key2: key2,
+        value1: value1,
+        value2: value2,
+        similarity: similarity.round(2)
+      }
+    end
+
+    def generate_report
+      puts "\nGenerating reports..."
+      report = {
+        generated_at: Time.now.iso8601,
+        similarity_threshold: SIMILARITY_THRESHOLD,
+        findings: @similarities
+      }
+
+      File.write(REPORT_FILE, JSON.pretty_generate(report))
+      puts "  - Generated #{REPORT_FILE}"
+      
+      summary = generate_summary(report)
+      File.write("translation_similarities_summary.txt", summary)
+      puts "  - Generated translation_similarities_summary.txt"
+    end
+
+    def generate_summary(report)
+      summary = ["Translation Similarities Report", "=" * 30, ""]
+      
+      report[:findings].each do |lang, similarities|
+        summary << "Language: #{lang}"
+        summary << "-" * 20
+        
+        similarities.each do |sim|
+          summary << "Similar translations found:"
+          summary << "  Key 1: #{sim[:key1]}"
+          summary << "  Value 1: #{sim[:value1]}"
+          summary << "  Key 2: #{sim[:key2]}"
+          summary << "  Value 2: #{sim[:value2]}"
+          summary << "  Similarity: #{(sim[:similarity] * 100).round(1)}%"
+          summary << ""
+        end
+        
+        summary << ""
+      end
+      
+      summary.join("\n")
+    end
+  end
+end

data/lib/better_translate/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module BetterTranslate
-  VERSION = "0.3.1"
+  VERSION = "0.4.1"
 end

data/lib/better_translate.rb

@@ -7,6 +7,7 @@ require "better_translate/translator"
 require "better_translate/service"
 require "better_translate/writer"
 require "better_translate/helper"
+require "better_translate/similarity_analyzer"
 
 require 'better_translate/providers/base_provider'
 require 'better_translate/providers/chatgpt_provider'

data/lib/generators/better_translate/analyze_generator.rb

@@ -0,0 +1,43 @@
+# frozen_string_literal: true
+
+require "rails/generators"
+
+module BetterTranslate
+  module Generators
+    class AnalyzeGenerator < Rails::Generators::Base
+      desc "Analyze translation files for similarities"
+
+      def analyze_translations
+        say "Starting translation similarity analysis...", :blue
+
+        # Find all YAML files in the locales directory
+        locale_dir = Rails.root.join("config", "locales")
+        yaml_files = Dir[locale_dir.join("*.yml")]
+
+        if yaml_files.empty?
+          say "No YAML files found in #{locale_dir}", :red
+          return
+        end
+
+        say "Found #{yaml_files.length} YAML files to analyze", :green
+
+        # Run analysis
+        analyzer = BetterTranslate::SimilarityAnalyzer.new(yaml_files)
+        analyzer.analyze
+
+        # Show results
+        say "\nAnalysis complete!", :green
+        say "Reports generated:", :green
+        say "  * #{BetterTranslate::SimilarityAnalyzer::REPORT_FILE} (detailed JSON report)"
+        say "  * translation_similarities_summary.txt (human-readable summary)"
+        
+        # Show quick summary from the text file
+        if File.exist?("translation_similarities_summary.txt")
+          summary = File.read("translation_similarities_summary.txt")
+          say "\nQuick Summary:", :blue
+          say summary
+        end
+      end
+    end
+  end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: better_translate
 version: !ruby/object:Gem::Version
-  version: 0.3.1
+  version: 0.4.1
 platform: ruby
 authors:
 - Alessio Bussolari
@@ -179,19 +179,19 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '3.19'
 - !ruby/object:Gem::Dependency
-  name: vcr
+  name: dotenv
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '6.1'
+        version: '2.8'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '6.1'
+        version: '2.8'
 description: |
   BetterTranslate is a powerful Ruby gem for translating YAML files using AI providers.
 
@@ -222,10 +222,12 @@ files:
 - lib/better_translate/providers/chatgpt_provider.rb
 - lib/better_translate/providers/gemini_provider.rb
 - lib/better_translate/service.rb
+- lib/better_translate/similarity_analyzer.rb
 - lib/better_translate/translator.rb
 - lib/better_translate/utils.rb
 - lib/better_translate/version.rb
 - lib/better_translate/writer.rb
+- lib/generators/better_translate/analyze_generator.rb
 - lib/generators/better_translate/install_generator.rb
 - lib/generators/better_translate/templates/better_translate.rb
 - lib/generators/better_translate/translate_generator.rb