llm_translate 0.1.0 → 0.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 584007a41c9b59041a1ecd9e42c96231c66fdc606edb19567c58fd342d838a15
-  data.tar.gz: 1b9afce81578be82bcfd4f81a7fa6050073d9eef8848a7ee43ef9110b38ead94
+  metadata.gz: b3d7bffb10cabd77729e1e806a256a16bd7a036faa0ea52b7024e3a521dada5e
+  data.tar.gz: 50fcf8a22940afb311387913d2455dc519812abe8618be54c47541808193afd6
 SHA512:
-  metadata.gz: a959c5dfcf20bfc2bc94f787cf8c3e49ee75417a5be461a0b2650d023ce964670dd566e118e764b3152c3afdc2a22d2ac33605fdfca7909e55848e0761b0515c
-  data.tar.gz: d56df324476f82d6deb7e4b28bba026349661e4b5c389e1068e09dae36ecae7c38a8d38cee8cbfa861c686db3f063f3715cc0c18430e462d86f4ebf69079ce72
+  metadata.gz: 76e635e0838f893377ed9ba05e8e4e04c36053cbb4c68b2ab09cbd8c69d8a7433400886e1f817647d9b322fc7dad1e4643cc53b6471fc7ed2b6538df015108d7
+  data.tar.gz: 713bb859602fd5f511444f8e491bcd556beb7954f2f03cb36ff79938ff1e07e60c5419dd2c9818dc4d1f95136af8d4722b649a267e4217128598a26036b50532

data/README.md

@@ -177,11 +177,6 @@ translation:
   default_prompt: "Your custom prompt with {content} placeholder"
   preserve_formatting: true
   translate_code_comments: false
-  preserve_patterns:
-    - "```[\\s\\S]*?```"    # Code blocks
-    - "`[^`]+`"              # Inline code
-    - "\\[.*?\\]\\(.*?\\)"   # Links
-    - "!\\[.*?\\]\\(.*?\\)"  # Images
 
 # File Processing
 files:

data/README.zh.md

@@ -1,21 +1,21 @@
 # LlmTranslate
 
-一个由AI驱动的Markdown翻译器，能够在翻译内容时保持格式不变，同时使用各种AI提供者。
+AI 驱动的 Markdown 翻译工具，在使用各种 AI 提供商翻译内容的同时保持格式不变。
 
-## 特性
+## 功能特性
 
-- 🤖 **AI驱动的翻译**：支持OpenAI、Anthropic和Ollama
-- 📝 **Markdown格式保留**：保持代码块、链接、图像和格式不变
-- 🔧 **灵活配置**：基于YAML的配置，支持环境变量
+- 🤖 **AI 驱动翻译**：支持 OpenAI、Anthropic 和 Ollama
+- 📝 **Markdown 格式保留**：保持代码块、链接、图片和格式不变
+- 🔧 **灵活配置**：基于 YAML 的配置，支持环境变量
 - 📁 **批量处理**：递归处理整个目录结构
-- 🚀 **CLI接口**：易于使用的命令行接口，使用Thor
-- 📊 **进度跟踪**：内置日志记录和报告
-- ⚡ **错误处理**：强大的错误处理机制，带有重试机制
-- 🎯 **可定制**：自定义提示、文件模式和输出策略
+- 🚀 **CLI 界面**：使用 Thor 实现的易用命令行界面
+- 📊 **进度跟踪**：内置日志记录和报告功能
+- ⚡ **错误处理**：具有重试机制的强大错误处理
+- 🎯 **可定制性**：自定义提示、文件模式和输出策略
 
 ## 安装
 
-将此行添加到您应用程序的Gemfile中：
+将以下行添加到您应用程序的 Gemfile 中：
 
 ```ruby
 gem 'llm_translate'
@@ -27,15 +27,15 @@ gem 'llm_translate'
 bundle install
 ```
 
-或者您也可以自己安装：
+或者自行安装：
 
 ```bash
 gem install llm_translate
 ```
 
-## 依赖
+## 依赖项
 
-该gem需要`rubyllm` gem进行AI集成：
+该 gem 需要 `rubyllm` gem 进行 AI 集成：
 
 ```bash
 gem install rubyllm
@@ -48,22 +48,22 @@ gem install rubyllm
    llm_translate init
    ```
 
-2. **设置您的API密钥**：
+2. **设置您的 API 密钥**：
    ```bash
    export LLM_TRANSLATE_API_KEY="your-api-key-here"
    ```
 
-3. **翻译您的markdown文件**：
+3. **翻译您的 markdown 文件**：
    ```bash
-   llm_translate translate --config ./translator.yml
+   llm_translate translate --config ./llm_translate.yml
    ```
 
 ## 配置
 
-翻译器使用YAML配置文件。以下是一个最小示例：
+翻译器使用 YAML 配置文件。这是一个最小示例：
 
 ```yaml
-# translator.yml
+# llm_translate.yml
 ai:
   api_key: ${LLM_TRANSLATE_API_KEY}
   provider: "openai"
@@ -91,7 +91,7 @@ logging:
   output: "console"
 ```
 
-### AI提供者
+### AI 提供商
 
 #### OpenAI
 ```yaml
@@ -109,101 +109,188 @@ ai:
   model: "claude-3-sonnet-20240229"
 ```
 
-#### Ollama（本地）
-```bash
-bundle install
-```0
+#### Ollama (本地)
+```yaml
+ai:
+  provider: "ollama"
+  model: "llama2"
+  # 如果不使用默认设置，请设置 OLLAMA_HOST 环境变量
+```
 
-## 使用
+## 使用方法
 
 ### 基本翻译
 
 #### 目录模式（默认）
 ```bash
-bundle install
-```1
+llm_translate translate --config ./llm_translate.yml
+```
 
 #### 单文件模式
-要翻译单个文件，请在配置中设置`input_file```bash
-   llm_translate init
-   ```7output_file`：
+要翻译单个文件，请在配置中设置 `input_file` 和 `output_file`：
 
-```bash
-bundle install
-```2
+```yaml
+files:
+  # 单文件模式
+  input_file: "./README.md"
+  output_file: "./README.zh.md"
+```
 
-当同时指定`input_file```bash
-   llm_translate init
-   ```7output_file`时，翻译器将以单文件模式运行，忽略与目录相关的设置。
+当同时指定 `input_file` 和 `output_file` 时，翻译器将以单文件模式运行，忽略与目录相关的设置。
 
 ### 命令行选项
 
 ```bash
-bundle install
-```3
+llm_translate translate [OPTIONS]
+
+Options:
+  -c, --config PATH      配置文件路径（默认：./llm_translate.yml）
+  -i, --input PATH       输入目录（覆盖配置）
+  -o, --output PATH      输出目录（覆盖配置）
+  -p, --prompt TEXT      自定义翻译提示（覆盖配置）
+  -v, --verbose          启用详细输出
+  -d, --dry-run          执行试运行而不进行实际翻译
+
+Other Commands:
+  llm_translate init        初始化新的配置文件
+  llm_translate version     显示版本信息
+```
 
 ### 配置文件结构
 
-```bash
-bundle install
-```4[\s\S]*?```bash
-bundle install
-```5
+```yaml
+# AI 配置
+ai:
+  api_key: ${LLM_TRANSLATE_API_KEY}
+  provider: "openai"  # openai, anthropic, ollama
+  model: "gpt-4"
+  temperature: 0.3
+  max_tokens: 4000
+  retry_attempts: 3
+  retry_delay: 2
+  timeout: 60
+
+# 翻译设置
+translation:
+  target_language: "zh-CN"
+  source_language: "auto"
+  default_prompt: "您的自定义提示，包含 {content} 占位符"
+  preserve_formatting: true
+  translate_code_comments: false
+
+# 文件处理
+files:
+  input_directory: "./docs"
+  output_directory: "./docs-translated"
+  filename_strategy: "suffix"  # suffix, replace, directory
+  filename_suffix: ".zh"
+  include_patterns:
+    - "**/*.md"
+    - "**/*.markdown"
+  exclude_patterns:
+    - "**/node_modules/**"
+    - "**/.*"
+  preserve_directory_structure: true
+  overwrite_policy: "ask"  # ask, overwrite, skip, backup
+  backup_directory: "./backups"
+
+# 日志记录
+logging:
+  level: "info"  # debug, info, warn, error
+  output: "console"  # console, file, both
+  file_path: "./logs/translator.log"
+  verbose_translation: false
+  error_log_path: "./logs/errors.log"
+
+# 错误处理
+error_handling:
+  on_error: "log_and_continue"  # stop, log_and_continue, skip_file
+  max_consecutive_errors: 5
+  retry_on_failure: 2
+  generate_error_report: true
+  error_report_path: "./logs/error_report.md"
+
+# 性能
+performance:
+  concurrent_files: 3
+  batch_size: 5
+  request_interval: 1  # 请求之间的秒数
+  max_memory_mb: 500
+
+# 输出
+output:
+  show_progress: true
+  show_statistics: true
+  generate_report: true
+  report_path: "./reports/translation_report.md"
+  format: "markdown"
+  include_metadata: true
+```
 
 ## 示例
 
 ### 翻译文档
 
 ```bash
-bundle install
-```6
+# 将 ./docs 中的所有 markdown 文件翻译为中文
+llm_translate translate --input ./docs --output ./docs-zh
+
+# 使用自定义提示
+llm_translate translate --prompt "翻译以下内容为中文，保持技术术语不变: {content}"
+
+# 试运行以查看将要翻译的内容
+llm_translate translate --dry-run --verbose
+```
 
 ### 批量翻译
 
 ```bash
-bundle install
-```7
+# 翻译多种语言版本
+for lang in zh-CN ja-JP ko-KR; do
+  llm_translate translate --config "./configs/llm_translate-${lang}.yml"
+done
+```
 
 ## 开发
 
-克隆代码库后，运行：
+检出仓库后，运行：
 
 ```bash
 bundle install
 ```
 
-要运行测试：
+运行测试：
 
 ```bash
-bundle install
-```9
+bundle exec rspec
+```
 
-要运行代码检查：
+运行代码检查：
 
 ```bash
-gem install llm_translate
-```0
+bundle exec rubocop
+```
 
-要将此gem安装到您的本地机器上：
+将此 gem 安装到您的本地机器：
 
 ```bash
-gem install llm_translate
-```1
+bundle exec rake install
+```
 
 ## 贡献
 
-欢迎在GitHub上提交错误报告和拉取请求，地址为 https://github.com/translator/translator。
+欢迎在 GitHub 上提交错误报告和拉取请求：https://github.com/llm_translate/llm_translate。
 
 ## 许可证
 
-该gem在[MIT许可证](https://opensource.org/licenses/MIT)条款下作为开源软件提供。
+该 gem 根据 [MIT 许可证](https://opensource.org/licenses/MIT) 的条款作为开源软件提供。
 
 ## 更新日志
 
 ### v0.1.0
-- 初始发布
-- 支持OpenAI、Anthropic和Ollama提供者
-- Markdown格式保留
+- 初始版本
+- 支持 OpenAI、Anthropic 和 Ollama 提供商
+- Markdown 格式保留
 - 可配置的翻译提示
 - 批量文件处理
-- 综合的错误处理和日志记录
+- 全面的错误处理和日志记录

data/content/llm_translate.yml

@@ -50,12 +50,7 @@ translation:
   # 是否翻译代码注释
   translate_code_comments: false
   
-  # 需要保留不翻译的内容模式
-  preserve_patterns:
-    - "```[\\s\\S]*?```"  # 代码块
-    - "`[^`]+`"            # 行内代码
-    - "\\[.*?\\]\\(.*?\\)" # 链接
-    - "!\\[.*?\\]\\(.*?\\)" # 图片
+
 
 # 文件处理配置
 files:

data/lib/llm_translate/ai_client.rb

@@ -1,6 +1,8 @@
 # frozen_string_literal: true
 
 require 'ruby_llm'
+# require 'pry'
+
 module LlmTranslate
   class AiClient
     attr_reader :config, :logger
@@ -56,16 +58,14 @@ module LlmTranslate
 
     def configure_ruby_llm
       RubyLLM.configure do |config_obj|
-        # For aihubmix.com or any custom host, use OpenAI-compatible API
         config_obj.openai_api_key = config.api_key
         config_obj.openai_api_base = config.ai_host
-        config_obj.default_model = config.ai_model
       end
     end
 
     def make_request(prompt)
       chat = RubyLLM.chat
-                    .with_model(config.ai_model)
+                    .with_model(config.ai_model, assume_exists: true, provider: config.ai_provider)
                     .with_temperature(config.temperature)
 
       response = chat.ask(prompt)

data/lib/llm_translate/cli.rb

@@ -70,24 +70,19 @@ module LlmTranslate
         end
 
         # Translate files
-        success_count = 0
-        error_count = 0
-
-        files.each_with_index do |file_path, index|
-          logger.info "[#{index + 1}/#{files.length}] Processing: #{file_path}"
-
-          translator_engine.translate_file(file_path) unless options[:dry_run]
+        if options[:dry_run]
+          logger.info "DRY RUN: Would translate #{files.length} files with #{config.concurrent_files} concurrent threads"
+          success_count = files.length
+          error_count = 0
+        else
+          logger.info "Starting translation with #{config.concurrent_files} concurrent files"
 
-          success_count += 1
-          logger.info "✓ Successfully processed: #{file_path}"
-        rescue StandardError => e
-          error_count += 1
-          logger.error "✗ Failed to process #{file_path}: #{e.message}"
+          results = translator_engine.translate_files_concurrently(files)
+          success_count = results[:success].length
+          error_count = results[:error].length
 
-          if config.should_stop_on_error?(error_count)
-            logger.error 'Stopping due to too many consecutive errors'
-            break
-          end
+          # Check if we should stop on too many errors
+          logger.error "Stopping due to too many errors (#{error_count})" if config.should_stop_on_error?(error_count)
         end
 
         # Summary

data/lib/llm_translate/config.rb

@@ -75,10 +75,6 @@ module LlmTranslate
       data.dig('translation', 'translate_code_comments') == true
     end
 
-    def preserve_patterns
-      data.dig('translation', 'preserve_patterns') || default_preserve_patterns
-    end
-
     # File Configuration
     def input_directory
       cli_options[:input] || data.dig('files', 'input_directory') || './docs'
@@ -124,10 +120,6 @@ module LlmTranslate
       data.dig('files', 'overwrite_policy') || 'ask'
     end
 
-    def backup_directory
-      data.dig('files', 'backup_directory') || './backups'
-    end
-
     # Logging Configuration
     def log_level
       cli_options[:verbose] ? 'debug' : (data.dig('logging', 'level') || 'info')
@@ -137,18 +129,10 @@ module LlmTranslate
       data.dig('logging', 'output') || 'console'
     end
 
-    def log_file_path
-      data.dig('logging', 'file_path') || './logs/llm_translate.log'
-    end
-
     def verbose_translation?
       cli_options[:verbose] || data.dig('logging', 'verbose_translation') == true
     end
 
-    def error_log_path
-      data.dig('logging', 'error_log_path') || './logs/errors.log'
-    end
-
     # Error Handling Configuration
     def on_error
       data.dig('error_handling', 'on_error') || 'log_and_continue'
@@ -166,10 +150,6 @@ module LlmTranslate
       data.dig('error_handling', 'generate_error_report') != false
     end
 
-    def error_report_path
-      data.dig('error_handling', 'error_report_path') || './logs/error_report.md'
-    end
-
     def should_stop_on_error?(error_count)
       on_error == 'stop' || error_count >= max_consecutive_errors
     end
@@ -179,27 +159,11 @@ module LlmTranslate
       data.dig('performance', 'concurrent_files') || 3
     end
 
-    def batch_size
-      data.dig('performance', 'batch_size') || 5
-    end
-
     def request_interval
       data.dig('performance', 'request_interval') || 1
     end
 
-    def max_memory_mb
-      data.dig('performance', 'max_memory_mb') || 500
-    end
-
     # Output Configuration
-    def show_progress?
-      data.dig('output', 'show_progress') != false
-    end
-
-    def show_statistics?
-      data.dig('output', 'show_statistics') != false
-    end
-
     def generate_report?
       data.dig('output', 'generate_report') != false
     end
@@ -208,14 +172,6 @@ module LlmTranslate
       data.dig('output', 'report_path') || './reports/translation_report.md'
     end
 
-    def output_format
-      data.dig('output', 'format') || 'markdown'
-    end
-
-    def include_metadata?
-      data.dig('output', 'include_metadata') != false
-    end
-
     private
 
     def load_config_file(config_path)
@@ -266,14 +222,5 @@ module LlmTranslate
         {content}
       PROMPT
     end
-
-    def default_preserve_patterns
-      [
-        '```[\\s\\S]*?```',      # Code blocks
-        '`[^`]+`',               # Inline code
-        '\\[.*?\\]\\(.*?\\)',    # Links
-        '!\\[.*?\\]\\(.*?\\)'    # Images
-      ]
-    end
   end
 end

data/lib/llm_translate/logger.rb

@@ -67,7 +67,7 @@ module LlmTranslate
       when 'console'
         create_console_logger
       when 'file'
-        create_file_logger(config.log_file_path)
+        create_file_logger('./logs/llm_translate.log')
       when 'both'
         create_multi_logger
       else
@@ -76,15 +76,8 @@ module LlmTranslate
     end
 
     def create_error_logger
-      return nil unless config.error_log_path
-
-      FileUtils.mkdir_p(File.dirname(config.error_log_path))
-      error_logger = ::Logger.new(config.error_log_path)
-      error_logger.level = log_level_constant
-      error_logger.formatter = proc do |severity, datetime, _progname, msg|
-        "[#{datetime.strftime('%Y-%m-%d %H:%M:%S')}] #{severity}: #{msg}\n"
-      end
-      error_logger
+      # Error logger is no longer supported, return nil
+      nil
     end
 
     def create_console_logger
@@ -119,7 +112,7 @@ module LlmTranslate
 
     def create_multi_logger
       console_logger = create_console_logger
-      file_logger = create_file_logger(config.log_file_path)
+      file_logger = create_file_logger('./logs/llm_translate.log')
 
       MultiLogger.new([console_logger, file_logger])
     end

data/lib/llm_translate/translator_engine.rb

@@ -2,6 +2,7 @@
 
 require 'pathname'
 require 'fileutils'
+require 'async'
 
 module LlmTranslate
   class TranslatorEngine
@@ -53,6 +54,66 @@ module LlmTranslate
       sleep(config.request_interval) if config.request_interval.positive?
     end
 
+    def translate_files_concurrently(file_paths)
+      return translate_files_sequentially(file_paths) if config.concurrent_files <= 1
+
+      results = { success: [], error: [] }
+
+      # Use Async to run concurrent translation tasks
+      Async do |task|
+        # Process files in batches to limit concurrency
+        file_paths.each_slice(config.concurrent_files) do |batch|
+          # Create async tasks for the current batch
+          batch_tasks = batch.map.with_index do |file_path, _batch_index|
+            # Calculate overall index
+            overall_index = file_paths.index(file_path) + 1
+
+            task.async do
+              logger.info "[#{overall_index}/#{file_paths.length}] Processing: #{file_path}"
+
+              # Translate the file
+              translate_file(file_path)
+
+              # Collect successful result
+              results[:success] << file_path
+
+              logger.info "✓ Successfully processed: #{file_path}"
+              { status: :success, file: file_path }
+            rescue StandardError => e
+              # Collect error result
+              results[:error] << { file: file_path, error: e.message }
+
+              logger.error "✗ Failed to process #{file_path}: #{e.message}"
+              { status: :error, file: file_path, error: e.message }
+            end
+          end
+
+          # Wait for all tasks in this batch to complete before starting the next batch
+          batch_tasks.each(&:wait)
+        end
+      end
+
+      results
+    end
+
+    def translate_files_sequentially(file_paths)
+      results = { success: [], error: [] }
+
+      file_paths.each_with_index do |file_path, index|
+        logger.info "[#{index + 1}/#{file_paths.length}] Processing: #{file_path}"
+
+        translate_file(file_path)
+
+        results[:success] << file_path
+        logger.info "✓ Successfully processed: #{file_path}"
+      rescue StandardError => e
+        results[:error] << { file: file_path, error: e.message }
+        logger.error "✗ Failed to process #{file_path}: #{e.message}"
+      end
+
+      results
+    end
+
     def translate_content(content, file_path = nil)
       if config.preserve_formatting?
         translate_with_format_preservation(content)
@@ -87,147 +148,8 @@ module LlmTranslate
     end
 
     def translate_with_format_preservation(content)
-      # Extract and preserve special markdown elements
-      preserved_elements = extract_preserved_elements(content)
-
-      # Replace preserved elements with placeholders
-      content_with_placeholders = replace_with_placeholders(content, preserved_elements)
-
       # Translate the content with placeholders
-      translated_content = ai_client.translate(content_with_placeholders)
-
-      # Restore preserved elements
-      restore_preserved_elements(translated_content, preserved_elements)
-    end
-
-    def extract_preserved_elements(content)
-      preserved = {}
-      pattern_index = 0
-
-      config.preserve_patterns.each do |pattern|
-        regex = Regexp.new(pattern, Regexp::MULTILINE)
-
-        content.scan(regex) do |match|
-          # Handle both single match and capture groups
-          match_text = match.is_a?(Array) ? match[0] : match
-          placeholder = "PRESERVED_ELEMENT_#{pattern_index}"
-          preserved[placeholder] = match_text
-          pattern_index += 1
-        end
-      end
-
-      preserved
-    end
-
-    def replace_with_placeholders(content, preserved_elements)
-      result = content.dup
-
-      preserved_elements.each do |placeholder, original_text|
-        # Escape special regex characters in the original text
-        escaped_text = Regexp.escape(original_text)
-        result = result.gsub(Regexp.new(escaped_text), placeholder)
-      end
-
-      result
-    end
-
-    def restore_preserved_elements(translated_content, preserved_elements)
-      result = translated_content.dup
-
-      preserved_elements.each do |placeholder, original_text|
-        result = result.gsub(placeholder, original_text)
-      end
-
-      result
-    end
-
-    # Additional helper methods for handling special cases
-
-    def split_large_content(content, max_size = 3000)
-      # Split content into chunks if it's too large for the AI model
-      return [content] if content.length <= max_size
-
-      chunks = []
-      lines = content.split("\n")
-      current_chunk = ''
-
-      lines.each do |line|
-        # If adding this line would exceed the limit, start a new chunk
-        if "#{current_chunk}#{line}\n".length > max_size && !current_chunk.empty?
-          chunks << current_chunk.strip
-          current_chunk = "#{line}\n"
-        else
-          current_chunk += "#{line}\n"
-        end
-      end
-
-      # Add the last chunk if it's not empty
-      chunks << current_chunk.strip unless current_chunk.strip.empty?
-
-      chunks
-    end
-
-    def translate_large_content(content)
-      chunks = split_large_content(content)
-
-      return ai_client.translate(content) if chunks.length == 1
-
-      logger.info "Splitting large content into #{chunks.length} chunks"
-
-      translated_chunks = chunks.map.with_index do |chunk, index|
-        logger.debug "Translating chunk #{index + 1}/#{chunks.length}"
-
-        translated = ai_client.translate(chunk)
-
-        # Add delay between chunks to avoid rate limiting
-        sleep(config.request_interval) if config.request_interval.positive? && index < chunks.length - 1
-
-        translated
-      end
-
-      translated_chunks.join("\n\n")
-    end
-
-    def detect_language(content)
-      # Simple language detection based on content
-      # This is a basic implementation - could be enhanced with a proper language detection library
-
-      # Check for common English words
-      english_indicators = %w[the and or but with from this that these those]
-      chinese_indicators = %w[的 在 是 和 或者 但是 这 那]
-
-      english_score = english_indicators.count { |word| content.downcase.include?(word) }
-      chinese_score = chinese_indicators.count { |word| content.include?(word) }
-
-      if chinese_score > english_score
-        'zh'
-      elsif english_score.positive?
-        'en'
-      else
-        config.source_language
-      end
-    end
-
-    def should_translate_content?(content)
-      # Skip translation if content is mostly code or already in target language
-
-      # Skip if content is mostly code blocks
-      code_block_pattern = /```[\s\S]*?```/m
-      code_blocks = content.scan(code_block_pattern)
-      code_length = code_blocks.join.length
-
-      if code_length > content.length * 0.8
-        logger.debug 'Skipping translation: content is mostly code blocks'
-        return false
-      end
-
-      # Skip if content is very short
-      if content.strip.length < 10
-        logger.debug 'Skipping translation: content too short'
-        return false
-      end
-
-      true
+      ai_client.translate(content)
     end
   end
 end

data/lib/llm_translate/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module LlmTranslate
-  VERSION = '0.1.0'
+  VERSION = '0.3.0'
 end

data/llm_translate.gemspec

@@ -31,6 +31,7 @@ Gem::Specification.new do |spec|
   spec.require_paths = ['lib']
 
   # Dependencies
+  spec.add_dependency 'async', '~> 2.0'
   spec.add_dependency 'ruby_llm', '~> 1.6'
   spec.add_dependency 'thor', '~> 1.3'
 

data/llm_translate.yml

@@ -50,12 +50,7 @@ translation:
   # 是否翻译代码注释
   translate_code_comments: false
   
-  # 需要保留不翻译的内容模式
-  preserve_patterns:
-    - "```[\\s\\S]*?```"  # 代码块
-    - "`[^`]+`"            # 行内代码
-    - "\\[.*?\\]\\(.*?\\)" # 链接
-    - "!\\[.*?\\]\\(.*?\\)" # 图片
+
 
 # 文件处理配置
 files:

data/test_llm_translate.yml

@@ -48,11 +48,7 @@ translation:
   translate_code_comments: false
   
   # 需要保留不翻译的内容模式
-  preserve_patterns:
-    - "```[\\s\\S]*?```"  # 代码块
-    - "`[^`]+`"            # 行内代码
-    - "\\[.*?\\]\\(.*?\\)" # 链接
-    - "!\\[.*?\\]\\(.*?\\)" # 图片
+
 
 # 文件处理配置
 files:

data/test_new_config.yml

@@ -50,12 +50,7 @@ translation:
   # 是否翻译代码注释
   translate_code_comments: false
   
-  # 需要保留不翻译的内容模式
-  preserve_patterns:
-    - "```[\\s\\S]*?```"  # 代码块
-    - "`[^`]+`"            # 行内代码
-    - "\\[.*?\\]\\(.*?\\)" # 链接
-    - "!\\[.*?\\]\\(.*?\\)" # 图片
+
 
 # 文件处理配置
 files:

metadata

@@ -1,15 +1,29 @@
 --- !ruby/object:Gem::Specification
 name: llm_translate
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.3.0
 platform: ruby
 authors:
 - LlmTranslate Team
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2025-08-19 00:00:00.000000000 Z
+date: 2025-09-01 00:00:00.000000000 Z
 dependencies:
+- !ruby/object:Gem::Dependency
+  name: async
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.0'
 - !ruby/object:Gem::Dependency
   name: ruby_llm
   requirement: !ruby/object:Gem::Requirement