chatwoot/lib/integrations/openai/processor_service.rb

class Integrations::Openai::ProcessorService < Integrations::OpenaiBaseService
  AGENT_INSTRUCTION = 'You are a helpful support agent.'.freeze
  LANGUAGE_INSTRUCTION = 'Ensure that the reply should be in user language.'.freeze
  def reply_suggestion_message
    make_api_call(reply_suggestion_body)
  end

  def summarize_message
    make_api_call(summarize_body)
  end

  def rephrase_message
    make_api_call(rephrase_body)
  end

  def fix_spelling_grammar_message
    make_api_call(fix_spelling_grammar_body)
  end

  def shorten_message
    make_api_call(shorten_body)
  end

  def expand_message
    make_api_call(expand_body)
  end

  def make_friendly_message
    make_api_call(make_friendly_body)
  end

  def make_formal_message
    make_api_call(make_formal_body)
  end

  def simplify_message
    make_api_call(simplify_body)
  end

  private

  def rephrase_body
    build_api_call_body("#{AGENT_INSTRUCTION} Please rephrase the following response. " \
                        "#{LANGUAGE_INSTRUCTION}")
  end

  def fix_spelling_grammar_body
    build_api_call_body("#{AGENT_INSTRUCTION} Please fix the spelling and grammar of the following response. " \
                        "#{LANGUAGE_INSTRUCTION}")
  end

  def shorten_body
    build_api_call_body("#{AGENT_INSTRUCTION} Please shorten the following response. #{LANGUAGE_INSTRUCTION}")
  end

  def expand_body
    build_api_call_body("#{AGENT_INSTRUCTION} Please expand the following response. #{LANGUAGE_INSTRUCTION}")
  end

  def make_friendly_body
    build_api_call_body("#{AGENT_INSTRUCTION} Please make the following response more friendly. #{LANGUAGE_INSTRUCTION}")
  end

  def make_formal_body
    build_api_call_body("#{AGENT_INSTRUCTION} Please make the following response more formal. #{LANGUAGE_INSTRUCTION}")
  end

  def simplify_body
    build_api_call_body("#{AGENT_INSTRUCTION} Please simplify the following response. #{LANGUAGE_INSTRUCTION}")
  end

  def build_api_call_body(system_content, user_content = event['data']['content'])
    {
      model: GPT_MODEL,
      messages: [
        { role: 'system', content: system_content },
        { role: 'user', content: user_content }
      ]
    }.to_json
  end

  def conversation_messages(in_array_format: false)
    conversation = find_conversation
    messages = init_messages_body(in_array_format)

    add_messages_until_token_limit(conversation, messages, in_array_format)
  end

  def add_messages_until_token_limit(conversation, messages, in_array_format, start_from = 0)
    character_count = start_from
    conversation.messages.chat.reorder('id desc').each do |message|
      character_count, message_added = add_message_if_within_limit(character_count, message, messages, in_array_format)
      break unless message_added
    end
    messages
  end

  def add_message_if_within_limit(character_count, message, messages, in_array_format)
    if valid_message?(message, character_count)
      add_message_to_list(message, messages, in_array_format)
      character_count += message.content.length
      [character_count, true]
    else
      [character_count, false]
    end
  end

  def valid_message?(message, character_count)
    message.content.present? && character_count + message.content.length <= TOKEN_LIMIT
  end

  def add_message_to_list(message, messages, in_array_format)
    formatted_message = format_message(message, in_array_format)
    messages.prepend(formatted_message)
  end

  def init_messages_body(in_array_format)
    in_array_format ? [] : ''
  end

  def format_message(message, in_array_format)
    in_array_format ? format_message_in_array(message) : format_message_in_string(message)
  end

  def format_message_in_array(message)
    { role: (message.incoming? ? 'user' : 'assistant'), content: message.content }
  end

  def format_message_in_string(message)
    sender_type = message.incoming? ? 'Customer' : 'Agent'
    "#{sender_type} #{message.sender&.name} : #{message.content}\n"
  end

  def summarize_body
    {
      model: GPT_MODEL,
      messages: [
        { role: 'system',
          content: 'Please summarize the key points from the following conversation between support agents and ' \
                   'customer as bullet points for the next support agent looking into the conversation. Reply in the user\'s language.' },
        { role: 'user', content: conversation_messages }
      ]
    }.to_json
  end

  def reply_suggestion_body
    {
      model: GPT_MODEL,
      messages: [
        { role: 'system',
          content: 'Please suggest a reply to the following conversation between support agents and customer. Reply in the user\'s language.' }
      ].concat(conversation_messages(in_array_format: true))
    }.to_json
  end
end

Integrations::Openai::ProcessorService.prepend_mod_with('Integrations::OpenaiProcessorService')