langchainrb 0.15.6 → 0.16.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: e5ee7b8ae1bd67027846f794a9d070ce4dcedc6e3c6c36eb169a42ba8a55f124
-  data.tar.gz: c5889ddebb21147551c33fe8be1df6ce1659018d6fb489579e497567fd744bf0
+  metadata.gz: 3f685910b0f3f1816c3822debc2ec470d72d85203b2695ab8ab780b5d0f1cb09
+  data.tar.gz: 7ec406ad7980e12739aa70e9710b21e1f7df0a1e46f66820d7003026e3bbc877
 SHA512:
-  metadata.gz: ef5d74c4e9f2a0e5f0f13b000dd5b82bbfa3ed9481d0bd1defce49d03da7071b31e02a5ec0237cce43b8ea8d76607541e59fe400d9db8803cf6961bb02b2f5bc
-  data.tar.gz: 01a816d66a313b387ce2b37789501636e7b7bede507bd61253bcea7cac3cd641fe74ab18498fe7ddff1bc5b8e38ec39bc0385912ae9e9f310cf94f4d5ec31e9c
+  metadata.gz: 1145ffbab814f09acb539df3662f0c4c5536ded25252a4db3e640d29cc550930b11ac9310b11436d591e1ec13449af59f125bdab3fe463f9a59683ea9d8f38ed
+  data.tar.gz: d9e76d70f24f964b3f17addda08ace46695149af48c7bd4aff1936a10888640f9cb222bef15bdb205ad7997028f88c81142dcb3ef49d96fcd37682c56409f4c6

data/CHANGELOG.md

@@ -1,5 +1,9 @@
 ## [Unreleased]
 
+## [0.16.0] - 2024-09-19
+- Remove `Langchain::Thread` class as it was not needed.
+- Support `cohere` provider for `Langchain::LLM::AwsBedrock#embed`
+
 ## [0.15.6] - 2024-09-16
 - Throw an error when `Langchain::Assistant#add_message_callback` is not a callable proc.
 - Resetting instructions on Langchain::Assistant with Google Gemini no longer throws an error.

data/lib/langchain/assistants/assistant.rb

@@ -12,27 +12,23 @@ module Langchain
   #       tools: [Langchain::Tool::NewsRetriever.new(api_key: ENV["NEWS_API_KEY"])]
   #     )
   class Assistant
-    extend Forwardable
-    def_delegators :thread, :messages
-
-    attr_reader :llm, :thread, :instructions, :state, :llm_adapter, :tool_choice
-    attr_reader :total_prompt_tokens, :total_completion_tokens, :total_tokens
+    attr_reader :llm, :instructions, :state, :llm_adapter, :tool_choice
+    attr_reader :total_prompt_tokens, :total_completion_tokens, :total_tokens, :messages
     attr_accessor :tools, :add_message_callback
 
     # Create a new assistant
     #
     # @param llm [Langchain::LLM::Base] LLM instance that the assistant will use
-    # @param thread [Langchain::Thread] The thread that'll keep track of the conversation
     # @param tools [Array<Langchain::Tool::Base>] Tools that the assistant has access to
-    # @param instructions [String] The system instructions to include in the thread
+    # @param instructions [String] The system instructions
     # @param tool_choice [String] Specify how tools should be selected. Options: "auto", "any", "none", or <specific function name>
     # @params add_message_callback [Proc] A callback function (Proc or lambda) that is called when any message is added to the conversation
     def initialize(
       llm:,
-      thread: nil,
       tools: [],
       instructions: nil,
       tool_choice: "auto",
+      messages: [],
       add_message_callback: nil
     )
       unless tools.is_a?(Array) && tools.all? { |tool| tool.class.singleton_class.included_modules.include?(Langchain::ToolDefinition) }
@@ -42,14 +38,13 @@ module Langchain
       @llm = llm
       @llm_adapter = LLM::Adapter.build(llm)
 
-      @thread = thread || Langchain::Thread.new
-
       # TODO: Validate that it is, indeed, a Proc or lambda
       if !add_message_callback.nil? && !add_message_callback.respond_to?(:call)
         raise ArgumentError, "add_message_callback must be a callable object, like Proc or lambda"
       end
-      @thread.add_message_callback = add_message_callback
+      @add_message_callback = add_message_callback
 
+      self.messages = messages
       @tools = tools
       self.tool_choice = tool_choice
       @instructions = instructions
@@ -59,41 +54,60 @@ module Langchain
       @total_completion_tokens = 0
       @total_tokens = 0
 
-      raise ArgumentError, "Thread must be an instance of Langchain::Thread" unless @thread.is_a?(Langchain::Thread)
-
-      # The first message in the thread should be the system instructions
+      # The first message in the messages array should be the system instructions
       # For Google Gemini, and Anthropic system instructions are added to the `system:` param in the `chat` method
       initialize_instructions
     end
 
-    # Add a user message to the thread
+    # Add a user message to the messages array
     #
     # @param content [String] The content of the message
     # @param role [String] The role attribute of the message. Default: "user"
     # @param tool_calls [Array<Hash>] The tool calls to include in the message
     # @param tool_call_id [String] The ID of the tool call to include in the message
-    # @return [Array<Langchain::Message>] The messages in the thread
+    # @return [Array<Langchain::Message>] The messages
     def add_message(content: nil, role: "user", tool_calls: [], tool_call_id: nil)
       message = build_message(role: role, content: content, tool_calls: tool_calls, tool_call_id: tool_call_id)
-      messages = thread.add_message(message)
+
+      # Call the callback with the message
+      add_message_callback.call(message) if add_message_callback # rubocop:disable Style/SafeNavigation
+
+      # Prepend the message to the messages array
+      messages << message
+
       @state = :ready
 
       messages
     end
 
-    # Set multiple messages to the thread
+    # Convert messages to an LLM APIs-compatible array of hashes
     #
-    # @param messages [Array<Hash>] The messages to set
-    # @return [Array<Langchain::Message>] The messages in the thread
+    # @return [Array<Hash>] Messages as an OpenAI API-compatible array of hashes
+    def array_of_message_hashes
+      messages
+        .map(&:to_hash)
+        .compact
+    end
+
+    # Only used by the Assistant when it calls the LLM#complete() method
+    def prompt_of_concatenated_messages
+      messages.map(&:to_s).join
+    end
+
+    # Set multiple messages
+    #
+    # @param messages [Array<Langchain::Message>] The messages to set
+    # @return [Array<Langchain::Message>] The messages
     def messages=(messages)
-      clear_thread!
-      add_messages(messages: messages)
+      raise ArgumentError, "messages array must only contain Langchain::Message instance(s)" unless messages.is_a?(Array) && messages.all? { |m| m.is_a?(Langchain::Messages::Base) }
+
+      @messages = messages
     end
 
-    # Add multiple messages to the thread
+    # Add multiple messages
     #
     # @param messages [Array<Hash>] The messages to add
-    # @return [Array<Langchain::Message>] The messages in the thread
+    # @return [Array<Langchain::Message>] The messages
     def add_messages(messages:)
       messages.each do |message_hash|
         add_message(**message_hash.slice(:content, :role, :tool_calls, :tool_call_id))
@@ -103,10 +117,10 @@ module Langchain
     # Run the assistant
     #
     # @param auto_tool_execution [Boolean] Whether or not to automatically run tools
-    # @return [Array<Langchain::Message>] The messages in the thread
+    # @return [Array<Langchain::Message>] The messages
     def run(auto_tool_execution: false)
-      if thread.messages.empty?
-        Langchain.logger.warn("No messages in the thread")
+      if messages.empty?
+        Langchain.logger.warn("No messages to process")
         @state = :completed
         return
       end
@@ -114,39 +128,39 @@ module Langchain
       @state = :in_progress
       @state = handle_state until run_finished?(auto_tool_execution)
 
-      thread.messages
+      messages
     end
 
     # Run the assistant with automatic tool execution
     #
-    # @return [Array<Langchain::Message>] The messages in the thread
+    # @return [Array<Langchain::Message>] The messages
     def run!
       run(auto_tool_execution: true)
     end
 
-    # Add a user message to the thread and run the assistant
+    # Add a user message and run the assistant
     #
     # @param content [String] The content of the message
     # @param auto_tool_execution [Boolean] Whether or not to automatically run tools
-    # @return [Array<Langchain::Message>] The messages in the thread
+    # @return [Array<Langchain::Message>] The messages
     def add_message_and_run(content:, auto_tool_execution: false)
       add_message(content: content, role: "user")
       run(auto_tool_execution: auto_tool_execution)
     end
 
-    # Add a user message to the thread and run the assistant with automatic tool execution
+    # Add a user message and run the assistant with automatic tool execution
     #
     # @param content [String] The content of the message
-    # @return [Array<Langchain::Message>] The messages in the thread
+    # @return [Array<Langchain::Message>] The messages
     def add_message_and_run!(content:)
       add_message_and_run(content: content, auto_tool_execution: true)
     end
 
-    # Submit tool output to the thread
+    # Submit tool output
     #
     # @param tool_call_id [String] The ID of the tool call to submit output for
     # @param output [String] The output of the tool
-    # @return [Array<Langchain::Message>] The messages in the thread
+    # @return [Array<Langchain::Message>] The messages
     def submit_tool_output(tool_call_id:, output:)
       tool_role = determine_tool_role
 
@@ -154,18 +168,21 @@ module Langchain
       add_message(role: tool_role, content: output, tool_call_id: tool_call_id)
     end
 
-    # Delete all messages in the thread
+    # Delete all messages
     #
     # @return [Array] Empty messages array
-    def clear_thread!
+    def clear_messages!
       # TODO: If this a bug? Should we keep the "system" message?
-      thread.messages = []
+      @messages = []
     end
 
+    # TODO: Remove in the next major release
+    alias_method :clear_thread!, :clear_messages!
+
     # Set new instructions
     #
     # @param new_instructions [String] New instructions that will be set as a system message
-    # @return [Array<Langchain::Message>] The messages in the thread
+    # @return [Array<Langchain::Message>] The messages
     def instructions=(new_instructions)
       @instructions = new_instructions
 
@@ -173,7 +190,7 @@ module Langchain
       if !llm.is_a?(Langchain::LLM::GoogleGemini) &&
           !llm.is_a?(Langchain::LLM::GoogleVertexAI) &&
           !llm.is_a?(Langchain::LLM::Anthropic)
-        # Find message with role: "system" in thread.messages and delete it from the thread.messages array
+        # Find message with role: "system" in messages and delete it from the messages array
         replace_system_message!(content: new_instructions)
       end
     end
@@ -192,12 +209,12 @@ module Langchain
     # Replace old system message with new one
     #
     # @param content [String] New system message content
-    # @return [Array<Langchain::Message>] The messages in the thread
+    # @return [Array<Langchain::Message>] The messages
     def replace_system_message!(content:)
-      thread.messages.delete_if(&:system?)
+      messages.delete_if(&:system?)
 
       message = build_message(role: "system", content: content)
-      thread.messages.unshift(message)
+      messages.unshift(message)
     end
 
     # TODO: If tool_choice = "tool_function_name" and then tool is removed from the assistant, should we set tool_choice back to "auto"?
@@ -231,11 +248,11 @@ module Langchain
       end
     end
 
-    # Process the latest message in the thread
+    # Process the latest message
     #
     # @return [Symbol] The next state
     def process_latest_message
-      last_message = thread.messages.last
+      last_message = messages.last
 
       case last_message.standard_role
       when :system
@@ -261,14 +278,14 @@ module Langchain
     #
     # @return [Symbol] The next state
     def handle_llm_message
-      thread.messages.last.tool_calls.any? ? :requires_action : :completed
+      messages.last.tool_calls.any? ? :requires_action : :completed
     end
 
     # Handle unexpected message scenario
     #
     # @return [Symbol] The failed state
     def handle_unexpected_message
-      Langchain.logger.error("Unexpected message role encountered: #{thread.messages.last.standard_role}")
+      Langchain.logger.error("Unexpected message role encountered: #{messages.last.standard_role}")
       :failed
     end
 
@@ -301,7 +318,7 @@ module Langchain
     #
     # @return [Symbol] The next state
     def execute_tools
-      run_tools(thread.messages.last.tool_calls)
+      run_tools(messages.last.tool_calls)
       :in_progress
     rescue => e
       Langchain.logger.error("Error running tools: #{e.message}; #{e.backtrace.join('\n')}")
@@ -340,7 +357,7 @@ module Langchain
 
       params = @llm_adapter.build_chat_params(
         instructions: @instructions,
-        messages: thread.array_of_message_hashes,
+        messages: array_of_message_hashes,
         tools: @tools,
         tool_choice: tool_choice
       )

data/lib/langchain/contextual_logger.rb

@@ -49,6 +49,8 @@ module Langchain
         "[#{for_class_name}]:"
       end
       log_line_parts << colorize(args.first, MESSAGE_COLOR_OPTIONS[method])
+      log_line_parts << kwargs if !!kwargs && kwargs.any?
+      log_line_parts << block.call if block
       log_line = log_line_parts.compact.join(" ")
 
       @logger.send(

data/lib/langchain/llm/aws_bedrock.rb

@@ -50,7 +50,7 @@ module Langchain::LLM
 
     SUPPORTED_COMPLETION_PROVIDERS = %i[anthropic ai21 cohere meta].freeze
     SUPPORTED_CHAT_COMPLETION_PROVIDERS = %i[anthropic].freeze
-    SUPPORTED_EMBEDDING_PROVIDERS = %i[amazon].freeze
+    SUPPORTED_EMBEDDING_PROVIDERS = %i[amazon cohere].freeze
 
     def initialize(completion_model: DEFAULTS[:completion_model_name], embedding_model: DEFAULTS[:embedding_model_name], aws_client_options: {}, default_options: {})
       depends_on "aws-sdk-bedrockruntime", req: "aws-sdk-bedrockruntime"
@@ -82,8 +82,7 @@ module Langchain::LLM
     def embed(text:, **params)
       raise "Completion provider #{embedding_provider} is not supported." unless SUPPORTED_EMBEDDING_PROVIDERS.include?(embedding_provider)
 
-      parameters = {inputText: text}
-      parameters = parameters.merge(params)
+      parameters = compose_embedding_parameters params.merge(text:)
 
       response = client.invoke_model({
         model_id: @defaults[:embedding_model_name],
@@ -92,7 +91,7 @@ module Langchain::LLM
         accept: "application/json"
       })
 
-      Langchain::LLM::AwsTitanResponse.new(JSON.parse(response.body.string))
+      parse_embedding_response response
     end
 
     #
@@ -214,6 +213,14 @@ module Langchain::LLM
       end
     end
 
+    def compose_embedding_parameters(params)
+      if embedding_provider == :amazon
+        compose_embedding_parameters_amazon params
+      elsif embedding_provider == :cohere
+        compose_embedding_parameters_cohere params
+      end
+    end
+
     def parse_response(response)
       if completion_provider == :anthropic
         Langchain::LLM::AnthropicResponse.new(JSON.parse(response.body.string))
@@ -226,6 +233,37 @@ module Langchain::LLM
       end
     end
 
+    def parse_embedding_response(response)
+      json_response = JSON.parse(response.body.string)
+
+      if embedding_provider == :amazon
+        Langchain::LLM::AwsTitanResponse.new(json_response)
+      elsif embedding_provider == :cohere
+        Langchain::LLM::CohereResponse.new(json_response)
+      end
+    end
+
+    def compose_embedding_parameters_amazon(params)
+      default_params = @defaults.merge(params)
+
+      {
+        inputText: default_params[:text],
+        dimensions: default_params[:dimensions],
+        normalize: default_params[:normalize]
+      }.compact
+    end
+
+    def compose_embedding_parameters_cohere(params)
+      default_params = @defaults.merge(params)
+
+      {
+        texts: [default_params[:text]],
+        truncate: default_params[:truncate],
+        input_type: default_params[:input_type],
+        embedding_types: default_params[:embedding_types]
+      }.compact
+    end
+
     def compose_parameters_cohere(params)
       default_params = @defaults.merge(params)
 

data/lib/langchain/llm/google_gemini.rb

@@ -74,7 +74,7 @@ module Langchain::LLM
       if wrapped_response.chat_completion || Array(wrapped_response.tool_calls).any?
         wrapped_response
       else
-        raise StandardError.new(response)
+        raise StandardError.new(parsed_response)
       end
     end
 

data/lib/langchain/llm/google_vertex_ai.rb

@@ -112,7 +112,7 @@ module Langchain::LLM
       if wrapped_response.chat_completion || Array(wrapped_response.tool_calls).any?
         wrapped_response
       else
-        raise StandardError.new(response)
+        raise StandardError.new(parsed_response)
       end
     end
   end

data/lib/langchain/llm/ollama.rb

@@ -270,7 +270,7 @@ module Langchain::LLM
         conn.request :json
         conn.response :json
         conn.response :raise_error
-        conn.response :logger, nil, {headers: true, bodies: true, errors: true}
+        conn.response :logger, Langchain.logger, {headers: true, bodies: true, errors: true}
       end
     end
 

data/lib/langchain/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Langchain
-  VERSION = "0.15.6"
+  VERSION = "0.16.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: langchainrb
 version: !ruby/object:Gem::Version
-  version: 0.15.6
+  version: 0.16.0
 platform: ruby
 authors:
 - Andrei Bondarev
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2024-09-16 00:00:00.000000000 Z
+date: 2024-09-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: baran
@@ -672,7 +672,6 @@ files:
 - lib/langchain/assistants/messages/mistral_ai_message.rb
 - lib/langchain/assistants/messages/ollama_message.rb
 - lib/langchain/assistants/messages/openai_message.rb
-- lib/langchain/assistants/thread.rb
 - lib/langchain/chunk.rb
 - lib/langchain/chunker/base.rb
 - lib/langchain/chunker/markdown.rb

data/lib/langchain/assistants/thread.rb

@@ -1,46 +0,0 @@
-# frozen_string_literal: true
-
-module Langchain
-  # Langchain::Thread keeps track of messages in a conversation.
-  # TODO: Add functionality to persist to the thread to disk, DB, storage, etc.
-  class Thread
-    attr_accessor :messages, :add_message_callback
-
-    # @param messages [Array<Langchain::Message>]
-    # @param add_message_callback [Proc] A callback to call when a message is added to the thread
-    def initialize(messages: [], add_message_callback: nil)
-      raise ArgumentError, "messages array must only contain Langchain::Message instance(s)" unless messages.is_a?(Array) && messages.all? { |m| m.is_a?(Langchain::Messages::Base) }
-
-      @add_message_callback = add_message_callback
-      @messages = messages
-    end
-
-    # Convert the thread to an LLM APIs-compatible array of hashes
-    #
-    # @return [Array<Hash>] The thread as an OpenAI API-compatible array of hashes
-    def array_of_message_hashes
-      messages
-        .map(&:to_hash)
-        .compact
-    end
-
-    # Only used by the Assistant when it calls the LLM#complete() method
-    def prompt_of_concatenated_messages
-      messages.map(&:to_s).join
-    end
-
-    # Add a message to the thread
-    #
-    # @param message [Langchain::Message] The message to add
-    # @return [Array<Langchain::Message>] The updated messages array
-    def add_message(message)
-      raise ArgumentError, "message must be a Langchain::Message instance" unless message.is_a?(Langchain::Messages::Base)
-
-      # Call the callback with the message
-      add_message_callback.call(message) if add_message_callback # rubocop:disable Style/SafeNavigation
-
-      # Prepend the message to the thread
-      messages << message
-    end
-  end
-end