langchainrb 0.16.1 → 0.17.1

data/lib/langchain/assistants/llm/adapters/ollama.rb

@@ -0,0 +1,57 @@
+module Langchain
+  class Assistant
+    module LLM
+      module Adapters
+        class Ollama < Base
+          def build_chat_params(tools:, instructions:, messages:, tool_choice:)
+            params = {messages: messages}
+            if tools.any?
+              params[:tools] = build_tools(tools)
+            end
+            params
+          end
+
+          def build_message(role:, content: nil, image_url: nil, tool_calls: [], tool_call_id: nil)
+            warn "Image URL is not supported by Ollama currently" if image_url
+
+            Langchain::Messages::OllamaMessage.new(role: role, content: content, tool_calls: tool_calls, tool_call_id: tool_call_id)
+          end
+
+          # Extract the tool call information from the OpenAI tool call hash
+          #
+          # @param tool_call [Hash] The tool call hash
+          # @return [Array] The tool call information
+          def extract_tool_call_args(tool_call:)
+            tool_call_id = tool_call.dig("id")
+
+            function_name = tool_call.dig("function", "name")
+            tool_name, method_name = function_name.split("__")
+
+            tool_arguments = tool_call.dig("function", "arguments")
+            tool_arguments = if tool_arguments.is_a?(Hash)
+              Langchain::Utils::HashTransformer.symbolize_keys(tool_arguments)
+            else
+              JSON.parse(tool_arguments, symbolize_names: true)
+            end
+
+            [tool_call_id, tool_name, method_name, tool_arguments]
+          end
+
+          def available_tool_names(tools)
+            build_tools(tools).map { |tool| tool.dig(:function, :name) }
+          end
+
+          def allowed_tool_choices
+            ["auto", "none"]
+          end
+
+          private
+
+          def build_tools(tools)
+            tools.map { |tool| tool.class.function_schemas.to_openai_format }.flatten
+          end
+        end
+      end
+    end
+  end
+end

data/lib/langchain/assistants/llm/adapters/openai.rb

@@ -0,0 +1,65 @@
+module Langchain
+  class Assistant
+    module LLM
+      module Adapters
+        class OpenAI < Base
+          def build_chat_params(tools:, instructions:, messages:, tool_choice:)
+            params = {messages: messages}
+            if tools.any?
+              params[:tools] = build_tools(tools)
+              params[:tool_choice] = build_tool_choice(tool_choice)
+            end
+            params
+          end
+
+          def build_message(role:, content: nil, image_url: nil, tool_calls: [], tool_call_id: nil)
+            Langchain::Messages::OpenAIMessage.new(role: role, content: content, image_url: image_url, tool_calls: tool_calls, tool_call_id: tool_call_id)
+          end
+
+          # Extract the tool call information from the OpenAI tool call hash
+          #
+          # @param tool_call [Hash] The tool call hash
+          # @return [Array] The tool call information
+          def extract_tool_call_args(tool_call:)
+            tool_call_id = tool_call.dig("id")
+
+            function_name = tool_call.dig("function", "name")
+            tool_name, method_name = function_name.split("__")
+
+            tool_arguments = tool_call.dig("function", "arguments")
+            tool_arguments = if tool_arguments.is_a?(Hash)
+              Langchain::Utils::HashTransformer.symbolize_keys(tool_arguments)
+            else
+              JSON.parse(tool_arguments, symbolize_names: true)
+            end
+
+            [tool_call_id, tool_name, method_name, tool_arguments]
+          end
+
+          def build_tools(tools)
+            tools.map { |tool| tool.class.function_schemas.to_openai_format }.flatten
+          end
+
+          def allowed_tool_choices
+            ["auto", "none"]
+          end
+
+          def available_tool_names(tools)
+            build_tools(tools).map { |tool| tool.dig(:function, :name) }
+          end
+
+          private
+
+          def build_tool_choice(choice)
+            case choice
+            when "auto"
+              choice
+            else
+              {"type" => "function", "function" => {"name" => choice}}
+            end
+          end
+        end
+      end
+    end
+  end
+end

data/lib/langchain/assistants/messages/base.rb

@@ -3,7 +3,11 @@
 module Langchain
   module Messages
     class Base
-      attr_reader :role, :content, :tool_calls, :tool_call_id
+      attr_reader :role,
+        :content,
+        :image_url,
+        :tool_calls,
+        :tool_call_id
 
       # Check if the message came from a user
       #

data/lib/langchain/assistants/messages/mistral_ai_message.rb

@@ -15,17 +15,20 @@ module Langchain
 
       # Initialize a new MistralAI message
       #
-      # @param [String] The role of the message
-      # @param [String] The content of the message
-      # @param [Array<Hash>] The tool calls made in the message
-      # @param [String] The ID of the tool call
-      def initialize(role:, content: nil, tool_calls: [], tool_call_id: nil) # TODO: Implement image_file: reference (https://platform.openai.com/docs/api-reference/messages/object#messages/object-content)
+      # @param role [String] The role of the message
+      # @param content [String] The content of the message
+      # @param image_url [String] The URL of the image
+      # @param tool_calls [Array<Hash>] The tool calls made in the message
+      # @param tool_call_id [String] The ID of the tool call
+      def initialize(role:, content: nil, image_url: nil, tool_calls: [], tool_call_id: nil) # TODO: Implement image_file: reference (https://platform.openai.com/docs/api-reference/messages/object#messages/object-content)
         raise ArgumentError, "Role must be one of #{ROLES.join(", ")}" unless ROLES.include?(role)
         raise ArgumentError, "Tool calls must be an array of hashes" unless tool_calls.is_a?(Array) && tool_calls.all? { |tool_call| tool_call.is_a?(Hash) }
 
         @role = role
         # Some Tools return content as a JSON hence `.to_s`
         @content = content.to_s
+        # Make sure you're using the Pixtral model if you want to send image_url
+        @image_url = image_url
         @tool_calls = tool_calls
         @tool_call_id = tool_call_id
       end
@@ -43,9 +46,28 @@ module Langchain
       def to_hash
         {}.tap do |h|
           h[:role] = role
-          h[:content] = content if content # Content is nil for tool calls
-          h[:tool_calls] = tool_calls if tool_calls.any?
-          h[:tool_call_id] = tool_call_id if tool_call_id
+
+          if tool_calls.any?
+            h[:tool_calls] = tool_calls
+          else
+            h[:tool_call_id] = tool_call_id if tool_call_id
+
+            h[:content] = []
+
+            if content && !content.empty?
+              h[:content] << {
+                type: "text",
+                text: content
+              }
+            end
+
+            if image_url
+              h[:content] << {
+                type: "image_url",
+                image_url: image_url
+              }
+            end
+          end
         end
       end
 

data/lib/langchain/assistants/messages/openai_message.rb

@@ -15,17 +15,25 @@ module Langchain
 
       # Initialize a new OpenAI message
       #
-      # @param [String] The role of the message
-      # @param [String] The content of the message
-      # @param [Array<Hash>] The tool calls made in the message
-      # @param [String] The ID of the tool call
-      def initialize(role:, content: nil, tool_calls: [], tool_call_id: nil) # TODO: Implement image_file: reference (https://platform.openai.com/docs/api-reference/messages/object#messages/object-content)
+      # @param role [String] The role of the message
+      # @param content [String] The content of the message
+      # @param image_url [String] The URL of the image
+      # @param tool_calls [Array<Hash>] The tool calls made in the message
+      # @param tool_call_id [String] The ID of the tool call
+      def initialize(
+        role:,
+        content: nil,
+        image_url: nil,
+        tool_calls: [],
+        tool_call_id: nil
+      )
         raise ArgumentError, "Role must be one of #{ROLES.join(", ")}" unless ROLES.include?(role)
         raise ArgumentError, "Tool calls must be an array of hashes" unless tool_calls.is_a?(Array) && tool_calls.all? { |tool_call| tool_call.is_a?(Hash) }
 
         @role = role
         # Some Tools return content as a JSON hence `.to_s`
         @content = content.to_s
+        @image_url = image_url
         @tool_calls = tool_calls
         @tool_call_id = tool_call_id
       end
@@ -43,9 +51,30 @@ module Langchain
       def to_hash
         {}.tap do |h|
           h[:role] = role
-          h[:content] = content if content # Content is nil for tool calls
-          h[:tool_calls] = tool_calls if tool_calls.any?
-          h[:tool_call_id] = tool_call_id if tool_call_id
+
+          if tool_calls.any?
+            h[:tool_calls] = tool_calls
+          else
+            h[:tool_call_id] = tool_call_id if tool_call_id
+
+            h[:content] = []
+
+            if content && !content.empty?
+              h[:content] << {
+                type: "text",
+                text: content
+              }
+            end
+
+            if image_url
+              h[:content] << {
+                type: "image_url",
+                image_url: {
+                  url: image_url
+                }
+              }
+            end
+          end
         end
       end
 

data/lib/langchain/llm/anthropic.rb

@@ -14,7 +14,7 @@ module Langchain::LLM
     DEFAULTS = {
       temperature: 0.0,
       completion_model_name: "claude-2.1",
-      chat_completion_model_name: "claude-3-sonnet-20240229",
+      chat_completion_model_name: "claude-3-5-sonnet-20240620",
       max_tokens_to_sample: 256
     }.freeze
 
@@ -109,6 +109,7 @@ module Langchain::LLM
       raise ArgumentError.new("model argument is required") if parameters[:model].empty?
       raise ArgumentError.new("max_tokens argument is required") if parameters[:max_tokens].nil?
 
+      binding.pry
       response = client.messages(parameters: parameters)
 
       Langchain::LLM::AnthropicResponse.new(response)

data/lib/langchain/llm/base.rb

@@ -11,9 +11,8 @@ module Langchain::LLM
   # - {Langchain::LLM::Anthropic}
   # - {Langchain::LLM::Azure}
   # - {Langchain::LLM::Cohere}
-  # - {Langchain::LLM::GooglePalm}
-  # - {Langchain::LLM::GoogleVertexAI}
   # - {Langchain::LLM::GoogleGemini}
+  # - {Langchain::LLM::GoogleVertexAI}
   # - {Langchain::LLM::HuggingFace}
   # - {Langchain::LLM::LlamaCpp}
   # - {Langchain::LLM::OpenAI}

data/lib/langchain/tool/database.rb

@@ -122,7 +122,7 @@ module Langchain::Tool
       end
       db.foreign_key_list(table).each do |fk|
         schema << ",\n" if fk == db.foreign_key_list(table).first
-        schema << "FOREIGN KEY (#{fk[:columns][0]}) REFERENCES #{fk[:table]}(#{fk[:key][0]})"
+        schema << "FOREIGN KEY (#{fk[:columns]&.first}) REFERENCES #{fk[:table]}(#{fk[:key]&.first})"
         schema << ",\n" unless fk == db.foreign_key_list(table).last
       end
       schema << ");\n"

data/lib/langchain/vectorsearch/elasticsearch.rb

@@ -37,7 +37,7 @@ module Langchain::Vectorsearch
       @options = {
         url: url,
         request_timeout: 20,
-        log: false
+        logger: Langchain.logger
       }.merge(es_options)
 
       @es_client = ::Elasticsearch::Client.new(**options)

data/lib/langchain/vectorsearch/milvus.rb

@@ -6,16 +6,18 @@ module Langchain::Vectorsearch
     # Wrapper around Milvus REST APIs.
     #
     # Gem requirements:
-    #     gem "milvus", "~> 0.9.3"
+    #     gem "milvus", "~> 0.10.3"
     #
     # Usage:
-    # milvus = Langchain::Vectorsearch::Milvus.new(url:, index_name:, llm:, api_key:)
+    #     milvus = Langchain::Vectorsearch::Milvus.new(url:, index_name:, llm:, api_key:)
     #
-
     def initialize(url:, index_name:, llm:, api_key: nil)
       depends_on "milvus"
 
-      @client = ::Milvus::Client.new(url: url)
+      @client = ::Milvus::Client.new(
+        url: url,
+        logger: Langchain.logger
+      )
       @index_name = index_name
 
       super(llm: llm)
@@ -24,33 +26,24 @@ module Langchain::Vectorsearch
     def add_texts(texts:)
       client.entities.insert(
         collection_name: index_name,
-        num_rows: Array(texts).size,
-        fields_data: [
-          {
-            field_name: "content",
-            type: ::Milvus::DATA_TYPES["varchar"],
-            field: Array(texts)
-          }, {
-            field_name: "vectors",
-            type: ::Milvus::DATA_TYPES["float_vector"],
-            field: Array(texts).map { |text| llm.embed(text: text).embedding }
-          }
-        ]
+        data: texts.map do |text|
+          {content: text, vector: llm.embed(text: text).embedding}
+        end
       )
     end
 
+    # TODO: Add update_texts method
+
     # Deletes a list of texts in the index
     #
     # @param ids [Array<Integer>] The ids of texts to delete
     # @return [Boolean] The response from the server
     def remove_texts(ids:)
       raise ArgumentError, "ids must be an array" unless ids.is_a?(Array)
-      # Convert ids to integers if strings are passed
-      ids = ids.map(&:to_i)
 
       client.entities.delete(
         collection_name: index_name,
-        expression: "id in #{ids}"
+        filter: "id in #{ids}"
       )
     end
 
@@ -62,33 +55,25 @@ module Langchain::Vectorsearch
       client.collections.create(
         auto_id: true,
         collection_name: index_name,
-        description: "Default schema created by langchain.rb",
         fields: [
           {
-            name: "id",
-            is_primary_key: true,
-            autoID: true,
-            data_type: ::Milvus::DATA_TYPES["int64"]
+            fieldName: "id",
+            isPrimary: true,
+            dataType: "Int64"
           }, {
-            name: "content",
-            is_primary_key: false,
-            data_type: ::Milvus::DATA_TYPES["varchar"],
-            type_params: [
-              {
-                key: "max_length",
-                value: "32768" # Largest allowed value
-              }
-            ]
+            fieldName: "content",
+            isPrimary: false,
+            dataType: "VarChar",
+            elementTypeParams: {
+              max_length: "32768" # Largest allowed value
+            }
           }, {
-            name: "vectors",
-            data_type: ::Milvus::DATA_TYPES["float_vector"],
-            is_primary_key: false,
-            type_params: [
-              {
-                key: "dim",
-                value: llm.default_dimensions.to_s
-              }
-            ]
+            fieldName: "vector",
+            isPrimary: false,
+            dataType: "FloatVector",
+            elementTypeParams: {
+              dim: llm.default_dimensions.to_s
+            }
           }
         ]
       )
@@ -97,13 +82,17 @@ module Langchain::Vectorsearch
     # Create the default index
     # @return [Boolean] The response from the server
     def create_default_index
-      client.indices.create(
+      client.indexes.create(
         collection_name: index_name,
-        field_name: "vectors",
-        extra_params: [
-          {key: "metric_type", value: "L2"},
-          {key: "index_type", value: "IVF_FLAT"},
-          {key: "params", value: "{\"nlist\":1024}"}
+        index_params: [
+          {
+            metricType: "L2",
+            fieldName: "vector",
+            indexName: "vector_idx",
+            indexConfig: {
+              index_type: "AUTOINDEX"
+            }
+          }
         ]
       )
     end
@@ -111,13 +100,13 @@ module Langchain::Vectorsearch
     # Get the default schema
     # @return [Hash] The response from the server
     def get_default_schema
-      client.collections.get(collection_name: index_name)
+      client.collections.describe(collection_name: index_name)
     end
 
     # Delete default schema
     # @return [Hash] The response from the server
     def destroy_default_schema
-      client.collections.delete(collection_name: index_name)
+      client.collections.drop(collection_name: index_name)
     end
 
     # Load default schema into memory
@@ -138,16 +127,12 @@ module Langchain::Vectorsearch
     def similarity_search_by_vector(embedding:, k: 4)
       load_default_schema
 
-      client.search(
+      client.entities.search(
         collection_name: index_name,
-        output_fields: ["id", "content"], # Add "vectors" if need to have full vectors returned.
-        top_k: k.to_s,
-        vectors: [embedding],
-        dsl_type: 1,
-        params: "{\"nprobe\": 10}",
-        anns_field: "vectors",
-        metric_type: "L2",
-        vector_type: ::Milvus::DATA_TYPES["float_vector"]
+        anns_field: "vector",
+        data: [embedding],
+        limit: k,
+        output_fields: ["content", "id", "vector"]
       )
     end
 
@@ -159,8 +144,7 @@ module Langchain::Vectorsearch
     def ask(question:, k: 4, &block)
       search_results = similarity_search(query: question, k: k)
 
-      content_field = search_results.dig("results", "fields_data").select { |field| field.dig("field_name") == "content" }
-      content_data = content_field.first.dig("Field", "Scalars", "Data", "StringData", "data")
+      content_data = search_results.dig("data").map { |result| result.dig("content") }
 
       context = content_data.join("\n---\n")
 

data/lib/langchain/vectorsearch/qdrant.rb

@@ -6,7 +6,7 @@ module Langchain::Vectorsearch
     # Wrapper around Qdrant
     #
     # Gem requirements:
-    #     gem "qdrant-ruby", "~> 0.9.3"
+    #     gem "qdrant-ruby", "~> 0.9.8"
     #
     # Usage:
     #     qdrant = Langchain::Vectorsearch::Qdrant.new(url:, api_key:, index_name:, llm:)
@@ -22,7 +22,8 @@ module Langchain::Vectorsearch
 
       @client = ::Qdrant::Client.new(
         url: url,
-        api_key: api_key
+        api_key: api_key,
+        logger: Langchain.logger
       )
       @index_name = index_name
 

data/lib/langchain/vectorsearch/weaviate.rb

@@ -6,7 +6,7 @@ module Langchain::Vectorsearch
     # Wrapper around Weaviate
     #
     # Gem requirements:
-    #     gem "weaviate-ruby", "~> 0.9.0"
+    #     gem "weaviate-ruby", "~> 0.9.2"
     #
     # Usage:
     #     weaviate = Langchain::Vectorsearch::Weaviate.new(url: ENV["WEAVIATE_URL"], api_key: ENV["WEAVIATE_API_KEY"], index_name: "Docs", llm: llm)
@@ -22,7 +22,8 @@ module Langchain::Vectorsearch
 
       @client = ::Weaviate::Client.new(
         url: url,
-        api_key: api_key
+        api_key: api_key,
+        logger: Langchain.logger
       )
 
       # Weaviate requires the class name to be Capitalized: https://weaviate.io/developers/weaviate/configuration/schema-configuration#create-a-class

data/lib/langchain/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Langchain
-  VERSION = "0.16.1"
+  VERSION = "0.17.1"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: langchainrb
 version: !ruby/object:Gem::Version
-  version: 0.16.1
+  version: 0.17.1
 platform: ruby
 authors:
 - Andrei Bondarev
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2024-09-30 00:00:00.000000000 Z
+date: 2024-10-07 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: baran
@@ -318,20 +318,6 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
-- !ruby/object:Gem::Dependency
-  name: google_palm_api
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: 0.1.3
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: 0.1.3
 - !ruby/object:Gem::Dependency
   name: google_search_results
   requirement: !ruby/object:Gem::Requirement
@@ -380,14 +366,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.9.3
+        version: 0.10.3
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.9.3
+        version: 0.10.3
 - !ruby/object:Gem::Dependency
   name: llama_cpp
   requirement: !ruby/object:Gem::Requirement
@@ -520,14 +506,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.9.4
+        version: 0.9.8
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.9.4
+        version: 0.9.8
 - !ruby/object:Gem::Dependency
   name: roo
   requirement: !ruby/object:Gem::Requirement
@@ -604,14 +590,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.8.10
+        version: 0.9.2
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.8.10
+        version: 0.9.2
 - !ruby/object:Gem::Dependency
   name: wikipedia-client
   requirement: !ruby/object:Gem::Requirement
@@ -652,6 +638,13 @@ files:
 - README.md
 - lib/langchain.rb
 - lib/langchain/assistants/assistant.rb
+- lib/langchain/assistants/llm/adapter.rb
+- lib/langchain/assistants/llm/adapters/_base.rb
+- lib/langchain/assistants/llm/adapters/anthropic.rb
+- lib/langchain/assistants/llm/adapters/google_gemini.rb
+- lib/langchain/assistants/llm/adapters/mistral_ai.rb
+- lib/langchain/assistants/llm/adapters/ollama.rb
+- lib/langchain/assistants/llm/adapters/openai.rb
 - lib/langchain/assistants/messages/anthropic_message.rb
 - lib/langchain/assistants/messages/base.rb
 - lib/langchain/assistants/messages/google_gemini_message.rb
@@ -683,7 +676,6 @@ files:
 - lib/langchain/llm/base.rb
 - lib/langchain/llm/cohere.rb
 - lib/langchain/llm/google_gemini.rb
-- lib/langchain/llm/google_palm.rb
 - lib/langchain/llm/google_vertex_ai.rb
 - lib/langchain/llm/hugging_face.rb
 - lib/langchain/llm/llama_cpp.rb
@@ -701,7 +693,6 @@ files:
 - lib/langchain/llm/response/base_response.rb
 - lib/langchain/llm/response/cohere_response.rb
 - lib/langchain/llm/response/google_gemini_response.rb
-- lib/langchain/llm/response/google_palm_response.rb
 - lib/langchain/llm/response/hugging_face_response.rb
 - lib/langchain/llm/response/llama_cpp_response.rb
 - lib/langchain/llm/response/mistral_ai_response.rb