lex-llm-ollama 0.1.6 → 0.1.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: a9b4ccf479be5fe7a785b15c5a8f5cc328c3e916736d75b22584af2a7e3488b5
-  data.tar.gz: 8ee2de3a028a8a877a4055e83019652feb8208f1d359db91584b9a795b5b20ed
+  metadata.gz: '04731865b67b0980013a81e3c1c076e39daa788b7e1136b78c7bb25ec5f3ac8b'
+  data.tar.gz: 7dad2cff87b97058266b5a96c8b7bdd5600f3436ca73115fd81c0db517bf2d0d
 SHA512:
-  metadata.gz: de11691a812a9a45ec23ee8b2391cdc902c77043acd368f6b48d3ace4e53acd20da23a8d8d2b66d90a6878542ec186d6b2bf809a86e660b545ff5fe2aeed8577
-  data.tar.gz: 8080ef3565c62e8c587036d70fce6f08ac15c4890b3d5b4316a1cf8f85ac9407483c0638292d1bd24857451490a9a49d4287db40c75f5a921d1a4d5dfffa6a6e
+  metadata.gz: 45071306f61626d92ff3037f713dab35426170010bdecdf26dee092553a2d07d52595e44e26d7c3ec692a350f024c01df2e56e5c6f2445a9cdfbedfeba8030c7
+  data.tar.gz: bc489708640e30d27c306433d5f1e2f82d335550c66bb758a49cfc0a02183a048f3667a1007240f3ff7ee8b2f69192bbb22e4cdf4f0ec4a62e57cae4eef36fb3

data/CHANGELOG.md

@@ -1,5 +1,15 @@
 # Changelog
 
+## 0.1.7 - 2026-04-30
+
+- Provider contract overhaul: adopt lex-llm >= 0.1.9 base contract.
+- Replace `default_settings` with full schema (enabled, base_url, default_model, whitelist/blacklist, model_cache_ttl, tls, instances).
+- Remove `Provider.register` call; providers are now discovered via the extension registry.
+- Delete local `RegistryPublisher`, `RegistryEventBuilder`, and `Transport/` directory; use shared classes from lex-llm base.
+- Replace `config.ollama_api_base` with `resolve_base_url` multi-host resolution from the base provider contract.
+- Enrich `parse_list_models_response` to infer embedding capabilities and modalities from model name and family.
+- Add `settings` and `config_base_url` accessors to the provider for whitelist/blacklist and base URL resolution support.
+
 ## 0.1.6 - 2026-04-30
 
 - Add `Legion::Logging::Helper` to Ollama module, RegistryPublisher, and RegistryEventBuilder.

data/README.md

@@ -2,22 +2,21 @@
 
 LegionIO LLM provider extension for [Ollama](https://ollama.ai).
 
-This gem lives under `Legion::Extensions::Llm::Ollama` and depends on `lex-llm` for shared provider-neutral routing, fleet, and schema primitives.
+This gem lives under `Legion::Extensions::Llm::Ollama` and depends on `lex-llm` (>= 0.1.9) for shared provider-neutral routing, fleet, transport, and registry primitives.
 
 Load it with `require 'legion/extensions/llm/ollama'`.
 
 ## What It Provides
 
-- `Legion::Extensions::Llm::Provider` registration as `:ollama`
 - Ollama-native chat requests through `POST /api/chat`
 - Streaming chat support
-- Model discovery through `GET /api/tags`
+- Model discovery through `GET /api/tags` with automatic embedding capability inference
 - Running model inspection through `GET /api/ps`
 - Model details through `POST /api/show`
 - Model download helper through `POST /api/pull`
 - Embeddings through `POST /api/embed`
-- Best-effort `llm.registry` availability events from readiness and model discovery when Legion Transport is loaded
-- Shared fleet/default settings via `Legion::Extensions::Llm.provider_settings`
+- Best-effort `llm.registry` availability events via the shared `Legion::Extensions::Llm::RegistryPublisher`
+- Full settings schema with model whitelist/blacklist, TLS, and multi-host base URL resolution
 - Full `Legion::Logging::Helper` integration with structured `handle_exception` in every rescue block
 
 ## Architecture
@@ -25,11 +24,10 @@ Load it with `require 'legion/extensions/llm/ollama'`.
 ```
 Legion::Extensions::Llm::Ollama
 ├── Provider               # Ollama provider (chat, stream, embed, models, readiness)
-├── RegistryPublisher      # Best-effort async llm.registry event publishing
-├── RegistryEventBuilder   # Sanitized lex-llm registry envelope construction
-└── Transport/
-    ├── Exchanges::LlmRegistry     # Topic exchange for llm.registry
-    └── Messages::RegistryEvent    # AMQP message wrapper for registry events
+└── (shared from lex-llm)
+    ├── RegistryPublisher      # Best-effort async llm.registry event publishing
+    ├── RegistryEventBuilder   # Sanitized registry envelope construction
+    └── Transport/             # Shared exchange and message classes
 ```
 
 ## Defaults
@@ -37,16 +35,14 @@ Legion::Extensions::Llm::Ollama
 ```ruby
 Legion::Extensions::Llm::Ollama.default_settings
 # {
-#   provider_family: :ollama,
-#   instances: {
-#     default: {
-#       endpoint: "http://localhost:11434",
-#       tier: :local,
-#       transport: :http,
-#       usage: { inference: true, embedding: true },
-#       limits: { concurrency: 1 }
-#     }
-#   }
+#   enabled: false,
+#   base_url: '127.0.0.1:11434',
+#   default_model: 'qwen3.5:latest',
+#   model_whitelist: [],
+#   model_blacklist: [],
+#   model_cache_ttl: 60,
+#   tls: { enabled: false, verify: :peer },
+#   instances: {}
 # }
 ```
 
@@ -54,7 +50,6 @@ Legion::Extensions::Llm::Ollama.default_settings
 
 ```ruby
 Legion::Extensions::Llm.configure do |config|
-  config.ollama_api_base = "http://localhost:11434"
   config.default_model = "qwen3.6:27b"
   config.default_embedding_model = "nomic-embed-text:latest"
 end

data/lex-llm-ollama.gemspec

@@ -26,5 +26,5 @@ Gem::Specification.new do |spec|
   spec.add_dependency 'legion-json', '>= 1.2.1'
   spec.add_dependency 'legion-logging', '>= 1.3.2'
   spec.add_dependency 'legion-settings', '>= 1.3.14'
-  spec.add_dependency 'lex-llm', '>= 0.1.5'
+  spec.add_dependency 'lex-llm', '>= 0.1.9'
 end

data/lib/legion/extensions/llm/ollama/provider.rb

@@ -13,12 +13,11 @@ module Legion
 
             def slug = 'ollama'
             def local? = true
-            def configuration_options = %i[ollama_api_base ollama_keep_alive]
             def configuration_requirements = []
             def capabilities = Capabilities
 
             def registry_publisher
-              @registry_publisher ||= RegistryPublisher.new
+              @registry_publisher ||= Ollama.registry_publisher
             end
           end
 
@@ -33,8 +32,16 @@ module Legion
             def embeddings?(_model) = true
           end
 
+          def settings
+            Ollama.default_settings
+          end
+
           def api_base
-            config.ollama_api_base || 'http://localhost:11434'
+            resolve_base_url || normalize_url(settings[:base_url] || '127.0.0.1:11434')
+          end
+
+          def config_base_url
+            settings[:base_url]
           end
 
           def completion_url = '/api/chat'
@@ -87,13 +94,17 @@ module Legion
 
           private
 
+          def ollama_keep_alive
+            settings[:keep_alive]
+          end
+
           def render_payload(messages, tools:, temperature:, model:, stream:, schema:, thinking:, tool_prefs:) # rubocop:disable Metrics/ParameterLists
             {
               model: model.id,
               messages: format_messages(messages),
               stream: stream,
               think: thinking ? true : nil,
-              keep_alive: config.ollama_keep_alive,
+              keep_alive: ollama_keep_alive,
               format: schema_format(schema),
               options: { temperature: temperature }.compact,
               tools: format_tools(tools),
@@ -193,17 +204,36 @@ module Legion
 
           def parse_list_models_response(response, provider, _capabilities)
             response.body.fetch('models', []).map do |model|
+              family = model.dig('details', 'family')
+              caps = infer_capabilities(model.fetch('name'), family, Array(model['capabilities']))
+              output_mods = embedding_model?(model.fetch('name'), family) ? [:embeddings] : [:text]
+
               Legion::Extensions::Llm::Model::Info.new(
                 id: model.fetch('name'),
                 name: model.fetch('name'),
                 provider: provider,
-                created_at: model['modified_at'],
-                capabilities: Array(model['capabilities']),
-                metadata: model
+                family: family,
+                capabilities: caps,
+                modalities_output: output_mods,
+                metadata: model.merge('created_at' => model['modified_at'])
               )
             end
           end
 
+          def infer_capabilities(name, family, api_caps)
+            return api_caps.map(&:to_sym) unless api_caps.empty?
+
+            if embedding_model?(name, family)
+              [:embedding]
+            else
+              %i[completion streaming tools vision]
+            end
+          end
+
+          def embedding_model?(name, family)
+            name.to_s.match?(/embed|embedding/i) || family.to_s.match?(/bert|nomic/i)
+          end
+
           def render_embedding_payload(text, model:, dimensions:)
             { model: model, input: text, dimensions: dimensions }.compact
           end

data/lib/legion/extensions/llm/ollama/version.rb

@@ -4,7 +4,7 @@ module Legion
   module Extensions
     module Llm
       module Ollama
-        VERSION = '0.1.6'
+        VERSION = '0.1.7'
       end
     end
   end

data/lib/legion/extensions/llm/ollama.rb

@@ -1,9 +1,7 @@
 # frozen_string_literal: true
 
 require 'legion/extensions/llm'
-require 'legion/extensions/llm/ollama/registry_event_builder'
 require 'legion/extensions/llm/ollama/provider'
-require 'legion/extensions/llm/ollama/registry_publisher'
 require 'legion/extensions/llm/ollama/version'
 
 module Legion
@@ -17,25 +15,26 @@ module Legion
         PROVIDER_FAMILY = :ollama
 
         def self.default_settings
-          ::Legion::Extensions::Llm.provider_settings(
-            family: PROVIDER_FAMILY,
-            instance: {
-              endpoint: 'http://localhost:11434',
-              tier: :local,
-              transport: :http,
-              usage: { inference: true, embedding: true },
-              limits: { concurrency: 1 }
-            }
-          )
+          {
+            enabled: false,
+            base_url: '127.0.0.1:11434',
+            default_model: 'qwen3.5:latest',
+            model_whitelist: [],
+            model_blacklist: [],
+            model_cache_ttl: 60,
+            tls: { enabled: false, verify: :peer },
+            instances: {}
+          }
         end
 
         def self.provider_class
           Provider
         end
+
+        def self.registry_publisher
+          @registry_publisher ||= Legion::Extensions::Llm::RegistryPublisher.new(provider_family: PROVIDER_FAMILY)
+        end
       end
     end
   end
 end
-
-Legion::Extensions::Llm::Provider.register(Legion::Extensions::Llm::Ollama::PROVIDER_FAMILY,
-                                           Legion::Extensions::Llm::Ollama::Provider)

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: lex-llm-ollama
 version: !ruby/object:Gem::Version
-  version: 0.1.6
+  version: 0.1.7
 platform: ruby
 authors:
 - LegionIO
@@ -57,14 +57,14 @@ dependencies:
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.1.5
+        version: 0.1.9
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.1.5
+        version: 0.1.9
 description: Ollama provider integration for the LegionIO LLM routing framework.
 email:
 - matthewdiverson@gmail.com
@@ -84,10 +84,6 @@ files:
 - lex-llm-ollama.gemspec
 - lib/legion/extensions/llm/ollama.rb
 - lib/legion/extensions/llm/ollama/provider.rb
-- lib/legion/extensions/llm/ollama/registry_event_builder.rb
-- lib/legion/extensions/llm/ollama/registry_publisher.rb
-- lib/legion/extensions/llm/ollama/transport/exchanges/llm_registry.rb
-- lib/legion/extensions/llm/ollama/transport/messages/registry_event.rb
 - lib/legion/extensions/llm/ollama/version.rb
 homepage: https://github.com/LegionIO/lex-llm-ollama
 licenses:

data/lib/legion/extensions/llm/ollama/registry_event_builder.rb

@@ -1,146 +0,0 @@
-# frozen_string_literal: true
-
-module Legion
-  module Extensions
-    module Llm
-      module Ollama
-        # Builds sanitized lex-llm registry envelopes for Ollama provider state.
-        class RegistryEventBuilder # rubocop:disable Metrics/ClassLength
-          include Legion::Logging::Helper
-
-          def readiness(readiness)
-            registry_event_class.public_send(
-              readiness[:ready] ? :available : :unavailable,
-              provider_offering(readiness),
-              runtime: runtime_metadata,
-              health: readiness_health(readiness),
-              metadata: readiness_metadata(readiness)
-            )
-          end
-
-          def model_available(model, readiness:)
-            registry_event_class.available(
-              model_offering(model),
-              runtime: runtime_metadata,
-              health: model_health(readiness),
-              metadata: model_metadata(model)
-            )
-          end
-
-          private
-
-          def provider_offering(readiness)
-            {
-              provider_family: :ollama,
-              provider_instance: provider_instance,
-              transport: :http,
-              model: 'provider-readiness',
-              usage_type: :inference,
-              capabilities: [],
-              health: readiness_health(readiness),
-              metadata: { lex: :llm_ollama, provider_readiness: true }
-            }
-          end
-
-          def model_offering(model)
-            {
-              provider_family: :ollama,
-              provider_instance: provider_instance,
-              transport: :http,
-              model: model.id,
-              usage_type: usage_type_for(model),
-              capabilities: capabilities_for(model),
-              limits: model_limits(model),
-              metadata: model_metadata(model)
-            }
-          end
-
-          def readiness_health(readiness)
-            health = {
-              ready: readiness[:ready] == true,
-              status: readiness[:ready] ? :available : :unavailable,
-              checked: readiness.dig(:health, :checked) != false
-            }
-            add_readiness_error(health, readiness[:health])
-          end
-
-          def add_readiness_error(health, source)
-            error = source.is_a?(Hash) ? source : {}
-            error_class = error[:error] || error['error']
-            error_message = error[:message] || error['message']
-            health[:error_class] = error_class if error_class
-            health[:error] = error_message if error_message
-            health
-          end
-
-          def model_health(readiness)
-            ready = readiness.fetch(:ready, true) == true
-            { ready:, status: ready ? :available : :degraded }
-          end
-
-          def readiness_metadata(readiness)
-            {
-              extension: :lex_llm_ollama,
-              provider: :ollama,
-              configured: readiness[:configured] == true,
-              live: readiness[:live] == true
-            }
-          end
-
-          def model_metadata(model)
-            metadata = model.metadata || {}
-            {
-              extension: :lex_llm_ollama,
-              provider: :ollama,
-              model_name: model.name,
-              family: metadata.dig('details', 'family'),
-              parameter_size: metadata.dig('details', 'parameter_size'),
-              quantization_level: metadata.dig('details', 'quantization_level')
-            }.compact
-          end
-
-          def runtime_metadata
-            { node: provider_instance }
-          end
-
-          def model_limits(model)
-            context_window = model.metadata&.dig('model_info', 'general.context_length') ||
-                             model.metadata&.dig('model_info', "#{model_family(model)}.context_length")
-            { context_window: context_window }.compact
-          end
-
-          def capabilities_for(model)
-            configured = Array(model.capabilities).map(&:to_sym)
-            return configured unless configured.empty?
-
-            usage_type_for(model) == :embedding ? [:embedding] : %i[chat streaming tools vision]
-          end
-
-          def usage_type_for(model)
-            return :embedding if model.id.to_s.match?(/embed|embedding/i)
-
-            family = model_family(model).to_s
-            family.match?(/bert|nomic/i) ? :embedding : :inference
-          end
-
-          def model_family(model)
-            model.metadata&.dig('details', 'family') || model.metadata&.dig(:details, :family)
-          end
-
-          def provider_instance
-            configured_node = (::Legion::Settings.dig(:node, :canonical_name) if defined?(::Legion::Settings))
-            value = configured_node.to_s.strip
-            value.empty? ? :ollama : value.to_sym
-          rescue StandardError => e
-            handle_exception(e, level: :debug, handled: true, operation: 'ollama.registry.provider_instance')
-            :ollama
-          end
-
-          def registry_event_class
-            ::Legion::Extensions::Llm::Routing::RegistryEvent
-          end
-        end
-      end
-    end
-  end
-end

data/lib/legion/extensions/llm/ollama/registry_publisher.rb

@@ -1,94 +0,0 @@
-# frozen_string_literal: true
-
-module Legion
-  module Extensions
-    module Llm
-      module Ollama
-        # Best-effort publisher for Ollama provider availability events.
-        class RegistryPublisher
-          include Legion::Logging::Helper
-
-          APP_ID = 'lex-llm-ollama'
-
-          def initialize(builder: RegistryEventBuilder.new)
-            @builder = builder
-          end
-
-          def publish_readiness_async(readiness)
-            log.info { 'publishing readiness event to llm.registry' }
-            schedule { publish_event(@builder.readiness(readiness)) }
-          end
-
-          def publish_models_async(models, readiness:)
-            log.info { "publishing #{Array(models).size} model event(s) to llm.registry" }
-            schedule do
-              Array(models).each do |model|
-                publish_event(@builder.model_available(model, readiness:))
-              end
-            end
-          end
-
-          private
-
-          def schedule(&)
-            return false unless publishing_available?
-
-            Thread.new do
-              Thread.current.abort_on_exception = false
-              yield
-            rescue StandardError => e
-              handle_exception(e, level: :debug, handled: true, operation: 'ollama.registry.schedule_thread')
-            end
-          rescue StandardError => e
-            handle_exception(e, level: :debug, handled: true, operation: 'ollama.registry.schedule')
-            false
-          end
-
-          def publish_event(event)
-            return false unless publishing_available?
-
-            message_class.new(event:, app_id: APP_ID).publish(spool: false)
-          rescue StandardError => e
-            handle_exception(e, level: :warn, handled: true, operation: 'ollama.registry.publish_event')
-            false
-          end
-
-          def publishing_available?
-            return false unless registry_event_available?
-            return false unless transport_message_available?
-            return true unless defined?(::Legion::Transport::Connection)
-            return true unless ::Legion::Transport::Connection.respond_to?(:session_open?)
-
-            ::Legion::Transport::Connection.session_open?
-          rescue StandardError => e
-            handle_exception(e, level: :debug, handled: true, operation: 'ollama.registry.publishing_available?')
-            false
-          end
-
-          def registry_event_available?
-            defined?(::Legion::Extensions::Llm::Routing::RegistryEvent)
-          end
-
-          def transport_message_available?
-            return true if message_class_defined?
-            return false unless defined?(::Legion::Transport::Message) && defined?(::Legion::Transport::Exchange)
-
-            require 'legion/extensions/llm/ollama/transport/messages/registry_event'
-            message_class_defined?
-          rescue LoadError => e
-            handle_exception(e, level: :debug, handled: true, operation: 'ollama.registry.transport_load')
-            false
-          end
-
-          def message_class_defined?
-            defined?(::Legion::Extensions::Llm::Ollama::Transport::Messages::RegistryEvent)
-          end
-
-          def message_class
-            ::Legion::Extensions::Llm::Ollama::Transport::Messages::RegistryEvent
-          end
-        end
-      end
-    end
-  end
-end

data/lib/legion/extensions/llm/ollama/transport/exchanges/llm_registry.rb

@@ -1,24 +0,0 @@
-# frozen_string_literal: true
-
-module Legion
-  module Extensions
-    module Llm
-      module Ollama
-        module Transport
-          module Exchanges
-            # Topic exchange for Ollama provider availability events.
-            class LlmRegistry < ::Legion::Transport::Exchange
-              def exchange_name
-                'llm.registry'
-              end
-
-              def default_type
-                'topic'
-              end
-            end
-          end
-        end
-      end
-    end
-  end
-end

data/lib/legion/extensions/llm/ollama/transport/messages/registry_event.rb

@@ -1,42 +0,0 @@
-# frozen_string_literal: true
-
-require 'legion/extensions/llm/ollama/transport/exchanges/llm_registry'
-
-module Legion
-  module Extensions
-    module Llm
-      module Ollama
-        module Transport
-          module Messages
-            # Publishes lex-llm RegistryEvent envelopes to the llm.registry exchange.
-            class RegistryEvent < ::Legion::Transport::Message
-              def initialize(event:, **options)
-                super(**event.to_h.merge(options))
-              end
-
-              def exchange
-                Transport::Exchanges::LlmRegistry
-              end
-
-              def routing_key
-                @options[:routing_key] || "llm.registry.#{@options.fetch(:event_type)}"
-              end
-
-              def type
-                'llm.registry.event'
-              end
-
-              def app_id
-                @options[:app_id] || RegistryPublisher::APP_ID
-              end
-
-              def persistent # rubocop:disable Naming/PredicateMethod
-                false
-              end
-            end
-          end
-        end
-      end
-    end
-  end
-end