lex-llm-ollama 0.2.11 → 0.2.12

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 98c1040c20dfe6ead193a404a6e34fda71e98188859f1e9c828a9778fff1b262
-  data.tar.gz: e5ce31cff8c62f5cab9ce6a59d33e90847bfbbe89e2d9445d8f51ce8540fe67c
+  metadata.gz: ae0169129db1cf8833dcbc252f5b06b6c3ed6e2ab106a262b9dc90789919da55
+  data.tar.gz: f8af5ef3ccea0f79638bbdf77dfb43392cc7332a602aedbc01a2514e229d8e03
 SHA512:
-  metadata.gz: cf3f73f8533cf2bd1145e3beffd12f602150eae3b4ad0031df24eaecac36bef6536d04e18a15b20cb2d6f529812a5cb63f2dfad28ed0638820a582f6e0ff37c9
-  data.tar.gz: da96da090f714b1d0811d475d091f637d4124931168039a2d0de2190b4022860e40a0c099819fb1d359c2ee055daba984d87a9748d9481e635fbe9673ea6eab6
+  metadata.gz: 208c7d4cd46c681bffd89ff263378beb2460af6e27554ca79c04e2850771bd69b9722e3fc547093af2dc20332ee29e17da72363aab8e92b435f1b93d3615a77f
+  data.tar.gz: 92333b402f15d5eb694a747a45cb6b37289d6d2d6fc43dcb2784647e59c12c4eca2b959491f3fe4d567f4b134f8a37d4589be7b76b3ad887b1630bcaa46662cc

data/.rubocop.yml

@@ -14,5 +14,21 @@ Metrics/BlockLength:
     - spec/**/*
 Metrics/MethodLength:
   Enabled: false
+Metrics/ParameterLists:
+  Enabled: false
+Metrics/AbcSize:
+  Enabled: false
+Metrics/CyclomaticComplexity:
+  Enabled: false
+Metrics/PerceivedComplexity:
+  Enabled: false
 RSpec/MultipleExpectations:
   Enabled: false
+RSpec/ExampleLength:
+  Enabled: false
+RSpec/LeakyConstantDeclaration:
+  Enabled: false
+RSpec/InstanceVariable:
+  Enabled: false
+Style/Documentation:
+  Enabled: false

data/CHANGELOG.md

@@ -1,5 +1,14 @@
 # Changelog
 
+## 0.2.12 - 2026-05-21
+
+- Add `default_transport`/`default_tier` class declarations, remove duplicate instance methods
+- Add `model_allowed?` filtering in `discover_offerings`
+- Add `DiscoveryRefresh` actor (Every, 30min, run_now) for non-blocking model discovery
+- Identity headers included via base provider
+- api_base reads from settings[:endpoint] fallback
+
+
 ## 0.2.10 - 2026-05-16
 
 - Stop assuming every non-embedding Ollama model supports tools; fallback chat discovery now advertises completion, streaming, and vision only.

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

@@ -0,0 +1,53 @@
+# frozen_string_literal: true
+
+begin
+  require 'legion/extensions/actors/every'
+rescue LoadError => e
+  warn(e.message) if $VERBOSE
+end
+
+return unless defined?(Legion::Extensions::Actors::Every)
+
+module Legion
+  module Extensions
+    module Llm
+      module Ollama
+        module Actor
+          class DiscoveryRefresh < Legion::Extensions::Actors::Every
+            include Legion::Logging::Helper
+
+            REFRESH_INTERVAL = 1800
+
+            def runner_class    = self.class
+            def runner_function = 'manual'
+            def run_now?        = true
+            def use_runner?     = false
+            def check_subtask?  = false
+            def generate_task?  = false
+
+            def time
+              return REFRESH_INTERVAL unless defined?(Legion::Settings)
+
+              Legion::Settings.dig(:extensions, :llm, :ollama, :discovery_interval) || REFRESH_INTERVAL
+            end
+
+            def manual
+              log.debug('[ollama][discovery_refresh] refreshing model list')
+              return unless defined?(Legion::LLM::Discovery)
+
+              Legion::LLM::Discovery.run
+              if defined?(Legion::LLM::Router) && Legion::LLM::Router.respond_to?(:populate_auto_rules)
+                Legion::LLM::Router.populate_auto_rules(Legion::LLM::Discovery.discovered_instances)
+              end
+              if defined?(Legion::LLM::Inventory) && Legion::LLM::Inventory.respond_to?(:invalidate_offerings_cache!)
+                Legion::LLM::Inventory.invalidate_offerings_cache!
+              end
+            rescue StandardError => e
+              handle_exception(e, level: :warn, handled: true, operation: 'ollama.actor.discovery_refresh')
+            end
+          end
+        end
+      end
+    end
+  end
+end

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

@@ -16,6 +16,8 @@ module Legion
 
             def slug = 'ollama'
             def local? = true
+            def default_transport = :http
+            def default_tier = :local
             def configuration_requirements = []
             def capabilities = Capabilities
 
@@ -40,7 +42,7 @@ module Legion
           end
 
           def api_base
-            resolve_base_url || normalize_url(settings[:base_url] || '127.0.0.1:11434')
+            resolve_base_url || normalize_url(settings[:base_url] || settings[:endpoint] || 'http://127.0.0.1:11434')
           end
 
           def config_base_url
@@ -110,9 +112,13 @@ module Legion
             log.debug do
               "ollama provider discovering offerings live=#{live} cached_model_count=#{Array(@cached_models).size}"
             end
-            resolve_models(live).map { |model_info| offering_from_model(model_info) }.tap do |offerings|
-              log.debug { "ollama provider built offering_count=#{offerings.size} live=#{live}" }
+            offerings = resolve_models(live).filter_map do |model_info|
+              next unless model_allowed?(model_info.id)
+
+              offering_from_model(model_info)
             end
+            log.debug { "ollama provider built offering_count=#{offerings.size} live=#{live}" }
+            offerings
           rescue Faraday::ConnectionFailed => e
             log.warn("[ollama] instance=#{provider_instance_id} unreachable: #{e.message}")
             []
@@ -167,14 +173,6 @@ module Legion
             )
           end
 
-          def offering_transport
-            config.respond_to?(:transport) ? config.transport : :http
-          end
-
-          def offering_tier
-            config.respond_to?(:tier) ? config.tier : :local
-          end
-
           def offering_usage_type(model_info)
             model_info.embedding? ? :embedding : :inference
           end
@@ -238,7 +236,7 @@ module Legion
             settings[:keep_alive]
           end
 
-          def render_payload(messages, tools:, temperature:, model:, stream:, schema:, thinking:, tool_prefs:) # rubocop:disable Metrics/ParameterLists
+          def render_payload(messages, tools:, temperature:, model:, stream:, schema:, thinking:, tool_prefs:)
             model_id = model.respond_to?(:id) ? model.id : model
             log.debug do
               "ollama provider rendering chat payload model=#{model_id} message_count=#{messages.size} " \

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

@@ -4,7 +4,7 @@ module Legion
   module Extensions
     module Llm
       module Ollama
-        VERSION = '0.2.11'
+        VERSION = '0.2.12'
       end
     end
   end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: lex-llm-ollama
 version: !ruby/object:Gem::Version
-  version: 0.2.11
+  version: 0.2.12
 platform: ruby
 authors:
 - LegionIO
@@ -97,6 +97,7 @@ files:
 - README.md
 - lex-llm-ollama.gemspec
 - lib/legion/extensions/llm/ollama.rb
+- lib/legion/extensions/llm/ollama/actors/discovery_refresh.rb
 - lib/legion/extensions/llm/ollama/actors/fleet_worker.rb
 - lib/legion/extensions/llm/ollama/provider.rb
 - lib/legion/extensions/llm/ollama/runners/fleet_worker.rb