lex-llm 0.6.2 → 0.6.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 92e15ead2801ddcfdb692c3135b0a8774bf6ffbc137e70562afaf107dcd182da
-  data.tar.gz: 79ac6bb87c7c57f22857982096fe7818f1d07de3c62b7dbd114588451ed494cb
+  metadata.gz: b1d6f2214c5ae415eab2e51068984dfdfb3cc7a35223439573d550d820bf864d
+  data.tar.gz: 23ba380421192e5320c2060fd89e357d26d2cb7958836697970eaa4bba5746fc
 SHA512:
-  metadata.gz: c46aaae88b383e6ffd8e24076d64287b7fd04bc0c3108b7f7b760807b5ce38d3b49b88294c5cf183d289255ccfecdf2c6513b63646ac3178cf79136f837f67d2
-  data.tar.gz: 4ca741e7ea3dcd3bfee9ad59a30a0379a130a3a885f344ad9b0e60509b0d4c18e5ef1e2e2bdf8c0c818e313a493357f475c60d7c27219df1c9302a77b2cebfbf
+  metadata.gz: e1fb762ee7588f593a75e81b3a6ebc15f131ba74d0d5a54a6cd030d0367c4e41a1a2240c4287c10de4b7bbf25838f27c7a4ae557f75c67aa0b1e481378f752d8
+  data.tar.gz: 9d29438b9d78732887e081b41b7ee792b2459284d654e80e7c70f12be62fcaa585c3e166213530d7bfa57fd09bc5b9de43e12530daf8a38574172452de30a727

data/CHANGELOG.md

@@ -1,5 +1,13 @@
 # Changelog
 
+## 0.6.3 - 2026-06-25
+
+### Fixed
+- `ContentBlock.from_hash` rescues `NoMethodError` when content arrays contain corrupted String elements (serialized `#inspect` output from prior storage bugs) — returns a text block instead of crashing with `undefined method 'transform_keys' for an instance of String`.
+- `ContentBlock.from_hash` normalizes `output_text`/`input_text` types to `:text` via `TEXT_TYPE_ALIASES` so Responses API content blocks are recognized by `text?` and extracted by `Message#text`.
+- `ContentBlock#to_s` returns clean text for all text-type blocks; `#inspect` returns a concise debug representation instead of the full 18-field Data.define dump.
+- `Canonical::Message#to_s` delegates to `#text` to prevent Array#inspect leaking struct internals into string contexts.
+
 ## 0.6.2 - 2026-06-20
 
 ### Fixed

data/lib/legion/extensions/llm/canonical/content_block.rb

@@ -6,6 +6,7 @@ module Legion
       module Canonical
         # Typed content block with media_type support per G20a.
         # Ports field vocabulary from Legion::LLM::Types::ContentBlock.
+        # rubocop:disable Lint/ConstantDefinitionInBlock -- required for Data.define block scope
         ContentBlock = ::Data.define(
           :type, :text, :data, :source_type, :media_type,
           :detail, :name, :file_id,
@@ -13,6 +14,8 @@ module Legion
           :source, :start_index, :end_index,
           :code, :message, :cache_control
         ) do
+          TEXT_TYPE_ALIASES = %i[text output_text input_text].freeze
+
           # Build a text content block.
           def self.text(content, cache_control: nil)
             new(
@@ -64,12 +67,17 @@ module Legion
           end
 
           # Build from a Hash (raw provider response or deserialized wire payload).
+          # Rescues NoMethodError from corrupted inputs (e.g. String elements from
+          # prior serialization bugs where ContentBlock#inspect leaked into storage).
           def self.from_hash(source)
             return nil if source.nil?
 
             h = source.transform_keys(&:to_sym)
             type_raw = h.delete(:type)
-            h[:type] = type_raw&.to_sym if type_raw
+            if type_raw
+              type_sym = type_raw.to_sym
+              h[:type] = TEXT_TYPE_ALIASES.include?(type_sym) ? :text : type_sym
+            end
 
             new(
               type: h[:type],
@@ -91,6 +99,10 @@ module Legion
               message: h[:message],
               cache_control: h[:cache_control]
             )
+          rescue NoMethodError => e
+            Legion::Logging.log.warn('[canonical][content_block] from_hash received non-Hash input ' \
+                                     "(#{source.class}): #{e.message}")
+            text(source.to_s)
           end
 
           # Serialize to a Hash for AMQP/fleet/wire transport.
@@ -98,9 +110,22 @@ module Legion
             super.compact
           end
 
+          # Human-readable string — prevents #inspect leaking into user-facing output.
+          def to_s
+            return "[tool_use:#{name}]" if type == :tool_use
+            return '[image]' if type == :image
+
+            text.to_s
+          end
+
+          # Concise inspect — prevents raw Data.define dump in Array#inspect output.
+          def inspect
+            "#<ContentBlock:#{type} #{to_s.slice(0, 80).inspect}>"
+          end
+
           # Whether this block carries textual content.
           def text?
-            type == :text
+            TEXT_TYPE_ALIASES.include?(type)
           end
 
           # Whether this block carries thinking/reasoning content.
@@ -120,6 +145,7 @@ module Legion
         end
 
         ContentBlock::CONTENT_BLOCK_TYPES = %i[text thinking tool_use tool_result image audio video].freeze
+        # rubocop:enable Lint/ConstantDefinitionInBlock
       end
     end
   end

data/lib/legion/extensions/llm/canonical/message.rb

@@ -123,6 +123,11 @@ module Legion
             super.compact
           end
 
+          # Human-readable string — prevents #inspect leaking into user-facing output.
+          def to_s
+            text
+          end
+
           # Minimal provider-facing hash (role + text content).
           def to_provider_hash
             { role: role.to_s, content: text }.compact

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

@@ -3,7 +3,7 @@
 module Legion
   module Extensions
     module Llm
-      VERSION = '0.6.2'
+      VERSION = '0.6.3'
     end
   end
 end

data/spec/legion/extensions/llm/canonical/content_block_spec.rb

@@ -166,6 +166,52 @@ RSpec.describe Legion::Extensions::Llm::Canonical::ContentBlock do
     end
   end
 
+  describe 'Responses API type normalization (output_text/input_text)' do
+    it 'normalizes output_text to :text via from_hash' do
+      block = described_class.from_hash(type: 'output_text', text: 'The seat templates')
+
+      expect(block.type).to eq(:text)
+      expect(block.text).to eq('The seat templates')
+      expect(block.text?).to be true
+    end
+
+    it 'normalizes input_text to :text via from_hash' do
+      block = described_class.from_hash(type: 'input_text', text: 'user message')
+
+      expect(block.type).to eq(:text)
+      expect(block.text?).to be true
+    end
+
+    it 'returns text content from to_s for output_text blocks' do
+      block = described_class.from_hash(type: 'output_text', text: "The seat templates don't")
+
+      expect(block.to_s).to eq("The seat templates don't")
+    end
+
+    it 'does not leak #inspect struct into Array#to_s' do
+      blocks = [described_class.from_hash(type: 'output_text', text: 'hello')]
+
+      expect(blocks.inspect).not_to include('data Legion::Extensions')
+      expect(blocks.inspect).not_to include('source_type=nil')
+    end
+  end
+
+  describe '#to_s' do
+    it 'returns text for text blocks' do
+      expect(described_class.text('hello').to_s).to eq('hello')
+    end
+
+    it 'returns placeholder for tool_use blocks' do
+      block = described_class.tool_use(id: '1', name: 'bash', input: {})
+      expect(block.to_s).to eq('[tool_use:bash]')
+    end
+
+    it 'returns placeholder for image blocks' do
+      block = described_class.image(data: 'x', media_type: 'image/png')
+      expect(block.to_s).to eq('[image]')
+    end
+  end
+
   describe 'round-trip' do
     it 'preserves text block through from_hash/to_h' do
       original = { type: 'text', text: 'hello world' }

data/spec/legion/extensions/llm/canonical/message_spec.rb

@@ -163,6 +163,29 @@ RSpec.describe Legion::Extensions::Llm::Canonical::Message do
 
       expect(msg.text).to eq('')
     end
+
+    it 'extracts text from output_text ContentBlock array (Responses API / Codex)' do
+      msg = described_class.from_hash(
+        role: :assistant,
+        content: [{ type: 'output_text', text: "The seat templates don't" }]
+      )
+
+      expect(msg.text).to eq("The seat templates don't")
+      expect(msg.text).not_to include('#<data')
+      expect(msg.text).not_to include('ContentBlock')
+    end
+
+    it 'extracts text from mixed output_text and text blocks' do
+      msg = described_class.from_hash(
+        role: :assistant,
+        content: [
+          { type: 'output_text', text: 'first ' },
+          { type: 'text', text: 'second' }
+        ]
+      )
+
+      expect(msg.text).to eq('first second')
+    end
   end
 
   describe '#to_h' do

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: lex-llm
 version: !ruby/object:Gem::Version
-  version: 0.6.2
+  version: 0.6.3
 platform: ruby
 authors:
 - LegionIO