ruby_llm-contract 0.10.5 → 1.0.0

data/lib/ruby_llm/contract/eval/pipeline_result_adapter.rb

@@ -3,8 +3,7 @@
 module RubyLLM
   module Contract
     module Eval
-      # Lightweight adapter that wraps a Pipeline::Result to look like a Step::Result.
-      # Replaces OpenStruct usage in Runner#normalize_pipeline_result.
+      # Replaces OpenStruct usage in StepResultNormalizer#normalize_pipeline_result.
       PipelineResultAdapter = Struct.new(:status, :ok_flag, :parsed_output, :validation_errors, :trace) do
         def ok?
           ok_flag

data/lib/ruby_llm/contract/eval/prompt_diff_comparator.rb

@@ -3,7 +3,6 @@
 module RubyLLM
   module Contract
     module Eval
-      # Encapsulates the safety and mismatch rules for prompt A/B comparison.
       class PromptDiffComparator
         def initialize(candidate_cases:, baseline_cases:, diff:)
           @candidate_cases = candidate_cases

data/lib/ruby_llm/contract/eval/prompt_diff_presenter.rb

@@ -3,7 +3,6 @@
 module RubyLLM
   module Contract
     module Eval
-      # Renders a prompt diff as a readable console summary.
       class PromptDiffPresenter
         VARIANT_LABEL_WIDTH = 12
         TABLE_WIDTH = 26

data/lib/ruby_llm/contract/eval/prompt_diff_serializer.rb

@@ -3,7 +3,6 @@
 module RubyLLM
   module Contract
     module Eval
-      # Normalizes report results into comparable prompt-diff case hashes.
       class PromptDiffSerializer
         def call(report)
           report.results.reject { |result| result.step_status == :skipped }.map do |result|

data/lib/ruby_llm/contract/eval/report_presenter.rb

@@ -3,7 +3,6 @@
 module RubyLLM
   module Contract
     module Eval
-      # Formats eval reports for console and string output.
       class ReportPresenter
         def initialize(report:, stats:)
           @report = report

data/lib/ruby_llm/contract/eval/report_stats.rb

@@ -3,7 +3,6 @@
 module RubyLLM
   module Contract
     module Eval
-      # Computes aggregate metrics for an eval report.
       class ReportStats
         def initialize(results:)
           @results = results

data/lib/ruby_llm/contract/eval/report_storage.rb

@@ -6,7 +6,6 @@ require "fileutils"
 module RubyLLM
   module Contract
     module Eval
-      # Persists eval reports as history entries and regression baselines.
       class ReportStorage
         def initialize(report:, stats:)
           @report = report

data/lib/ruby_llm/contract/eval/retry_optimizer.rb

@@ -13,6 +13,10 @@ module RubyLLM
       #   result.print_summary
       #   result.to_dsl  # => copy-paste retry_policy
       class RetryOptimizer
+        # Documented CANDIDATES= shorthand, parsed by
+        # OptimizeRakeTask#parse_candidates and rendered in this table's headers.
+        EFFORT_SEPARATOR = "@"
+
         Result = Struct.new(:step_name, :eval_names, :candidate_labels, :score_matrix,
                             :constraining_eval, :chain, :chain_details, keyword_init: true) do
           # Terminology alias — `hardest_eval` is the narrative name used in docs;
@@ -80,11 +84,10 @@ module RubyLLM
           end
 
           def short_candidate_label(label)
-            label
-              .sub("gpt-5-", "")
-              .sub("gpt-4.1", "4.1")
-              .sub(" (effort: ", "@")
-              .sub(")", "")
+            config = CandidateLabel.parse(label)
+            model = config[:model].sub("gpt-5-", "").sub("gpt-4.1", "4.1")
+            effort = config[:reasoning_effort]
+            effort ? "#{model}#{EFFORT_SEPARATOR}#{effort}" : model
           end
 
           def print_dsl(io)
@@ -176,11 +179,9 @@ module RubyLLM
         def build_chain(matrix, labels, evals)
           total = evals.size
 
-          # Find cheapest model that passes every eval — the safe fallback.
           safe_fallback = labels.find { |l| evals.all? { |e| (matrix.dig(e, l) || 0) >= @min_score } }
           return [[], []] unless safe_fallback
 
-          # Prepend cheaper models that pass a strict subset.
           chain = []
           details = []
           covered_evals = Set.new
@@ -193,27 +194,16 @@ module RubyLLM
             next if new_additions.empty?
 
             covered_evals.merge(new_additions)
-            chain << parse_label_to_config(label)
+            chain << CandidateLabel.parse(label)
             details << { label: label, passes: new_additions.size, cost: label }
           end
 
-          # Always end with the safe fallback.
-          chain << parse_label_to_config(safe_fallback)
+          chain << CandidateLabel.parse(safe_fallback)
           details << { label: safe_fallback, passes: total, cost: safe_fallback }
 
           [chain, details]
         end
 
-        def parse_label_to_config(label)
-          if label.match?(/\(effort: (\w+)\)/)
-            model = label.sub(/\s*\(effort:.*/, "").strip
-            effort = label.match(/\(effort: (\w+)\)/)[1]
-            { model: model, reasoning_effort: effort }
-          else
-            { model: label }
-          end
-        end
-
         def empty_result(evals)
           Result.new(
             step_name: @step.name || @step.to_s,

data/lib/ruby_llm/contract/eval/trait_evaluator.rb

@@ -3,8 +3,6 @@
 module RubyLLM
   module Contract
     module Eval
-      # Extracted from Runner to reduce class length.
-      # Evaluates expected_traits against parsed output.
       module TraitEvaluator
         private
 

data/lib/ruby_llm/contract/eval.rb

@@ -23,6 +23,7 @@ require_relative "eval/report_storage"
 require_relative "eval/report"
 require_relative "eval/aggregated_report"
 require_relative "eval/eval_definition"
+require_relative "eval/candidate_label"
 require_relative "eval/model_comparison"
 require_relative "eval/baseline_diff"
 require_relative "eval/prompt_diff_serializer"

data/lib/ruby_llm/contract/pipeline/runner.rb

@@ -95,7 +95,6 @@ module RubyLLM
           ((Process.clock_gettime(Process::CLOCK_MONOTONIC) - start_time) * 1000).round
         end
 
-        # Encapsulates mutable state during pipeline execution
         class ExecutionState
           attr_reader :trace_id, :step_results, :step_traces, :outputs_by_step,
                       :current_input, :status, :failed_step

data/lib/ruby_llm/contract/rake_task/suite_gate.rb

@@ -3,18 +3,6 @@
 module RubyLLM
   module Contract
     class RakeTask < ::Rake::TaskLib
-      # Encapsulates the pass/fail gate that runs after `RakeTask#define_task`
-      # has collected eval reports. Extracted from the prior `define_task`
-      # god-method so each gating dimension (cost, score, regression) is
-      # testable in isolation.
-      #
-      # Returns a `Verdict` value object with:
-      #   - `passed?`         — overall gate verdict
-      #   - `abort_reason`    — String for `abort` when `passed? == false`, nil otherwise
-      #   - `passed_reports`  — [[host, report], ...] of reports that individually passed
-      #                         (used to decide which baselines to save)
-      #   - `suite_cost`      — total cost across all reports
-      #
       # Gate ordering (preserved from pre-refactor behaviour):
       #   1. cost gate runs FIRST — if `maximum_cost` set and exceeded, the
       #      suite aborts before any score check; passed_reports is empty.

data/lib/ruby_llm/contract/rake_task.rb

@@ -164,7 +164,7 @@ module RubyLLM
                     Array(JSON.parse(raw))
                   else
                     raw.split(",").map(&:strip).reject(&:empty?).map do |entry|
-                      model, effort = entry.split("@", 2)
+                      model, effort = entry.split(Eval::RetryOptimizer::EFFORT_SEPARATOR, 2)
                       config = { model: model.strip }
                       config[:reasoning_effort] = effort.strip if effort && !effort.empty?
                       config
@@ -191,7 +191,6 @@ module RubyLLM
       end
     end
 
-    # Auto-register the optimize task when this file is loaded
     OptimizeRakeTask.new
   end
 end

data/lib/ruby_llm/contract/step/base.rb

@@ -6,6 +6,11 @@ module RubyLLM
       class Base
         DEFAULT_OUTPUT_TOKENS = 256
 
+        # Eval::CaseExecutor substring-matches this to choose skip over raise.
+        # Reworded without the consumer, every skipped eval case becomes a hard
+        # failure of the whole run, so both ends read it from here.
+        NO_ADAPTER_MESSAGE = "No adapter configured"
+
         def self.inherited(subclass)
           super
           Contract.register_eval_host(subclass) if respond_to?(:eval_defined?) && eval_defined?
@@ -141,7 +146,10 @@ module RubyLLM
           def estimate_eval_cost_for_model(cases, model_name)
             cases.sum do |test_case|
               estimate = estimate_cost(input: test_case.input, model: model_name)
-              estimate ? estimate[:estimated_cost] : 0.0
+              # Two misses floor to 0.0, not one: model absent from the registry
+              # (nil estimate), or present with unreadable pricing (hash whose
+              # estimated_cost is nil). Documented as a floor, not a fail-closed.
+              (estimate && estimate[:estimated_cost]) || 0.0
             end.round(6)
           end
 
@@ -225,8 +233,9 @@ module RubyLLM
             adapter = context[:adapter] || RubyLLM::Contract.configuration.default_adapter
             return adapter if adapter
 
-            raise RubyLLM::Contract::Error, "No adapter configured. Set one with RubyLLM::Contract.configure " \
-                                            "{ |c| c.default_adapter = ... } or pass context: { adapter: ... }"
+            raise RubyLLM::Contract::Error,
+                  "#{NO_ADAPTER_MESSAGE}. Set one with RubyLLM::Contract.configure " \
+                  "{ |c| c.default_adapter = ... } or pass context: { adapter: ... }"
           end
 
           # ADR-0021 deliverable 2: narrow ArgumentError rescue to DSL-setup phase only.

data/lib/ruby_llm/contract/step/dsl.rb

@@ -3,8 +3,6 @@
 module RubyLLM
   module Contract
     module Step
-      # Extracted from Base to reduce class length.
-      # DSL accessor methods for step definition (input_type, output_type, prompt, etc.).
       module Dsl # rubocop:disable Metrics/ModuleLength
         # Sentinel signalling "explicitly reset" (`some_attr(:default)`).
         # Distinguishes reset (lookup stops at this class, returns nil) from
@@ -65,8 +63,8 @@ module RubyLLM
 
         def output_schema(&block)
           if block
-            require "ruby_llm/schema"
-            @output_schema = ::RubyLLM::Schema.create(&block)
+            require "schematist"
+            @output_schema = ::Schematist::Schema.create(&block)
           elsif defined?(@output_schema)
             @output_schema
           elsif superclass.respond_to?(:output_schema)

data/lib/ruby_llm/contract/step/limit_checker.rb

@@ -3,8 +3,6 @@
 module RubyLLM
   module Contract
     module Step
-      # Extracted from Runner to reduce class length.
-      # Handles input token limit and cost limit checks.
       module LimitChecker
         private
 

data/lib/ruby_llm/contract/step/retry_executor.rb

@@ -3,8 +3,6 @@
 module RubyLLM
   module Contract
     module Step
-      # Extracted from Base to reduce class length.
-      # Handles retry logic: run_with_retry, build_retry_result, aggregate usage, build attempt entries.
       module RetryExecutor
         include Concerns::UsageAggregator
 

data/lib/ruby_llm/contract/token_estimator.rb

@@ -23,8 +23,6 @@ module RubyLLM
     module TokenEstimator
       CHARS_PER_TOKEN = 4
 
-      # Heuristic estimate. Returns an integer token count.
-      # See module docstring for accuracy caveats.
       def self.estimate(messages)
         return 0 unless messages.is_a?(Array)
 

data/lib/ruby_llm/contract/version.rb

@@ -2,6 +2,6 @@
 
 module RubyLLM
   module Contract
-    VERSION = "0.10.5"
+    VERSION = "1.0.0"
   end
 end

data/lib/ruby_llm/contract.rb

@@ -21,8 +21,6 @@ module RubyLLM
         step_adapter_overrides.clear
       end
 
-      # --- Eval host registry ---
-
       def register_eval_host(klass)
         eval_hosts << klass unless eval_hosts.include?(klass)
       end
@@ -101,9 +99,7 @@ module RubyLLM
 
       private
 
-      # Filter stale hosts, deduplicate by name (last wins), prune registry in-place
       def live_eval_hosts
-        # Remove hosts without evals
         @eval_hosts&.reject! { |h| !h.respond_to?(:eval_defined?) || !h.eval_defined? }
 
         # Deduplicate: if two classes share a name (reload), keep the latest

data/ruby_llm-contract.gemspec

@@ -16,7 +16,6 @@ Gem::Specification.new do |spec|
   spec.license = "MIT"
   spec.required_ruby_version = ">= 3.2.0"
 
-  spec.metadata["homepage_uri"] = spec.homepage
   spec.metadata["source_code_uri"] = spec.homepage
   spec.metadata["changelog_uri"] = "#{spec.homepage}/blob/main/CHANGELOG.md"
   spec.metadata["documentation_uri"] = "#{spec.homepage}#readme"
@@ -25,16 +24,18 @@ Gem::Specification.new do |spec|
   spec.files = Dir.chdir(__dir__) do
     # Internal trackers + dev configs excluded so the published gem
     # contains only what adopters actually need at runtime.
-    excluded_files = %w[TODO.md .rspec .rubycritic.yml .simplecov]
+    excluded_files = %w[.rspec .simplecov .rubocop.yml .rubocop_todo.yml Gemfile Rakefile]
     `git ls-files -z`.split("\x0").reject do |f|
       (File.expand_path(f) == __FILE__) ||
-        f.start_with?("spec/", "docs/ideas/", "doc/", ".ai/", ".claude/", ".git", ".revive/") ||
+        f.start_with?("spec/", "docs/ideas/", "doc/", ".ai/", ".claude/", ".git",
+                      ".revive/", "gemfiles/") ||
         excluded_files.include?(f)
     end
   end
   spec.require_paths = ["lib"]
 
   spec.add_dependency "dry-types", "~> 1.7"
-  spec.add_dependency "ruby_llm", "~> 1.12"
-  spec.add_dependency "ruby_llm-schema", "~> 0.3"
+  spec.add_dependency "ruby_llm", "~> 2.0"
+  # schematist is what ruby_llm-schema became; ruby_llm 2.x depends on it too.
+  spec.add_dependency "schematist", "~> 1.1"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: ruby_llm-contract
 version: !ruby/object:Gem::Version
-  version: 0.10.5
+  version: 1.0.0
 platform: ruby
 authors:
 - Justyna
@@ -29,28 +29,28 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.12'
+        version: '2.0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.12'
+        version: '2.0'
 - !ruby/object:Gem::Dependency
-  name: ruby_llm-schema
+  name: schematist
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.3'
+        version: '1.1'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.3'
+        version: '1.1'
 description: Wraps RubyLLM::Chat with input/output contracts, business-rule validation,
   retry with model escalation on validation failure, pre-flight cost ceilings, and
   an evaluation framework. Sibling abstraction to RubyLLM::Agent — same niche (reusable
@@ -59,13 +59,9 @@ executables: []
 extensions: []
 extra_rdoc_files: []
 files:
-- ".rubocop.yml"
 - CHANGELOG.md
-- Gemfile
-- Gemfile.lock
 - LICENSE
 - README.md
-- Rakefile
 - docs/architecture.md
 - docs/guide/best_practices.md
 - docs/guide/eval_first.md
@@ -124,6 +120,7 @@ files:
 - lib/ruby_llm/contract/eval.rb
 - lib/ruby_llm/contract/eval/aggregated_report.rb
 - lib/ruby_llm/contract/eval/baseline_diff.rb
+- lib/ruby_llm/contract/eval/candidate_label.rb
 - lib/ruby_llm/contract/eval/case_executor.rb
 - lib/ruby_llm/contract/eval/case_result.rb
 - lib/ruby_llm/contract/eval/case_result_builder.rb
@@ -200,7 +197,6 @@ homepage: https://github.com/justi/ruby_llm-contract
 licenses:
 - MIT
 metadata:
-  homepage_uri: https://github.com/justi/ruby_llm-contract
   source_code_uri: https://github.com/justi/ruby_llm-contract
   changelog_uri: https://github.com/justi/ruby_llm-contract/blob/main/CHANGELOG.md
   documentation_uri: https://github.com/justi/ruby_llm-contract#readme

data/.rubocop.yml

@@ -1,58 +0,0 @@
-AllCops:
-  TargetRubyVersion: 3.2
-  NewCops: enable
-  SuggestExtensions: false
-
-Style/Documentation:
-  Enabled: false
-
-Style/StringLiterals:
-  EnforcedStyle: double_quotes
-
-Style/StringLiteralsInInterpolation:
-  EnforcedStyle: double_quotes
-
-Metrics/BlockLength:
-  Exclude:
-    - 'spec/**/*'
-    - '*.gemspec'
-
-Metrics/MethodLength:
-  Max: 25
-
-Layout/LineLength:
-  Max: 120
-
-Style/OneClassPerFile:
-  Exclude:
-    - 'spec/**/*'
-    - 'examples/**/*'
-
-Lint/UnusedBlockArgument:
-  Exclude:
-    - 'spec/**/*'
-
-Naming/VariableNumber:
-  Exclude:
-    - 'spec/**/*'
-    - 'examples/**/*'
-
-AllCops:
-  Exclude:
-    - 'internal/**/*'
-
-Metrics/ClassLength:
-  Max: 140
-
-Metrics/ModuleLength:
-  Max: 150
-
-Metrics/AbcSize:
-  Max: 30
-
-Metrics/ParameterLists:
-  Max: 12
-  MaxOptionalParameters: 10
-
-Style/FormatStringToken:
-  Enabled: false

data/Gemfile

@@ -1,13 +0,0 @@
-# frozen_string_literal: true
-
-source "https://rubygems.org"
-
-gemspec
-
-group :development, :test do
-  gem "rake", "~> 13.0"
-  gem "rspec", "~> 3.13"
-  gem "rubocop", "~> 1.75"
-  gem "rubycritic", "~> 4.9"
-  gem "simplecov", "~> 0.22"
-end