split-test-rb 1.0.2 → 1.0.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 9cb166ceb6a788adcf47b71d1ef238cbe94ac18798c2a3a170a99dd44f9715e6
-  data.tar.gz: 208cc97019fe0efc571d5f6b69f8b4f75ff6b93d89fb204c367de1a97b53fb08
+  metadata.gz: ffaee1d0d9211e912e20044c5fbf90b4d6c2ad643321d3c5dc94f2cbf5a8921f
+  data.tar.gz: bae21fcdfab3ff5d94e96a1bfa8d00e391ff5387b20e51570ee1bfe3f169ae83
 SHA512:
-  metadata.gz: 734254624a19e1241eb9f08621947f6a1ea7ff292308c6ac87ab0a0a6b356abd2f93cfa5cb5c26bc6510cd49b3c42f208d7fe8c930abd1ec65354827003adf8f
-  data.tar.gz: c50254eab36456759c70d091f67aeca40153b658e0d8533bbafa8245af89b718c1cd4bc9f8faaf0c1d98925cada0dd12a9447f79d65297fd4ef9a98453ceb08f
+  metadata.gz: 56d1f3f3ddaa7e24be47904d499e01d4aac169f4fdfde9678a7c72644d2a5ac065d1bb990e44a1e579c6ff402f469fd9f2867eedbf92f68289b9b88c9deccd50
+  data.tar.gz: 365fbe88adab8a447848389ec8039521d7488a974079fb2ed91086f35d111ce8a74bd465a5b4835a631edd64650396800dcd8828250e0a4c6970b2d3257bd0d7

data/README.md

@@ -49,6 +49,7 @@ Options:
   --test-pattern PATTERN      Test file pattern (default: **/*_spec.rb)
   --split-by-example-threshold SECONDS
                               Split files with execution time >= threshold into individual examples
+  --dry-run-json PATH         RSpec JSON from `rspec --dry-run --format json`, used to list current examples of heavy files
   --debug                     Show debug information
   -h, --help                  Show help message
 ```
@@ -103,6 +104,29 @@ This is useful when:
 
 **Note:** The JSON report must contain the `id` field for each example (RSpec's default JSON formatter includes this). The tool uses these IDs to generate the example-specific paths that RSpec can run.
 
+#### Keeping heavy files up to date with `--dry-run-json`
+
+By default, heavy files are assigned only by the example IDs found in the cached JSON reports. Examples that are not in the cache are not assigned to any node:
+
+- Examples newly added to a heavy file
+- Examples whose IDs shifted because other examples were inserted or moved (RSpec example IDs are position based, e.g. `[1:2:1]`)
+
+Pass the output of `rspec --dry-run --format json` to use the examples that currently exist:
+
+```bash
+bundle exec rspec --dry-run --format json --out tmp/dry-run.json
+split-test-rb --json-path tmp/test-results \
+  --node-index $CI_NODE_INDEX \
+  --node-total $CI_NODE_TOTAL \
+  --split-by-example-threshold 10.0 \
+  --dry-run-json tmp/dry-run.json
+```
+
+With this option, for each heavy file:
+- Examples in the dry-run JSON are assigned, using cached timings when available and the default timing (1.0s) otherwise
+- Cached example IDs that no longer exist are ignored
+- If the dry-run JSON has no examples for the file, the whole file is assigned
+
 ## How It Works
 
 1. **Parse RSpec JSON**: Extracts test file paths and execution times from the JSON report
@@ -160,11 +184,14 @@ bundle exec rspec --format json --out tmp/rspec-results/results.json
 
 ## Release
 
-To release a new version:
+To release a new version, use the [Release workflow](.github/workflows/release.yml):
+
+1. Go to the [Actions](https://github.com/naofumi-fujii/split-test-rb/actions/workflows/release.yml) tab
+2. Click "Run workflow"
+3. Select the version bump type (patch/minor/major)
+4. Click "Run workflow"
 
-1. Update the version number in `lib/split_test_rb/version.rb`
-2. Commit the change and push to `main`
-3. The [Release workflow](.github/workflows/release.yml) automatically creates a git tag (`v*`) and publishes the gem to RubyGems via trusted publishing
+The workflow will automatically update `lib/split_test_rb/version.rb`, commit it to `main`, create a git tag (`v*`), publish the gem to RubyGems via trusted publishing, and create a GitHub Release.
 
 ## License
 

data/lib/split_test_rb/version.rb

@@ -1,3 +1,3 @@
 module SplitTestRb
-  VERSION = '1.0.2'.freeze
+  VERSION = '1.0.4'.freeze
 end

data/lib/split_test_rb.rb

@@ -121,8 +121,10 @@ module SplitTestRb
     # Distributes test files across nodes based on execution times
     # Uses greedy algorithm: assign each file to the node with lowest cumulative time
     def self.balance(timings, total_nodes)
-      # Sort files by execution time (descending) for better balance
-      sorted_files = timings.sort_by { |_file, time| -time }
+      # Sort files by execution time (descending) for better balance.
+      # Ties are broken by file name so that every node computes the same split
+      # even when the input order differs (e.g. a dry-run JSON listed in random order)
+      sorted_files = timings.sort_by { |file, time| [-time, file] }
 
       # Initialize nodes with empty arrays and zero cumulative time
       nodes = Array.new(total_nodes) { { files: [], total_time: 0 } }
@@ -139,6 +141,58 @@ module SplitTestRb
     end
   end
 
+  # Splits heavy test files into individual examples for finer-grained balancing
+  # Uses current_example_ids (from `rspec --dry-run --format json`) as the source of truth when given,
+  # so that examples missing from the cached JSON (new or shifted IDs) are still assigned
+  class ExampleSplitter
+    DEFAULT_TIMING = 1.0
+
+    def initialize(json_files, current_example_ids: nil, default_files: nil)
+      @example_timings = JsonParser.parse_files_with_examples(json_files)
+      @current_example_ids = current_example_ids
+      @default_files = default_files
+    end
+
+    # Returns timings where files >= threshold are replaced by their examples
+    def split(file_timings, threshold)
+      heavy_files = file_timings.select { |_file, time| time >= threshold }
+      return file_timings if heavy_files.empty?
+
+      # Start with light files (below threshold)
+      timings = file_timings.reject { |file, _| heavy_files.key?(file) }
+      heavy_files.each { |heavy_file, file_time| timings.merge!(examples_for(heavy_file, file_time)) }
+      timings
+    end
+
+    private
+
+    def examples_for(heavy_file, file_time)
+      return cached_examples_for(heavy_file) unless @current_example_ids
+
+      ids = @current_example_ids.select { |example_id| example_of?(example_id, heavy_file) }
+      # Assign the whole file when the dry-run JSON has no examples for it, so no tests are dropped
+      return { heavy_file => file_time } if ids.empty?
+
+      ids.to_h { |example_id| [example_id, timing_for(example_id)] }
+    end
+
+    def cached_examples_for(heavy_file)
+      @example_timings.select { |example_id, _| example_of?(example_id, heavy_file) }
+    end
+
+    def timing_for(example_id)
+      return @example_timings[example_id] if @example_timings.key?(example_id)
+
+      @default_files&.add(example_id)
+      DEFAULT_TIMING
+    end
+
+    # Returns true if example_id (e.g. "spec/a_spec.rb[1:1]") belongs to file_path
+    def example_of?(example_id, file_path)
+      example_id.start_with?("#{file_path}[")
+    end
+  end
+
   # Command-line interface
   class CLI
     def self.run(argv)
@@ -186,7 +240,9 @@ module SplitTestRb
       # Apply example-level splitting if threshold is set
       threshold = options[:split_by_example_threshold]
       timings = if threshold
-                  apply_example_splitting(file_timings, json_files, threshold)
+                  current_example_ids = load_current_example_ids(options[:dry_run_json])
+                  apply_example_splitting(file_timings, json_files, threshold,
+                                          current_example_ids: current_example_ids, default_files: default_files)
                 else
                   file_timings
                 end
@@ -194,24 +250,22 @@ module SplitTestRb
       [timings, default_files, json_files]
     end
 
-    # Splits heavy files (>= threshold) into individual examples
-    def self.apply_example_splitting(file_timings, json_files, threshold)
-      heavy_files = file_timings.select { |_file, time| time >= threshold }
-      return file_timings if heavy_files.empty?
+    # Splits heavy files (>= threshold) into individual examples (delegates to ExampleSplitter)
+    def self.apply_example_splitting(file_timings, json_files, threshold, current_example_ids: nil, default_files: nil)
+      ExampleSplitter.new(json_files, current_example_ids: current_example_ids, default_files: default_files)
+                     .split(file_timings, threshold)
+    end
 
-      example_timings = JsonParser.parse_files_with_examples(json_files)
+    # Loads example IDs from an RSpec dry-run JSON report, or returns nil when no path is given
+    def self.load_current_example_ids(dry_run_json)
+      return nil unless dry_run_json
 
-      # Start with light files (below threshold)
-      timings = file_timings.reject { |file, _| heavy_files.key?(file) }
-
-      # Add individual examples from heavy files
-      heavy_files.each_key do |heavy_file|
-        example_timings.each do |example_id, time|
-          timings[example_id] = time if example_id.start_with?(heavy_file)
-        end
+      unless File.exist?(dry_run_json)
+        warn "Error: dry-run JSON not found: #{dry_run_json}"
+        exit 1
       end
 
-      timings
+      JsonParser.parse_with_examples(dry_run_json).keys
     end
 
     # Adds test files missing from JSON results with default timing (1.0s)
@@ -249,7 +303,8 @@ module SplitTestRb
       debug: false,
       test_dir: 'spec',
       test_pattern: '**/*_spec.rb',
-      split_by_example_threshold: nil
+      split_by_example_threshold: nil,
+      dry_run_json: nil
     }.freeze
 
     # Parses command-line arguments and returns options hash
@@ -284,10 +339,7 @@ module SplitTestRb
     def self.define_test_options(opts, options)
       opts.on('--test-dir DIR', 'Test directory (default: spec)') { |v| options[:test_dir] = v }
       opts.on('--test-pattern PATTERN', 'Test file pattern (default: **/*_spec.rb)') { |v| options[:test_pattern] = v }
-      opts.on('--split-by-example-threshold SECONDS', Float,
-              'Split files with execution time >= threshold into individual examples') do |v|
-        options[:split_by_example_threshold] = v
-      end
+      define_split_options(opts, options)
       opts.on('--debug', 'Show debug information') { options[:debug] = true }
       opts.on('-h', '--help', 'Show this help message') do
         puts opts
@@ -299,6 +351,18 @@ module SplitTestRb
       end
     end
 
+    # Defines example-level splitting related CLI options
+    def self.define_split_options(opts, options)
+      opts.on('--split-by-example-threshold SECONDS', Float,
+              'Split files with execution time >= threshold into individual examples') do |v|
+        options[:split_by_example_threshold] = v
+      end
+      opts.on('--dry-run-json PATH',
+              'RSpec JSON from `rspec --dry-run --format json`, used to list current examples of heavy files') do |v|
+        options[:dry_run_json] = v
+      end
+    end
+
     def self.find_all_spec_files(test_dir = 'spec', test_pattern = '**/*_spec.rb')
       # Find all test files in the specified directory with the given pattern
       glob_pattern = File.join(test_dir, test_pattern)

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: split-test-rb
 version: !ruby/object:Gem::Version
-  version: 1.0.2
+  version: 1.0.4
 platform: ruby
 authors:
 - Naofumi Fujii
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-03-23 00:00:00.000000000 Z
+date: 2026-10-09 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rspec
@@ -58,14 +58,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.22'
+        version: '1.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.22'
+        version: '1.0'
 description: A simple CLI tool to balance RSpec tests across parallel CI nodes using
   RSpec JSON reports
 email:
@@ -92,7 +92,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 3.2.0
+      version: 3.3.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="