serialbench 0.12.0 → 0.12.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: fb1d328cda41b77014e07bcca03086cf860a81a9f02c8863918e37d41ab1e201
-  data.tar.gz: 889ba7216bdc343cf5886f2fdaf6257452633161cd291c66e54823caa4cfbb7d
+  metadata.gz: '06086099f5bb25ceb1e784b26ef5a2e66bc0529b4d96beae9a19fa33c8039ef4'
+  data.tar.gz: 0f3ab5f9d4fddac28db8f239ba0e2aec955133a91251906177aac8139291d248
 SHA512:
-  metadata.gz: 64d0c977c7efffaa614ac03425eae1cf5c3e83d44f06bc2ba98da8b03671644887ec8813d8c506987e6d0f0ce74424af5bf74bf025ad9a09303f7d57a0c3ffa4
-  data.tar.gz: 4db8775576b87ff4eb268f063e323527f5f639d3c65c55f12f02078d7772ff552dc67578f048179a7c17d0a33a08058527959f96f2b0403595ce349c4bb9c79c
+  metadata.gz: 8eb041f1fce782a7a1ba78ac8bb73ecce811174da90b5aadcebf763669013de8925b8970cc76e75d9338ef1ea9f304db53f1b9d4a660712b6fc57883b57370b8
+  data.tar.gz: 47257d11443f97212634770660f8a3e2f80e268c48712d3d26d22470d9530f3fe7ab99924a17a1922306d5e8649f0d65aaf60b31d01aa09be969300aa1fb355d

data/.github/scripts/push-to-data.sh

@@ -31,28 +31,49 @@ echo "Pushing ${TARGET_PATH} to serialbench/data..."
 
 # Try create; 409 (branch race) and 422 (exists) both retry with fresh sha
 put_file() {
-  local sha_arg=()
-  [ -n "${1:-}" ] && sha_arg=(-d "{\"message\": \"${PLATFORM} ruby-${RUBY_VERSION} ${FMT} (update)\", \"content\": \"${CONTENT}\", \"sha\": \"$1\", \"branch\": \"main\"}")
-  [ -n "${1:-}" ] || sha_arg=(-d "{\"message\": \"${PLATFORM} ruby-${RUBY_VERSION} ${FMT}\", \"content\": \"${CONTENT}\", \"branch\": \"main\"}")
+  # The body goes through a file: passing it as a curl argument hits the
+  # ~32KB exec limit on windows/msys once a results.yaml grows ("Argument
+  # list too long" -> curl never runs -> the format silently drops).
+  local body
+  body=$(mktemp /tmp/push-body.XXXXXX)
+  if [ -n "${1:-}" ]; then
+    printf '{"message": "%s (update)", "content": "%s", "sha": "%s", "branch": "main"}' \
+      "${PLATFORM} ruby-${RUBY_VERSION} ${FMT}" "$CONTENT" "$1" > "$body"
+  else
+    printf '{"message": "%s", "content": "%s", "branch": "main"}' \
+      "${PLATFORM} ruby-${RUBY_VERSION} ${FMT}" "$CONTENT" > "$body"
+  fi
   curl -s -o /tmp/data-push-resp.json -w "%{http_code}" -X PUT \
     -H "Authorization: Bearer ${DATA_TOKEN}" \
     -H "Accept: application/vnd.github+json" \
     "https://api.github.com/repos/serialbench/data/contents/${TARGET_PATH}" \
-    "${sha_arg[@]}"
+    --data-binary @"$body" || echo 000
+  rm -f "$body"
+}
+
+# Sha of the existing data file, empty on any failure — guarded so a dead
+# pipe (curl -f, grep miss, SIGPIPE under pipefail) cannot kill the leg.
+fetch_sha() {
+  curl -s -H "Authorization: Bearer ${DATA_TOKEN}" \
+    "https://api.github.com/repos/serialbench/data/contents/${TARGET_PATH}" \
+    | grep -m1 -oE '"sha": ?"[a-f0-9]{40}"' | grep -oE '[a-f0-9]{40}' || true
 }
 
 HTTP_CODE=$(put_file "")
 for TRY in 1 2 3 4 5; do
   if [ "$HTTP_CODE" = "200" ] || [ "$HTTP_CODE" = "201" ]; then break; fi
   if [ "$HTTP_CODE" = "409" ] || [ "$HTTP_CODE" = "422" ]; then
-    SHA=$(curl -sf \
-      -H "Authorization: Bearer ${DATA_TOKEN}" \
-      "https://api.github.com/repos/serialbench/data/contents/${TARGET_PATH}" \
-      | grep -oE '"sha":\s*"[a-f0-9]+"' | head -1 | grep -oE '[a-f0-9]{40}')
+    SHA=$(fetch_sha)
     [ -z "$SHA" ] && break
     echo "attempt $TRY got $HTTP_CODE — refetching sha and retrying"
     sleep $((TRY * 3))
     HTTP_CODE=$(put_file "$SHA")
+  elif [ "$HTTP_CODE" = "000" ]; then
+    # 000 = no HTTP response at all (connection-level blip). Retrying is
+    # the only recovery — breaking here silently drops the format.
+    echo "attempt $TRY got 000 (no response) — retrying"
+    sleep $((TRY * 5))
+    HTTP_CODE=$(put_file "")
   else
     break
   fi
@@ -65,19 +86,18 @@ elif [ "$HTTP_CODE" = "422" ]; then
   # moves between fetch and put) is retried with a fresh sha.
   UPDATE_CODE=409
   for TRY in 1 2 3; do
-    SHA=$(curl -sf \
-      -H "Authorization: Bearer ${DATA_TOKEN}" \
-      "https://api.github.com/repos/serialbench/data/contents/${TARGET_PATH}" \
-      | grep -oE '"sha":\s*"[a-f0-9]+"' | head -1 | grep -oE '[a-f0-9]{40}')
+    SHA=$(fetch_sha)
     if [ -z "$SHA" ]; then
       echo "::warning::422 but couldn't get sha for ${TARGET_PATH}"
       break
     fi
+    printf '{"message": "%s (update)", "content": "%s", "sha": "%s", "branch": "main"}' \
+      "${PLATFORM} ruby-${RUBY_VERSION} ${FMT}" "$CONTENT" "${SHA}" > /tmp/push-body.json
     UPDATE_CODE=$(curl -s -o /dev/null -w "%{http_code}" -X PUT \
       -H "Authorization: Bearer ${DATA_TOKEN}" \
       -H "Accept: application/vnd.github+json" \
       "https://api.github.com/repos/serialbench/data/contents/${TARGET_PATH}" \
-      -d "{\"message\": \"${PLATFORM} ruby-${RUBY_VERSION} ${FMT} (update)\", \"content\": \"${CONTENT}\", \"sha\": \"${SHA}\", \"branch\": \"main\"}")
+      --data-binary @/tmp/push-body.json)
     [ "$UPDATE_CODE" = "200" ] && break
     echo "update attempt $TRY got $UPDATE_CODE — refetching sha"
     sleep 2

data/.github/workflows/benchmark.yml

@@ -179,6 +179,41 @@ jobs:
         run: .github/scripts/push-to-data.sh json ${{ matrix.ruby-version }} ${{ matrix.platform }}
         shell: bash
 
+      - name: Benchmark cbor
+        continue-on-error: true
+        shell: bash
+        env:
+          GITHUB_RUNNER_PLATFORM: ${{ matrix.platform }}
+          DATA_REPO_TOKEN: ${{ secrets.DATA_REPO_TOKEN }}
+          DATA_RUN_DATE: ${{ needs.setup.outputs.run-date }}
+        run: |
+          if [ "${{ github.run_attempt }}" -gt 1 ] && .github/scripts/format-status.sh cbor ${{ matrix.ruby-version }} ${{ matrix.platform }}; then
+            echo "cbor already in the data repo — skipping (gap-filling rerun)"
+            exit 0
+          fi
+          bundle exec serialbench environment execute \
+            config/environments/ci-ruby-${{ matrix.ruby-version }}.yml \
+            config/benchmarks/full-cbor.yml \
+            "results/runs/ci-ruby-${{ matrix.ruby-version }}-${{ matrix.platform }}/cbor"
+
+      - name: Upload cbor
+        if: ${{ !cancelled() }}
+        uses: actions/upload-artifact@v4
+        with:
+          name: benchmark-results-${{ matrix.platform }}-ruby-${{ matrix.ruby-version }}-cbor
+          path: results/runs/ci-ruby-${{ matrix.ruby-version }}-${{ matrix.platform }}/cbor
+          retention-days: 30
+          if-no-files-found: warn
+
+      - name: Push cbor to data repo
+        if: ${{ !cancelled() }}
+        continue-on-error: true
+        env:
+          DATA_REPO_TOKEN: ${{ secrets.DATA_REPO_TOKEN }}
+          DATA_RUN_DATE: ${{ needs.setup.outputs.run-date }}
+        run: .github/scripts/push-to-data.sh cbor ${{ matrix.ruby-version }} ${{ matrix.platform }}
+        shell: bash
+
       - name: Benchmark yaml
         continue-on-error: true
         shell: bash
@@ -331,7 +366,7 @@ jobs:
           DATA_RUN_DATE: ${{ needs.setup.outputs.run-date }}
         run: |
           MISSING=""
-          for FMT in json yaml toml html xml; do
+          for FMT in json yaml toml html xml cbor; do
             if .github/scripts/format-status.sh "$FMT" "${{ matrix.ruby-version }}" "${{ matrix.platform }}"; then
               echo "$FMT: in data repo"
             else

data/Gemfile

@@ -12,12 +12,13 @@ unless Gem.win_platform? && RUBY_PLATFORM.include?('aarch64')
   gem 'libxml-ruby'
 end
 
-gem 'leptris', '1.9.193.4'  # html5 conformance campaign + memchr tokenizer perf
-gem 'yeptris', '0.6.5.4'    # compiled plan walk; dump round-trip fixed (plain keys)
-gem 'teptris', '0.2.36'     # full platform matrix incl. mingw - windows coverage unblocked
+gem 'leptris', '1.9.380.1'  # parse_html defaults :html4 (nokogiri-parity); :whatwg priced separately
+gem 'yeptris', '0.6.36.0'   # JSON + CBOR load/dump; faces load standalone (#300, fixed 0.6.35.2)
+gem 'teptris', '0.3.6.3'    # 0.3.x line (native-doc API); Teptris::TOML.load/dump unchanged
+gem 'cbor'                  # RFC 7049 CBOR codec
 gem 'benchmark'  # Removed from stdlib in Ruby 4.0
 gem 'base64'  # Required for Ruby 3.4+
-gem 'lutaml-model', '~> 0.7'
+gem 'lutaml-model', '~> 0.8'
 gem 'octokit'
 gem 'rake'
 gem 'rspec'

data/README.adoc

@@ -9,7 +9,7 @@ image:https://img.shields.io/github/issues-pr-raw/serialbench/serialbench.svg["P
 
 Serialbench is a comprehensive benchmarking suite that evaluates the performance of popular Ruby serialization libraries across multiple formats. It provides detailed performance comparisons and analysis to help developers make informed decisions when choosing serialization libraries for their Ruby applications.
 
-**Supported Formats**: XML, JSON, YAML, TOML, and more
+**Supported Formats**: XML, HTML, JSON, YAML, TOML, and CBOR
 
 **Key Metrics**: Parsing speed, generation speed, memory usage, streaming capabilities, and feature completeness
 
@@ -23,74 +23,129 @@ Serialbench is a comprehensive benchmarking suite that evaluates the performance
 
 | XML
 | https://github.com/ohler55/ox[Ox]
-| v2.14.23
+| 2.14.29
 | C extension XML parser
 
 | XML
 | https://github.com/xml4r/libxml-ruby[LibXML]
-| v4.1.2
+| 6.0.0
 | Ruby bindings for libxml2
 
 | XML
 | https://github.com/sparklemotion/nokogiri[Nokogiri]
-| v1.18.8
+| 1.19.4
 | XML/HTML parser with XPath and CSS selectors
 
 | XML
 | https://github.com/YorickPeterse/oga[Oga]
-| v3.4
+| 3.5
 | Pure Ruby XML parser with XPath support
 
 | XML
 | https://github.com/leptris/leptris-ruby[Leptris]
-| v1.1.0
-| FFI binding to libleptris (pure-C XML 1.0, XPath 1.0, SAX)
+| 1.9.379.0
+| FFI binding to libleptris (pure-C XML 1.0, XPath/XQuery/XSLT, SAX)
 
 | XML
 | https://github.com/ruby/rexml[REXML]
-| v3.4.1
+| 3.4.4
 | Ruby's standard library XML parser
 
+| XML
+| https://github.com/Saxonica/Saxon-HE[Saxon-HE]
+| 12.10
+| Reference XSLT 2.0/3.0 engine, invoked through its Java CLI (`SAXON_JAR`)
+
+| HTML
+| Nokogiri
+| 1.19.4
+| HTML parsing through the same gem as the XML adapter
+
+| HTML
+| Oga
+| 3.5
+| Pure Ruby HTML parsing
+
+| HTML
+| https://github.com/leptris/leptris-ruby[Leptris]
+| 1.9.379.0
+| HTML 4 engine (libxml2/Nokogiri-compatible shape)
+
+| HTML
+| https://github.com/leptris/leptris-ruby[Leptris-whatwg]
+| 1.9.379.0
+| WHATWG-conformant HTML engine, priced separately
+
 | JSON
 | https://github.com/ohler55/oj[Oj]
-| v3.16.11
+| 3.17.6
 | JSON parser with multiple parsing modes
 
 | JSON
 | https://github.com/brianmario/yajl-ruby[YAJL]
-| v1.4.3
+| 1.4.3
 | JSON library with streaming capabilities
 
 | JSON
 | https://github.com/flori/json[JSON]
-| v2.12.2
+| 3.1.0
 | Ruby's standard library JSON parser
 
+| JSON
+| https://github.com/glycerine/rapidjson[rapidjson]
+| 0.4.0
+| Ruby bindings for RapidJSON
+
+| JSON
+| https://github.com/leptris/yeptris-ruby[yeptris-json]
+| 0.6.34.3
+| FFI binding to libyeptris' native JSON materializer
+
 | YAML
 | https://github.com/ruby/psych[Psych]
-| v5.1.2
+| bundled
 | Ruby's standard library YAML parser
 
 | YAML
 | https://github.com/ruby/syck[Syck]
-| v1.5.1.1
+| 1.6.2
 | Legacy YAML parser
 
+| YAML
+| https://github.com/leptris/yeptris-ruby[yeptris-yaml]
+| 0.6.34.3
+| FFI binding to libyeptris' YAML face
+
+| CBOR
+| https://github.com/cbor-ruby/cbor[cbor]
+| 0.5.10.3
+| RFC 7049 CBOR codec
+
+| CBOR
+| https://github.com/leptris/yeptris-ruby[yeptris-cbor]
+| 0.6.34.3
+| FFI binding to libyeptris' CBOR face (parse-only)
+
 | TOML
 | https://github.com/fbernier/tomlib[Tomlib]
-| v0.7.3
+| 0.7.3
 | TOML parser implemented in C
 
 | TOML
 | https://github.com/emancu/toml-rb[TOML-RB]
-| v2.2.0
+| 4.2.1
 | Pure Ruby TOML parser
 
 | TOML
 | https://github.com/fbernier/tomlrb[tomlrb]
-| v2.0.3
+| 2.0.4
 | A Racc based TOML Ruby parser (Only supports parsing, no support for dumping/writing.)
 
+| TOML
+| https://github.com/leptris/teptris-ruby[Teptris]
+| 0.3.6.3
+| Native C extension over libteptris (C99 TOML parser/emitter)
+
 |===
 
 
@@ -101,9 +156,8 @@ different formats for single-environment and multi-environment runs.
 
 The data formats include:
 
-* **Single benchmark results**: Individual benchmark run output
-* **Result set data structure**: Multi-platform benchmark aggregation
-* **JSON schema specification**: Complete schema validation rules
+* **Single benchmark results**: One YAML file per platform × Ruby × format
+* **JSON schema specification**: Schema validation via `serialbench validate`
 * **Configuration file formats**: Docker and ASDF configuration examples
 
 == Prerequisites
@@ -192,8 +246,8 @@ $ gem install serialbench
 
 == Command line interface
 
-Serialbench provides a comprehensive Thor-based CLI with four main subcommands
-for managing environments, benchmarks, result sets, and Ruby builds.
+Serialbench provides a Thor-based CLI with subcommands for managing
+environments, benchmarks, Ruby builds, and schema validation.
 
 === Main Commands Overview
 
@@ -208,8 +262,8 @@ USAGE:
 COMMANDS:
   environment   Manage benchmark environments (Docker, ASDF, Local)
   benchmark     Manage individual benchmark runs
-  resultset     Manage benchmark resultsets (collections of runs)
-  ruby-build    Manage Ruby-Build definitions for validation
+  ruby_build    Manage Ruby-Build definitions for validation
+  validate      Validate YAML files against schemas
   version       Show version information
   help          Show this help message
 
@@ -217,21 +271,16 @@ EXAMPLES:
   # Create a Docker environment
   serialbench environment new docker-test docker
 
-  # Run multi-environment benchmarks
-  serialbench environment multi-execute asdf --config=serialbench-asdf.yml
-  serialbench environment multi-execute docker --config=serialbench-docker.yml
+  # Run a benchmark in an environment
+  serialbench environment execute config/environments/docker-test.yml config/benchmarks/full-xml.yml results/runs/docker-test/xml
 
   # Create and execute a benchmark
   serialbench benchmark create my-benchmark
-  serialbench benchmark execute my-benchmark.yml
-
-  # Create a result set for comparison
-  serialbench resultset create comparison-set
-  serialbench resultset add-result comparison-set results/my-benchmark
+  serialbench benchmark execute config/environments/local-dev.yml config/runs/my-benchmark.yml
 
-  # Generate static sites
-  serialbench benchmark build-site results/my-benchmark
-  serialbench resultset build-site resultsets/comparison-set
+  # Validate YAML files against the bundled schemas
+  serialbench validate result results/runs/my-run/xml/results.yaml
+  serialbench validate config config/benchmarks/full.yml
 ----
 
 === Environment management
@@ -270,23 +319,6 @@ The `_docker_execute` command is a private command used internally by the
 `execute` command to run benchmarks in Docker environments.
 
 
-=== Result set management
-
-The `resultset` subcommand manages collections of benchmark runs for comparison analysis.
-
-[source]
-----
-$ serialbench resultset help
-Commands:
-  serialbench resultset add-result RESULT_PATH RESULTSET_PATH     # Add a run to a resultset
-  serialbench resultset build-site RESULTSET_PATH [OUTPUT_DIR]    # Generate HTML site for a resultset
-  serialbench resultset create NAME PATH                          # Create a new resultset
-  serialbench resultset help [COMMAND]                            # Describe subcommands or one specific subcommand
-  serialbench resultset list                                      # List all available resultsets
-  serialbench resultset remove-result RESULTSET_PATH RESULT_PATH  # Remove a run from a resultset
-----
-
-
 === ruby-build management
 
 The `ruby-build` subcommand manages Ruby build definitions and version information.
@@ -322,20 +354,11 @@ $ bundle exec serialbench environment prepare config/environments/docker-ruby-3.
 # 2. Run benchmark
 $ bundle exec serialbench environment execute config/environments/docker-ruby-3.1.yml config/benchmarks/short.yml results/runs/docker-ruby-3.1-results
 
-# 3. Create a resultset
-$ bundle exec serialbench resultset create docker-comparison results/sets/docker-comparison
-
-# 3a. (Optional) Build the site from the result if you want to visualize results
-$ bundle exec serialbench benchmark build-site results/runs/docker-ruby-3.1-results/ --output_dir=_site_result
+# 3. Inspect the per-format results
+$ ls results/runs/docker-ruby-3.1-results/xml/   # one results.yaml per format
 
-# 4. Add the result to the resultset
-$ bundle exec serialbench resultset add-result results/sets/docker-comparison/ results/runs/docker-ruby-3.1-results/
-
-# 5. Build the site from the resultset
-$ bundle exec serialbench resultset build-site results/sets/docker-comparison/
-
-# 6. Open the generated site
-$ open _site/index.html
+# CI pushes each format to the serialbench/data repo as it completes; the
+# dashboard (serialbench/serialbench.github.io) builds from that repo.
 ----
 
 === ASDF-based testing
@@ -456,33 +479,6 @@ warmup: 3
 
 == Results structure
 
-=== Individual run results
-
-Results are stored in a structured directory format, with each run containing
-raw benchmark data and execution logs.
-
-The directory is located at `results/runs/{name}/`, where `{name}` is the name
-of the environment used for the benchmark.
-
-[source]
-----
-results/runs/docker-ruby-33-results/
-├── results.yaml                    # Raw benchmark data
-└── benchmark.log                   # Execution log
-----
-
-=== ResultSet structure
-
-ResultSets aggregate multiple benchmark runs for comparison. They are stored in
-a structured directory format at `results/sets/{name}/`, where `{name}` is the
-name of the result set.
-
-[source]
-----
-results/sets/ruby-version-comparison/
-└── resultset.yml                  # Result set configuration
-----
-
 == Benchmark categories
 
 === Parsing performance

data/config/benchmarks/full-cbor.yml

@@ -0,0 +1,23 @@
+# Per-format split of full.yml: each format runs in its own process so
+# memory profiling of large documents cannot compound across formats.
+name: full-cbor
+
+data_sizes:
+- small
+- medium
+- large
+
+formats:
+- cbor
+
+iterations:
+  small: 10
+  medium: 3
+  large: 1
+
+operations:
+- parsing
+- generation
+- memory
+
+warmup: 2

data/data/schemas/benchmark_config.yml

@@ -0,0 +1,34 @@
+type: object
+title: Serialbench Benchmark Config
+description: Schema for benchmark configuration files (basic structure validation)
+properties:
+  name:
+    type: string
+  description:
+    type: string
+  data_sizes:
+    type: array
+    items:
+      type: string
+      enum: [small, medium, large]
+  formats:
+    type: array
+    items:
+      type: string
+      enum: [xml, html, json, yaml, toml, cbor]
+  iterations:
+    type: object
+    properties:
+      small:
+        type: integer
+      medium:
+        type: integer
+      large:
+        type: integer
+  warmup:
+    type: integer
+  operations:
+    type: array
+    items:
+      type: string
+      enum: [parsing, generation, xpath, xquery, xslt, xslt30, validation, streaming, memory]

data/data/schemas/environment_config.yml

@@ -0,0 +1,27 @@
+type: object
+title: Serialbench Environment Config
+description: Schema for benchmark environment configuration files (basic structure validation)
+properties:
+  name:
+    type: string
+  kind:
+    type: string
+    enum: [local, docker, asdf]
+  created_at:
+    type: string
+  ruby_build_tag:
+    type: string
+  description:
+    type: string
+  docker:
+    type: object
+    properties:
+      image:
+        type: string
+      dockerfile:
+        type: string
+  asdf:
+    type: object
+    properties:
+      auto_install:
+        type: boolean

data/data/schemas/resultset.yml

@@ -0,0 +1,16 @@
+type: object
+title: Serialbench Resultset
+description: Schema for benchmark resultset files (basic structure validation)
+properties:
+  name:
+    type: string
+  description:
+    type: string
+  created_at:
+    type: string
+  updated_at:
+    type: string
+  results:
+    type: array
+    items:
+      type: object

data/lib/serialbench/benchmark_runner.rb

@@ -170,7 +170,7 @@ _DEFAULT_XSLT30_STYLESHEET = <<'XSL30'.freeze
 </xsl:stylesheet>
 XSL30
 
-    XPATH_QUERIES = ['//book', "//book[@id='101']", '//book[price > 30]/title'].freeze
+    XPATH_QUERIES = ['//user | //record', "//user[@id='101']", '//preferences/theme'].freeze
     XQUERY_EXPRESSIONS = ['count(//user | //record)', "//record[@id='101']/data/field1", '//user[profile/age > 40]/name'].freeze
     RNG_SCHEMA = File.expand_path('test_data/schema.rng', Dir.pwd).then { |p| File.exist?(p) ? File.read(p) : _DEFAULT_RNG_SCHEMA }
     XSLT_STYLESHEET = File.expand_path('test_data/transform.xsl', Dir.pwd).then { |p| File.exist?(p) ? File.read(p) : _DEFAULT_XSLT_STYLESHEET }

data/lib/serialbench/models/benchmark_config.rb

@@ -48,7 +48,7 @@ module Serialbench
       attribute :name, :string
       attribute :description, :string
       attribute :data_sizes, :string, collection: true, values: %w[small medium large]
-      attribute :formats, :string, collection: true, values: %w[xml html json yaml toml]
+      attribute :formats, :string, collection: true, values: %w[xml html json yaml toml cbor]
       attribute :iterations, BenchmarkIteration
       attribute :warmup, :integer, default: -> { 1 }
       attribute :operations, :string, collection: true, values: %w[parsing generation xpath xquery xslt xslt30 validation streaming memory]

data/lib/serialbench/models/benchmark_result.rb

@@ -5,7 +5,7 @@ require 'lutaml/model'
 module Serialbench
   module Models
     class SerializerInformation < Lutaml::Model::Serializable
-      attribute :format, :string, values: %w[xml html json yaml toml]
+      attribute :format, :string, values: %w[xml html json yaml toml cbor]
       attribute :name, :string
       attribute :version, :string
       attribute :features, :hash
@@ -20,7 +20,7 @@ module Serialbench
 
     class AdapterPerformance < Lutaml::Model::Serializable
       attribute :adapter, :string
-      attribute :format, :string, values: %w[xml html json yaml toml]
+      attribute :format, :string, values: %w[xml html json yaml toml cbor]
       attribute :data_size, :string, values: %w[small medium large]
 
       key_value do

data/lib/serialbench/serializers/cbor/base_cbor_serializer.rb

@@ -0,0 +1,37 @@
+# frozen_string_literal: true
+
+require_relative '../base_serializer'
+
+module Serialbench
+  module Serializers
+    module Cbor
+      class BaseCborSerializer < BaseSerializer
+        def self.format
+          :cbor
+        end
+
+        def features
+          {
+            streaming: supports?(:streaming),
+            canonical: supports?(:canonical)
+          }
+        end
+
+        def library_require_name
+          raise NotImplementedError, 'Subclasses must implement #library_require_name'
+        end
+
+        def available?
+          return @available if defined?(@available)
+
+          @available = begin
+            require library_require_name
+            true
+          rescue LoadError
+            false
+          end
+        end
+      end
+    end
+  end
+end

data/lib/serialbench/serializers/cbor/cbor_serializer.rb

@@ -0,0 +1,37 @@
+# frozen_string_literal: true
+
+require_relative 'base_cbor_serializer'
+
+module Serialbench
+  module Serializers
+    module Cbor
+      # The `cbor` gem (RFC 7049 codec, to_cbor/CBOR.load)
+      class CborSerializer < BaseCborSerializer
+        def name
+          'cbor'
+        end
+
+        def parse(cbor_bytes)
+          require 'cbor'
+          CBOR.load(cbor_bytes)
+        end
+
+        def generate(object, _options = {})
+          require 'cbor'
+          object.to_cbor
+        end
+
+        def version
+          return 'unknown' unless available?
+
+          require 'cbor'
+          Gem.loaded_specs['cbor']&.version.to_s
+        end
+
+        def library_require_name
+          'cbor'
+        end
+      end
+    end
+  end
+end

data/lib/serialbench/serializers/cbor/yeptris_serializer.rb

@@ -0,0 +1,61 @@
+# frozen_string_literal: true
+
+require_relative 'base_cbor_serializer'
+
+module Serialbench
+  module Serializers
+    module Cbor
+      # Yeptris CBOR side: load + dump. dump round-trips the canonical
+      # fixtures exactly as of 0.6.35.1 (yeptris-ruby#293 — before that,
+      # strings shaped like exponent notation encoded as floats; before
+      # 0.6.34, dump segfaulted outright, yeptris-ruby#152).
+      class YeptrisSerializer < BaseCborSerializer
+        def name
+          'yeptris-cbor'
+        end
+
+        def parse(cbor_bytes)
+          require 'yeptris/cbor'
+          Yeptris::CBOR.load(cbor_bytes)
+        end
+
+        def generate(object, _options = {})
+          require 'yeptris/cbor'
+          Yeptris::CBOR.dump(object)
+        end
+
+        def capabilities
+          Set.new(%i[dom parse generate])
+        end
+
+        def available?
+          return @available if defined?(@available)
+
+          @available = begin
+            # Faces bootstrap themselves since yeptris 0.6.35.2
+            # (yeptris-ruby#300) — standalone requires are safe.
+            require 'yeptris/cbor'
+            raise LoadError, 'libyeptris has no CBOR support' unless Yeptris::CBOR.available?
+
+            Yeptris::CBOR.load("\xf5")
+            true
+          rescue StandardError, LoadError => e
+            warn "#{name} unavailable: #{e.class}: #{e.message}"
+            false
+          end
+        end
+
+        def version
+          return 'unknown' unless available?
+
+          require 'yeptris'
+          Yeptris::VERSION
+        end
+
+        def library_require_name
+          'yeptris'
+        end
+      end
+    end
+  end
+end

data/lib/serialbench/serializers/html/leptris_serializer.rb

@@ -10,8 +10,11 @@ module Serialbench
           'leptris'
         end
 
+        # 1.9.201 defaults parse_html to the :html4 engine — the
+        # libxml2/Nokogiri-compatible shape. The conformant engine is
+        # benchmarked separately as leptris-whatwg.
         def capabilities
-          super | Set.new(%i[xpath html5])
+          super | Set.new(%i[xpath])
         end
 
         def parse(html_string)

data/lib/serialbench/serializers/html/leptris_whatwg_serializer.rb

@@ -0,0 +1,28 @@
+# frozen_string_literal: true
+
+require_relative 'leptris_serializer'
+
+module Serialbench
+  module Serializers
+    module Html
+      # The WHATWG-conformant HTML engine (:whatwg mode, 1.9.104+):
+      # implied-<head> set and foster parenting — html5lib corpus
+      # 294/1555. Priced separately from the :html4 parity engine.
+      class LeptrisWhatwgSerializer < LeptrisSerializer
+        def name
+          'leptris-whatwg'
+        end
+
+        def capabilities
+          super | Set.new(%i[html5])
+        end
+
+        def parse(html_string)
+          require 'leptris'
+          require 'leptris/xml'
+          Leptris::XML.parse_html(html_string, mode: :whatwg)
+        end
+      end
+    end
+  end
+end

data/lib/serialbench/serializers/json/yeptris_serializer.rb

@@ -5,7 +5,8 @@ require_relative 'base_json_serializer'
 module Serialbench
   module Serializers
     module Json
-      # Yeptris JSON side: parsing only — no dump in 0.2.x
+      # Yeptris JSON side: 0.2.x was parse-only; 0.6.x ships a native
+      # dump that round-trips the canonical fixtures exactly.
       class YeptrisSerializer < BaseJsonSerializer
         def name
           'yeptris-json'
@@ -16,8 +17,13 @@ module Serialbench
           Yeptris::JSON.load(json_string)
         end
 
+        def generate(object, _options = {})
+          require 'yeptris'
+          Yeptris::JSON.dump(object)
+        end
+
         def capabilities
-          Set.new(%i[dom parse])
+          Set.new(%i[dom parse generate])
         end
 
         def available?

data/lib/serialbench/serializers.rb

@@ -18,6 +18,7 @@ require_relative 'serializers/html/base_html_serializer'
 require_relative 'serializers/html/nokogiri_serializer'
 require_relative 'serializers/html/oga_serializer'
 require_relative 'serializers/html/leptris_serializer'
+require_relative 'serializers/html/leptris_whatwg_serializer'
 
 # JSON Serializers
 require_relative 'serializers/json/base_json_serializer'
@@ -33,6 +34,11 @@ require_relative 'serializers/yaml/psych_serializer'
 require_relative 'serializers/yaml/syck_serializer'
 require_relative 'serializers/yaml/yeptris_serializer'
 
+# CBOR Serializers
+require_relative 'serializers/cbor/base_cbor_serializer'
+require_relative 'serializers/cbor/cbor_serializer'
+require_relative 'serializers/cbor/yeptris_serializer'
+
 # TOML Serializers
 require_relative 'serializers/toml/base_toml_serializer'
 require_relative 'serializers/toml/toml_rb_serializer'
@@ -63,7 +69,8 @@ module Serialbench
       html: [
         Html::NokogiriSerializer,
         Html::OgaSerializer,
-        Html::LeptrisSerializer
+        Html::LeptrisSerializer,
+        Html::LeptrisWhatwgSerializer
       ],
       yaml: [
         Yaml::PsychSerializer,
@@ -75,6 +82,10 @@ module Serialbench
         Toml::TomlibSerializer,
         Toml::TomlrbSerializer,
         Toml::TeptrisSerializer
+      ],
+      cbor: [
+        Cbor::CborSerializer,
+        Cbor::YeptrisSerializer
       ]
     }.freeze
 

data/lib/serialbench/test_data.rb

@@ -9,6 +9,7 @@ module Serialbench
 
     def initialize(config)
       @config = config
+      require 'cbor'
     end
 
     def load
@@ -23,7 +24,7 @@ module Serialbench
 
     def fixture_file(size, format)
       path = "test_data/#{size}.#{format}"
-      File.read(path) if File.exist?(path)
+      File.binread(path) if File.exist?(path)
     end
 
     def generate(key)
@@ -43,10 +44,25 @@ module Serialbench
       when 'medium.html' then generate_medium_html
       when 'large.toml' then generate_large_toml
       when 'large.html' then generate_large_html
+      when 'small.cbor' then generate_small_cbor
+      when 'medium.cbor' then generate_medium_cbor
+      when 'large.cbor' then generate_large_cbor
       else raise ArgumentError, "no test data generator for #{key}"
       end
     end
 
+    def generate_small_cbor
+      small_test_data_structure.to_cbor
+    end
+
+    def generate_medium_cbor
+      medium_test_data_structure.to_cbor
+    end
+
+    def generate_large_cbor
+      large_test_data_structure.to_cbor
+    end
+
     # Shared data structure generators
     def small_test_data_structure
       {

data/lib/serialbench/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Serialbench
-  VERSION = '0.12.0'
+  VERSION = '0.12.2'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: serialbench
 version: !ruby/object:Gem::Version
-  version: 0.12.0
+  version: 0.12.2
 platform: ruby
 authors:
 - Ribose
@@ -270,6 +270,7 @@ files:
 - Gemfile
 - README.adoc
 - Rakefile
+- config/benchmarks/full-cbor.yml
 - config/benchmarks/full-html.yml
 - config/benchmarks/full-json.yml
 - config/benchmarks/full-toml.yml
@@ -284,7 +285,10 @@ files:
 - config/environments/docker-ruby-3.2.yml
 - config/environments/docker-ruby-3.3.yml
 - config/environments/docker-ruby-3.4.yml
+- data/schemas/benchmark_config.yml
+- data/schemas/environment_config.yml
 - data/schemas/result.yml
+- data/schemas/resultset.yml
 - docker/Dockerfile.alpine
 - docker/Dockerfile.ubuntu
 - docker/README.md
@@ -324,8 +328,12 @@ files:
 - lib/serialbench/runners/local_runner.rb
 - lib/serialbench/serializers.rb
 - lib/serialbench/serializers/base_serializer.rb
+- lib/serialbench/serializers/cbor/base_cbor_serializer.rb
+- lib/serialbench/serializers/cbor/cbor_serializer.rb
+- lib/serialbench/serializers/cbor/yeptris_serializer.rb
 - lib/serialbench/serializers/html/base_html_serializer.rb
 - lib/serialbench/serializers/html/leptris_serializer.rb
+- lib/serialbench/serializers/html/leptris_whatwg_serializer.rb
 - lib/serialbench/serializers/html/nokogiri_serializer.rb
 - lib/serialbench/serializers/html/oga_serializer.rb
 - lib/serialbench/serializers/json/base_json_serializer.rb