serialbench 0.9.2 → 0.11.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 99cdf688cd6a4bd4662e8f0182911ec87f6b2ecf184045c6514cee6959c09e2d
-  data.tar.gz: 3e1ba2b330008ef51fe28c7fad6b0571bad06c491ca9c6a7e46634c558b20a23
+  metadata.gz: b2f980d28503e644be4c2f0807637b531c91640ad50af4eeb72c5cf41b26f3eb
+  data.tar.gz: 2e45051b18ab56d9a4fe2d569d3abc3206f151398dca094aa0a28fecd3ceb07f
 SHA512:
-  metadata.gz: 30fc0f72e78b350c78b918a6366793be940a35ab6b826e3a74c08547d11b0251f90d8be9ec4822a7b1c197656805e6c7e2142c483b123981c7e7e5ba64c1e030
-  data.tar.gz: a44b14c1638c2c450b89d8f8a032041cdfb14eb1a1ccb57eb9a6b88d248f2f18d1be127095babef559f938918048b89b1c7d950cf9fb9f64eabcbfd29787de61
+  metadata.gz: 35d305480351a282cf7cccc5c18874d1d5f053f2ee914452082c0abed7f1b92e66a63ba582a90ed898320d55c499a43b5207c3e4f12c529111223f7841e5d89b
+  data.tar.gz: 7441471432ba4b498ac91d7a4bdf4b21338c36789a231014f3f0bc49bb3858952f3627eeb2209442b9c266d4664e91515e677dda472003bb20614c25648c5010

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

@@ -29,12 +29,34 @@ CONTENT=$(base64 < "$RESULTS_FILE" | tr -d '\r\n')
 
 echo "Pushing ${TARGET_PATH} to serialbench/data..."
 
-# Try create; if file exists (422), get sha and update
-HTTP_CODE=$(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}" \
-  -d "{\"message\": \"${PLATFORM} ruby-${RUBY_VERSION} ${FMT}\", \"content\": \"${CONTENT}\", \"branch\": \"main\"}")
+# 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\"}")
+  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}" \
+    "${sha_arg[@]}"
+}
+
+HTTP_CODE=$(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}')
+    [ -z "$SHA" ] && break
+    echo "attempt $TRY got $HTTP_CODE — refetching sha and retrying"
+    sleep $((TRY * 3))
+    HTTP_CODE=$(put_file "$SHA")
+  else
+    break
+  fi
+done
 
 if [ "$HTTP_CODE" = "200" ] || [ "$HTTP_CODE" = "201" ]; then
   echo "Pushed ${TARGET_PATH}"

data/.github/workflows/benchmark.yml

@@ -116,6 +116,21 @@ jobs:
       - name: Install gems
         run: bundle install
 
+      - name: Fetch Saxon-HE (XSLT 2.0/3.0 comparison)
+        run: |
+          mkdir -p vendor
+          curl -sL -o vendor/saxonhe.jar "https://repo1.maven.org/maven2/net/sf/saxon/Saxon-HE/12.10/Saxon-HE-12.10.jar"
+          curl -sL -o vendor/xmlresolver.jar "https://repo1.maven.org/maven2/org/xmlresolver/xmlresolver/5.2.2/xmlresolver-5.2.2.jar"
+          echo "SAXON_JAR=$PWD/vendor/saxonhe.jar" >> "$GITHUB_ENV"
+          echo "XMLRESOLVER_JAR=$PWD/vendor/xmlresolver.jar" >> "$GITHUB_ENV"
+        shell: bash
+
+      - name: Load canonical fixtures
+        run: |
+          git clone --depth 1 https://github.com/serialbench/fixtures.git test_data
+          (cd test_data && sha256sum -c SHA256SUMS) || (cd test_data && shasum -a 256 -c SHA256SUMS)
+        shell: bash
+
       - name: Point teptris at its vendored DLL (Windows)
         if: startsWith(matrix.platform, 'windows-')
         shell: bash
@@ -249,6 +264,41 @@ jobs:
         run: .github/scripts/push-to-data.sh toml ${{ matrix.ruby-version }} ${{ matrix.platform }}
         shell: bash
 
+      - name: Benchmark html
+        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 html ${{ matrix.ruby-version }} ${{ matrix.platform }}; then
+            echo "html already in the data repo — skipping (gap-filling rerun)"
+            exit 0
+          fi
+          bundle exec serialbench environment execute \
+            config/environments/ci-ruby-${{ matrix.ruby-version }}.yml \
+            config/benchmarks/full-html.yml \
+            "results/runs/ci-ruby-${{ matrix.ruby-version }}-${{ matrix.platform }}/html"
+
+      - name: Upload html
+        if: ${{ !cancelled() }}
+        uses: actions/upload-artifact@v4
+        with:
+          name: benchmark-results-${{ matrix.platform }}-ruby-${{ matrix.ruby-version }}-html
+          path: results/runs/ci-ruby-${{ matrix.ruby-version }}-${{ matrix.platform }}/html
+          retention-days: 30
+          if-no-files-found: warn
+
+      - name: Push html 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 html ${{ matrix.ruby-version }} ${{ matrix.platform }}
+        shell: bash
+
       # Each format is an independent checkpoint with its own upload.
       # A failure in one format doesn't prevent the others from uploading.
       - name: Benchmark xml
@@ -296,7 +346,7 @@ jobs:
           DATA_RUN_DATE: ${{ needs.setup.outputs.run-date }}
         run: |
           MISSING=""
-          for FMT in json yaml toml xml; do
+          for FMT in json yaml toml html xml; do
             if .github/scripts/format-status.sh "$FMT" "${{ matrix.ruby-version }}" "${{ matrix.platform }}"; then
               echo "$FMT: in data repo"
             else

data/.gitignore

@@ -30,3 +30,4 @@ results/
 
 # Ruby version manager files
 .ruby-version
+test_data/

data/README.adoc

@@ -1,9 +1,9 @@
 = Serialbench: Ruby serialization library performance benchmarker
 
 image:https://img.shields.io/gem/v/serialbench.svg["Gem Version", link="https://rubygems.org/gems/serialbench"]
-image:https://github.com/metanorma/serialbench/actions/workflows/ci.yml/badge.svg["Build Status", link="https://github.com/metanorma/serialbench/actions/workflows/ci.yml"]
-image:https://github.com/metanorma/serialbench/actions/workflows/benchmark.yml/badge.svg["Benchmark Status", link="https://github.com/metanorma/serialbench/actions/workflows/benchmark.yml"]
-image:https://img.shields.io/github/issues-pr-raw/metanorma/serialbench.svg["Pull Requests", link="https://github.com/metanorma/serialbench/pulls"]
+image:https://github.com/serialbench/serialbench/actions/workflows/ci.yml/badge.svg["Build Status", link="https://github.com/serialbench/serialbench/actions/workflows/ci.yml"]
+image:https://github.com/serialbench/serialbench/actions/workflows/benchmark.yml/badge.svg["Benchmark Status", link="https://github.com/serialbench/serialbench/actions/workflows/benchmark.yml"]
+image:https://img.shields.io/github/issues-pr-raw/serialbench/serialbench.svg["Pull Requests", link="https://github.com/serialbench/serialbench/pulls"]
 
 == Overview
 

data/config/benchmarks/full-html.yml

@@ -0,0 +1,26 @@
+# Per-format split of full.yml: each format runs in its own process so
+# memory profiling of large documents cannot compound across formats
+# (windows runners OOMed with all four formats in one process).
+name: full-html
+
+data_sizes:
+- small
+- medium
+- large
+
+formats:
+- html
+
+iterations:
+  small: 10
+  medium: 3
+  large: 1
+
+operations:
+- parsing
+- generation
+- streaming
+- xpath
+- memory
+
+warmup: 2

data/config/benchmarks/full-xml.yml

@@ -21,6 +21,10 @@ operations:
 - generation
 - streaming
 - xpath
+- xquery
+- xslt
+- xslt30
+- validation
 - memory
 
 warmup: 2

data/lib/serialbench/benchmark_runner.rb

@@ -26,15 +26,170 @@ module Serialbench
     end
 
     # Each operation maps to a lambda. Adding an operation = one entry.
+_DEFAULT_RNG_SCHEMA = <<'RNG'.freeze
+<?xml version="1.0" encoding="UTF-8"?>
+<grammar xmlns="http://relaxng.org/ns/structure/1.0" datatypeLibrary="http://www.w3.org/2001/XMLSchema-datatypes">
+  <start>
+    <choice>
+      <ref name="config"/>
+      <ref name="users"/>
+      <ref name="dataset"/>
+    </choice>
+  </start>
+
+  <define name="config">
+    <element name="config">
+      <element name="database">
+        <element name="host"><text/></element>
+        <element name="port"><data type="integer"/></element>
+        <element name="name"><text/></element>
+        <element name="user"><text/></element>
+        <element name="password"><text/></element>
+      </element>
+      <element name="cache">
+        <element name="enabled"><data type="boolean"/></element>
+        <element name="ttl"><data type="integer"/></element>
+      </element>
+    </element>
+  </define>
+
+  <define name="users">
+    <element name="users">
+      <oneOrMore>
+        <ref name="user"/>
+      </oneOrMore>
+    </element>
+  </define>
+
+  <define name="user">
+    <element name="user">
+      <attribute name="id"><data type="integer"/></attribute>
+      <element name="name"><text/></element>
+      <element name="email"><text/></element>
+      <element name="created_at"><text/></element>
+      <element name="profile">
+        <element name="age"><data type="integer"/></element>
+        <element name="city"><text/></element>
+        <element name="preferences">
+          <element name="theme"><text/></element>
+          <element name="notifications"><data type="boolean"/></element>
+        </element>
+      </element>
+    </element>
+  </define>
+
+  <define name="dataset">
+    <element name="dataset">
+      <element name="header">
+        <element name="created"><text/></element>
+        <element name="count"><data type="integer"/></element>
+        <element name="format"><text/></element>
+      </element>
+      <element name="records">
+        <oneOrMore>
+          <ref name="record"/>
+        </oneOrMore>
+      </element>
+    </element>
+  </define>
+
+  <define name="record">
+    <element name="record">
+      <attribute name="id"><data type="integer"/></attribute>
+      <element name="timestamp"><text/></element>
+      <element name="data">
+        <element name="field1"><text/></element>
+        <element name="field2"><data type="integer"/></element>
+        <element name="field3"><text/></element>
+        <element name="nested">
+          <oneOrMore><element name="item"><text/></element></oneOrMore>
+        </element>
+      </element>
+      <element name="metadata">
+        <element name="source"><text/></element>
+        <element name="version"><text/></element>
+        <element name="checksum"><text/></element>
+      </element>
+    </element>
+  </define>
+</grammar>
+RNG
+
+_DEFAULT_XSLT_STYLESHEET = <<'XSL'.freeze
+<?xml version="1.0" encoding="UTF-8"?>
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
+  <xsl:output method="xml" indent="yes"/>
+  <xsl:template match="/">
+    <report>
+      <xsl:apply-templates select="//user | //record | //database | //cache"/>
+      <total>
+        <xsl:value-of select="count(//user) + count(//record)"/>
+      </total>
+    </report>
+  </xsl:template>
+  <xsl:template match="user">
+    <row kind="user" id="{@id}" label="{name}"/>
+  </xsl:template>
+  <xsl:template match="record">
+    <row kind="record" id="{@id}" label="{data/field1}"/>
+  </xsl:template>
+  <xsl:template match="database">
+    <row kind="database" label="{name}"/>
+  </xsl:template>
+  <xsl:template match="cache">
+    <row kind="cache" label="{ttl}"/>
+  </xsl:template>
+</xsl:stylesheet>
+XSL
+
+_DEFAULT_XSLT30_STYLESHEET = <<'XSL30'.freeze
+<?xml version="1.0" encoding="UTF-8"?>
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+                xmlns:xs="http://www.w3.org/2001/XMLSchema"
+                exclude-result-prefixes="xs"
+                version="3.0">
+  <xsl:output method="xml" indent="yes"/>
+  <xsl:template match="/">
+    <report30>
+      <xsl:iterate select="//user | //record | //database | //cache">
+        <xsl:param name="labels" as="xs:string*" select="()"/>
+        <xsl:on-completion>
+          <joined><xsl:value-of select="string-join($labels, ', ')"/></joined>
+          <count><xsl:value-of select="count($labels)"/></count>
+        </xsl:on-completion>
+        <xsl:variable name="label" select="string((name, data/field1, ttl)[1])"/>
+        <xsl:if test="$label != ''">
+          <row kind="{local-name()}" label="{upper-case($label)}"/>
+        </xsl:if>
+        <xsl:next-iteration>
+          <xsl:with-param name="labels" select="if ($label != '') then ($labels, $label) else $labels"/>
+        </xsl:next-iteration>
+      </xsl:iterate>
+    </report30>
+  </xsl:template>
+</xsl:stylesheet>
+XSL30
+
+    XPATH_QUERIES = ['//book', "//book[@id='101']", '//book[price > 30]/title'].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 }
+    XSLT30_STYLESHEET = File.expand_path('test_data/transform30.xsl', Dir.pwd).then { |p| File.exist?(p) ? File.read(p) : _DEFAULT_XSLT30_STYLESHEET }
+
     OPERATIONS = {
       'parsing' => ->(s, data) { s.parse(data) },
       'generation' => ->(s, data) { s.generate(s.parse(data)) },
       'xpath' => lambda { |s, data|
         doc = s.parse(data)
-        s.xpath_query(doc, '//book')
-        s.xpath_query(doc, "//book[@id='101']")
-        s.xpath_query(doc, '//book[price > 30]/title')
+        XPATH_QUERIES.each { |q| s.xpath_query(doc, q) }
+      },
+      'xquery' => lambda { |s, data|
+        doc = s.parse(data)
+        XQUERY_EXPRESSIONS.each { |x| s.xquery_eval(doc, x) }
       },
+      'xslt' => ->(s, data) { s.xslt_apply(data, XSLT_STYLESHEET) },
+      'xslt30' => ->(s, data) { s.xslt_apply(data, XSLT30_STYLESHEET) },
+      'validation' => ->(s, data) { s.validate(s.parse(data), RNG_SCHEMA) },
       'streaming' => ->(s, data) { s.stream_parse(data) { |_event, _data| } },
     }.freeze
 
@@ -155,12 +310,22 @@ module Serialbench
       serializers = Serializers.for_format(format)
 
       case type_name
+      when 'parsing', 'memory'
+        serializers.select { |s| s.supports?(:parse) }
       when 'generation'
         serializers.select { |s| s.supports?(:generate) }
       when 'streaming'
         serializers.select { |s| s.supports?(:sax) || s.supports?(:streaming) }
       when 'xpath'
         serializers.select { |s| s.supports?(:xpath) }
+      when 'xquery'
+        serializers.select { |s| s.supports?(:xquery) }
+      when 'xslt'
+        serializers.select { |s| s.supports?(:xslt) }
+      when 'validation'
+        serializers.select { |s| s.supports?(:validation) }
+      when 'xslt30'
+        serializers.select { |s| s.supports?(:xslt30) }
       else
         serializers
       end

data/lib/serialbench/models/benchmark_config.rb

@@ -48,10 +48,10 @@ 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 json yaml toml]
+      attribute :formats, :string, collection: true, values: %w[xml html json yaml toml]
       attribute :iterations, BenchmarkIteration
       attribute :warmup, :integer, default: -> { 1 }
-      attribute :operations, :string, collection: true, values: %w[parsing generation xpath streaming memory]
+      attribute :operations, :string, collection: true, values: %w[parsing generation xpath xquery xslt xslt30 validation streaming memory]
 
       key_value do
         map 'name', to: :name

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 json yaml toml]
+      attribute :format, :string, values: %w[xml html json yaml toml]
       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 json yaml toml]
+      attribute :format, :string, values: %w[xml html json yaml toml]
       attribute :data_size, :string, values: %w[small medium large]
 
       key_value do
@@ -71,6 +71,10 @@ module Serialbench
       attribute :memory, MemoryPerformance, collection: true
       attribute :streaming, IterationPerformance, collection: true
       attribute :xpath, IterationPerformance, collection: true
+      attribute :xquery, IterationPerformance, collection: true
+      attribute :xslt, IterationPerformance, collection: true
+      attribute :xslt30, IterationPerformance, collection: true
+      attribute :validation, IterationPerformance, collection: true
 
       key_value do
         map 'serializers', to: :serializers
@@ -79,6 +83,10 @@ module Serialbench
         map 'memory', to: :memory
         map 'streaming', to: :streaming
         map 'xpath', to: :xpath
+        map 'xquery', to: :xquery
+        map 'xslt', to: :xslt
+        map 'xslt30', to: :xslt30
+        map 'validation', to: :validation
       end
     end
   end

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

@@ -0,0 +1,40 @@
+# frozen_string_literal: true
+
+require_relative '../base_serializer'
+
+module Serialbench
+  module Serializers
+    module Html
+      class BaseHtmlSerializer < BaseSerializer
+        def self.format
+          :html
+        end
+
+        def capabilities
+          super | Set.new(%i[dom parse generate])
+        end
+
+        def features
+          {
+            html5: supports?(:html5),
+            xpath: supports?(:xpath)
+          }
+        end
+
+        def generate(data)
+          data.is_a?(Hash) ? from_hash(data) : serialize_document(data)
+        end
+
+        def from_hash(hash)
+          rows = hash.flat_map do |key, value|
+            case value
+            when Hash then value.map { |k, v| "<tr><td>#{k}</td><td>#{v}</td></tr>" }
+            else ["<tr><td>#{key}</td><td>#{value}</td></tr>"]
+            end
+          end
+          serialize_document(parse("<html><body><table>#{rows.join}</table></body></html>"))
+        end
+      end
+    end
+  end
+end

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

@@ -0,0 +1,58 @@
+# frozen_string_literal: true
+
+require_relative 'base_html_serializer'
+
+module Serialbench
+  module Serializers
+    module Html
+      class LeptrisSerializer < BaseHtmlSerializer
+        def name
+          'leptris'
+        end
+
+        def capabilities
+          super | Set.new(%i[xpath html5])
+        end
+
+        def parse(html_string)
+          require 'leptris'
+          require 'leptris/xml'
+          Leptris::XML.parse_html(html_string)
+        end
+
+        def xpath_query(document, expression)
+          document.xpath(expression).size
+        end
+
+        def serialize_document(document)
+          document.to_xml
+        end
+
+        def available?
+          return @available if defined?(@available)
+
+          @available = begin
+            require 'leptris'
+            require 'leptris/xml'
+            Leptris::XML.parse_html('<p>probe</p>')
+            true
+          rescue StandardError, LoadError => e
+            warn "#{name} unavailable: #{e.class}: #{e.message}"
+            false
+          end
+        end
+
+        def version
+          return 'unknown' unless available?
+
+          require 'leptris'
+          Leptris::VERSION
+        end
+
+        def library_require_name
+          'leptris'
+        end
+      end
+    end
+  end
+end

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

@@ -0,0 +1,56 @@
+# frozen_string_literal: true
+
+require_relative 'base_html_serializer'
+
+module Serialbench
+  module Serializers
+    module Html
+      class NokogiriSerializer < BaseHtmlSerializer
+        def name
+          'nokogiri'
+        end
+
+        def capabilities
+          super | Set.new(%i[xpath html5])
+        end
+
+        def parse(html_string)
+          require 'nokogiri'
+          Nokogiri::HTML5(html_string)
+        end
+
+        def xpath_query(document, expression)
+          document.xpath(expression).size
+        end
+
+        def serialize_document(document)
+          document.to_html
+        end
+
+        def available?
+          return @available if defined?(@available)
+
+          @available = begin
+            require 'nokogiri'
+            Nokogiri::HTML5('<p>probe</p>')
+            true
+          rescue StandardError, LoadError => e
+            warn "#{name} unavailable: #{e.class}: #{e.message}"
+            false
+          end
+        end
+
+        def version
+          return 'unknown' unless available?
+
+          require 'nokogiri'
+          Nokogiri::VERSION
+        end
+
+        def library_require_name
+          'nokogiri'
+        end
+      end
+    end
+  end
+end

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

@@ -0,0 +1,48 @@
+# frozen_string_literal: true
+
+require_relative 'base_html_serializer'
+
+module Serialbench
+  module Serializers
+    module Html
+      class OgaSerializer < BaseHtmlSerializer
+        def name
+          'oga'
+        end
+
+        def parse(html_string)
+          require 'oga'
+          Oga.parse_html(html_string)
+        end
+
+        def serialize_document(document)
+          document.to_xml
+        end
+
+        def available?
+          return @available if defined?(@available)
+
+          @available = begin
+            require 'oga'
+            Oga.parse_html('<p>probe</p>')
+            true
+          rescue StandardError, LoadError => e
+            warn "#{name} unavailable: #{e.class}: #{e.message}"
+            false
+          end
+        end
+
+        def version
+          return 'unknown' unless available?
+
+          require 'oga'
+          Oga::VERSION
+        end
+
+        def library_require_name
+          'oga'
+        end
+      end
+    end
+  end
+end

data/lib/serialbench/serializers/xml/base_xml_serializer.rb

@@ -27,6 +27,10 @@ module Serialbench
           super | Set.new(%i[namespaces])
         end
 
+        def xslt_apply(source_xml, stylesheet)
+          xslt_transform(parse(source_xml), stylesheet)
+        end
+
         # XML-specific features derive from the capability set
         def features
           {

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

@@ -37,7 +37,26 @@ module Serialbench
         end
 
         def capabilities
-          super | Set.new(%i[xpath sax stax])
+          super | Set.new(%i[xpath xquery xslt xslt30 validation sax stax])
+        end
+
+        def xquery_eval(document, expression)
+          require 'leptris'
+          require 'leptris/xml'
+          result = Leptris::XML::XQuery.parse(expression).eval(document)
+          result.respond_to?(:length) ? result.length : 1
+        end
+
+        def xslt_transform(document, stylesheet)
+          require 'leptris'
+          require 'leptris/xml'
+          Leptris::XML::XSLT.parse(stylesheet).apply_to(document).serialize
+        end
+
+        def validate(document, schema)
+          require 'leptris'
+          require 'leptris/xml'
+          Leptris::XML::RelaxNG.parse(schema).valid?(document)
         end
 
         def xpath_query(document, expression)

data/lib/serialbench/serializers/xml/nokogiri_serializer.rb

@@ -39,7 +39,17 @@ module Serialbench
         end
 
         def capabilities
-          super | Set.new(%i[xpath sax stax])
+          super | Set.new(%i[xpath xslt validation sax stax])
+        end
+
+        def xslt_transform(document, stylesheet)
+          require 'nokogiri'
+          Nokogiri::XSLT(stylesheet).transform(document).to_s
+        end
+
+        def validate(document, schema)
+          require 'nokogiri'
+          Nokogiri::XML::RelaxNG(schema).valid?(document)
         end
 
         def xpath_query(document, expression)

data/lib/serialbench/serializers/xml/saxon_serializer.rb

@@ -0,0 +1,72 @@
+# frozen_string_literal: true
+
+require_relative 'base_xml_serializer'
+require 'open3'
+require 'tmpdir'
+
+module Serialbench
+  module Serializers
+    module Xml
+      # Saxon-HE via its Java CLI: the reference XSLT 2.0/3.0 engine.
+      # Cold-process semantics: every iteration is a full JVM + compile +
+      # transform, so ips is comparable to leptris only end-to-end, never
+      # per-transform — features record cold_process: true.
+      class SaxonSerializer < BaseXmlSerializer
+        def name
+          'saxon-he'
+        end
+
+        def capabilities
+          Set.new(%i[xslt xslt30])
+        end
+
+        def features
+          { 'xslt_version' => '3.0', 'cold_process' => true }
+        end
+
+        def classpath
+          jars = [ENV['SAXON_JAR'], ENV['XMLRESOLVER_JAR']].compact
+          jars.empty? ? nil : jars.join(File::PATH_SEPARATOR)
+        end
+
+        def xslt_apply(source_xml, stylesheet)
+          Dir.mktmpdir('saxonbench') do |dir|
+            src = File.join(dir, 's.xml')
+            xsl = File.join(dir, 's.xsl')
+            File.write(src, source_xml)
+            File.write(xsl, stylesheet)
+            out, _err, _st = Open3.capture3('java', '-cp', classpath, 'net.sf.saxon.Transform',
+                                            "-xsl:#{xsl}", "-s:#{src}")
+            raise Error, "saxon transform failed: #{out}#{_err}"[0, 200] unless _st.success?
+
+            out
+          end
+        end
+
+        def available?
+          return @available if defined?(@available)
+
+          @available = begin
+            cp = classpath
+            raise LoadError, 'SAXON_JAR / XMLRESOLVER_JAR not set' unless cp
+
+            out, _err, st = Open3.capture3('java', '-cp', cp, 'net.sf.saxon.Version')
+            raise LoadError, 'java failed' unless st.success?
+
+            true
+          rescue StandardError, LoadError => e
+            warn "#{name} unavailable: #{e.class}: #{e.message}"
+            false
+          end
+        end
+
+        def version
+          return 'unknown' unless available?
+
+          out, = Open3.capture3('java', '-cp', classpath, 'net.sf.saxon.Version')
+          out[/Saxon-HE (\S+)/, 1] || out.strip[0, 40]
+        end
+      end
+    end
+  end
+end

data/lib/serialbench/serializers.rb

@@ -11,6 +11,13 @@ require_relative 'serializers/xml/nokogiri_serializer'
 require_relative 'serializers/xml/oga_serializer'
 require_relative 'serializers/xml/libxml_serializer'
 require_relative 'serializers/xml/leptris_serializer'
+require_relative 'serializers/xml/saxon_serializer'
+
+# HTML Serializers
+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'
 
 # JSON Serializers
 require_relative 'serializers/json/base_json_serializer'
@@ -43,7 +50,8 @@ module Serialbench
         Xml::NokogiriSerializer,
         Xml::OgaSerializer,
         Xml::LibxmlSerializer,
-        Xml::LeptrisSerializer
+        Xml::LeptrisSerializer,
+        Xml::SaxonSerializer
       ],
       json: [
         Json::JsonSerializer,
@@ -52,6 +60,11 @@ module Serialbench
         Json::YajlSerializer,
         Json::YeptrisSerializer
       ],
+      html: [
+        Html::NokogiriSerializer,
+        Html::OgaSerializer,
+        Html::LeptrisSerializer
+      ],
       yaml: [
         Yaml::PsychSerializer,
         Yaml::SyckSerializer,

data/lib/serialbench/test_data.rb

@@ -38,8 +38,11 @@ module Serialbench
       when 'medium.yaml' then generate_medium_yaml
       when 'large.yaml' then generate_large_yaml
       when 'small.toml' then generate_small_toml
+      when 'small.html' then generate_small_html
       when 'medium.toml' then generate_medium_toml
+      when 'medium.html' then generate_medium_html
       when 'large.toml' then generate_large_toml
+      when 'large.html' then generate_large_html
       else raise ArgumentError, "no test data generator for #{key}"
       end
     end
@@ -205,6 +208,50 @@ module Serialbench
       XML
     end
 
+    # HTML test data generators (same dataset as XML, as tables)
+    def generate_small_html
+      data = small_test_data_structure
+      rows = data[:config].flat_map do |_section, entries|
+        entries.flat_map do |_k, v|
+          case v
+          when Hash then v.map { |k2, v2| "<tr><td>#{k2}</td><td>#{v2}</td></tr>" }
+          else []
+          end
+        end
+      end
+      html_page('config', rows)
+    end
+
+    def generate_medium_html
+      data = medium_test_data_structure
+      rows = data[:users].map do |u|
+        "<tr><td>#{u[:id]}</td><td>#{u[:name][0, 30]}</td></tr>"
+      end
+      html_page('users', rows)
+    end
+
+    def generate_large_html
+      data = large_test_data_structure
+      rows = data[:dataset][:records].map do |r|
+        "<tr><td>#{r[:id]}</td><td>#{r[:data][:field1][0, 30]}</td></tr>"
+      end
+      html_page('dataset', rows)
+    end
+
+    def html_page(title, rows)
+      <<~HTML
+        <!DOCTYPE html>
+        <html>
+        <head><meta charset="utf-8"><title>serialbench #{title}</title></head>
+        <body>
+        <table>
+        #{rows.join("\n")}
+        </table>
+        </body>
+        </html>
+      HTML
+    end
+
     # JSON test data generators
     def generate_small_json
       JSON.generate(small_test_data_structure)

data/lib/serialbench/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Serialbench
-  VERSION = '0.9.2'
+  VERSION = '0.11.0'
 end

data/serialbench.gemspec

@@ -14,7 +14,7 @@ Gem::Specification.new do |spec|
 
   spec.summary = 'Comprehensive serialization benchmarking tool for Ruby'
   spec.description = 'A benchmarking suite for comparing performance of various serialization libraries in Ruby, including XML, JSON, and TOML parsers/generators.'
-  spec.homepage = 'https://github.com/serialbench/serialbench'
+  spec.homepage = 'https://github.com/serialbench/serialbench-ruby'
   spec.license = 'BSD-2-Clause'
   spec.required_ruby_version = Gem::Requirement.new('>= 2.7.0')
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: serialbench
 version: !ruby/object:Gem::Version
-  version: 0.9.2
+  version: 0.11.0
 platform: ruby
 authors:
 - Ribose
@@ -270,6 +270,7 @@ files:
 - Gemfile
 - README.adoc
 - Rakefile
+- config/benchmarks/full-html.yml
 - config/benchmarks/full-json.yml
 - config/benchmarks/full-toml.yml
 - config/benchmarks/full-xml.yml
@@ -323,6 +324,10 @@ files:
 - lib/serialbench/runners/local_runner.rb
 - lib/serialbench/serializers.rb
 - lib/serialbench/serializers/base_serializer.rb
+- lib/serialbench/serializers/html/base_html_serializer.rb
+- lib/serialbench/serializers/html/leptris_serializer.rb
+- lib/serialbench/serializers/html/nokogiri_serializer.rb
+- lib/serialbench/serializers/html/oga_serializer.rb
 - lib/serialbench/serializers/json/base_json_serializer.rb
 - lib/serialbench/serializers/json/json_serializer.rb
 - lib/serialbench/serializers/json/oj_serializer.rb
@@ -341,6 +346,7 @@ files:
 - lib/serialbench/serializers/xml/oga_serializer.rb
 - lib/serialbench/serializers/xml/ox_serializer.rb
 - lib/serialbench/serializers/xml/rexml_serializer.rb
+- lib/serialbench/serializers/xml/saxon_serializer.rb
 - lib/serialbench/serializers/yaml/base_yaml_serializer.rb
 - lib/serialbench/serializers/yaml/psych_serializer.rb
 - lib/serialbench/serializers/yaml/syck_serializer.rb
@@ -349,13 +355,13 @@ files:
 - lib/serialbench/version.rb
 - lib/serialbench/yaml_validator.rb
 - serialbench.gemspec
-homepage: https://github.com/serialbench/serialbench
+homepage: https://github.com/serialbench/serialbench-ruby
 licenses:
 - BSD-2-Clause
 metadata:
-  homepage_uri: https://github.com/serialbench/serialbench
-  source_code_uri: https://github.com/serialbench/serialbench
-  bug_tracker_uri: https://github.com/serialbench/serialbench/issues
+  homepage_uri: https://github.com/serialbench/serialbench-ruby
+  source_code_uri: https://github.com/serialbench/serialbench-ruby
+  bug_tracker_uri: https://github.com/serialbench/serialbench-ruby/issues
 rdoc_options: []
 require_paths:
 - lib