rxerces 0.5.0 → 0.6.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 301821063fa998d1a4522029ef825aaa7bddc45370ed0c1bb904f8b31fd9f036
-  data.tar.gz: cabdd59089a44485d847b7ff3637197d2c24e38abb3e5c68423c397a9869e843
+  metadata.gz: 3a1129744191c92a7a15cb783fc74cbccfd2dbeedfbcf5741e295686c38b8108
+  data.tar.gz: 0b03758412cff12acccf3232f0f135ec69e0eaabcb6aa015091a44ed75e53081
 SHA512:
-  metadata.gz: 9edd86524fc0ac49a51a9b2658d723a84f85c295d362e8dad500bcd8a47e95c3ad4b41d073f7bdcee0927bd38a6b1710537ee190693ec739de20e65b22e04d8a
-  data.tar.gz: 035ed74771b3d3e6125c1567d5452ac773f80a2f9ae220efd063358f7fcc93326f1beb027bf67e45ad042a950a8372b208426f07ab5320c58bb453f40c4d4c19
+  metadata.gz: e2ba8796e66c0163d1f3bd9ae91d2fde804b462b39fd750e8f600b073e1ad02e36a2978cd533ea7256a9ad7b7d906c4462927bc609b1f59ac7c33f39b5423b83
+  data.tar.gz: 9b7b522fc2c5f203a2de65ba2d112ef644cf84fb18128503432ab37240f4b31cdc7acde58c83d00e5b17e006433e595aa19f46f2a5a11345c1778c55a76c1820

data/CHANGES.md

@@ -1,3 +1,10 @@
+## 0.6.0 - 17-Dec-2025
+* Some internal refactoring, better initialization, some better string
+  handling, that sort of stuff.
+* Added the Document#errors method with more detailed information. Also
+  helps with Nokogiri compatibility.
+* Added some benchmarks (they're not great compared to others, oh well).
+
 ## 0.5.0 - 16-Dec-2025
 * Implemented a real css method. Requires Xalan to be installed.
 * Added text/content methods for most classes.

data/benchmarks/README.md

@@ -0,0 +1,68 @@
+# RXerces Performance Benchmarks
+
+This directory contains performance benchmarks comparing RXerces with other popular Ruby XML libraries (Nokogiri and Ox).
+
+## Setup
+
+Install the required gems:
+
+```bash
+gem install benchmark-ips nokogiri ox
+```
+
+## Running Benchmarks
+
+Run all benchmarks:
+
+```bash
+ruby benchmarks/parse_benchmark.rb
+ruby benchmarks/xpath_benchmark.rb
+ruby benchmarks/css_benchmark.rb
+ruby benchmarks/traversal_benchmark.rb
+ruby benchmarks/serialization_benchmark.rb
+```
+
+Or run a specific benchmark:
+
+```bash
+ruby benchmarks/parse_benchmark.rb
+```
+
+## Benchmark Categories
+
+### 1. Parse Benchmark (`parse_benchmark.rb`)
+Tests XML document parsing performance with small, medium, and large documents.
+
+### 2. XPath Benchmark (`xpath_benchmark.rb`)
+Tests XPath query performance including:
+- Simple queries (`//book`)
+- Attribute-based queries (`//book[@category='fiction']`)
+- Complex queries with predicates
+- `at_xpath` for first-match queries
+
+### 3. CSS Benchmark (`css_benchmark.rb`)
+Tests CSS selector performance including:
+- Simple selectors (`div`)
+- Class selectors (`.title`)
+- ID selectors (`#div100`)
+- Descendant combinators (`div.content p.text`)
+- `at_css` for first-match queries
+
+### 4. Traversal Benchmark (`traversal_benchmark.rb`)
+Tests DOM traversal operations:
+- `.children` access
+- `.element_children` access
+- `.parent` access
+- `.ancestors` access
+- `.next_sibling` access
+- `.text` extraction
+
+### 5. Serialization Benchmark (`serialization_benchmark.rb`)
+Tests document serialization (`to_s`/`to_xml`) with various document sizes.
+
+## Notes
+
+- All benchmarks use `benchmark-ips` for accurate iterations-per-second measurements
+- Each benchmark runs with a 2-second warmup and 5-second measurement period
+- Nokogiri and Ox tests are skipped if not installed
+- RXerces requires Xalan-C for full XPath 1.0 support (CSS selectors need XPath)

data/benchmarks/css_benchmark.rb

@@ -0,0 +1,115 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+require 'benchmark/ips'
+require 'rxerces'
+
+begin
+  require 'nokogiri'
+  NOKOGIRI_AVAILABLE = true
+rescue LoadError
+  NOKOGIRI_AVAILABLE = false
+  puts "Nokogiri not available - install with: gem install nokogiri"
+end
+
+# Sample HTML/XML for CSS selector queries
+HTML_DATA = begin
+  divs = (1..200).map do |i|
+    "<div class=\"content\" id=\"div#{i}\">
+        <h2 class=\"title\">Heading #{i}</h2>
+        <p class=\"text\">Paragraph #{i} with <span class=\"highlight\">highlighted</span> text.</p>
+        <ul class=\"list\">
+          <li>Item 1</li>
+          <li>Item 2</li>
+          <li class=\"special\">Special Item</li>
+        </ul>
+      </div>"
+  end
+  "<html><body>#{divs.join}</body></html>"
+end
+
+puts "=" * 80
+puts "CSS Selector Benchmarks"
+puts "=" * 80
+puts "Document Size: #{HTML_DATA.bytesize} bytes"
+puts
+
+# Parse documents once
+rxerces_doc = RXerces::XML::Document.parse(HTML_DATA)
+nokogiri_doc = Nokogiri::HTML(HTML_DATA) if NOKOGIRI_AVAILABLE
+
+# Simple CSS selector
+puts "Simple CSS: div"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_doc.css('div') }
+  x.report("nokogiri") { nokogiri_doc.css('div') } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Class selector
+puts "Class CSS: .title"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_doc.css('.title') }
+  x.report("nokogiri") { nokogiri_doc.css('.title') } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# ID selector
+puts "ID CSS: #div100"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_doc.css('#div100') }
+  x.report("nokogiri") { nokogiri_doc.css('#div100') } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Descendant combinator
+puts "Descendant CSS: div.content p.text"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_doc.css('div.content p.text') }
+  x.report("nokogiri") { nokogiri_doc.css('div.content p.text') } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# at_css (first match)
+puts "at_css: .title (first match only)"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_doc.at_css('.title') }
+  x.report("nokogiri") { nokogiri_doc.at_css('.title') } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+puts "=" * 80

data/benchmarks/parse_benchmark.rb

@@ -0,0 +1,103 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+require 'benchmark/ips'
+require 'rxerces'
+
+# Try to load Nokogiri and Ox
+begin
+  require 'nokogiri'
+  NOKOGIRI_AVAILABLE = true
+rescue LoadError
+  NOKOGIRI_AVAILABLE = false
+  puts "Nokogiri not available - install with: gem install nokogiri"
+end
+
+begin
+  require 'ox'
+  OX_AVAILABLE = true
+rescue LoadError
+  OX_AVAILABLE = false
+  puts "Ox not available - install with: gem install ox"
+end
+
+# Sample XML documents of varying sizes
+SMALL_XML = <<~XML
+  <root>
+    <person id="1" name="Alice">
+      <age>30</age>
+      <city>New York</city>
+    </person>
+  </root>
+XML
+
+MEDIUM_XML = begin
+  people = (1..100).map do |i|
+    "<person id=\"#{i}\" name=\"Person#{i}\"><age>#{20 + (i % 50)}</age><city>City#{i % 20}</city></person>"
+  end
+  "<root>#{people.join}</root>"
+end
+
+# Generate a large XML document
+def generate_large_xml(count = 1000)
+  people = (1..count).map do |i|
+    "<person id=\"#{i}\" name=\"Person#{i}\"><age>#{20 + (i % 50)}</age><city>City#{i % 20}</city></person>"
+  end
+  "<root>#{people.join}</root>"
+end
+
+LARGE_XML = generate_large_xml(1000)
+
+puts "=" * 80
+puts "XML Parsing Benchmarks"
+puts "=" * 80
+puts
+
+# Small XML parsing
+puts "Small XML Parsing (#{SMALL_XML.bytesize} bytes)"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { RXerces::XML::Document.parse(SMALL_XML) }
+  x.report("nokogiri") { Nokogiri::XML(SMALL_XML) } if NOKOGIRI_AVAILABLE
+  x.report("ox") { Ox.parse(SMALL_XML) } if OX_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Medium XML parsing
+puts "Medium XML Parsing (#{MEDIUM_XML.bytesize} bytes)"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { RXerces::XML::Document.parse(MEDIUM_XML) }
+  x.report("nokogiri") { Nokogiri::XML(MEDIUM_XML) } if NOKOGIRI_AVAILABLE
+  x.report("ox") { Ox.parse(MEDIUM_XML) } if OX_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Large XML parsing
+puts "Large XML Parsing (#{LARGE_XML.bytesize} bytes)"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { RXerces::XML::Document.parse(LARGE_XML) }
+  x.report("nokogiri") { Nokogiri::XML(LARGE_XML) } if NOKOGIRI_AVAILABLE
+  x.report("ox") { Ox.parse(LARGE_XML) } if OX_AVAILABLE
+
+  x.compare!
+end
+
+puts
+puts "=" * 80

data/benchmarks/run_all.rb

@@ -0,0 +1,25 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+# Run all benchmarks in sequence
+
+benchmarks = %w[
+  parse_benchmark.rb
+  xpath_benchmark.rb
+  css_benchmark.rb
+  traversal_benchmark.rb
+  serialization_benchmark.rb
+]
+
+puts "Running all RXerces benchmarks..."
+puts "=" * 80
+puts
+
+benchmarks.each do |benchmark|
+  puts "\nRunning #{benchmark}...\n\n"
+  system("ruby -Ilib benchmarks/#{benchmark}")
+  puts "\n"
+end
+
+puts "=" * 80
+puts "All benchmarks complete!"

data/benchmarks/serialization_benchmark.rb

@@ -0,0 +1,93 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+require 'benchmark/ips'
+require 'rxerces'
+
+begin
+  require 'nokogiri'
+  NOKOGIRI_AVAILABLE = true
+rescue LoadError
+  NOKOGIRI_AVAILABLE = false
+  puts "Nokogiri not available - install with: gem install nokogiri"
+end
+
+puts "=" * 80
+puts "Document Serialization Benchmarks"
+puts "=" * 80
+puts
+
+# Small document
+SMALL_XML = <<~XML
+  <root>
+    <person id="1" name="Alice">
+      <age>30</age>
+      <city>New York</city>
+    </person>
+  </root>
+XML
+
+rxerces_small = RXerces::XML::Document.parse(SMALL_XML)
+nokogiri_small = Nokogiri::XML(SMALL_XML) if NOKOGIRI_AVAILABLE
+
+puts "Small document to_s (#{SMALL_XML.bytesize} bytes)"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_small.to_s }
+  x.report("nokogiri") { nokogiri_small.to_xml } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Medium document
+def generate_xml(count)
+  people = (1..count).map do |i|
+    "<person id=\"#{i}\" name=\"Person#{i}\"><age>#{20 + (i % 50)}</age><city>City#{i % 20}</city></person>"
+  end
+  "<root>#{people.join}</root>"
+end
+
+MEDIUM_XML = generate_xml(100)
+
+rxerces_medium = RXerces::XML::Document.parse(MEDIUM_XML)
+nokogiri_medium = Nokogiri::XML(MEDIUM_XML) if NOKOGIRI_AVAILABLE
+
+puts "Medium document to_s (#{MEDIUM_XML.bytesize} bytes)"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_medium.to_s }
+  x.report("nokogiri") { nokogiri_medium.to_xml } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Large document
+LARGE_XML = generate_xml(1000)
+
+rxerces_large = RXerces::XML::Document.parse(LARGE_XML)
+nokogiri_large = Nokogiri::XML(LARGE_XML) if NOKOGIRI_AVAILABLE
+
+puts "Large document to_s (#{LARGE_XML.bytesize} bytes)"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_large.to_s }
+  x.report("nokogiri") { nokogiri_large.to_xml } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+puts "=" * 80

data/benchmarks/traversal_benchmark.rb

@@ -0,0 +1,149 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+require 'benchmark/ips'
+require 'rxerces'
+
+begin
+  require 'nokogiri'
+  NOKOGIRI_AVAILABLE = true
+rescue LoadError
+  NOKOGIRI_AVAILABLE = false
+  puts "Nokogiri not available - install with: gem install nokogiri"
+end
+
+# Sample XML for DOM traversal
+XML_DATA = begin
+  sections = (1..100).map do |i|
+    "<section id=\"s#{i}\">
+      <article id=\"a#{i}\">
+        <header>
+          <title>Title #{i}</title>
+          <author>Author #{i}</author>
+        </header>
+        <content>
+          <paragraph>Paragraph 1</paragraph>
+          <paragraph>Paragraph 2</paragraph>
+          <paragraph>Paragraph 3</paragraph>
+        </content>
+        <footer>
+          <date>2024-01-01</date>
+        </footer>
+      </article>
+    </section>"
+  end
+  "<root>#{sections.join}</root>"
+end
+
+puts "=" * 80
+puts "DOM Traversal Benchmarks"
+puts "=" * 80
+puts "Document Size: #{XML_DATA.bytesize} bytes"
+puts
+
+# Parse documents once
+rxerces_doc = RXerces::XML::Document.parse(XML_DATA)
+nokogiri_doc = Nokogiri::XML(XML_DATA) if NOKOGIRI_AVAILABLE
+
+rxerces_root = rxerces_doc.root
+nokogiri_root = nokogiri_doc.root if NOKOGIRI_AVAILABLE
+
+# Children access
+puts "Access .children"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_root.children }
+  x.report("nokogiri") { nokogiri_root.children } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Element children access
+puts "Access .element_children (elements only)"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_root.element_children }
+  x.report("nokogiri") { nokogiri_root.element_children } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Parent access
+puts "Access .parent on deep node"
+puts "-" * 80
+
+rxerces_deep = rxerces_doc.xpath('//paragraph').first
+nokogiri_deep = nokogiri_doc.xpath('//paragraph').first if NOKOGIRI_AVAILABLE
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_deep.parent }
+  x.report("nokogiri") { nokogiri_deep.parent } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Ancestors access
+puts "Access .ancestors on deep node"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_deep.ancestors }
+  x.report("nokogiri") { nokogiri_deep.ancestors } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Next sibling
+puts "Access .next_sibling"
+puts "-" * 80
+
+rxerces_para = rxerces_doc.xpath('//paragraph').first
+nokogiri_para = nokogiri_doc.xpath('//paragraph').first if NOKOGIRI_AVAILABLE
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_para.next_sibling }
+  x.report("nokogiri") { nokogiri_para.next_sibling } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Text extraction
+puts "Extract .text from element"
+puts "-" * 80
+
+rxerces_section = rxerces_doc.xpath('//section').first
+nokogiri_section = nokogiri_doc.xpath('//section').first if NOKOGIRI_AVAILABLE
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_section.text }
+  x.report("nokogiri") { nokogiri_section.text } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+puts "=" * 80

data/benchmarks/xpath_benchmark.rb

@@ -0,0 +1,100 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+require 'benchmark/ips'
+require 'rxerces'
+
+begin
+  require 'nokogiri'
+  NOKOGIRI_AVAILABLE = true
+rescue LoadError
+  NOKOGIRI_AVAILABLE = false
+  puts "Nokogiri not available - install with: gem install nokogiri"
+end
+
+# Sample XML for XPath queries
+XML_DATA = begin
+  books = (1..500).map do |i|
+    category = ['fiction', 'science', 'biography', 'history'][i % 4]
+    year = 1990 + (i % 30)
+    price = 10.0 + (i % 50)
+    "<book id=\"book#{i}\" category=\"#{category}\">
+      <title lang=\"en\">Title #{i}</title>
+      <author>Author #{i}</author>
+      <year>#{year}</year>
+      <price>#{'%.2f' % price}</price>
+    </book>"
+  end
+  "<catalog>#{books.join}</catalog>"
+end
+
+puts "=" * 80
+puts "XPath Query Benchmarks"
+puts "=" * 80
+puts "XML Size: #{XML_DATA.bytesize} bytes"
+puts
+
+# Parse documents once
+rxerces_doc = RXerces::XML::Document.parse(XML_DATA)
+nokogiri_doc = Nokogiri::XML(XML_DATA) if NOKOGIRI_AVAILABLE
+
+# Simple XPath queries
+puts "Simple XPath: //book"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_doc.xpath('//book') }
+  x.report("nokogiri") { nokogiri_doc.xpath('//book') } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Attribute-based XPath queries
+puts "Attribute XPath: //book[@category='fiction']"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_doc.xpath("//book[@category='fiction']") }
+  x.report("nokogiri") { nokogiri_doc.xpath("//book[@category='fiction']") } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# Complex XPath queries
+puts "Complex XPath: //book[year > 2000 and price < 30]/title"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_doc.xpath('//book[year > 2000 and price < 30]/title') }
+  x.report("nokogiri") { nokogiri_doc.xpath('//book[year > 2000 and price < 30]/title') } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+
+# at_xpath (first match)
+puts "at_xpath: //book (first match only)"
+puts "-" * 80
+
+Benchmark.ips do |x|
+  x.config(time: 5, warmup: 2)
+
+  x.report("rxerces") { rxerces_doc.root.at_xpath('//book') }
+  x.report("nokogiri") { nokogiri_doc.at_xpath('//book') } if NOKOGIRI_AVAILABLE
+
+  x.compare!
+end
+
+puts
+puts "=" * 80