leptris 1.9.36-x86_64-linux → 1.9.37-x86_64-linux

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: e4be54118ce35832b98391483d01500bd26bb78627f6f22b8c7f2aafd9fd5828
-  data.tar.gz: 35b928a020f94c23c8fb53b7c44836fc91fc0edca78e842ff5d128b0d5668d79
+  metadata.gz: 87d2effb07e2fb9f93d586b12433c01ce390c04303d1ef061bd8c7c1f7b70c3f
+  data.tar.gz: ebc3a9b7a9acb3f477088e8922f87e1387857f0cd80970f93c3e5b462503d7ae
 SHA512:
-  metadata.gz: ec5af59d3374938a0646ddafd3849b0387cc59d4b2c1c4f85efd159cfd3b120027b0f28f3287110901046919c83b70b08705f0620c358a37ec16971a182b5815
-  data.tar.gz: 25e80c6c256cc6f57ba49be772728bb72d335d57e2c04d85d7ecc03b6fd1cfd6dcfea8abe47ae3ca552039011732a314e9bc931077fc0c20ab8594c88c46dc67
+  metadata.gz: f8f517db0f999dfade73cdaedee26f0b0ba52fc963010232d72a02b0bb73202c0337b1c9213f9edb4dd7652492542066ecfe90fb85d5eb9c535a90a7445ed9e2
+  data.tar.gz: cd2083d13f71dbd42ebf8c3c6326e8ab98f51439ee3b4b89415eda00cf3e2b09cbff6f1e67ca6f446ca71abb9f37e5c918d32d93ab21d09ad40f85047e0383a9

data/CHANGELOG.md

@@ -5,6 +5,35 @@ All notable changes to Leptris will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [1.9.37] - 2026-08-29
+
+### Added
+
+- **`Element#inner_html`** (Nokogiri parity): the serialized
+  children — elements through the engine serializer, text
+  XML-escaped at the seam (`& < >` and CR → `&#xD;`, libxml2's
+  rules), comments/CDATA/PIs in literal forms. Well-formed by
+  construction: a spec re-parses the output to the same children.
+  (Nokogiri's inner_html HTML-serializes XML documents — SGML-style
+  PI closes, bare CDATA content — and its output does not
+  re-parse.) Measured 21 µs/element vs Nokogiri 15 — a first cut
+  on a new API with correct forms as the contract.
+- **libleptris 1.9.12 lockstep** (with 1.9.11): pure XSLT
+  conformance upstream (libxslt suite 152 -> 180/205 — template
+  priority, xsl:number, attribute sets, namespace-declaration
+  ordering, strip/preserve-space, copy semantics); audit 250/250,
+  no new C surface.
+
+### Changed
+
+- **Serialize byte-scratch**: the size+fill serialization pair
+  allocated and freed a MemoryPointer per call — inner_html
+  serializes each child, and the allocation dominated per-child
+  cost. A grow-only thread-local byte scratch (the pointer-scratch
+  discipline) now serves every serialize call. Serialized output
+  is also forced UTF-8 at the seam (previously true only by
+  accident of ASCII-only content comparisons).
+
 ## [1.9.36] - 2026-08-29
 
 ### Fixed

data/Rakefile

@@ -8,7 +8,7 @@ RSpec::Core::RakeTask.new(:spec)
 # Pin for `rake compile` and the platform-gem builds. Keep in lockstep
 # with .github/workflows/build.yml (which calls `rake compile`) and the
 # CHANGELOG when libleptris releases.
-LIBLEPTRIS_VERSION = "1.9.10"
+LIBLEPTRIS_VERSION = "1.9.12"
 
 CMAKE_FLAGS = %w[
   -DCMAKE_BUILD_TYPE=Release

data/lib/leptris/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Leptris
-  VERSION = "1.9.36"
+  VERSION = "1.9.37"
 end

data/lib/leptris/xml/element.rb

@@ -336,6 +336,28 @@ class Leptris::XML::Element < Leptris::XML::Node
     scopes
   end
 
+  # The serialized children (Nokogiri parity): elements serialize
+  # through the engine (correct escaping), text is XML-escaped,
+  # comments/CDATA/PIs render their literal forms. The complement
+  # of #inner_text, which returns the unescaped text content.
+  def inner_html
+    children.map do |child|
+      case child
+      when Leptris::XML::Element then child.to_xml
+      when Leptris::XML::CDATA # before Text: CDATA subclasses it
+        "<![CDATA[#{child.content}]]>"
+      when Leptris::XML::Text
+        Leptris::XML::Serialization.escape_text(child.content)
+      when Leptris::XML::Comment then "<!--#{child.content}-->"
+      when Leptris::XML::ProcessingInstruction
+        data = child.content
+        data.empty? ? "<?#{child.name}?>" : "<?#{child.name} #{data}?>"
+      else
+        Leptris::XML::Serialization.escape_text(child.content.to_s)
+      end
+    end.join
+  end
+
   def to_xml(indent: 0, no_decl: false, encoding: nil)
     Leptris::XML::Serialization.to_xml(
       Leptris::XML::FFI.method(:leptris_element_serialize_into), @c_ptr,

data/lib/leptris/xml/ffi.rb

@@ -865,13 +865,24 @@ module Leptris
       def self.serialize_into_string(ffi_function, c_ptr, options)
         need = ffi_function.call(c_ptr, nil, 0, nil, options)
         return "" if need.zero?
-        buffer = ::FFI::MemoryPointer.new(:char, need)
-        begin
-          ffi_function.call(c_ptr, buffer, need, nil, options)
-          buffer.read_string
-        ensure
-          buffer.free
+        buffer = scratch_bytes(need)
+        ffi_function.call(c_ptr, buffer, need, nil, options)
+        buffer.read_string.force_encoding(Encoding::UTF_8)
+      end
+
+      # Byte scratch for serialize-into cycles: the size+fill pair
+      # allocated and freed a buffer per call — inner_html
+      # serializes each child, and the allocation was most of the
+      # per-child cost. Grows to the largest serialization seen;
+      # no Ruby reentrancy during a serialize call.
+      def self.scratch_bytes(need)
+        scratch = (Thread.current[:leptris_scratch] ||= {})
+        ptr = scratch[:bytes]
+        if ptr.nil? || ptr.size < need
+          ptr&.free
+          ptr = scratch[:bytes] = ::FFI::MemoryPointer.new(:char, need)
         end
+        ptr
       end
 
       # All-kind child handle batch (libleptris 1.7.0). Opportunistic

data/lib/leptris/xml/serialization.rb

@@ -62,6 +62,19 @@ module Leptris::XML::Serialization
     Leptris::XML::FFI.read_owned_string(str_ptr)
   end
 
+  # XML-escapes text content for inner_html: & first, then < >
+  # and CR (libxml2 emits &#xD; for a bare carriage return).
+  ESCAPE_TEXT = {
+    "&" => "&amp;", "<" => "&lt;", ">" => "&gt;", "\r" => "&#xD;",
+  }.freeze
+  private_constant :ESCAPE_TEXT
+  ESCAPE_TEXT_RE = /[&<>\r]/.freeze
+  private_constant :ESCAPE_TEXT_RE
+
+  def self.escape_text(string)
+    string.gsub(ESCAPE_TEXT_RE, ESCAPE_TEXT)
+  end
+
   # Builds a SerializeOptions struct. Returns [opts, encoding_anchor];
   # the anchor must stay referenced while opts is in an FFI call.
   def self.build_options(indent: 0, no_decl: false, encoding: nil)

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: leptris
 version: !ruby/object:Gem::Version
-  version: 1.9.36
+  version: 1.9.37
 platform: x86_64-linux
 authors:
 - Ribose Inc.
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-08-28 00:00:00.000000000 Z
+date: 2026-08-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ffi