leptris 1.9.220.0 → 1.9.221.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 4ee8a29e686e3a4079398135f42f0bfc4f9362d92238863ec93dba31b5875bc0
-  data.tar.gz: 390e5a0b87b57a77818207491d5cc96ea41941bdc04c6edd57fb5942f1d39c2e
+  metadata.gz: abec97f5dc76127e52554fe9254f7b9a93676c0583dde964e474a7e662884878
+  data.tar.gz: d2d30b60d8ad656c6ed3564e63e9ea7ecc006ebe17cdb026075dfd8407311dd8
 SHA512:
-  metadata.gz: 5c6f747575b65cb29526fa1ac86cfe9856684345444c0c29cbb4b9c1f3850e02bf938f21d86f604adeafcc2ebf14f76b7b028bf6fdbf15d955cb04217abe94dc
-  data.tar.gz: a871ecf4f747071d6b6aca705d0bd2ef037f63f2d009bc2d35dcdeefc8f269632e1a1c596296848e8cc78268e4aa6c913d0094140e0543f387505a8f8e3f4533
+  metadata.gz: fe37f35a8359f66544471ea70405b3895ce07acc64d9e92067fd26d9945b1c6ec2f2b82ca7a4b18628f59ca012c8347adbea24a173a8fc52bf725dacdd761e36
+  data.tar.gz: d2531467d1c319ff967ab0eb7b6cafc54beead04048d3d788300e8bc042cd1ce20784caf98d9e1646944d6db346cec66368bece7f1e1ace2adbf05d779039906

data/CHANGELOG.md

@@ -5,6 +5,14 @@ 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).
 
+## [1.9.221.0] - 2026-09-22
+
+### Performance — engine sync (1.9.221)
+
+- Vendored libleptris 1.9.221: per-document overflow teardown in
+  the dispatch row (#682). No binding-facing surface changes;
+  audit 342/342, suite 770/0.
+
 ## [1.9.220.0] - 2026-09-22
 
 ### Performance — engine sync (1.9.217–1.9.220)

data/README.adoc

@@ -504,10 +504,69 @@ Row kinds: `:scalar`, `:collection`, `:nested` (via `plan:`), `:raw`
 offset + `type_tag` echo, matching `Node#byte_offset`). Namespace
 binding lives on plans: `ns:` is `:none` (default), `:any`, or
 `{ exact: "urn:..." }`; `flags:` also accepts `:cdata`, `:ordered`,
-`:ns_lenient` (#754 out-of-namespace adoption). Children the plan
-does not describe are skipped; undescribed document order within a
-row is preserved. `#walk` returns a lazy `PlanValue` tree (the
-result outlives the document); `#to_ruby` materializes it.
+`:ns_lenient` (#754 out-of-namespace adoption), and `:order_spine`
+(#1273 — unmatched sibling text/comment/PI runs emit as positioned
+scalars so ordered hosts rebuild element order without re-parsing).
+Children the plan does not describe are skipped; undescribed document
+order within a row is preserved. `#walk` returns a lazy `PlanValue`
+tree (the result outlives the document); `#to_ruby` materializes it.
+
+=== Typed scalars and the fused entry (#230/#298)
+
+Plan rows accept `type: :integer | :float | :boolean` — the walk
+parses the values natively (in-pass, libleptris #1269a) and
+`#to_ruby` / the bulk face return `Integer` / `Float` / `true` /
+`false`. Unparseable values fall back to the raw String;
+`PlanValue#string_value` stays the raw escape, and
+`#int_value` / `#float_value` / `#bool_value` expose the parsed
+accessors directly.
+
+`Descriptor#materialize(source)` is the fused entry: source bytes →
+typed rows in ONE C call — parse, walk, and free happen natively
+(#1269b); the standalone `PlanValue` tree is all that surfaces.
+
+[source,ruby]
+----
+descriptor = Leptris::XML::Descriptor.build(
+  name: "iso",
+  children: [{ name: "row", kind: :collection, plan: {
+    name: "row",
+    attributes: [{ name: "id", kind: :scalar, type: :integer }],
+    children: [
+      { name: "price", kind: :scalar, type: :float },
+      { name: "active", kind: :scalar, type: :boolean },
+    ] } },
+  ])
+
+rows = descriptor.materialize(xml_source)   # one C call
+rows.as_kwarg_hash[:children]["row"]        # typed, hydrator-ready
+----
+
+=== Predicate rows (#1272) and order identity (#1273)
+
+Same-wire-name rows can partition on attribute values with `when:` —
+AND across pairs, exclusive routing (first matching row wins per
+occurrence):
+
+[source,ruby]
+----
+Leptris::XML::Descriptor.build(
+  name: "r",
+  children: [
+    { name: "item", kind: :collection, when: { "kind" => "a" },
+      plan: { name: "item", children: [{ name: "name", kind: :scalar }] } },
+    { name: "item", kind: :collection, when: { "kind" => "b" },
+      plan: { name: "item", children: [{ name: "name", kind: :scalar }] } },
+  ])
+----
+
+`PlanValue#node_kind` and `#order_index` carry document-order
+identity for matched rows (and, under `:order_spine`, for the
+interleaved text runs) — ordered/mixed content reconstructs
+without fragment re-parsing. `PlanValue#as_kwarg_hash` is the
+bulk hydrator face: one Ruby call per node returns the typed
+attribute/children hash (the crossings floor on the 5k-row ISO
+fixture is ~4 FFI crossings per row, pinned by spec).
 
 == XQuery
 

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 = "1.9.220"
+LIBLEPTRIS_VERSION = "1.9.221"
 # Vendored alongside libleptris for fn:normalize-unicode (TODO
 # .restructure/20): built per platform with a RELOCATABLE @rpath
 # install name, loaded by ffi.rb before libleptris so the

data/ext/build_windows_native.rb

@@ -22,8 +22,15 @@ root = File.expand_path("..", __dir__)
 ext_dir = File.join(root, "ext", "leptris", "native")
 minor = RUBY_VERSION[/\A\d+\.\d+/]
 minor_us = minor.tr(".", "_")
+INIT_PREFIX = "Init_native_"
 
 Dir.chdir(ext_dir) do
+  # aarch64-ucrt 1.9.221.0 shipped a DLL exporting only Init_native
+  # because make reused a stale native.obj from a failed earlier
+  # attempt (the log shows "creating Makefile -> linking" with no
+  # compile). Force a clean compile on every invocation.
+  FileUtils.rm_f(Dir.glob("#{ext_dir}/native.{o,obj,so,dll}"))
+  FileUtils.rm_f("#{ext_dir}/leptris/xml/native.so")
   system(RbConfig.ruby, "extconf.rb") or abort "extconf failed under #{RUBY_VERSION}"
   # mkmf's link binds the CURRENT Ruby's runtime DLL — exactly
   # what the versioned naming is for.
@@ -31,6 +38,18 @@ Dir.chdir(ext_dir) do
   abort "make failed under #{RUBY_VERSION}" unless success
   so = Dir.glob("native.{so,dll}").first
   abort "native bundle not produced under #{RUBY_VERSION}" unless so
+  # Functional export gate: dlopen the built DLL and resolve the
+  # MINOR-SUFFIXED init symbol. A PE DLL can carry the symbol in its
+  # string table while exporting only mkmf's Init_native (the arm64
+  # toolchain ignores our .def when its own wins the link) — only a
+  # real lookup proves native_<minor>.so will load.
+  require "fiddle"
+  begin
+    Fiddle.dlopen(File.expand_path(so))[INIT_PREFIX + minor_us]
+  rescue Fiddle::DLError => e
+    abort "#{so} does not export #{INIT_PREFIX + minor_us} — refusing to " \
+          "install a DLL that would 127 at require (#{e.message})"
+  end
   dest = File.join(root, "lib", "leptris", "xml", "native_#{minor_us}.so")
   # The artifact must reference only this minor's Ruby DLL.
   imported = `strings #{so} 2>/dev/null`[/[a-z0-9-]*ruby\d{3,}\.dll/i]

data/ext/leptris/native/native.c

@@ -2524,7 +2524,23 @@ static VALUE leptris_native_resolve_rb(VALUE self, VALUE lib_path)
     return Qnil;
 }
 
-void Init_native(void)
+/*
+ * Windows export discipline: a PE DLL exports only what a .def file or
+ * __declspec(dllexport) names. native_layer requires the MINOR-SUFFIXED
+ * symbol (native_3_4.so -> Init_native_3_4), but mkmf generates its own
+ * .def listing just Init_native, and depending on toolchain that def
+ * can win over ours (aarch64-ucrt ld does not auto-export the rest, so
+ * the lookup 127s and the platform silently degrades to the FFI
+ * surface). dllexport pins all four entry points into the export table
+ * on every Windows toolchain; it is a no-op elsewhere.
+ */
+#if defined(_WIN32)
+#  define LEPTRIS_INIT_EXPORT __declspec(dllexport)
+#else
+#  define LEPTRIS_INIT_EXPORT
+#endif
+
+LEPTRIS_INIT_EXPORT void Init_native(void)
 {
     VALUE m_leptris, m_xml, m_native;
 
@@ -2683,7 +2699,7 @@ void Init_native(void)
  * each needs its exact init name exported. All are defined here
  * unconditionally: a DLL only ever loads under the Ruby minor it
  * was built against (PE binds its build Ruby's runtime DLL). */
-void Init_native_3_3(void) { Init_native(); }
-void Init_native_3_4(void) { Init_native(); }
-void Init_native_4_0(void) { Init_native(); }
+LEPTRIS_INIT_EXPORT void Init_native_3_3(void) { Init_native(); }
+LEPTRIS_INIT_EXPORT void Init_native_3_4(void) { Init_native(); }
+LEPTRIS_INIT_EXPORT void Init_native_4_0(void) { Init_native(); }
 #endif

data/lib/leptris/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Leptris
-  VERSION = "1.9.220.0"
+  VERSION = "1.9.221.1"
 end

data/vendor-src/README.md

@@ -1,7 +1,7 @@
 Sources vendored in this gem (recompile rights; the ruby
 variant builds them at install via extconf.rb):
 
-- libleptris v1.9.220
+- libleptris v1.9.221
 - utf8proc v2.11.0
 
 Rebuild by hand:

data/vendor-src/libleptris/CHANGELOG.md

@@ -1,5 +1,13 @@
 ## [Unreleased]
 
+## [1.9.221] - 2026-09-22
+
+### Performance
+
+- per-document overflow teardown (#682 dispatch row) (dom)
+
+
+
 ## [1.9.220] - 2026-09-21
 
 ### Performance

data/vendor-src/libleptris/CMakeLists.txt

@@ -3,7 +3,7 @@
 cmake_minimum_required(VERSION 3.20)
 
 project(leptris
-    VERSION 1.9.220
+    VERSION 1.9.221
     DESCRIPTION "Fast XML parser and XPath evaluator in pure C"
     LANGUAGES C CXX
 )

data/vendor-src/libleptris/src/leptris/dom/compact.c

@@ -9,6 +9,7 @@
  */
 
 #include "compact.h"
+#include "../leptris_internal.h"   /* struct leptris_document (overflow_entries) */
 #include "../common/port.h"
 #include <stdlib.h>
 #include <string.h>
@@ -158,6 +159,14 @@ int leptris_compact_overflow_set(LeptrisCompactOverflowTable* table,
     table->buckets[index] = entry;
     table->entry_count++;
 
+    /* #682: parallel per-document chain — teardown walks exactly
+     * this doc's entries instead of scanning the whole table. */
+    if (doc) {
+        entry->doc_next =
+            (struct leptris_compact_overflow_entry*)doc->overflow_entries;
+        doc->overflow_entries = entry;
+    }
+
     if (table->entry_count >= table->bucket_count) {
         overflow_table_grow(table);
     }
@@ -208,26 +217,40 @@ void leptris_compact_cleanup(void) {
 void leptris_compact_cleanup_document(struct leptris_document* doc) {
     if (!doc || !g_overflow_table) return;
 
-    size_t removed_count = 0;
+    /* #682: walk THIS document's entry chain (linked at set time)
+     * instead of scanning every bucket of the global table — the
+     * old full-table sweep ran on every document free and dominated
+     * apply-heavy XSLT profiles (result docs register thousands of
+     * cross-block edge entries; the sampler pinned 35% of the
+     * dispatch bench in this loop). */
+    LeptrisCompactOverflowEntry* e =
+        (LeptrisCompactOverflowEntry*)doc->overflow_entries;
+    doc->overflow_entries = NULL;
 
-    for (size_t i = 0; i < g_overflow_table->bucket_count; i++) {
-        LeptrisCompactOverflowEntry** entry_ptr = &g_overflow_table->buckets[i];
-        LeptrisCompactOverflowEntry* entry = *entry_ptr;
-
-        while (entry) {
-            if (entry->doc == doc) {
-                /* Unlink only — entries live in SLABS owned by the
-                 * table; free()ing them corrupted the heap whenever
-                 * a document had overflow entries (common since the
-                 * #450 sibling-edge work routed far links here). */
-                *entry_ptr = entry->next;
-                entry = *entry_ptr;
-                removed_count++;
-            } else {
-                entry_ptr = &entry->next;
-                entry = entry->next;
+    size_t removed_count = 0;
+    while (e) {
+        LeptrisCompactOverflowEntry* next = e->doc_next;
+        if (e->doc == doc) {
+            /* Stale chain links (entry re-registered under another
+             * doc) are skipped — the ->doc check keeps them alive
+             * for their true owner. */
+            size_t index = hash_pointer(e->key,
+                                        g_overflow_table->bucket_count);
+            LeptrisCompactOverflowEntry** entry_ptr =
+                &g_overflow_table->buckets[index];
+            while (*entry_ptr) {
+                if (*entry_ptr == e) {
+                    /* Unlink only — entries live in SLABS owned by
+                     * the table; free()ing them corrupted the heap
+                     * (#450 sibling-edge follow-up). */
+                    *entry_ptr = e->next;
+                    removed_count++;
+                    break;
+                }
+                entry_ptr = &(*entry_ptr)->next;
             }
         }
+        e = next;
     }
 
     g_overflow_table->entry_count -= removed_count;

data/vendor-src/libleptris/src/leptris/dom/compact.h

@@ -60,6 +60,9 @@ typedef struct leptris_compact_overflow_entry {
     void* value;                  /* Actual pointer value */
     struct leptris_document* doc;   /* Document that owns this entry */
     struct leptris_compact_overflow_entry* next;
+    /* #682: per-document chain — teardown walks THIS doc's entries
+     * instead of scanning every bucket of the global table. */
+    struct leptris_compact_overflow_entry* doc_next;
 } LeptrisCompactOverflowEntry;
 
 /**

data/vendor-src/libleptris/vcpkg.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://raw.githubusercontent.com/microsoft/vcpkg/master/scripts/vcpkg.schema.json",
   "name": "leptris",
-  "version": "1.9.220",
+  "version": "1.9.221",
   "description": "High-performance XML parser and XPath 1.0 engine in C (libleptris)",
   "homepage": "https://github.com/leptris/leptris",
   "license": "MIT",

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: leptris
 version: !ruby/object:Gem::Version
-  version: 1.9.220.0
+  version: 1.9.221.1
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-09-21 00:00:00.000000000 Z
+date: 2026-09-22 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ffi