nokogiri 1.12.3 → 1.13.1

data/lib/nokogiri/xslt.rb

@@ -1,4 +1,6 @@
+# coding: utf-8
 # frozen_string_literal: true
+
 module Nokogiri
   class << self
     ###
@@ -8,7 +10,7 @@ module Nokogiri
     #
     #   xslt = Nokogiri::XSLT(File.read(ARGV[0]))
     #
-    def XSLT stylesheet, modules = {}
+    def XSLT(stylesheet, modules = {})
       XSLT.parse(stylesheet, modules)
     end
   end
@@ -33,22 +35,28 @@ module Nokogiri
         end
       end
 
-      ###
-      # Quote parameters in +params+ for stylesheet safety
+      # :call-seq:
+      #   quote_params(params) → Array
+      #
+      # Quote parameters in +params+ for stylesheet safety.
+      # See Nokogiri::XSLT::Stylesheet.transform for example usage.
+      #
+      # [Parameters]
+      # - +params+ (Hash, Array) XSLT parameters (key->value, or tuples of [key, value])
+      #
+      # [Returns] Array of string parameters, with quotes correctly escaped for use with XSLT::Stylesheet.transform
+      #
       def quote_params(params)
-        parray = (params.instance_of?(Hash) ? params.to_a.flatten : params).dup
-        parray.each_with_index do |v, i|
-          parray[i] = if i % 2 > 0
-            if v =~ /'/
-              "concat('#{v.gsub(/'/, %q{', "'", '})}')"
-            else
-              "'#{v}'"
-            end
+        params.flatten.each_slice(2).each_with_object([]) do |kv, quoted_params|
+          key, value = kv.map(&:to_s)
+          value = if /'/.match?(value)
+            "concat('#{value.gsub(/'/, %q{', "'", '})}')"
           else
-            v.to_s
+            "'#{value}'"
           end
+          quoted_params << key
+          quoted_params << value
         end
-        parray.flatten
       end
     end
   end

data/lib/nokogiri.rb

@@ -1,6 +1,5 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 # frozen_string_literal: true
-# Modify the PATH on windows so that the external DLLs will get loaded.
 
 require "rbconfig"
 
@@ -19,30 +18,32 @@ require_relative "nokogiri/extension"
 #
 # Here is an example:
 #
-#   require 'nokogiri'
-#   require 'open-uri'
+#     require 'nokogiri'
+#     require 'open-uri'
 #
-#   # Get a Nokogiri::HTML4::Document for the page we’re interested in...
+#     # Get a Nokogiri::HTML4::Document for the page we’re interested in...
 #
-#   doc = Nokogiri::HTML4(URI.open('http://www.google.com/search?q=tenderlove'))
+#     doc = Nokogiri::HTML4(URI.open('http://www.google.com/search?q=tenderlove'))
 #
-#   # Do funky things with it using Nokogiri::XML::Node methods...
+#     # Do funky things with it using Nokogiri::XML::Node methods...
 #
-#   ####
-#   # Search for nodes by css
-#   doc.css('h3.r a.l').each do |link|
-#     puts link.content
-#   end
+#     ####
+#     # Search for nodes by css
+#     doc.css('h3.r a.l').each do |link|
+#       puts link.content
+#     end
 #
-# See Nokogiri::XML::Searchable#css for more information about CSS searching.
-# See Nokogiri::XML::Searchable#xpath for more information about XPath searching.
+# See also:
+#
+# - Nokogiri::XML::Searchable#css for more information about CSS searching
+# - Nokogiri::XML::Searchable#xpath for more information about XPath searching
 module Nokogiri
   class << self
     ###
     # Parse an HTML or XML document.  +string+ contains the document.
     def parse(string, url = nil, encoding = nil, options = nil)
       if string.respond_to?(:read) ||
-          /^\s*<(?:!DOCTYPE\s+)?html[\s>]/i === string[0, 512]
+          /^\s*<(?:!DOCTYPE\s+)?html[\s>]/i.match?(string[0, 512])
         # Expect an HTML indicator to appear within the first 512
         # characters of a document. (<?xml ?> + <?xml-stylesheet ?>
         # shouldn't be that long)
@@ -85,6 +86,7 @@ module Nokogiri
       Nokogiri(*args, &block).slop!
     end
 
+    # :nodoc:
     def install_default_aliases
       # Make sure to support some popular encoding aliases not known by
       # all iconv implementations.
@@ -105,7 +107,7 @@ end
 #
 # To specify the type of document, use {Nokogiri.XML}, {Nokogiri.HTML4}, or {Nokogiri.HTML5}.
 def Nokogiri(*args, &block)
-  if block_given?
+  if block
     Nokogiri::HTML4::Builder.new(&block).doc.root
   else
     Nokogiri.parse(*args)
@@ -113,6 +115,7 @@ def Nokogiri(*args, &block)
 end
 
 require_relative "nokogiri/version"
+require_relative "nokogiri/class_resolver"
 require_relative "nokogiri/syntax_error"
 require_relative "nokogiri/xml"
 require_relative "nokogiri/xslt"

data/lib/xsd/xmlparser/nokogiri.rb

@@ -1,8 +1,9 @@
 # frozen_string_literal: true
-require 'nokogiri'
 
-module XSD # :nodoc:
-  module XMLParser # :nodoc:
+require "nokogiri"
+
+module XSD
+  module XMLParser
     ###
     # Nokogiri XML parser for soap4r.
     #
@@ -27,40 +28,40 @@ module XSD # :nodoc:
     class Nokogiri < XSD::XMLParser::Parser
       ###
       # Create a new XSD parser with +host+ and +opt+
-      def initialize host, opt = {}
+      def initialize(host, opt = {})
         super
-        @parser = ::Nokogiri::XML::SAX::Parser.new(self, @charset || 'UTF-8')
+        @parser = ::Nokogiri::XML::SAX::Parser.new(self, @charset || "UTF-8")
       end
 
       ###
       # Start parsing +string_or_readable+
-      def do_parse string_or_readable
+      def do_parse(string_or_readable)
         @parser.parse(string_or_readable)
       end
 
       ###
       # Handle the start_element event with +name+ and +attrs+
-      def start_element name, attrs = []
+      def start_element(name, attrs = [])
         super(name, Hash[*attrs.flatten])
       end
 
       ###
       # Handle the end_element event with +name+
-      def end_element name
+      def end_element(name)
         super
       end
 
       ###
       # Handle errors with message +msg+
-      def error msg
-        raise ParseError.new(msg)
+      def error(msg)
+        raise ParseError, msg
       end
-      alias :warning :error
+      alias_method :warning, :error
 
       ###
       # Handle cdata_blocks containing +string+
-      def cdata_block string
-        characters string
+      def cdata_block(string)
+        characters(string)
       end
 
       ###
@@ -70,16 +71,16 @@ module XSD # :nodoc:
       # +prefix+ is the namespace prefix for the element
       # +uri+ is the associated namespace URI
       # +ns+ is a hash of namespace prefix:urls associated with the element
-      def start_element_namespace name, attrs = [], prefix = nil, uri = nil, ns = []
+      def start_element_namespace(name, attrs = [], prefix = nil, uri = nil, ns = [])
         ###
         # Deal with SAX v1 interface
-        name = [prefix, name].compact.join(':')
-        attributes = ns.map { |ns_prefix,ns_uri|
-          [['xmlns', ns_prefix].compact.join(':'), ns_uri]
-        } + attrs.map { |attr|
-          [[attr.prefix, attr.localname].compact.join(':'), attr.value]
-        }.flatten
-        start_element name, attributes
+        name = [prefix, name].compact.join(":")
+        attributes = ns.map do |ns_prefix, ns_uri|
+          [["xmlns", ns_prefix].compact.join(":"), ns_uri]
+        end + attrs.map do |attr|
+          [[attr.prefix, attr.localname].compact.join(":"), attr.value]
+        end.flatten
+        start_element(name, attributes)
       end
 
       ###
@@ -87,13 +88,13 @@ module XSD # :nodoc:
       # +name+ is the element's name
       # +prefix+ is the namespace prefix associated with the element
       # +uri+ is the associated namespace URI
-      def end_element_namespace name, prefix = nil, uri = nil
+      def end_element_namespace(name, prefix = nil, uri = nil)
         ###
         # Deal with SAX v1 interface
-        end_element [prefix, name].compact.join(':')
+        end_element([prefix, name].compact.join(":"))
       end
 
-      %w{ xmldecl start_document end_document comment }.each do |name|
+      ["xmldecl", "start_document", "end_document", "comment"].each do |name|
         class_eval %{ def #{name}(*args); end }
       end
 

data/patches/libxml2/0008-htmlParseComment-handle-abruptly-closed-comments.patch

@@ -0,0 +1,61 @@
+From 3ea8d08da310b645e37940eaae5cc28e251b155b Mon Sep 17 00:00:00 2001
+From: Mike Dalessio <mike.dalessio@gmail.com>
+Date: Sat, 17 Jul 2021 14:36:53 -0400
+Subject: [PATCH] htmlParseComment: handle abruptly-closed comments
+
+See guidance provided on abrutply-closed comments here:
+
+https://html.spec.whatwg.org/multipage/parsing.html#parse-error-abrupt-closing-of-empty-comment
+---
+ HTMLparser.c              | 11 +++++++++++
+ include/libxml/xmlerror.h |  1 +
+ 2 files changed, 12 insertions(+)
+
+diff --git a/HTMLparser.c b/HTMLparser.c
+index b56363a..f0bf294 100644
+--- a/HTMLparser.c
++++ b/HTMLparser.c
+@@ -3485,10 +3485,20 @@ htmlParseComment(htmlParserCtxtPtr ctxt) {
+     q = CUR_CHAR(ql);
+     if (q == 0)
+         goto unfinished;
++    if (q == '>') {
++        htmlParseErr(ctxt, XML_ERR_COMMENT_ABRUPTLY_ENDED, "Comment abruptly ended", NULL, NULL);
++        cur = '>';
++        goto finished;
++    }
+     NEXTL(ql);
+     r = CUR_CHAR(rl);
+     if (r == 0)
+         goto unfinished;
++    if (q == '-' && r == '>') {
++        htmlParseErr(ctxt, XML_ERR_COMMENT_ABRUPTLY_ENDED, "Comment abruptly ended", NULL, NULL);
++        cur = '>';
++        goto finished;
++    }
+     NEXTL(rl);
+     cur = CUR_CHAR(l);
+     while ((cur != 0) &&
+@@ -3536,6 +3546,7 @@ htmlParseComment(htmlParserCtxtPtr ctxt) {
+ 	cur = next;
+ 	l = nl;
+     }
++finished:
+     buf[len] = 0;
+     if (cur == '>') {
+         NEXT;
+diff --git a/include/libxml/xmlerror.h b/include/libxml/xmlerror.h
+index c101997..7b68e40 100644
+--- a/include/libxml/xmlerror.h
++++ b/include/libxml/xmlerror.h
+@@ -209,6 +209,7 @@ typedef enum {
+     XML_ERR_VERSION_MISMATCH, /* 109 */
+     XML_ERR_NAME_TOO_LONG, /* 110 */
+     XML_ERR_USER_STOP, /* 111 */
++    XML_ERR_COMMENT_ABRUPTLY_ENDED, /* 112 */
+     XML_NS_ERR_XML_NAMESPACE = 200,
+     XML_NS_ERR_UNDEFINED_NAMESPACE, /* 201 */
+     XML_NS_ERR_QNAME, /* 202 */
+-- 
+2.31.0
+

data/patches/libxml2/0009-allow-wildcard-namespaces.patch

@@ -0,0 +1,77 @@
+From 74c95ec5932c737d4fcb06b8646b0017364ada14 Mon Sep 17 00:00:00 2001
+From: Mike Dalessio <mike.dalessio@gmail.com>
+Date: Fri, 24 Dec 2021 19:08:01 -0500
+Subject: [PATCH] attempt to hack in wildcard namespaces to xpath
+
+I'm not confident this is a bulletproof patch.
+---
+ xpath.c | 24 ++++++++++++++++++------
+ 1 file changed, 18 insertions(+), 6 deletions(-)
+
+diff --git a/xpath.c b/xpath.c
+index 1aa2f1a..c7f0885 100644
+--- a/xpath.c
++++ b/xpath.c
+@@ -146,6 +146,9 @@
+ #define XPATH_MAX_RECURSION_DEPTH 5000
+ #endif
+ 
++#define WILDCARD_PREFIX "*"
++#define IS_WILDCARD_PREFIX(p) xmlStrEqual((xmlChar*)WILDCARD_PREFIX, p)
++
+ /*
+  * TODO:
+  * There are a few spots where some tests are done which depend upon ascii
+@@ -11073,12 +11076,15 @@ xmlXPathCompNodeTest(xmlXPathParserContextPtr ctxt, xmlXPathTestVal *test,
+     SKIP_BLANKS;
+ 
+     if ((name == NULL) && (CUR == '*')) {
+-	/*
+-	 * All elements
+-	 */
+ 	NEXT;
+-	*test = NODE_TEST_ALL;
+-	return(NULL);
++	if (CUR != ':') {
++	    /*
++	     * All elements
++	     */
++	    *test = NODE_TEST_ALL;
++	    return(NULL);
++	}
++	name = xmlCharStrdup(WILDCARD_PREFIX);
+     }
+ 
+     if (name == NULL)
+@@ -11327,6 +11333,10 @@ xmlXPathCompStep(xmlXPathParserContextPtr ctxt) {
+ 	}
+ #endif
+ 	if (CUR == '*') {
++	    if (NXT(1) == ':') {
++		NEXT;
++		name = xmlCharStrdup(WILDCARD_PREFIX);
++	    }
+ 	    axis = AXIS_CHILD;
+ 	} else {
+ 	    if (name == NULL)
+@@ -12030,7 +12040,7 @@ xmlXPathNodeCollectAndTest(xmlXPathParserContextPtr ctxt,
+     /*
+     * Setup namespaces.
+     */
+-    if (prefix != NULL) {
++    if (prefix != NULL && !IS_WILDCARD_PREFIX(prefix)) {
+         URI = xmlXPathNsLookup(xpctxt, prefix);
+         if (URI == NULL) {
+ 	    xmlXPathReleaseObject(xpctxt, obj);
+@@ -12369,6 +12379,8 @@ xmlXPathNodeCollectAndTest(xmlXPathParserContextPtr ctxt,
+ 				    {
+ 					XP_TEST_HIT
+                                     }
++                                } else if (IS_WILDCARD_PREFIX(prefix)) {
++                                    XP_TEST_HIT
+                                 } else {
+                                     if ((cur->ns != NULL) &&
+                                         (xmlStrEqual(URI, cur->ns->href)))
+-- 
+2.31.0
+

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: nokogiri
 version: !ruby/object:Gem::Version
-  version: 1.12.3
+  version: 1.13.1
 platform: ruby
 authors:
 - Mike Dalessio
@@ -20,36 +20,36 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2021-08-10 00:00:00.000000000 Z
+date: 2022-01-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
-  name: racc
+  name: mini_portile2
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.4'
+        version: 2.7.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.4'
+        version: 2.7.0
 - !ruby/object:Gem::Dependency
-  name: mini_portile2
+  name: racc
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 2.6.1
+        version: '1.4'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 2.6.1
+        version: '1.4'
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
@@ -84,14 +84,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '5.8'
+        version: '5.15'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '5.8'
+        version: '5.15'
 - !ruby/object:Gem::Dependency
   name: minitest-reporters
   requirement: !ruby/object:Gem::Requirement
@@ -122,88 +122,158 @@ dependencies:
         version: '13.0'
 - !ruby/object:Gem::Dependency
   name: rake-compiler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '='
+      - !ruby/object:Gem::Version
+        version: 1.1.7
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '='
+      - !ruby/object:Gem::Version
+        version: 1.1.7
+- !ruby/object:Gem::Dependency
+  name: rake-compiler-dock
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.1'
+        version: '1.2'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.1'
+        version: '1.2'
 - !ruby/object:Gem::Dependency
-  name: rake-compiler-dock
+  name: rdoc
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.1'
+        version: '6.3'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.1'
+        version: '6.3'
 - !ruby/object:Gem::Dependency
   name: rexical
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 1.0.5
+        version: 1.0.7
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 1.0.5
+        version: 1.0.7
 - !ruby/object:Gem::Dependency
   name: rubocop
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.7'
+        version: '1.23'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.7'
+        version: '1.23'
 - !ruby/object:Gem::Dependency
-  name: simplecov
+  name: rubocop-minitest
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.17'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.17'
+- !ruby/object:Gem::Dependency
+  name: rubocop-performance
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.20'
+        version: '1.12'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.20'
+        version: '1.12'
 - !ruby/object:Gem::Dependency
-  name: yard
+  name: rubocop-rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.6'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.6'
+- !ruby/object:Gem::Dependency
+  name: rubocop-shopify
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.3'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.3'
+- !ruby/object:Gem::Dependency
+  name: ruby_memcheck
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+- !ruby/object:Gem::Dependency
+  name: simplecov
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.9'
+        version: '0.21'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.9'
+        version: '0.21'
 description: |
   Nokogiri (鋸) makes it easy and painless to work with XML and HTML from Ruby. It provides a
   sensible, easy-to-understand API for reading, writing, modifying, and querying documents. It is
@@ -341,6 +411,7 @@ files:
 - gumbo-parser/src/vector.c
 - gumbo-parser/src/vector.h
 - lib/nokogiri.rb
+- lib/nokogiri/class_resolver.rb
 - lib/nokogiri/css.rb
 - lib/nokogiri/css/node.rb
 - lib/nokogiri/css/parser.rb
@@ -420,6 +491,8 @@ files:
 - patches/libxml2/0005-avoid-isnan-isinf.patch
 - patches/libxml2/0006-update-automake-files-for-arm64.patch
 - patches/libxml2/0007-Fix-XPath-recursion-limit.patch
+- patches/libxml2/0008-htmlParseComment-handle-abruptly-closed-comments.patch
+- patches/libxml2/0009-allow-wildcard-namespaces.patch
 - patches/libxslt/0001-update-automake-files-for-arm64.patch
 - patches/libxslt/0002-Fix-xml2-config-check-in-configure-script.patch
 - ports/archives/libxml2-2.9.12.tar.gz
@@ -433,6 +506,7 @@ metadata:
   documentation_uri: https://nokogiri.org/rdoc/index.html
   changelog_uri: https://nokogiri.org/CHANGELOG.html
   source_code_uri: https://github.com/sparklemotion/nokogiri
+  rubygems_mfa_required: 'true'
 post_install_message: 
 rdoc_options:
 - "--main"
@@ -443,14 +517,14 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 2.5.0
+      version: 2.6.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.2.15
+rubygems_version: 3.3.3
 signing_key: 
 specification_version: 4
 summary: Nokogiri (鋸) makes it easy and painless to work with XML and HTML from Ruby.