nokogiri 1.10.9 → 1.18.3

data/lib/nokogiri/xslt/stylesheet.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Nokogiri
   module XSLT
     ###
@@ -8,16 +10,38 @@ module Nokogiri
     #   doc   = Nokogiri::XML(File.read('some_file.xml'))
     #   xslt  = Nokogiri::XSLT(File.read('some_transformer.xslt'))
     #
-    #   puts xslt.transform(doc)
+    #   xslt.transform(doc) # => Nokogiri::XML::Document
     #
-    # See Nokogiri::XSLT::Stylesheet#transform for more transformation
-    # information.
+    # Many XSLT transformations include serialization behavior to emit a non-XML document. For these
+    # cases, please take care to invoke the #serialize method on the result of the transformation:
+    #
+    #   doc   = Nokogiri::XML(File.read('some_file.xml'))
+    #   xslt  = Nokogiri::XSLT(File.read('some_transformer.xslt'))
+    #   xslt.serialize(xslt.transform(doc)) # => String
+    #
+    # or use the #apply_to method, which is a shortcut for `serialize(transform(document))`:
+    #
+    #   doc   = Nokogiri::XML(File.read('some_file.xml'))
+    #   xslt  = Nokogiri::XSLT(File.read('some_transformer.xslt'))
+    #   xslt.apply_to(doc) # => String
+    #
+    # See Nokogiri::XSLT::Stylesheet#transform for more information and examples.
     class Stylesheet
-      ###
-      # Apply an XSLT stylesheet to an XML::Document.
-      # +params+ is an array of strings used as XSLT parameters.
-      # returns serialized document
-      def apply_to document, params = []
+      # :call-seq:
+      #   apply_to(document, params = []) -> String
+      #
+      # Apply an XSLT stylesheet to an XML::Document and serialize it properly. This method is
+      # equivalent to calling #serialize on the result of #transform.
+      #
+      # [Parameters]
+      # - +document+ is an instance of XML::Document to transform
+      # - +params+ is an array of strings used as XSLT parameters, passed into #transform
+      #
+      # [Returns]
+      #   A string containing the serialized result of the transformation.
+      #
+      # See Nokogiri::XSLT::Stylesheet#transform for more information and examples.
+      def apply_to(document, params = [])
         serialize(transform(document, params))
       end
     end

data/lib/nokogiri/xslt.rb

@@ -1,16 +1,11 @@
-require 'nokogiri/xslt/stylesheet'
+# coding: utf-8
+# frozen_string_literal: true
 
 module Nokogiri
   class << self
-    ###
-    # Create a Nokogiri::XSLT::Stylesheet with +stylesheet+.
-    #
-    # Example:
-    #
-    #   xslt = Nokogiri::XSLT(File.read(ARGV[0]))
-    #
-    def XSLT stylesheet, modules = {}
-      XSLT.parse(stylesheet, modules)
+    # Convenience method for Nokogiri::XSLT.parse
+    def XSLT(...)
+      XSLT.parse(...)
     end
   end
 
@@ -19,38 +14,116 @@ module Nokogiri
   # Stylesheet object.
   module XSLT
     class << self
-      ###
-      # Parse the stylesheet in +string+, register any +modules+
-      def parse string, modules = {}
+      # :call-seq:
+      #   parse(xsl) → Nokogiri::XSLT::Stylesheet
+      #   parse(xsl, modules) → Nokogiri::XSLT::Stylesheet
+      #
+      # Parse the stylesheet in +xsl+, registering optional +modules+ as custom class handlers.
+      #
+      # [Parameters]
+      # - +xsl+ (String) XSL content to be parsed into a stylesheet
+      # - +modules+ (Hash<String ⇒ Class>) A hash of URI-to-handler relations for linking a
+      #   namespace to a custom function handler.
+      #
+      # ⚠ The XSLT handler classes are registered *globally*.
+      #
+      # Also see Nokogiri::XSLT.register
+      #
+      # *Example*
+      #
+      #   xml = Nokogiri.XML(<<~XML)
+      #     <nodes>
+      #       <node>Foo</node>
+      #       <node>Bar</node>
+      #     </nodes>
+      #   XML
+      #
+      #   handler = Class.new do
+      #     def reverse(node)
+      #       node.text.reverse
+      #     end
+      #   end
+      #
+      #   xsl = <<~XSL
+      #     <xsl:stylesheet version="1.0"
+      #       xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+      #       xmlns:myfuncs="http://nokogiri.org/xslt/myfuncs"
+      #       extension-element-prefixes="myfuncs">
+      #       <xsl:template match="/">
+      #         <reversed>
+      #           <xsl:for-each select="nodes/node">
+      #             <reverse><xsl:copy-of select="myfuncs:reverse(.)"/></reverse>
+      #           </xsl:for-each>
+      #         </reversed>
+      #       </xsl:template>
+      #     </xsl:stylesheet>
+      #   XSL
+      #
+      #   xsl = Nokogiri.XSLT(xsl, "http://nokogiri.org/xslt/myfuncs" => handler)
+      #   xsl.transform(xml).to_xml
+      #   # => "<?xml version=\"1.0\"?>\n" +
+      #   #    "<reversed>\n" +
+      #   #    "  <reverse>ooF</reverse>\n" +
+      #   #    "  <reverse>raB</reverse>\n" +
+      #   #    "</reversed>\n"
+      #
+      def parse(string, modules = {})
         modules.each do |url, klass|
-          XSLT.register url, klass
+          XSLT.register(url, klass)
         end
 
+        doc = XML::Document.parse(string, nil, nil, XML::ParseOptions::DEFAULT_XSLT)
         if Nokogiri.jruby?
-          Stylesheet.parse_stylesheet_doc(XML.parse(string), string)
+          Stylesheet.parse_stylesheet_doc(doc, string)
         else
-          Stylesheet.parse_stylesheet_doc(XML.parse(string))
+          Stylesheet.parse_stylesheet_doc(doc)
         end
       end
 
-      ###
-      # Quote parameters in +params+ for stylesheet safety
-      def quote_params params
-        parray = (params.instance_of?(Hash) ? params.to_a.flatten : params).dup
-        parray.each_with_index do |v,i|
-          if i % 2 > 0
-            parray[i]=
-              if v =~ /'/
-                "concat('#{ v.gsub(/'/, %q{', "'", '}) }')"
-              else
-                "'#{v}'";
-              end
+      # :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)
+        params.flatten.each_slice(2).with_object([]) do |kv, quoted_params|
+          key, value = kv.map(&:to_s)
+          value = if value.include?("'")
+            "concat('#{value.gsub("'", %q{', "'", '})}')"
           else
-            parray[i] = v.to_s
+            "'#{value}'"
           end
+          quoted_params << key
+          quoted_params << value
         end
-        parray.flatten
       end
+
+      #  call-seq:
+      #    register(uri, custom_handler_class)
+      #
+      #  Register a class that implements custom XSLT transformation functions.
+      #
+      #  ⚠ The XSLT handler classes are registered *globally*.
+      #
+      #  [Parameters}
+      #  - +uri+ (String) The namespace for the custom handlers
+      #  - +custom_handler_class+ (Class) A class with ruby methods that can be called during
+      #    transformation
+      #
+      #  See Nokogiri::XSLT.parse for usage.
+      #
+      def register(uri, custom_handler_class)
+        # NOTE: this is implemented in the C extension, see ext/nokogiri/xslt_stylesheet.c
+        raise NotImplementedError, "Nokogiri::XSLT.register is not implemented on JRuby"
+      end if Nokogiri.jruby?
     end
   end
 end
+
+require_relative "xslt/stylesheet"

data/lib/nokogiri.rb

@@ -1,96 +1,73 @@
-# -*- coding: utf-8 -*-
-# Modify the PATH on windows so that the external DLLs will get loaded.
-
-require 'rbconfig'
+# coding: utf-8
+# frozen_string_literal: true
 
 if defined?(RUBY_ENGINE) && RUBY_ENGINE == "jruby"
-  # The line below caused a problem on non-GAE rack environment.
-  # unless defined?(JRuby::Rack::VERSION) || defined?(AppEngine::ApiProxy)
-  #
-  # However, simply cutting defined?(JRuby::Rack::VERSION) off resulted in
-  # an unable-to-load-nokogiri problem. Thus, now, Nokogiri checks the presense
-  # of appengine-rack.jar in $LOAD_PATH. If Nokogiri is on GAE, Nokogiri
-  # should skip loading xml jars. This is because those are in WEB-INF/lib and
-  # already set in the classpath.
-  unless $LOAD_PATH.to_s.include?("appengine-rack")
-    require 'stringio'
-    require 'isorelax.jar'
-    require 'jing.jar'
-    require 'nekohtml.jar'
-    require 'nekodtd.jar'
-    require 'xercesImpl.jar'
-    require 'serializer.jar'
-    require 'xalan.jar'
-    require 'xml-apis.jar'
-  end
+  require_relative "nokogiri/jruby/dependencies"
 end
 
-begin
-  RUBY_VERSION =~ /(\d+\.\d+)/
-  require "nokogiri/#{$1}/nokogiri"
-rescue LoadError
-  require 'nokogiri/nokogiri'
-end
-require 'nokogiri/version'
-require 'nokogiri/syntax_error'
-require 'nokogiri/xml'
-require 'nokogiri/xslt'
-require 'nokogiri/html'
-require 'nokogiri/decorators/slop'
-require 'nokogiri/css'
-require 'nokogiri/html/builder'
+require_relative "nokogiri/extension"
 
 # Nokogiri parses and searches XML/HTML very quickly, and also has
 # correctly implemented CSS3 selector support as well as XPath 1.0
 # support.
 #
 # Parsing a document returns either a Nokogiri::XML::Document, or a
-# Nokogiri::HTML::Document depending on the kind of document you parse.
+# Nokogiri::HTML4::Document depending on the kind of document you parse.
 #
 # 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::HTML:Document for the page we’re interested in...
+#     doc = Nokogiri::HTML4(URI.open('http://www.google.com/search?q=tenderlove'))
 #
-#   doc = Nokogiri::HTML(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 also:
 #
-# See Nokogiri::XML::Searchable#css for more information about CSS searching.
-# See Nokogiri::XML::Searchable#xpath for more information about XPath searching.
+# - 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
+    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)
-        Nokogiri.HTML(string, url, encoding,
-          options || XML::ParseOptions::DEFAULT_HTML)
+        Nokogiri.HTML4(
+          string,
+          url,
+          encoding,
+          options || XML::ParseOptions::DEFAULT_HTML,
+        )
       else
-        Nokogiri.XML(string, url, encoding,
-          options || XML::ParseOptions::DEFAULT_XML)
-      end.tap { |doc|
+        Nokogiri.XML(
+          string,
+          url,
+          encoding,
+          options || XML::ParseOptions::DEFAULT_XML,
+        )
+      end.tap do |doc|
         yield doc if block_given?
-      }
+      end
     end
 
     ###
     # Create a new Nokogiri::XML::DocumentFragment
-    def make input = nil, opts = {}, &blk
+    def make(input = nil, opts = {}, &blk)
       if input
-        Nokogiri::HTML.fragment(input).children.first
+        Nokogiri::HTML4.fragment(input).children.first
       else
         Nokogiri(&blk)
       end
@@ -115,30 +92,37 @@ 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.
-      {
-        'Windows-31J' => 'CP932',	# Windows-31J is the IANA registered name of CP932.
-      }.each { |alias_name, name|
-        EncodingHandler.alias(name, alias_name) if EncodingHandler[alias_name].nil?
-      }
+      warn("Nokogiri.install_default_aliases is deprecated. Please call Nokogiri::EncodingHandler.install_default_aliases instead. This will become an error in Nokogiri v1.17.0.", uplevel: 1, category: :deprecated) # deprecated in v1.14.0, remove in v1.17.0
+      Nokogiri::EncodingHandler.install_default_aliases
     end
   end
-
-  Nokogiri.install_default_aliases
 end
 
 ###
-# Parser a document contained in +args+.  Nokogiri will try to guess what
-# type of document you are attempting to parse.  For more information, see
-# Nokogiri.parse
+# Parse a document contained in +args+.  Nokogiri will try to guess what type of document you are
+# attempting to parse.  For more information, see Nokogiri.parse
 #
-# To specify the type of document, use Nokogiri.XML or Nokogiri.HTML.
+# To specify the type of document, use {Nokogiri.XML}, {Nokogiri.HTML4}, or {Nokogiri.HTML5}.
 def Nokogiri(*args, &block)
-  if block_given?
-    Nokogiri::HTML::Builder.new(&block).doc.root
+  if block
+    Nokogiri::HTML4::Builder.new(&block).doc.root
   else
     Nokogiri.parse(*args)
   end
 end
+
+require_relative "nokogiri/version"
+require_relative "nokogiri/class_resolver"
+require_relative "nokogiri/syntax_error"
+require_relative "nokogiri/xml"
+require_relative "nokogiri/xslt"
+require_relative "nokogiri/html4"
+require_relative "nokogiri/html"
+require_relative "nokogiri/decorators/slop"
+require_relative "nokogiri/css"
+require_relative "nokogiri/html4/builder"
+require_relative "nokogiri/encoding_handler"
+
+require_relative "nokogiri/html5" if Nokogiri.uses_gumbo?

data/lib/xsd/xmlparser/nokogiri.rb

@@ -1,14 +1,15 @@
-require 'nokogiri'
+# frozen_string_literal: true
 
-module XSD # :nodoc:
-  module XMLParser # :nodoc:
+require "nokogiri"
+
+module XSD
+  module XMLParser
     ###
     # Nokogiri XML parser for soap4r.
     #
-    # Nokogiri may be used as the XML parser in soap4r.  Simply require
-    # 'xsd/xmlparser/nokogiri' in your soap4r applications, and soap4r
-    # will use Nokogiri as it's XML parser.  No other changes should be
-    # required to use Nokogiri as the XML parser.
+    # Nokogiri may be used as the XML parser in soap4r. Require 'xsd/xmlparser/nokogiri' in your
+    # soap4r applications, and soap4r will use Nokogiri as its XML parser. No other changes should
+    # be required to use Nokogiri as the XML parser.
     #
     # Example (using UW ITS Web Services):
     #
@@ -26,40 +27,40 @@ 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
 
       ###
@@ -69,16 +70,16 @@ 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 = []) # rubocop:disable Metrics/ParameterLists
         ###
         # 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
 
       ###
@@ -86,14 +87,16 @@ 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|
-        class_eval %{ def #{name}(*args); end }
+      ["xmldecl", "start_document", "end_document", "comment"].each do |name|
+        class_eval <<~RUBY, __FILE__, __LINE__ + 1
+          def #{name}(*args); end
+        RUBY
       end
 
       add_factory(self)

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
+