RubyGems - nokogiri - Versions diffs - 1.16.0 → 1.18.1 - Mend

nokogiri 1.16.0 → 1.18.1

Potentially problematic release.

This version of nokogiri might be problematic. Click here for more details.

Files changed (97) hide show

checksums.yaml +4 -4
data/Gemfile +14 -16
data/LICENSE-DEPENDENCIES.md +6 -6
data/README.md +8 -5
data/dependencies.yml +9 -9
data/ext/nokogiri/extconf.rb +188 -142
data/ext/nokogiri/gumbo.c +69 -53
data/ext/nokogiri/html4_document.c +10 -4
data/ext/nokogiri/html4_element_description.c +18 -18
data/ext/nokogiri/html4_sax_parser.c +40 -0
data/ext/nokogiri/html4_sax_parser_context.c +48 -58
data/ext/nokogiri/html4_sax_push_parser.c +25 -24
data/ext/nokogiri/libxml2_polyfill.c +114 -0
data/ext/nokogiri/nokogiri.c +9 -2
data/ext/nokogiri/nokogiri.h +18 -33
data/ext/nokogiri/xml_attr.c +1 -1
data/ext/nokogiri/xml_cdata.c +2 -10
data/ext/nokogiri/xml_comment.c +3 -8
data/ext/nokogiri/xml_document.c +167 -156
data/ext/nokogiri/xml_document_fragment.c +10 -25
data/ext/nokogiri/xml_dtd.c +1 -1
data/ext/nokogiri/xml_element_content.c +9 -9
data/ext/nokogiri/xml_encoding_handler.c +4 -4
data/ext/nokogiri/xml_namespace.c +6 -6
data/ext/nokogiri/xml_node.c +141 -104
data/ext/nokogiri/xml_node_set.c +46 -44
data/ext/nokogiri/xml_reader.c +74 -57
data/ext/nokogiri/xml_relax_ng.c +35 -56
data/ext/nokogiri/xml_sax_parser.c +156 -88
data/ext/nokogiri/xml_sax_parser_context.c +219 -131
data/ext/nokogiri/xml_sax_push_parser.c +68 -49
data/ext/nokogiri/xml_schema.c +50 -85
data/ext/nokogiri/xml_syntax_error.c +19 -11
data/ext/nokogiri/xml_text.c +2 -4
data/ext/nokogiri/xml_xpath_context.c +103 -100
data/ext/nokogiri/xslt_stylesheet.c +16 -11
data/gumbo-parser/Makefile +3 -0
data/gumbo-parser/src/ascii.c +2 -2
data/gumbo-parser/src/error.c +76 -48
data/gumbo-parser/src/error.h +5 -1
data/gumbo-parser/src/nokogiri_gumbo.h +11 -2
data/gumbo-parser/src/parser.c +66 -25
data/gumbo-parser/src/tokenizer.c +6 -6
data/lib/nokogiri/class_resolver.rb +1 -1
data/lib/nokogiri/css/node.rb +6 -2
data/lib/nokogiri/css/parser.rb +6 -4
data/lib/nokogiri/css/parser.y +2 -2
data/lib/nokogiri/css/parser_extras.rb +6 -66
data/lib/nokogiri/css/selector_cache.rb +38 -0
data/lib/nokogiri/css/tokenizer.rb +4 -4
data/lib/nokogiri/css/tokenizer.rex +9 -8
data/lib/nokogiri/css/xpath_visitor.rb +42 -6
data/lib/nokogiri/css.rb +86 -20
data/lib/nokogiri/decorators/slop.rb +3 -5
data/lib/nokogiri/encoding_handler.rb +2 -2
data/lib/nokogiri/html4/document.rb +44 -23
data/lib/nokogiri/html4/document_fragment.rb +124 -12
data/lib/nokogiri/html4/encoding_reader.rb +1 -1
data/lib/nokogiri/html4/sax/parser.rb +23 -38
data/lib/nokogiri/html4/sax/parser_context.rb +4 -9
data/lib/nokogiri/html4.rb +9 -14
data/lib/nokogiri/html5/builder.rb +40 -0
data/lib/nokogiri/html5/document.rb +61 -30
data/lib/nokogiri/html5/document_fragment.rb +130 -20
data/lib/nokogiri/html5/node.rb +4 -4
data/lib/nokogiri/html5.rb +114 -72
data/lib/nokogiri/version/constant.rb +1 -1
data/lib/nokogiri/xml/builder.rb +8 -1
data/lib/nokogiri/xml/document.rb +70 -26
data/lib/nokogiri/xml/document_fragment.rb +84 -13
data/lib/nokogiri/xml/node.rb +82 -11
data/lib/nokogiri/xml/node_set.rb +9 -7
data/lib/nokogiri/xml/parse_options.rb +1 -1
data/lib/nokogiri/xml/pp/node.rb +6 -1
data/lib/nokogiri/xml/reader.rb +51 -17
data/lib/nokogiri/xml/relax_ng.rb +57 -20
data/lib/nokogiri/xml/sax/document.rb +174 -83
data/lib/nokogiri/xml/sax/parser.rb +115 -41
data/lib/nokogiri/xml/sax/parser_context.rb +116 -8
data/lib/nokogiri/xml/sax/push_parser.rb +3 -0
data/lib/nokogiri/xml/sax.rb +48 -0
data/lib/nokogiri/xml/schema.rb +112 -45
data/lib/nokogiri/xml/searchable.rb +38 -42
data/lib/nokogiri/xml/syntax_error.rb +22 -0
data/lib/nokogiri/xml/xpath_context.rb +14 -3
data/lib/nokogiri/xml.rb +13 -24
data/lib/nokogiri/xslt/stylesheet.rb +29 -7
data/lib/nokogiri/xslt.rb +3 -9
data/lib/xsd/xmlparser/nokogiri.rb +3 -4
data/patches/libxml2/0019-xpath-Use-separate-static-hash-table-for-standard-fu.patch +244 -0
data/ports/archives/libxml2-2.13.5.tar.xz +0 -0
data/ports/archives/libxslt-1.1.42.tar.xz +0 -0
metadata +13 -14
data/ext/nokogiri/libxml2_backwards_compat.c +0 -121
data/patches/libxml2/0003-libxml2.la-is-in-top_builddir.patch +0 -25
data/ports/archives/libxml2-2.12.3.tar.xz +0 -0
data/ports/archives/libxslt-1.1.39.tar.xz +0 -0

data/lib/nokogiri/xml/node.rb CHANGED Viewed

@@ -127,6 +127,42 @@ module Nokogiri
         # This is intentionally empty, and sets the method signature for subclasses.
       end
+      #
+      # :call-seq:
+      #   dup → Nokogiri::XML::Node
+      #   dup(level) → Nokogiri::XML::Node
+      #   dup(level, new_parent_doc) → Nokogiri::XML::Node
+      #
+      # Duplicate this node.
+      #
+      # [Parameters]
+      # - +level+ (optional Integer). 0 is a shallow copy, 1 (the default) is a deep copy.
+      # - +new_parent_doc+ (optional Nokogiri::XML::Document)
+      #   The new node's parent Document. Defaults to the the Document of the current node.
+      # [Returns] The new Nokogiri::XML::Node
+      #
+      def dup(level = 1, new_parent_doc = document)
+        super().initialize_copy_with_args(self, level, new_parent_doc)
+      end
+      #
+      # :call-seq:
+      #   clone → Nokogiri::XML::Node
+      #   clone(level) → Nokogiri::XML::Node
+      #   clone(level, new_parent_doc) → Nokogiri::XML::Node
+      #
+      # Clone this node.
+      #
+      # [Parameters]
+      # - +level+ (optional Integer). 0 is a shallow copy, 1 (the default) is a deep copy.
+      # - +new_parent_doc+
+      #   The new node's parent Document. Defaults to the the Document of the current node.
+      # [Returns] The new Nokogiri::XML::Node
+      #
+      def clone(level = 1, new_parent_doc = document)
+        super().initialize_copy_with_args(self, level, new_parent_doc)
+      end
       ###
       # Decorate this node with the decorators set up in this node's Document
       def decorate!
@@ -228,7 +264,7 @@ module Nokogiri
           if new_parent.nil?
             raise "Failed to parse '#{node_or_tags}' in the context of a '#{context_node.name}' element"
           end
-        when XML::Node
+        when Node
           new_parent = node_or_tags.dup
         else
           raise ArgumentError, "Requires a String or Node argument, and cannot accept a #{node_or_tags.class}"
@@ -406,8 +442,48 @@ module Nokogiri
       end
       ####
-      # Set the Node's content to a Text node containing +string+. The string gets XML escaped, not
-      # interpreted as markup.
+      # call-seq:
+      #   content=(input)
+      #
+      # Set the content of this node to +input+.
+      #
+      # [Parameters]
+      # - +input+ (String) The new content for this node. Input is considered to be raw content, and
+      #   so will be entity-escaped in the final DOM string.
+      #
+      # [Example]
+      # Note how entities are handled:
+      #
+      #   doc = Nokogiri::HTML::Document.parse(<<~HTML)
+      #     <html>
+      #       <body>
+      #         <div id="first">asdf</div>
+      #         <div id="second">asdf</div>
+      #   HTML
+      #
+      #   text_node = doc.at_css("div#first").children.first
+      #   div_node = doc.at_css("div#second")
+      #
+      #   value = "You &amp; Me"
+      #
+      #   text_node.content = value
+      #   div_node.content = value
+      #
+      #   doc.css("div").to_html
+      #   # => "<div id=\"first\">You &amp;amp; Me</div>
+      #   #     <div id=\"second\">You &amp;amp; Me</div>"
+      #
+      # For content that is already entity-escaped, use CGI::unescapeHTML to decode it:
+      #
+      #   text_node.content = CGI::unescapeHTML(value)
+      #   div_node.content = CGI::unescapeHTML(value)
+      #
+      #   doc.css("div").to_html
+      #   # => "<div id=\"first\">You &amp; Me</div>
+      #   #     <div id=\"second\">You &amp; Me</div>"
+      #
+      # See also: #native_content=
+      #
       def content=(string)
         self.native_content = encode_special_chars(string.to_s)
       end
@@ -474,7 +550,6 @@ module Nokogiri
       alias_method :to_str, :content
       alias_method :name, :node_name
       alias_method :type, :node_type
-      alias_method :clone, :dup
       alias_method :elements, :element_children
       # :section: Working With Node Attributes
@@ -1051,9 +1126,11 @@ module Nokogiri
         error_count = document.errors.length
         node_set = in_context(contents, options.to_i)
         if document.errors.length > error_count
           raise document.errors[error_count] unless options.recover?
+          # TODO: remove this block when libxml2 < 2.13 is no longer supported
           if node_set.empty?
             # libxml2 < 2.13 does not obey the +recover+ option after encountering errors during
             # +in_context+ parsing, and so this horrible hack is here to try to emulate recovery
@@ -1080,6 +1157,7 @@ module Nokogiri
             node_set = fragment.children
           end
         end
         node_set
       end
@@ -1542,19 +1620,12 @@ module Nokogiri
         node_or_tags
       end
-      USING_LIBXML_WITH_BROKEN_SERIALIZATION = Nokogiri.uses_libxml?("~> 2.6.0").freeze
-      private_constant :USING_LIBXML_WITH_BROKEN_SERIALIZATION
       def to_format(save_option, options)
-        return dump_html if USING_LIBXML_WITH_BROKEN_SERIALIZATION
         options[:save_with] = save_option unless options[:save_with]
         serialize(options)
       end
       def write_format_to(save_option, io, options)
-        return (io << dump_html) if USING_LIBXML_WITH_BROKEN_SERIALIZATION
         options[:save_with] ||= save_option
         write_to(io, options)
       end

data/lib/nokogiri/xml/node_set.rb CHANGED Viewed

@@ -4,9 +4,13 @@
 module Nokogiri
   module XML
     ####
-    # A NodeSet contains a list of Nokogiri::XML::Node objects.  Typically
-    # a NodeSet is return as a result of searching a Document via
-    # Nokogiri::XML::Searchable#css or Nokogiri::XML::Searchable#xpath
+    # A NodeSet is an Enumerable that contains a list of Nokogiri::XML::Node objects.
+    #
+    # Typically a NodeSet is returned as a result of searching a Document via
+    # Nokogiri::XML::Searchable#css or Nokogiri::XML::Searchable#xpath.
+    #
+    # Note that the `#dup` and `#clone` methods perform shallow copies; these methods do not copy
+    # the Nodes contained in the NodeSet (similar to how Array and other Enumerable classes work).
     class NodeSet
       include Nokogiri::XML::Searchable
       include Enumerable
@@ -14,8 +18,6 @@ module Nokogiri
       # The Document this NodeSet is associated with
       attr_accessor :document
-      alias_method :clone, :dup
       # Create a NodeSet with +document+ defaulting to +list+
       def initialize(document, list = [])
         @document = document
@@ -121,7 +123,7 @@ module Nokogiri
           return self[args.first]
         end
-        super(*args)
+        super
       end
       alias_method :%, :at
@@ -423,7 +425,7 @@ module Nokogiri
       end
       ###
-      # Return a nicely formated string representation
+      # Return a nicely formatted string representation
       def inspect
         "[#{map(&:inspect).join(", ")}]"
       end

data/lib/nokogiri/xml/parse_options.rb CHANGED Viewed

@@ -140,7 +140,7 @@ module Nokogiri
       # Relax any hardcoded limit from the parser. Off by default.
       #
-      # ⚠ There may be a performance penalty when this option is set.
+      # ⚠ <b>It is UNSAFE to set this option</b> when parsing untrusted documents.
       HUGE        = 1 << 19
       # Support line numbers up to <code>long int</code> (default is a <code>short int</code>). On

data/lib/nokogiri/xml/pp/node.rb CHANGED Viewed

@@ -8,6 +8,11 @@ module Nokogiri
         COLLECTIONS = [:attribute_nodes, :children]
         def inspect
+          # handle the case where an exception is thrown during object construction
+          if respond_to?(:data_ptr?) && !data_ptr?
+            return "#<#{self.class}:#{format("0x%x", object_id)} (no data)>"
+          end
           attributes = inspect_attributes.reject do |x|
             attribute = send(x)
             !attribute || (attribute.respond_to?(:empty?) && attribute.empty?)
@@ -21,7 +26,7 @@ module Nokogiri
               "#{attribute}=#{send(attribute).inspect}"
             end.join(" ")
           end
-          "#<#{self.class.name}:#{format("0x%x", object_id)} #{attributes}>"
+          "#<#{self.class}:#{format("0x%x", object_id)} #{attributes}>"
         end
         def pretty_print(pp)

data/lib/nokogiri/xml/reader.rb CHANGED Viewed

@@ -3,32 +3,34 @@
 module Nokogiri
   module XML
     ###
-    # Nokogiri::XML::Reader parses an XML document similar to the way a cursor
-    # would move.  The Reader is given an XML document, and yields nodes
-    # to an each block.
+    # The Reader parser allows you to effectively pull parse an \XML document.  Once instantiated,
+    # call Nokogiri::XML::Reader#each to iterate over each node.
+    #
+    # Nokogiri::XML::Reader parses an \XML document similar to the way a cursor would move. The
+    # Reader is given an \XML document, and yields nodes to an each block.
+    #
+    # The Reader parser might be good for when you need the speed and low memory usage of a \SAX
+    # parser, but do not want to write a SAX::Document handler.
     #
     # Here is an example of usage:
     #
-    #     reader = Nokogiri::XML::Reader(<<-eoxml)
+    #     reader = Nokogiri::XML::Reader.new <<~XML
     #       <x xmlns:tenderlove='http://tenderlovemaking.com/'>
     #         <tenderlove:foo awesome='true'>snuggles!</tenderlove:foo>
     #       </x>
-    #     eoxml
+    #     XML
     #
     #     reader.each do |node|
-    #
     #       # node is an instance of Nokogiri::XML::Reader
     #       puts node.name
-    #
     #     end
     #
-    # Note that Nokogiri::XML::Reader#each can only be called once!!  Once
-    # the cursor moves through the entire document, you must parse the
-    # document again.  So make sure that you capture any information you
-    # need during the first iteration.
+    # ⚠ Nokogiri::XML::Reader#each can only be called once! Once the cursor moves through the entire
+    # document, you must parse the document again. It may be better to capture all information you
+    # need during a single iteration.
     #
-    # The Reader parser is good for when you need the speed of a SAX parser,
-    # but do not want to write a Document handler.
+    # ⚠ libxml2 does not support error recovery in the Reader parser. The +RECOVER+ ParseOption is
+    # ignored. If a syntax error is encountered during parsing, an exception will be raised.
     class Reader
       include Enumerable
@@ -65,23 +67,55 @@ module Nokogiri
       TYPE_END_ELEMENT = 15
       # Entity end node type
       TYPE_END_ENTITY = 16
-      # XML Declaration node type
+      # \XML Declaration node type
       TYPE_XML_DECLARATION = 17
       # A list of errors encountered while parsing
       attr_accessor :errors
-      # The XML source
+      # The \XML source
       attr_reader :source
       alias_method :self_closing?, :empty_element?
-      def initialize(source, url = nil, encoding = nil) # :nodoc:
+      # :call-seq:
+      #   Reader.new(input) { |options| ... } → Reader
+      #   Reader.new(input, url:, encoding:, options:) { |options| ... } → Reader
+      #
+      # Create a new Reader to parse an \XML document.
+      #
+      # [Required Parameters]
+      # - +input+ (String | IO): The \XML document to parse.
+      #
+      # [Optional Parameters]
+      # - +url:+ (String) The base URL of the document.
+      # - +encoding:+ (String) The name of the encoding of the document.
+      # - +options:+ (Integer | ParseOptions) Options to control the parser behavior.
+      #   Defaults to +ParseOptions::STRICT+.
+      #
+      # [Yields]
+      # If present, the block will be passed a Nokogiri::XML::ParseOptions object to modify before
+      # the fragment is parsed. See Nokogiri::XML::ParseOptions for more information.
+      def self.new(
+        string_or_io,
+        url_ = nil, encoding_ = nil, options_ = ParseOptions::STRICT,
+        url: url_, encoding: encoding_, options: options_
+      )
+        options = Nokogiri::XML::ParseOptions.new(options) if Integer === options
+        yield options if block_given?
+        if string_or_io.respond_to?(:read)
+          return Reader.from_io(string_or_io, url, encoding, options.to_i)
+        end
+        Reader.from_memory(string_or_io, url, encoding, options.to_i)
+      end
+      private def initialize(source, url = nil, encoding = nil) # :nodoc:
         @source   = source
         @errors   = []
         @encoding = encoding
       end
-      private :initialize
       # Get the attributes and namespaces of the current node as a Hash.
       #

data/lib/nokogiri/xml/relax_ng.rb CHANGED Viewed

@@ -3,36 +3,73 @@
 module Nokogiri
   module XML
     class << self
-      ###
-      # Create a new Nokogiri::XML::RelaxNG document from +string_or_io+.
-      # See Nokogiri::XML::RelaxNG for an example.
-      def RelaxNG(string_or_io, options = ParseOptions::DEFAULT_SCHEMA)
-        RelaxNG.new(string_or_io, options)
+      # :call-seq:
+      #   RelaxNG(input) → Nokogiri::XML::RelaxNG
+      #   RelaxNG(input, options:) → Nokogiri::XML::RelaxNG
+      #
+      # Convenience method for Nokogiri::XML::RelaxNG.new
+      def RelaxNG(...)
+        RelaxNG.new(...)
       end
     end
-    ###
-    # Nokogiri::XML::RelaxNG is used for validating XML against a
-    # RelaxNG schema.
+    # Nokogiri::XML::RelaxNG is used for validating \XML against a RELAX NG schema definition.
     #
-    # == Synopsis
+    # 🛡 <b>Do not use this class for untrusted schema documents.</b> RELAX NG input is always
+    # treated as *trusted*, meaning that the underlying parsing libraries <b>will access network
+    # resources</b>. This is counter to Nokogiri's "untrusted by default" security policy, but is an
+    # unfortunate limitation of the underlying libraries.
     #
-    # Validate an XML document against a RelaxNG schema.  Loop over the errors
-    # that are returned and print them out:
+    # *Example:* Determine whether an \XML document is valid.
     #
-    #   schema  = Nokogiri::XML::RelaxNG(File.open(ADDRESS_SCHEMA_FILE))
-    #   doc     = Nokogiri::XML(File.open(ADDRESS_XML_FILE))
+    #   schema = Nokogiri::XML::RelaxNG.new(File.read(RELAX_NG_FILE))
+    #   doc = Nokogiri::XML::Document.parse(File.read(XML_FILE))
+    #   schema.valid?(doc) # Boolean
     #
-    #   schema.validate(doc).each do |error|
-    #     puts error.message
-    #   end
+    # *Example:* Validate an \XML document against a \RelaxNG schema, and capture any errors that are found.
     #
-    # The list of errors are Nokogiri::XML::SyntaxError objects.
+    #   schema = Nokogiri::XML::RelaxNG.new(File.open(RELAX_NG_FILE))
+    #   doc = Nokogiri::XML::Document.parse(File.open(XML_FILE))
+    #   errors = schema.validate(doc) # Array<SyntaxError>
+    #
+    # *Example:* Validate an \XML document using a Document containing a RELAX NG schema definition.
+    #
+    #   schema_doc = Nokogiri::XML::Document.parse(File.read(RELAX_NG_FILE))
+    #   schema = Nokogiri::XML::RelaxNG.from_document(schema_doc)
+    #   doc = Nokogiri::XML::Document.parse(File.open(XML_FILE))
+    #   schema.valid?(doc) # Boolean
     #
-    # NOTE: RelaxNG input is always treated as TRUSTED documents, meaning that they will cause the
-    # underlying parsing libraries to access network resources. This is counter to Nokogiri's
-    # "untrusted by default" security policy, but is a limitation of the underlying libraries.
     class RelaxNG < Nokogiri::XML::Schema
+      # :call-seq:
+      #   new(input) → Nokogiri::XML::RelaxNG
+      #   new(input, options:) → Nokogiri::XML::RelaxNG
+      #
+      # Parse a RELAX NG schema definition from a String or IO to create a new Nokogiri::XML::RelaxNG.
+      #
+      # [Parameters]
+      # - +input+ (String | IO) RELAX NG schema definition
+      # - +options:+ (Nokogiri::XML::ParseOptions)
+      #   Defaults to Nokogiri::XML::ParseOptions::DEFAULT_SCHEMA ⚠ Unused
+      #
+      # [Returns] Nokogiri::XML::RelaxNG
+      #
+      # ⚠ +parse_options+ is currently unused by this method and is present only as a placeholder for
+      # future functionality.
+      #
+      # Also see convenience method Nokogiri::XML::RelaxNG()
+      def self.new(input, parse_options_ = ParseOptions::DEFAULT_SCHEMA, options: parse_options_)
+        from_document(Nokogiri::XML::Document.parse(input), options)
+      end
+      # :call-seq:
+      #   read_memory(input) → Nokogiri::XML::RelaxNG
+      #   read_memory(input, options:) → Nokogiri::XML::RelaxNG
+      #
+      # Convenience method for Nokogiri::XML::RelaxNG.new.
+      def self.read_memory(...)
+        # TODO deprecate this method
+        new(...)
+      end
     end
   end
 end