nokogiri 1.11.0.rc4 → 1.11.3

data/lib/nokogiri/xml/document.rb

@@ -1,3 +1,4 @@
+# coding: utf-8
 # frozen_string_literal: true
 
 require 'pathname'
@@ -13,11 +14,12 @@ module Nokogiri
     # Nokogiri::XML::Searchable#xpath
     #
     class Document < Nokogiri::XML::Node
-      # I'm ignoring unicode characters here.
-      # See http://www.w3.org/TR/REC-xml-names/#ns-decl for more details.
+      # See http://www.w3.org/TR/REC-xml-names/#ns-decl for more details. Note that we're not
+      # attempting to handle unicode characters partly because libxml2 doesn't handle unicode
+      # characters in NCNAMEs.
       NCNAME_START_CHAR = "A-Za-z_"
-      NCNAME_CHAR       = NCNAME_START_CHAR + "\\-.0-9"
-      NCNAME_RE         = /^xmlns(:[#{NCNAME_START_CHAR}][#{NCNAME_CHAR}]*)?$/
+      NCNAME_CHAR       = NCNAME_START_CHAR + "\\-\\.0-9"
+      NCNAME_RE         = /^xmlns(?::([#{NCNAME_START_CHAR}][#{NCNAME_CHAR}]*))?$/
 
       ##
       # Parse an XML file.
@@ -79,6 +81,35 @@ module Nokogiri
         return doc
       end
 
+      ##
+      # @!method wrap(java_document)
+      # @!scope class
+      #
+      # Create a {Document} using an existing Java DOM document object.
+      #
+      # The returned {Document} shares the same underlying data structure as the Java object, so
+      # changes in one are reflected in the other.
+      #
+      # @param java_document [Java::OrgW3cDom::Document]
+      # @return [Nokogiri::XML::Document]
+      # @note This method is only available when running JRuby.
+      # @note The class +Java::OrgW3cDom::Document+ is also accessible as +org.w3c.dom.Document+.
+      # @see #to_java
+
+      ##
+      # @!method to_java()
+      #
+      # Returns the underlying Java DOM document object for the {Document}.
+      #
+      # The returned Java object shares the same underlying data structure as the {Document}, so
+      # changes in one are reflected in the other.
+      #
+      # @return [Java::OrgW3cDom::Document]
+      # @note This method is only available when running JRuby.
+      # @note The class +Java::OrgW3cDom::Document+ is also accessible as +org.w3c.dom.Document+.
+      # @see .wrap
+
+
       # A list of Nokogiri::XML::SyntaxError found when parsing a document
       attr_accessor :errors
 
@@ -88,33 +119,58 @@ module Nokogiri
       end
 
       ##
-      # Create an element with +name+, and optionally setting the content and attributes.
+      # Create a new +Element+ with +name+ sharing GC lifecycle with the document, optionally
+      # setting contents or attributes.
+      #
+      # Arguments may be passed to initialize the element:
+      # - a +Hash+ argument will be used to set attributes
+      # - a non-Hash object that responds to +#to_s+ will be used to set the new node's contents
       #
-      #   doc.create_element "div" # <div></div>
-      #   doc.create_element "div", :class => "container" # <div class='container'></div>
-      #   doc.create_element "div", "contents" # <div>contents</div>
-      #   doc.create_element "div", "contents", :class => "container" # <div class='container'>contents</div>
-      #   doc.create_element "div" { |node| node['class'] = "container" } # <div class='container'></div>
+      # A block may be passed to mutate the node.
       #
-      def create_element name, *args, &block
+      # @param name [String]
+      # @param contents_or_attrs [#to_s,Hash]
+      # @yieldparam node [Nokogiri::XML::Element]
+      # @return [Nokogiri::XML::Element]
+      #
+      # @example An empty element without attributes
+      #   doc.create_element("div")
+      #   # => <div></div>
+      #
+      # @example An element with contents
+      #   doc.create_element("div", "contents")
+      #   # => <div>contents</div>
+      #
+      # @example An element with attributes
+      #   doc.create_element("div", {"class" => "container"})
+      #   # => <div class='container'></div>
+      #
+      # @example An element with contents and attributes
+      #   doc.create_element("div", "contents", {"class" => "container"})
+      #   # => <div class='container'>contents</div>
+      #
+      # @example Passing a block to mutate the element
+      #   doc.create_element("div") { |node| node["class"] = "blue" if before_noon? }
+      #
+      def create_element(name, *contents_or_attrs, &block)
         elm = Nokogiri::XML::Element.new(name, self, &block)
-        args.each do |arg|
+        contents_or_attrs.each do |arg|
           case arg
           when Hash
-            arg.each { |k,v|
+            arg.each do |k, v|
               key = k.to_s
               if key =~ NCNAME_RE
-                ns_name = key.split(":", 2)[1]
-                elm.add_namespace_definition ns_name, v
+                ns_name = Regexp.last_match(1)
+                elm.add_namespace_definition(ns_name, v)
               else
                 elm[k.to_s] = v.to_s
               end
-            }
+            end
           else
             elm.content = arg
           end
         end
-        if ns = elm.namespace_definitions.find { |n| n.prefix.nil? or n.prefix == '' }
+        if ns = elm.namespace_definitions.find { |n| n.prefix.nil? || (n.prefix == '') }
           elm.namespace = ns
         end
         elm
@@ -262,24 +318,14 @@ module Nokogiri
       end
       alias :<< :add_child
 
-      ##
-      # +JRuby+
-      # Wraps Java's org.w3c.dom.document and returns Nokogiri::XML::Document
-      def self.wrap(document) end if false # native-ext provides Document.wrap
-
-      ##
-      # +JRuby+
-      # Returns Java's org.w3c.dom.document of this Document.
-      def to_java; end if false # JRuby provides #to_java
-
       private
+
       def self.empty_doc? string_or_io
         string_or_io.nil? ||
           (string_or_io.respond_to?(:empty?) && string_or_io.empty?) ||
           (string_or_io.respond_to?(:eof?) && string_or_io.eof?)
       end
 
-      # @private
       IMPLIED_XPATH_CONTEXTS = [ '//'.freeze ].freeze # :nodoc:
 
       def inspect_attributes

data/lib/nokogiri/xml/node.rb

@@ -5,64 +5,54 @@ require "nokogiri/xml/node/save_options"
 
 module Nokogiri
   module XML
-    ####
-    # Nokogiri::XML::Node is your window to the fun filled world of dealing
-    # with XML and HTML tags.  A Nokogiri::XML::Node may be treated similarly
-    # to a hash with regard to attributes.  For example (from irb):
+    ##
+    # {Nokogiri::XML::Node} is your window to the fun filled world of dealing with XML and HTML
+    # tags. A {Nokogiri::XML::Node} may be treated similarly to a hash with regard to attributes. For
+    # example:
     #
-    #   irb(main):004:0> node
-    #   => <a href="#foo" id="link">link</a>
-    #   irb(main):005:0> node['href']
-    #   => "#foo"
-    #   irb(main):006:0> node.keys
-    #   => ["href", "id"]
-    #   irb(main):007:0> node.values
-    #   => ["#foo", "link"]
-    #   irb(main):008:0> node['class'] = 'green'
-    #   => "green"
-    #   irb(main):009:0> node
-    #   => <a href="#foo" id="link" class="green">link</a>
-    #   irb(main):010:0>
+    #   node = Nokogiri::XML::DocumentFragment.parse("<a href='#foo' id='link'>link</a>").at_css("a")
+    #   node.to_html # => "<a href=\"#foo\" id=\"link\">link</a>"
+    #   node['href'] # => "#foo"
+    #   node.keys # => ["href", "id"]
+    #   node.values # => ["#foo", "link"]
+    #   node['class'] = 'green' # => "green"
+    #   node.to_html # => "<a href=\"#foo\" id=\"link\" class=\"green\">link</a>"
     #
-    # See Nokogiri::XML::Node#[] and Nokogiri::XML#[]= for more information.
+    # See the method group entitled "Working With Node Attributes" for the full set of methods.
     #
-    # Nokogiri::XML::Node also has methods that let you move around your
+    # {Nokogiri::XML::Node} also has methods that let you move around your
     # tree.  For navigating your tree, see:
     #
-    # * Nokogiri::XML::Node#parent
-    # * Nokogiri::XML::Node#children
-    # * Nokogiri::XML::Node#next
-    # * Nokogiri::XML::Node#previous
-    #
+    # * {#parent}
+    # * {#children}
+    # * {#next}
+    # * {#previous}
     #
     # When printing or otherwise emitting a document or a node (and
     # its subtree), there are a few methods you might want to use:
     #
-    # * content, text, inner_text, to_str: emit plaintext
-    #
-    #   These methods will all emit the plaintext version of your
-    #   document, meaning that entities will be replaced (e.g., "&lt;"
-    #   will be replaced with "<"), meaning that any sanitizing will
-    #   likely be un-done in the output.
+    # * {#content}, {#text}, {#inner_text}, {#to_str}: These methods will all <b>emit plaintext</b>,
+    #   meaning that entities will be replaced (e.g., "&lt;" will be replaced with "<"), meaning
+    #   that any sanitizing will likely be un-done in the output.
     #
-    # * to_s, to_xml, to_html, inner_html: emit well-formed markup
+    # * {#to_s}, {#to_xml}, {#to_html}, {#inner_html}: These methods will all <b>emit
+    #   properly-escaped markup</b>, meaning that it's suitable for consumption by browsers,
+    #   parsers, etc.
     #
-    #   These methods will all emit properly-escaped markup, meaning
-    #   that it's suitable for consumption by browsers, parsers, etc.
+    # You may search this node's subtree using {#xpath} and {#css}
     #
-    # You may search this node's subtree using Searchable#xpath and Searchable#css
     class Node
       include Nokogiri::XML::PP::Node
       include Nokogiri::XML::Searchable
       include Enumerable
 
-      # Element node type, see Nokogiri::XML::Node#element?
+      # Element node type, see {Nokogiri::XML::Node#element?}
       ELEMENT_NODE = 1
       # Attribute node type
       ATTRIBUTE_NODE = 2
-      # Text node type, see Nokogiri::XML::Node#text?
+      # Text node type, see {Nokogiri::XML::Node#text?}
       TEXT_NODE = 3
-      # CDATA node type, see Nokogiri::XML::Node#cdata?
+      # CDATA node type, see {Nokogiri::XML::Node#cdata?}
       CDATA_SECTION_NODE = 4
       # Entity reference node type
       ENTITY_REF_NODE = 5
@@ -70,9 +60,9 @@ module Nokogiri
       ENTITY_NODE = 6
       # PI node type
       PI_NODE = 7
-      # Comment node type, see Nokogiri::XML::Node#comment?
+      # Comment node type, see {Nokogiri::XML::Node#comment?}
       COMMENT_NODE = 8
-      # Document node type, see Nokogiri::XML::Node#xml?
+      # Document node type, see {Nokogiri::XML::Node#xml?}
       DOCUMENT_NODE = 9
       # Document type node type
       DOCUMENT_TYPE_NODE = 10
@@ -80,7 +70,7 @@ module Nokogiri
       DOCUMENT_FRAG_NODE = 11
       # Notation node type
       NOTATION_NODE = 12
-      # HTML document node type, see Nokogiri::XML::Node#html?
+      # HTML document node type, see {Nokogiri::XML::Node#html?}
       HTML_DOCUMENT_NODE = 13
       # DTD node type
       DTD_NODE = 14
@@ -99,8 +89,15 @@ module Nokogiri
       # DOCB document node type
       DOCB_DOCUMENT_NODE = 21
 
-      def initialize(name, document) # :nodoc:
-        # ... Ya.  This is empty on purpose.
+      ##
+      # Create a new node with +name+ sharing GC lifecycle with +document+.
+      # @param name [String]
+      # @param document [Nokogiri::XML::Document]
+      # @yieldparam node [Nokogiri::XML::Node]
+      # @return [Nokogiri::XML::Node]
+      # @see Nokogiri::XML::Node.new
+      def initialize(name, document)
+        # This is intentionally empty.
       end
 
       ###

data/lib/nokogiri/xml/reader.rb

@@ -86,7 +86,8 @@ module Nokogiri
       private :initialize
 
       ###
-      # Get a list of attributes for the current node.
+      # Get the attributes of the current node as a Hash
+      # @return [Hash<String, String>] Attribute names and values
       def attributes
         attrs_hash = attribute_nodes.each_with_object({}) do |node, hash|
                        hash[node.name] = node.to_s
@@ -96,14 +97,6 @@ module Nokogiri
         attrs_hash
       end
 
-      ###
-      # Get a list of attributes for the current node
-      def attribute_nodes
-        nodes = attr_nodes
-        nodes.each { |v| v.instance_variable_set(:@_r, self) }
-        nodes
-      end
-
       ###
       # Move the cursor through the document yielding the cursor to the block
       def each

data/lib/nokogiri/xml/xpath/syntax_error.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 module Nokogiri
   module XML
-    class XPath
+    module XPath
       class SyntaxError < XML::SyntaxError
         def to_s
           [super.chomp, str1].compact.join(': ')

data/lib/nokogiri/xml/xpath.rb

@@ -3,9 +3,7 @@ require 'nokogiri/xml/xpath/syntax_error'
 
 module Nokogiri
   module XML
-    class XPath
-      # The Nokogiri::XML::Document tied to this XPath instance
-      attr_accessor :document
+    module XPath
     end
   end
 end

data/lib/nokogiri.rb

@@ -8,12 +8,8 @@ if defined?(RUBY_ENGINE) && RUBY_ENGINE == "jruby"
   require 'nokogiri/jruby/dependencies'
 end
 
-begin
-  RUBY_VERSION =~ /(\d+\.\d+)/
-  require "nokogiri/#{$1}/nokogiri"
-rescue LoadError
-  require 'nokogiri/nokogiri'
-end
+require 'nokogiri/extension'
+
 require 'nokogiri/version'
 require 'nokogiri/syntax_error'
 require 'nokogiri/xml'

data/patches/libxml2/0009-avoid-isnan-isinf.patch

@@ -0,0 +1,81 @@
+This patch is a result of rake-compiler-dock using centos 7 (manylinux2014) to cross-compile.
+
+Centos, for reasons I have not been able to discern, implements `isnan` and `isinf` as a function
+and not as a macro. Debian knows how to resolve that function at dynamic-link time (despite using a
+macro at compile time), but musl-based systems (like alpine) do not. Running `nm` on nokogiri.so
+created on such a centos system shows:
+
+```
+                 U __isinf@@GLIBC_2.2.5
+                 U __isnan@@GLIBC_2.2.5
+```
+
+(see https://github.com/sparklemotion/nokogiri/pull/2142 for more info)
+
+This patch avoids using glibc's `isnan` and `isinf` calls, instead using libxml2's fallback
+implementation. There's history here, see libxml2 commit 8813f39:
+
+    commit 8813f39
+    Author: Nick Wellnhofer <wellnhofer@aevum.de>
+    Date:   2017-09-21 00:11:26 +0200
+    
+        Simplify XPath NaN, inf and -0 handling
+        
+        Use C99 macros NAN, INFINITY, isnan, isinf. If they're not available:
+        
+        - Assume that (0.0 / 0.0) generates a NaN and !(x == x) tests for NaN.
+        - Use C89's HUGE_VAL for INFINITY.
+        
+        Remove manual handling of NaN, infinity and negative zero in functions
+        xmlXPathValueFlipSign and xmlXPathDivValues.
+        
+        Remove xmlXPathGetSign. All the tests for negative zero can be replaced
+        with a test for negative or positive zero.
+        
+        Simplify xmlXPathRoundFunction.
+        
+        Remove Trio dependency.
+        
+        This should work on IEEE 754 compliant implementations even if the C99
+        macros aren't available, but will likely break some ancient platforms.
+        If problems arise, my plan is to port the relevant trionan.c solution
+        to xpath.c. Note that non-compliant implementations are impossible
+        to fully support, anyway, since XPath requires IEEE 754.
+
+This patch would be unnecessary if any of the following was true:
+
+* centos implements these as macros, and doesn't generate an unresolved symbol for either in the shared library
+* we had a way to ensure `__isinf` and `__isnan` resolve on musl (e.g., we implement them locally)
+
+diff --git a/xpath.c b/xpath.c
+index 9f64ab9..5b6d999 100644
+--- a/xpath.c
++++ b/xpath.c
+@@ -509,11 +509,7 @@ xmlXPathInit(void) {
+  */
+ int
+ xmlXPathIsNaN(double val) {
+-#ifdef isnan
+-    return isnan(val);
+-#else
+     return !(val == val);
+-#endif
+ }
+ 
+ /**
+@@ -524,15 +520,11 @@ xmlXPathIsNaN(double val) {
+  */
+ int
+ xmlXPathIsInf(double val) {
+-#ifdef isinf
+-    return isinf(val) ? (val > 0 ? 1 : -1) : 0;
+-#else
+     if (val >= INFINITY)
+         return 1;
+     if (val <= -INFINITY)
+         return -1;
+     return 0;
+-#endif
+ }
+ 
+ #endif /* SCHEMAS or XPATH */

data/patches/libxml2/0010-parser.c-shrink-the-input-buffer-when-appropriate.patch

@@ -0,0 +1,70 @@
+From ca565c1edef9a455453fa8564270cc9c5813e1b9 Mon Sep 17 00:00:00 2001
+From: Mike Dalessio <mike.dalessio@gmail.com>
+Date: Sun, 31 Jan 2021 09:53:56 -0500
+Subject: [PATCH] parser.c: shrink the input buffer when appropriate
+
+Fixes GNOME/libxml2#200
+
+Also see discussions at:
+- GNOME/libxml2#192
+- https://gitlab.gnome.org/nwellnhof/libxml2/-/commit/99bda1e
+- https://github.com/sparklemotion/nokogiri/issues/2132
+---
+ parser.c | 6 ++++++
+ 1 file changed, 6 insertions(+)
+
+diff --git a/parser.c b/parser.c
+index a7bdc7f..efde672 100644
+--- a/parser.c
++++ b/parser.c
+@@ -4204,6 +4204,7 @@ xmlParseSystemLiteral(xmlParserCtxtPtr ctxt) {
+ 	}
+ 	count++;
+ 	if (count > 50) {
++	    SHRINK;
+ 	    GROW;
+ 	    count = 0;
+             if (ctxt->instate == XML_PARSER_EOF) {
+@@ -4291,6 +4292,7 @@ xmlParsePubidLiteral(xmlParserCtxtPtr ctxt) {
+ 	buf[len++] = cur;
+ 	count++;
+ 	if (count > 50) {
++	    SHRINK;
+ 	    GROW;
+ 	    count = 0;
+             if (ctxt->instate == XML_PARSER_EOF) {
+@@ -4571,6 +4573,7 @@ xmlParseCharDataComplex(xmlParserCtxtPtr ctxt, int cdata) {
+ 	}
+ 	count++;
+ 	if (count > 50) {
++	    SHRINK;
+ 	    GROW;
+ 	    count = 0;
+             if (ctxt->instate == XML_PARSER_EOF)
+@@ -4776,6 +4779,7 @@ xmlParseCommentComplex(xmlParserCtxtPtr ctxt, xmlChar *buf,
+ 
+ 	count++;
+ 	if (count > 50) {
++	    SHRINK;
+ 	    GROW;
+ 	    count = 0;
+             if (ctxt->instate == XML_PARSER_EOF) {
+@@ -5186,6 +5190,7 @@ xmlParsePI(xmlParserCtxtPtr ctxt) {
+ 		}
+ 		count++;
+ 		if (count > 50) {
++		    SHRINK;
+ 		    GROW;
+                     if (ctxt->instate == XML_PARSER_EOF) {
+                         xmlFree(buf);
+@@ -9783,6 +9788,7 @@ xmlParseCDSect(xmlParserCtxtPtr ctxt) {
+ 	sl = l;
+ 	count++;
+ 	if (count > 50) {
++	    SHRINK;
+ 	    GROW;
+             if (ctxt->instate == XML_PARSER_EOF) {
+ 		xmlFree(buf);
+-- 
+2.25.1
+