nokogiri 1.10.10 → 1.12.0

data/gumbo-parser/src/util.h

@@ -0,0 +1,30 @@
+#ifndef GUMBO_UTIL_H_
+#define GUMBO_UTIL_H_
+
+#include <stdbool.h>
+#include <stddef.h>
+#include "macros.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+// Utility function for allocating & copying a null-terminated string into a
+// freshly-allocated buffer. This is necessary for proper memory management; we
+// have the convention that all const char* in parse tree structures are
+// freshly-allocated, so if we didn't copy, we'd try to delete a literal string
+// when the parse tree is destroyed.
+char* gumbo_strdup(const char* str) XMALLOC NONNULL_ARGS;
+
+void* gumbo_alloc(size_t size) XMALLOC;
+void* gumbo_realloc(void* ptr, size_t size) RETURNS_NONNULL;
+void gumbo_free(void* ptr);
+
+// Debug wrapper for printf
+void gumbo_debug(const char* format, ...) PRINTF(1);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif // GUMBO_UTIL_H_

data/gumbo-parser/src/vector.c

@@ -0,0 +1,111 @@
+/*
+ Copyright 2018 Craig Barnes.
+ Copyright 2010 Google Inc.
+
+ Licensed under the Apache License, Version 2.0 (the "License");
+ you may not use this file except in compliance with the License.
+ You may obtain a copy of the License at
+
+    https://www.apache.org/licenses/LICENSE-2.0
+
+ Unless required by applicable law or agreed to in writing, software
+ distributed under the License is distributed on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ See the License for the specific language governing permissions and
+ limitations under the License.
+*/
+
+#include <assert.h>
+#include <stdlib.h>
+#include <string.h>
+#include "vector.h"
+#include "util.h"
+
+void gumbo_vector_init(unsigned int initial_capacity, GumboVector* vector) {
+  vector->length = 0;
+  vector->capacity = initial_capacity;
+  if (initial_capacity > 0) {
+    vector->data = gumbo_alloc(sizeof(void*) * initial_capacity);
+  } else {
+    vector->data = NULL;
+  }
+}
+
+void gumbo_vector_destroy(GumboVector* vector) {
+  if (vector->capacity > 0) {
+    gumbo_free(vector->data);
+  }
+}
+
+static void enlarge_vector_if_full(GumboVector* vector) {
+  if (vector->length >= vector->capacity) {
+    if (vector->capacity) {
+      vector->capacity *= 2;
+      size_t num_bytes = sizeof(void*) * vector->capacity;
+      vector->data = gumbo_realloc(vector->data, num_bytes);
+    } else {
+      // 0-capacity vector; no previous array to deallocate.
+      vector->capacity = 2;
+      vector->data = gumbo_alloc(sizeof(void*) * vector->capacity);
+    }
+  }
+}
+
+void gumbo_vector_add(void* element, GumboVector* vector) {
+  enlarge_vector_if_full(vector);
+  assert(vector->data);
+  assert(vector->length < vector->capacity);
+  vector->data[vector->length++] = element;
+}
+
+void* gumbo_vector_pop(GumboVector* vector) {
+  if (vector->length == 0) {
+    return NULL;
+  }
+  return vector->data[--vector->length];
+}
+
+int gumbo_vector_index_of(GumboVector* vector, const void* element) {
+  for (unsigned int i = 0; i < vector->length; ++i) {
+    if (vector->data[i] == element) {
+      return i;
+    }
+  }
+  return -1;
+}
+
+void gumbo_vector_insert_at (
+  void* element,
+  unsigned int index,
+  GumboVector* vector
+) {
+  assert(index <= vector->length);
+  enlarge_vector_if_full(vector);
+  ++vector->length;
+  memmove (
+    &vector->data[index + 1],
+    &vector->data[index],
+    sizeof(void*) * (vector->length - index - 1)
+  );
+  vector->data[index] = element;
+}
+
+void gumbo_vector_remove(void* node, GumboVector* vector) {
+  int index = gumbo_vector_index_of(vector, node);
+  if (index == -1) {
+    return;
+  }
+  gumbo_vector_remove_at(index, vector);
+}
+
+void* gumbo_vector_remove_at(unsigned int index, GumboVector* vector) {
+  assert(index < vector->length);
+  void* result = vector->data[index];
+  memmove (
+    &vector->data[index],
+    &vector->data[index + 1],
+    sizeof(void*) * (vector->length - index - 1)
+  );
+  --vector->length;
+  return result;
+}

data/gumbo-parser/src/vector.h

@@ -0,0 +1,45 @@
+#ifndef GUMBO_VECTOR_H_
+#define GUMBO_VECTOR_H_
+
+#include "gumbo.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+// Initializes a new GumboVector with the specified initial capacity.
+void gumbo_vector_init(unsigned int initial_capacity, GumboVector* vector);
+
+// Frees the memory used by a GumboVector. Does not free the contained
+// pointers.
+void gumbo_vector_destroy(GumboVector* vector);
+
+// Adds a new element to a GumboVector.
+void gumbo_vector_add(void* element, GumboVector* vector);
+
+// Removes and returns the element most recently added to the GumboVector.
+// Ownership is transferred to caller. Capacity is unchanged. If the vector is
+// empty, NULL is returned.
+void* gumbo_vector_pop(GumboVector* vector);
+
+// Inserts an element at a specific index. This is potentially O(N) time, but
+// is necessary for some of the spec's behavior.
+void gumbo_vector_insert_at (
+  void* element,
+  unsigned int index,
+  GumboVector* vector
+);
+
+// Removes an element from the vector, or does nothing if the element is not in
+// the vector.
+void gumbo_vector_remove(void* element, GumboVector* vector);
+
+// Removes and returns an element at a specific index. Note that this is
+// potentially O(N) time and should be used sparingly.
+void* gumbo_vector_remove_at(unsigned int index, GumboVector* vector);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif // GUMBO_VECTOR_H_

data/lib/nokogiri.rb

@@ -1,60 +1,30 @@
 # -*- coding: utf-8 -*-
+# frozen_string_literal: true
 # Modify the PATH on windows so that the external DLLs will get loaded.
 
-require 'rbconfig'
+require "rbconfig"
 
 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'
 #
-#   # Get a Nokogiri::HTML:Document for the page we’re interested in...
+#   # Get a Nokogiri::HTML4::Document for the page we’re interested in...
 #
-#   doc = Nokogiri::HTML(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...
 #
@@ -70,27 +40,27 @@ 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]
         # 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,
+        Nokogiri.HTML4(string, url, encoding,
           options || XML::ParseOptions::DEFAULT_HTML)
       else
         Nokogiri.XML(string, url, encoding,
           options || XML::ParseOptions::DEFAULT_XML)
-      end.tap { |doc|
+      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
@@ -119,10 +89,10 @@ module Nokogiri
       # 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|
+        "Windows-31J" => "CP932",	# Windows-31J is the IANA registered name of CP932.
+      }.each do |alias_name, name|
         EncodingHandler.alias(name, alias_name) if EncodingHandler[alias_name].nil?
-      }
+      end
     end
   end
 
@@ -130,15 +100,26 @@ module Nokogiri
 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
+    Nokogiri::HTML4::Builder.new(&block).doc.root
   else
     Nokogiri.parse(*args)
   end
 end
+
+require_relative "nokogiri/version"
+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/html5" if Nokogiri.uses_gumbo?

data/lib/nokogiri/css.rb

@@ -1,27 +1,28 @@
-require 'nokogiri/css/node'
-require 'nokogiri/css/xpath_visitor'
-x = $-w
-$-w = false
-require 'nokogiri/css/parser'
-$-w = x
-
-require 'nokogiri/css/tokenizer'
-require 'nokogiri/css/syntax_error'
-
+# frozen_string_literal: true
 module Nokogiri
   module CSS
     class << self
       ###
       # Parse this CSS selector in +selector+.  Returns an AST.
-      def parse selector
-        Parser.new.parse selector
+      def parse(selector)
+        Parser.new.parse(selector)
       end
 
       ###
       # Get the XPath for +selector+.
-      def xpath_for selector, options={}
-        Parser.new(options[:ns] || {}).xpath_for selector, options
+      def xpath_for(selector, options = {})
+        Parser.new(options[:ns] || {}).xpath_for(selector, options)
       end
     end
   end
 end
+
+require_relative "css/node"
+require_relative "css/xpath_visitor"
+x = $-w
+$-w = false
+require_relative "css/parser"
+$-w = x
+
+require_relative "css/tokenizer"
+require_relative "css/syntax_error"

data/lib/nokogiri/css/node.rb

@@ -1,3 +1,4 @@
+# frozen_string_literal: true
 module Nokogiri
   module CSS
     class Node

data/lib/nokogiri/css/parser.rb

@@ -1,13 +1,14 @@
+# frozen_string_literal: true
 #
 # DO NOT MODIFY!!!!
-# This file is automatically generated by Racc 1.4.14
-# from Racc grammer file "".
+# This file is automatically generated by Racc 1.5.2
+# from Racc grammar file "".
 #
 
 require 'racc/parser.rb'
 
 
-require 'nokogiri/css/parser_extras'
+require_relative "parser_extras"
 
 module Nokogiri
   module CSS
@@ -333,42 +334,42 @@ Racc_debug_parser = false
 
 def _reduce_1(val, _values, result)
         result = [val.first, val.last].flatten
-      
+
     result
 end
 
 def _reduce_2(val, _values, result)
- result = val.flatten 
+ result = val.flatten
     result
 end
 
 def _reduce_3(val, _values, result)
- result = [val.last].flatten 
+ result = [val.last].flatten
     result
 end
 
 def _reduce_4(val, _values, result)
- result = :DIRECT_ADJACENT_SELECTOR 
+ result = :DIRECT_ADJACENT_SELECTOR
     result
 end
 
 def _reduce_5(val, _values, result)
- result = :CHILD_SELECTOR 
+ result = :CHILD_SELECTOR
     result
 end
 
 def _reduce_6(val, _values, result)
- result = :FOLLOWING_SELECTOR 
+ result = :FOLLOWING_SELECTOR
     result
 end
 
 def _reduce_7(val, _values, result)
- result = :DESCENDANT_SELECTOR 
+ result = :DESCENDANT_SELECTOR
     result
 end
 
 def _reduce_8(val, _values, result)
- result = :CHILD_SELECTOR 
+ result = :CHILD_SELECTOR
     result
 end
 
@@ -378,7 +379,7 @@ def _reduce_9(val, _values, result)
                   else
                     Node.new(:CONDITIONAL_SELECTOR, [val.first, val[1]])
                   end
-      
+
     result
 end
 
@@ -386,13 +387,13 @@ end
 
 def _reduce_11(val, _values, result)
         result = Node.new(:CONDITIONAL_SELECTOR, val)
-      
+
     result
 end
 
 def _reduce_12(val, _values, result)
         result = Node.new(:CONDITIONAL_SELECTOR, val)
-      
+
     result
 end
 
@@ -400,39 +401,39 @@ def _reduce_13(val, _values, result)
         result = Node.new(:CONDITIONAL_SELECTOR,
           [Node.new(:ELEMENT_NAME, ['*']), val.first]
         )
-      
+
     result
 end
 
 def _reduce_14(val, _values, result)
         result = Node.new(val.first, [nil, val.last])
-      
+
     result
 end
 
 def _reduce_15(val, _values, result)
         result = Node.new(val[1], [val.first, val.last])
-      
+
     result
 end
 
 def _reduce_16(val, _values, result)
         result = Node.new(:DESCENDANT_SELECTOR, [val.first, val.last])
-      
+
     result
 end
 
 # reduce 17 omitted
 
 def _reduce_18(val, _values, result)
- result = Node.new(:CLASS_CONDITION, [unescape_css_identifier(val[1])]) 
+ result = Node.new(:CLASS_CONDITION, [unescape_css_identifier(val[1])])
     result
 end
 
 # reduce 19 omitted
 
 def _reduce_20(val, _values, result)
- result = Node.new(:ELEMENT_NAME, val) 
+ result = Node.new(:ELEMENT_NAME, val)
     result
 end
 
@@ -440,19 +441,19 @@ def _reduce_21(val, _values, result)
         result = Node.new(:ELEMENT_NAME,
           [[val.first, val.last].compact.join(':')]
         )
-      
+
     result
 end
 
 def _reduce_22(val, _values, result)
         name = @namespaces.key?('xmlns') ? "xmlns:#{val.first}" : val.first
         result = Node.new(:ELEMENT_NAME, [name])
-      
+
     result
 end
 
 def _reduce_23(val, _values, result)
- result = val[0] 
+ result = val[0]
     result
 end
 
@@ -462,7 +463,7 @@ def _reduce_25(val, _values, result)
         result = Node.new(:ATTRIBUTE_CONDITION,
           [val[1]] + (val[2] || [])
         )
-      
+
     result
 end
 
@@ -470,16 +471,16 @@ def _reduce_26(val, _values, result)
         result = Node.new(:ATTRIBUTE_CONDITION,
           [val[1]] + (val[2] || [])
         )
-      
+
     result
 end
 
 def _reduce_27(val, _values, result)
-        # Non standard, but hpricot supports it.
+        # non-standard, from hpricot
         result = Node.new(:PSEUDO_CLASS,
           [Node.new(:FUNCTION, ['nth-child(', val[1]])]
         )
-      
+
     result
 end
 
@@ -487,7 +488,7 @@ def _reduce_28(val, _values, result)
         result = Node.new(:ELEMENT_NAME,
           [[val.first, val.last].compact.join(':')]
         )
-      
+
     result
 end
 
@@ -495,52 +496,52 @@ def _reduce_29(val, _values, result)
         # Default namespace is not applied to attributes.
         # So we don't add prefix "xmlns:" as in namespaced_ident.
         result = Node.new(:ELEMENT_NAME, [val.first])
-      
+
     result
 end
 
 def _reduce_30(val, _values, result)
         result = Node.new(:FUNCTION, [val.first.strip])
-      
+
     result
 end
 
 def _reduce_31(val, _values, result)
         result = Node.new(:FUNCTION, [val.first.strip, val[1]].flatten)
-      
+
     result
 end
 
 def _reduce_32(val, _values, result)
         result = Node.new(:FUNCTION, [val.first.strip, val[1]].flatten)
-      
+
     result
 end
 
 def _reduce_33(val, _values, result)
         result = Node.new(:FUNCTION, [val.first.strip, val[1]].flatten)
-      
+
     result
 end
 
 def _reduce_34(val, _values, result)
         result = Node.new(:FUNCTION, [val.first.strip, val[1]].flatten)
-      
+
     result
 end
 
 def _reduce_35(val, _values, result)
- result = [val.first, val.last] 
+ result = [val.first, val.last]
     result
 end
 
 def _reduce_36(val, _values, result)
- result = [val.first, val.last] 
+ result = [val.first, val.last]
     result
 end
 
 def _reduce_37(val, _values, result)
- result = [val.first, val.last] 
+ result = [val.first, val.last]
     result
 end
 
@@ -557,13 +558,13 @@ def _reduce_40(val, _values, result)
         when 'n'
           result = Node.new(:NTH, ['1','n','+','0'])
         else
-          # This is not CSS standard.  It allows us to support this:
+          # non-standard to support custom functions:
           # assert_xpath("//a[foo(., @href)]", @parser.parse('a:foo(@href)'))
           # assert_xpath("//a[foo(., @a, b)]", @parser.parse('a:foo(@a, b)'))
           # assert_xpath("//a[foo(., a, 10)]", @parser.parse('a:foo(a, 10)'))
           result = val
         end
-      
+
     result
 end
 
@@ -573,7 +574,7 @@ def _reduce_41(val, _values, result)
         else
           raise Racc::ParseError, "parse error on IDENT '#{val[1]}'"
         end
-      
+
     result
 end
 
@@ -589,7 +590,7 @@ def _reduce_42(val, _values, result)
         else
           raise Racc::ParseError, "parse error on IDENT '#{val[1]}'"
         end
-      
+
     result
 end
 
@@ -609,18 +610,18 @@ def _reduce_43(val, _values, result)
         else
           raise Racc::ParseError, "parse error on IDENT '#{val[1]}'"
         end
-      
+
     result
 end
 
 def _reduce_44(val, _values, result)
         result = Node.new(:PSEUDO_CLASS, [val[1]])
-      
+
     result
 end
 
 def _reduce_45(val, _values, result)
- result = Node.new(:PSEUDO_CLASS, [val[1]]) 
+ result = Node.new(:PSEUDO_CLASS, [val[1]])
     result
 end
 
@@ -630,31 +631,31 @@ end
 
 def _reduce_48(val, _values, result)
         result = Node.new(:COMBINATOR, val)
-      
+
     result
 end
 
 def _reduce_49(val, _values, result)
         result = Node.new(:COMBINATOR, val)
-      
+
     result
 end
 
 def _reduce_50(val, _values, result)
         result = Node.new(:COMBINATOR, val)
-      
+
     result
 end
 
 def _reduce_51(val, _values, result)
         result = Node.new(:COMBINATOR, val)
-      
+
     result
 end
 
 def _reduce_52(val, _values, result)
         result = Node.new(:COMBINATOR, val)
-      
+
     result
 end
 
@@ -669,65 +670,65 @@ end
 # reduce 57 omitted
 
 def _reduce_58(val, _values, result)
- result = Node.new(:ID, [unescape_css_identifier(val.first)]) 
+ result = Node.new(:ID, [unescape_css_identifier(val.first)])
     result
 end
 
 def _reduce_59(val, _values, result)
- result = [val.first, unescape_css_identifier(val[1])] 
+ result = [val.first, unescape_css_identifier(val[1])]
     result
 end
 
 def _reduce_60(val, _values, result)
- result = [val.first, unescape_css_string(val[1])] 
+ result = [val.first, unescape_css_string(val[1])]
     result
 end
 
 def _reduce_61(val, _values, result)
- result = [val.first, val[1]] 
+ result = [val.first, val[1]]
     result
 end
 
 # reduce 62 omitted
 
 def _reduce_63(val, _values, result)
- result = :equal 
+ result = :equal
     result
 end
 
 def _reduce_64(val, _values, result)
- result = :prefix_match 
+ result = :prefix_match
     result
 end
 
 def _reduce_65(val, _values, result)
- result = :suffix_match 
+ result = :suffix_match
     result
 end
 
 def _reduce_66(val, _values, result)
- result = :substring_match 
+ result = :substring_match
     result
 end
 
 def _reduce_67(val, _values, result)
- result = :not_equal 
+ result = :not_equal
     result
 end
 
 def _reduce_68(val, _values, result)
- result = :includes 
+ result = :includes
     result
 end
 
 def _reduce_69(val, _values, result)
- result = :dash_match 
+ result = :dash_match
     result
 end
 
 def _reduce_70(val, _values, result)
         result = Node.new(:NOT, [val[1]])
-      
+
     result
 end
 
@@ -746,5 +747,5 @@ def _reduce_none(val, _values, result)
 end
 
     end   # class Parser
-    end   # module CSS
-  end   # module Nokogiri
+  end   # module CSS
+end   # module Nokogiri