nokogiri 1.11.7-java → 1.12.0.rc1-java

data/ext/nokogiri/xslt_stylesheet.c

@@ -64,7 +64,6 @@ parse_stylesheet_doc(VALUE klass, VALUE xmldocobj)
   VALUE errstr, exception;
   xsltStylesheetPtr ss ;
   Data_Get_Struct(xmldocobj, xmlDoc, xml);
-  exsltRegisterAll();
 
   errstr = rb_str_new(0, 0);
   xsltSetGenericErrorFunc((void *)errstr, xslt_generic_error_handler);
@@ -257,6 +256,8 @@ noko_init_xslt_stylesheet()
 
   cNokogiriXsltStylesheet = rb_define_class_under(mNokogiriXslt, "Stylesheet", rb_cObject);
 
+  rb_undef_alloc_func(cNokogiriXsltStylesheet);
+
   rb_define_singleton_method(cNokogiriXsltStylesheet, "parse_stylesheet_doc", parse_stylesheet_doc, 1);
   rb_define_method(cNokogiriXsltStylesheet, "serialize", serialize, 1);
   rb_define_method(cNokogiriXsltStylesheet, "transform", transform, -1);

data/gumbo-parser/CHANGES.md

@@ -0,0 +1,63 @@
+## Gumbo 0.10.1 (2015-04-30)
+
+Same as 0.10.0, but with the version number bumped because the last version-number commit to v0.9.4 makes GitHub think that v0.9.4 is the latest version and so it's not highlighted on the webpage.
+
+## Gumbo 0.10.0 (2015-04-30)
+
+* Full support for `<template>` tag (kevinhendricks, nostrademons).
+* Some fixes for `<rtc>`/`<rt>` handling (kevinhendricks, vmg).
+* All html5lib-trunk tests pass now! (kevinhendricks, vmg, nostrademons)
+* Support for fragment parsing (vmg)
+* A couple additional example programs (kevinhendricks)
+* Performance improvements totaling an estimated 30-40% total improvement (vmg, nostrademons).
+
+## Gumbo 0.9.4 (2015-04-30)
+
+* Additional Visual Studio fixes (lowjoel, nostrademons)
+* Fixed some unused variable warnings.
+* Fix for glibtoolize vs. libtoolize build errors on Mac.
+* Fixed `CDATA` end tag handling.
+
+## Gumbo 0.9.3 (2015-02-17)
+
+* Bugfix for `&AElig;` entities (rgrove)
+* Fix `CDATA` handling; `CDATA` sections now generate a `GUMBO_NODE_CDATA` node rather
+than plain text.
+* Fix `get_title example` to handle whitespace nodes (gsnedders)
+* Visual Studio compilation fixes (fishioon)
+* Take the namespace into account when determining whether a node matches a
+certain tag (aroben)
+* Replace the varargs tag functions with a tagset bytevector, for a 20-30%
+speedup in overall parse time (kevinhendricks, vmg)
+* Add MacOS X support to Travis CI, and fix the deployment/DLL issues this
+uncovered (nostrademons, kevinhendricks, vmg)
+
+## Gumbo 0.9.2 (2014-09-21)
+
+* Performance improvements: Ragel-based char ref decoder and DFA-based UTF8
+decoder, totaling speedups of up to 300%.
+* Added benchmarking program and some sample data.
+* Fixed a compiler error under Visual Studio.
+* Fix an error in the ctypes bindings that could lead to memory corruption in
+the Python bindings.
+* Fix duplicate attributes when parsing `<isindex>` tags.
+* Don't leave semicolons behind when consuming entity references (rgrove)
+* Internally rename some functions in preparation for an amalgamation file
+(jdeng)
+* Add proper cflags for gyp builds (skabbes)
+
+## Gumbo 0.9.1 (2014-08-07)
+
+* First version listed on PyPi.
+* Autotools files excluded from GitHub and generated via autogen.sh. (endgame)
+* Numerous compiler warnings fixed. (bnoordhuis, craigbarnes)
+* Google security audit passed.
+* Gyp support (tfarina)
+* Naming convention for structs changed to avoid C reserved words.
+* Fix several integer and buffer overflows (Maxime2)
+* Some Visual Studio compiler support (bugparty)
+* Python3 compatibility for the ctypes bindings.
+
+## Gumbo 0.9.0 (2013-08-13)
+
+* Initial release open-sourced by Google.

data/gumbo-parser/Makefile

@@ -0,0 +1,101 @@
+.PHONY: all clean check coverage
+
+gumbo_objs := $(patsubst %.c,build/%.o,$(wildcard src/*.c))
+test_objs := $(patsubst %.cc,build/%.o,$(wildcard test/*.cc))
+gtest_lib := googletest/make/gtest_main.a
+
+# make SANITIZEFLAGS='-fsanitize=undefined -fsanitize=address'
+SANITIZEFLAGS :=
+CPPFLAGS := -Isrc
+CFLAGS := -std=c99 -Os -Wall
+CXXFLAGS := -isystem googletest/include -std=c++11 -Os -Wall
+LDFLAGS := -pthread
+
+all: check
+
+src/%.c: src/%.rl
+	ragel -F1 -o $@ $<
+
+build/src:
+	mkdir -p $@
+
+build/test:
+	mkdir -p $@
+
+build/src/%.o: src/%.c build/src/flags | build/src
+	$(CC) -MMD $(CPPFLAGS) $(CFLAGS) $(SANITIZEFLAGS) -c -o $@ $<
+
+build/test/%.o: test/%.cc build/test/flags | build/test
+	$(CXX) -MMD $(CPPFLAGS) $(CXXFLAGS) $(SANITIZEFLAGS) -c -o $@ $<
+
+build/run_tests: $(gumbo_objs) $(test_objs) $(gtest_lib)
+	$(CXX) -o $@ $+ $(LDFLAGS) $(SANITIZEFLAGS)
+
+check: build/run_tests
+	./build/run_tests
+
+coverage:
+	$(RM) build/{src,test}/*.gcda
+	$(RM) build/*.info
+	$(MAKE) CPPFLAGS='-Isrc -DNDEBUG=1' \
+		CFLAGS='-std=c99 --coverage -g -O0' \
+		CXXFLAGS='-isystem googletest/include -std=c++11 --coverage -g -O0' \
+		LDFLAGS='--coverage' \
+		build/run_tests
+	lcov	--no-external \
+		--initial \
+		--capture \
+		--base-directory . \
+		--directory build \
+		--output-file build/coverage-pre.info
+	awk	-F '[:,]' \
+		'/^SF:/ { delete defs } /^FN:/ { defs[$$2]=1 } /^DA:/ { if ($$3 == 0 && $$2 in defs) next } { print }' \
+		build/coverage-pre.info > build/coverage-initial.info
+	./build/run_tests
+	lcov	--no-external \
+		--capture \
+		--base-directory . \
+		--directory build \
+		--rc lcov_branch_coverage=1 \
+		--output-file build/coverage-test.info
+	lcov	--add-tracefile build/coverage-initial.info \
+		--add-tracefile build/coverage-test.info \
+		--rc lcov_branch_coverage=1 \
+		--output-file build/coverage.info
+	lcov	--remove build/coverage.info '$(CURDIR)/googletest/*' \
+		--rc lcov_branch_coverage=1 \
+		--output-file build/coverage.info
+	genhtml --branch-coverage \
+		--output-directory build/coverage \
+		build/coverage.info
+
+clean:
+	$(RM) -r build
+
+build/src/flags: | build/src
+	@echo 'old_CC := $(CC)' > $@
+	@echo 'old_CPPFLAGS := $(CPPFLAGS)' >> $@
+	@echo 'old_CFLAGS := $(CFLAGS)' >>$@
+	@echo 'old_SANITIZEFLAGS := $(SANITIZEFLAGS)' >> $@
+	@echo 'old_LDFLAGS := $(LDFLAGS)' >> $@
+
+build/test/flags: | build/test
+	@echo 'old_CXX := $(CXX)' > $@
+	@echo 'old_CPPFLAGS := $(CPPFLAGS)' >> $@
+	@echo 'old_CXXFLAGS := $(CXXFLAGS)' >> $@
+	@echo 'old_SANITIZEFLAGS := $(SANITIZEFLAGS)' >> $@
+	@echo 'old_LDFLAGS := $(LDFLAGS)' >> $@
+
+ifeq (,$(filter clean coverage,$(MAKECMDGOALS)))
+# Ensure that the flags are up to date.
+-include build/src/flags build/test/flags
+ifneq ($(old_CC) | $(old_CPPFLAGS) | $(old_CFLAGS) | $(old_SANITIZEFLAGS) | $(old_LDFLAGS),$(CC) | $(CPPFLAGS) | $(CFLAGS) | $(SANITIZEFLAGS) | $(LDFLAGS))
+.PHONY: build/src/flags
+endif
+ifneq ($(old_CXX) | $(old_CPPFLAGS) | $(old_CXXFLAGS) | $(old_SANITIZEFLAGS) | $(old_LDFLAGS),$(CXX) | $(CPPFLAGS) | $(CXXFLAGS) | $(SANITIZEFLAGS) | $(LDFLAGS))
+.PHONY: build/test/flags
+endif
+
+# Include dependencies.
+-include $(test_objs:.o=.d) $(gumbo_objs:.o=.d)
+endif

data/gumbo-parser/THANKS

@@ -0,0 +1,27 @@
+Gumbo HTML parser THANKS file
+
+Gumbo was originally written by Jonathan Tang, but many people helped out through suggestions, question-answering, code reviews, bugfixes, and organizational support.  Here is a list of these people.  Help me keep it complete and exempt of errors.
+
+Adam Barth
+Adam Roben
+Ben Noordhuis
+Bowen Han
+Constantinos Michael
+Craig Barnes
+Geoffrey Sneddon
+Ian Hickson
+Jack Deng
+Joel Low
+Jonathan Shneier
+Kevin Hendricks
+Mason Tang
+Maxim Zakharov
+Michal Zalewski
+Neal Norwitz
+Othar Hansson
+Ryan Grove
+Stefan Haustein
+Steffen Meschkat
+Steven Kabbes
+Thiago Farina
+Vicent Marti

data/lib/nokogiri.rb

@@ -2,38 +2,29 @@
 # 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"
-  require 'nokogiri/jruby/dependencies'
+  require_relative "nokogiri/jruby/dependencies"
 end
 
-require 'nokogiri/extension'
-
-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(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...
 #
@@ -49,27 +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
@@ -98,10 +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
 
@@ -109,15 +100,26 @@ module Nokogiri
 end
 
 ###
-# 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
+# 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

@@ -1,28 +1,28 @@
 # frozen_string_literal: true
-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'
-
 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/parser.rb

@@ -8,7 +8,7 @@
 require 'racc/parser.rb'
 
 
-require 'nokogiri/css/parser_extras'
+require_relative "parser_extras"
 
 module Nokogiri
   module CSS

data/lib/nokogiri/css/parser.y

@@ -253,7 +253,7 @@ end
 
 ---- header
 
-require 'nokogiri/css/parser_extras'
+require_relative "parser_extras"
 
 ---- inner
 

data/lib/nokogiri/css/syntax_error.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
-require 'nokogiri/syntax_error'
+require_relative "../syntax_error"
 module Nokogiri
   module CSS
     class SyntaxError < ::Nokogiri::SyntaxError

data/lib/nokogiri/extension.rb

@@ -3,7 +3,7 @@
 # load the C or Java extension
 begin
   ::RUBY_VERSION =~ /(\d+\.\d+)/
-  require "nokogiri/#{Regexp.last_match(1)}/nokogiri"
+  require_relative "#{Regexp.last_match(1)}/nokogiri"
 rescue LoadError => e
   if e.message =~ /GLIBC/
     warn(<<~EOM)
@@ -22,5 +22,5 @@ rescue LoadError => e
     EOM
     raise e
   end
-  require 'nokogiri/nokogiri'
+  require_relative "nokogiri"
 end

data/lib/nokogiri/gumbo.rb

@@ -0,0 +1,14 @@
+# frozen_string_literal: true
+module Nokogiri
+  module Gumbo
+    # The default maximum number of attributes per element.
+    DEFAULT_MAX_ATTRIBUTES = 400
+
+    # The default maximum number of errors for parsing a document or a fragment.
+    DEFAULT_MAX_ERRORS = 0
+
+    # The default maximum depth of the DOM tree produced by parsing a document
+    # or fragment.
+    DEFAULT_MAX_TREE_DEPTH = 400
+  end
+end

data/lib/nokogiri/html.rb

@@ -1,38 +1,42 @@
 # frozen_string_literal: true
-require 'nokogiri/html/entity_lookup'
-require 'nokogiri/html/document'
-require 'nokogiri/html/document_fragment'
-require 'nokogiri/html/sax/parser_context'
-require 'nokogiri/html/sax/parser'
-require 'nokogiri/html/sax/push_parser'
-require 'nokogiri/html/element_description'
-require 'nokogiri/html/element_description_defaults'
+require_relative "html4"
 
 module Nokogiri
-  class << self
-    ###
-    # Parse HTML.  Convenience method for Nokogiri::HTML::Document.parse
-    def HTML thing, url = nil, encoding = nil, options = XML::ParseOptions::DEFAULT_HTML, &block
-      Nokogiri::HTML::Document.parse(thing, url, encoding, options, &block)
-    end
-  end
+  HTML = Nokogiri::HTML4
+
+  # @!method HTML(input, url = nil, encoding = nil, options = XML::ParseOptions::DEFAULT_HTML, &block)
+  # Parse HTML. Convenience method for Nokogiri::HTML4::Document.parse
+  # @!scope class
+  define_singleton_method(:HTML, Nokogiri.method(:HTML4))
 
+  # @note This module/namespace is an alias for {Nokogiri::HTML4} as of v1.12.0. Before v1.12.0,
+  #   {Nokogiri::HTML4} did not exist, and this was the module/namespace for all HTML-related
+  #   classes.
   module HTML
-    class << self
-      ###
-      # Parse HTML.  Convenience method for Nokogiri::HTML::Document.parse
-      def parse thing, url = nil, encoding = nil, options = XML::ParseOptions::DEFAULT_HTML, &block
-        Document.parse(thing, url, encoding, options, &block)
+    # @note This class is an alias for {Nokogiri::HTML4::Document} as of v1.12.0.
+    class Document < Nokogiri::XML::Document
+    end
+
+    # @note This class is an alias for {Nokogiri::HTML4::DocumentFragment} as of v1.12.0.
+    class DocumentFragment < Nokogiri::XML::DocumentFragment
+    end
+
+    # @note This class is an alias for {Nokogiri::HTML4::Builder} as of v1.12.0.
+    class Builder < Nokogiri::XML::Builder
+    end
+
+    module SAX
+      # @note This class is an alias for {Nokogiri::HTML4::SAX::Parser} as of v1.12.0.
+      class Parser < Nokogiri::XML::SAX::Parser
       end
 
-      ####
-      # Parse a fragment from +string+ in to a NodeSet.
-      def fragment string, encoding = nil
-        HTML::DocumentFragment.parse string, encoding
+      # @note This class is an alias for {Nokogiri::HTML4::SAX::ParserContext} as of v1.12.0.
+      class ParserContext < Nokogiri::XML::SAX::ParserContext
       end
-    end
 
-    # Instance of Nokogiri::HTML::EntityLookup
-    NamedCharacters = EntityLookup.new
+      # @note This class is an alias for {Nokogiri::HTML4::SAX::PushParser} as of v1.12.0.
+      class PushParser
+      end
+    end
   end
 end