nokogiri 1.5.11 → 1.6.0.rc1

data/CHANGELOG.rdoc

@@ -1,23 +1,13 @@
-=== 1.5.11 / 2013-12-14
+=== 1.6.0.rc1 / 2013-04-14
 
-* Bugfixes
-
-  * (JRuby) Fix out of memory bug when certain invalid documents are parsed.
-  * (JRuby) Fix regression of billion-laughs vulnerability. #586
-
-=== 1.5.10 / 2013-06-07
-
-* Bugfixes
-
-  * (JRuby) Fix "null document" error when parsing an empty IO in jruby 1.7.3. #883
-  * (JRuby) Fix schema validation when XSD has DOCTYPE set to DTD. #861 (Thanks, Patrick Cheng!)
-  * (MRI) Fix segfault when there is no default subelement for an HTML node. #917
-
-
-* Notes
+* Features
 
-  * Use rb_ary_entry instead of RARRAY_PTR (you know, for Rubinius). #877 (Thanks, Dirkjan Bussink!)
-  * Fix TypeError when running tests. #900 (Thanks, Cédric Boutillier!)
+  * (MRI) Source code for libxml 2.8.0 and libxslt 1.2.26 is packaged
+    with the gem. These libraries are compiled at gem install time
+    unless the environment variable NOKOGIRI_USE_SYSTEM_LIBRARIES is
+    set. VERSION_INFO (also `nokogiri -v`) exposes whether libxml was
+    compiled from packaged source, or the system library was used.
+  * (Windows) libxml upgraded to 2.8.0
 
 
 === 1.5.9 / 2013-03-21
@@ -72,7 +62,7 @@
   * XML::Node#[]= stringifies values. #729 (Thanks, Ben Langfeld.)
   * bin/nokogiri will process a document from $stdin
   * bin/nokogiri -e will execute a program from the command line
-  * (JRuby) bin/nokogiri --version will print the Xerces and NekoHTML versions.
+  * bin/nokogiri --version will print the Xerces and NekoHTML versions when ran with JRuby.
 
 
 * Bugfixes

data/Rakefile

@@ -32,22 +32,23 @@ HOE = Hoe.spec 'nokogiri' do
 
   self.extra_rdoc_files = FileList['*.rdoc','ext/nokogiri/*.c']
 
-  self.licenses = ['MIT']
-
   self.clean_globs += [
     'nokogiri.gemspec',
     'lib/nokogiri/nokogiri.{bundle,jar,rb,so}',
-    'lib/nokogiri/{1.8,1.9,2.0}',
+    'lib/nokogiri/{1.9,2.0}',
     # GENERATED_PARSER,
     # GENERATED_TOKENIZER
   ]
 
+  self.extra_deps += [
+    ["mini_portile",    "~> 0.5.0"],
+  ]
+
   self.extra_dev_deps += [
     ["hoe-bundler",     ">= 1.1"],
     ["hoe-debugging",   ">= 1.0.3"],
     ["hoe-gemspec",     ">= 1.0"],
     ["hoe-git",         ">= 1.4"],
-    ["mini_portile",    ">= 0.2.2"],
     ["minitest",        "~> 2.2.2"],
     ["rake",            ">= 0.9"],
     ["rake-compiler",   "~> 0.8.0"],
@@ -60,7 +61,7 @@ HOE = Hoe.spec 'nokogiri' do
   else
     self.spec_extras = {
       :extensions => ["ext/nokogiri/extconf.rb"],
-      :required_ruby_version => '>= 1.8.7'
+      :required_ruby_version => '>= 1.9.2'
     }
   end
 
@@ -108,6 +109,21 @@ else
 
   HOE.spec.files.reject! { |f| f =~ %r{\.(java|jar)$} }
 
+  windows_p = RbConfig::CONFIG['target_os'] == 'mingw32' || RbConfig::CONFIG['target_os'] =~ /mswin/
+
+  unless windows_p || java?
+    task gem_build_path do
+      add_file_to_gem "dependencies.yml"
+
+      dependencies = YAML.load_file("dependencies.yml")
+      %w[libxml2 libxslt].each do |lib|
+        version = dependencies[lib]
+        archive = File.join("ports", "archives", "#{lib}-#{version}.tar.gz")
+        add_file_to_gem archive
+      end
+    end
+  end
+
   Rake::ExtensionTask.new("nokogiri", HOE.spec) do |ext|
     ext.lib_dir = File.join(*['lib', 'nokogiri', ENV['FAT_DIR']].compact)
     ext.config_options << ENV['EXTOPTS']
@@ -196,7 +212,7 @@ end
 
 desc "build a windows gem without all the ceremony."
 task "gem:windows" => "gem" do
-  cross_rubies = ["1.8.7-p358", "1.9.3-p194", "2.0.0-p0"]
+  cross_rubies = ["1.9.3-p194", "2.0.0-p0"]
   ruby_cc_version = cross_rubies.collect { |_| _.split("-").first }.join(":") # e.g., "1.8.7:1.9.2"
   rake_compiler_config_path = "#{ENV['HOME']}/.rake-compiler/config.yml"
 

data/build_all

@@ -8,7 +8,7 @@
 #  here's what I recommend for building all the gems:
 #
 #   1. set up a vagrant VM guest running ubuntu lucid 32-bit.
-#   2. install rvm, and install 1.8.7, 1.9.3 and jruby.
+#   2. install rvm, and install 1.9.3, 2.0.0 and jruby.
 #   3. `sudo apt-get install mingw32`
 #
 #  as you build, you may run into these problems:

data/dependencies.yml

@@ -0,0 +1,4 @@
+libxml2: "2.8.0"
+libxslt: "1.1.26"
+zlib: "1.2.7"
+libiconv: "1.13.1"

data/ext/nokogiri/extconf.rb

@@ -18,7 +18,9 @@ end
 $CFLAGS << " #{ENV["CFLAGS"]}"
 $LIBS << " #{ENV["LIBS"]}"
 
-if RbConfig::CONFIG['target_os'] == 'mingw32' || RbConfig::CONFIG['target_os'] =~ /mswin/
+windows_p = RbConfig::CONFIG['target_os'] == 'mingw32' || RbConfig::CONFIG['target_os'] =~ /mswin/
+
+if windows_p
   $CFLAGS << " -DXP_WIN -DXP_WIN32 -DUSE_INCLUDED_VASPRINTF"
 elsif RbConfig::CONFIG['target_os'] =~ /solaris/
   $CFLAGS << " -DUSE_INCLUDED_VASPRINTF"
@@ -36,59 +38,107 @@ if RbConfig::MAKEFILE_CONFIG['CC'] =~ /gcc/
   $CFLAGS << " -Wall -Wcast-qual -Wwrite-strings -Wconversion -Wmissing-noreturn -Winline"
 end
 
-if RbConfig::CONFIG['target_os'] =~ /mswin/
-  lib_prefix = 'lib'
-
-  # There's no default include/lib dir on Windows. Let's just add the Ruby ones
-  # and resort on the search path specified by INCLUDE and LIB environment
-  # variables
+if windows_p
+  # I'm cross compiling!
   HEADER_DIRS = [INCLUDEDIR]
   LIB_DIRS = [LIBDIR]
   XML2_HEADER_DIRS = [File.join(INCLUDEDIR, "libxml2"), INCLUDEDIR]
 
 else
-  lib_prefix = ''
-
-  HEADER_DIRS = [
-    # First search /opt/local for macports
-    '/opt/local/include',
-
-    # Then search /usr/local for people that installed from source
-    '/usr/local/include',
-
-    # Check the ruby install locations
-    INCLUDEDIR,
-
-    # Finally fall back to /usr
-    '/usr/include',
-    '/usr/include/libxml2',
-  ]
-
-  LIB_DIRS = [
-    # First search /opt/local for macports
-    '/opt/local/lib',
-
-    # Then search /usr/local for people that installed from source
-    '/usr/local/lib',
-
-    # Check the ruby install locations
-    LIBDIR,
-
-    # Finally fall back to /usr
-    '/usr/lib',
-  ]
-
-  XML2_HEADER_DIRS = [
-    '/opt/local/include/libxml2',
-    '/usr/local/include/libxml2',
-    File.join(INCLUDEDIR, "libxml2")
-  ] + HEADER_DIRS
-
-  # If the user has homebrew installed, use the libxml2 inside homebrew
-  brew_prefix = `brew --prefix libxml2 2> /dev/null`.chomp
-  unless brew_prefix.empty?
-    LIB_DIRS.unshift File.join(brew_prefix, 'lib')
-    XML2_HEADER_DIRS.unshift File.join(brew_prefix, 'include/libxml2')
+  if ENV['NOKOGIRI_USE_SYSTEM_LIBRARIES']
+    HEADER_DIRS = [
+      # First search /opt/local for macports
+      '/opt/local/include',
+
+      # Then search /usr/local for people that installed from source
+      '/usr/local/include',
+
+      # Check the ruby install locations
+      INCLUDEDIR,
+
+      # Finally fall back to /usr
+      '/usr/include',
+      '/usr/include/libxml2',
+    ]
+
+    LIB_DIRS = [
+      # First search /opt/local for macports
+      '/opt/local/lib',
+
+      # Then search /usr/local for people that installed from source
+      '/usr/local/lib',
+
+      # Check the ruby install locations
+      LIBDIR,
+
+      # Finally fall back to /usr
+      '/usr/lib',
+    ]
+
+    XML2_HEADER_DIRS = [
+      '/opt/local/include/libxml2',
+      '/usr/local/include/libxml2',
+      File.join(INCLUDEDIR, "libxml2")
+    ] + HEADER_DIRS
+
+    # If the user has homebrew installed, use the libxml2 inside homebrew
+    brew_prefix = `brew --prefix libxml2 2> /dev/null`.chomp
+    unless brew_prefix.empty?
+      LIB_DIRS.unshift File.join(brew_prefix, 'lib')
+      XML2_HEADER_DIRS.unshift File.join(brew_prefix, 'include/libxml2')
+    end
+
+  else
+    require 'mini_portile'
+    require 'yaml'
+
+    common_recipe = lambda do |recipe|
+      recipe.target = File.join(ROOT, "ports")
+      recipe.files = ["ftp://ftp.xmlsoft.org/libxml2/#{recipe.name}-#{recipe.version}.tar.gz"]
+
+      checkpoint = "#{recipe.target}/#{recipe.name}-#{recipe.version}-#{recipe.host}.installed"
+      unless File.exist?(checkpoint)
+        recipe.cook
+        FileUtils.touch checkpoint
+      end
+      recipe.activate
+    end
+
+    dependencies = YAML.load_file(File.join(ROOT, "dependencies.yml"))
+
+    libxml2_recipe = MiniPortile.new("libxml2", dependencies["libxml2"]).tap do |recipe|
+      recipe.configure_options = [
+        "--enable-shared",
+        "--disable-static",
+        "--without-python",
+        "--without-readline",
+        "--with-c14n",
+        "--with-debug",
+        "--with-threads"
+      ]
+      common_recipe.call recipe
+    end
+
+    libxslt_recipe = MiniPortile.new("libxslt", dependencies["libxslt"]).tap do |recipe|
+      recipe.configure_options = [
+        "--enable-shared",
+        "--disable-static",
+        "--without-python",
+        "--without-crypto",
+        "--with-debug",
+        "--with-libxml-prefix=#{libxml2_recipe.path}"
+      ]
+      common_recipe.call recipe
+    end
+
+    $LDFLAGS << " -Wl,-rpath,#{libxml2_recipe.path}/lib"
+    $LDFLAGS << " -Wl,-rpath,#{libxslt_recipe.path}/lib"
+
+    $CFLAGS << " -DNOKOGIRI_USE_PACKAGED_LIBRARIES -DNOKOGIRI_LIBXML2_PATH='\"#{libxml2_recipe.path}\"' -DNOKOGIRI_LIBXSLT_PATH='\"#{libxslt_recipe.path}\"'"
+
+    HEADER_DIRS = [libxml2_recipe, libxslt_recipe].map { |_| File.join(_.path, "include") }
+    LIB_DIRS = [libxml2_recipe, libxslt_recipe].map { |_| File.join(_.path, "lib") }
+    XML2_HEADER_DIRS = HEADER_DIRS + [File.join(libxml2_recipe.path, "include", "libxml2")]
   end
 end
 
@@ -117,9 +167,9 @@ asplode "libxml2"  unless find_header('libxml/parser.h')
 asplode "libxslt"  unless find_header('libxslt/xslt.h')
 asplode "libexslt" unless find_header('libexslt/exslt.h')
 asplode "libiconv" unless have_iconv?
-asplode "libxml2"  unless find_library("#{lib_prefix}xml2", 'xmlParseDoc')
-asplode "libxslt"  unless find_library("#{lib_prefix}xslt", 'xsltParseStylesheetDoc')
-asplode "libexslt" unless find_library("#{lib_prefix}exslt", 'exsltFuncRegister')
+asplode "libxml2"  unless find_library("xml2", 'xmlParseDoc')
+asplode "libxslt"  unless find_library("xslt", 'xsltParseStylesheetDoc')
+asplode "libexslt" unless find_library("exslt", 'exsltFuncRegister')
 
 unless have_func('xmlHasFeature')
   abort "-----\nThe function 'xmlHasFeature' is missing from your installation of libxml2.  Likely this means that your installed version of libxml2 is old enough that nokogiri will not work well.  To get around this problem, please upgrade your installation of libxml2.
@@ -141,4 +191,5 @@ if ENV['CPUPROFILE']
 end
 
 create_makefile('nokogiri/nokogiri')
+
 # :startdoc:

data/ext/nokogiri/html_element_description.c

@@ -86,10 +86,7 @@ static VALUE default_sub_element(VALUE self)
   htmlElemDesc * description;
   Data_Get_Struct(self, htmlElemDesc, description);
 
-  if (description->defaultsubelt)
-    return NOKOGIRI_STR_NEW2(description->defaultsubelt);
-
-  return Qnil;
+  return NOKOGIRI_STR_NEW2(description->defaultsubelt);
 }
 
 /*

data/ext/nokogiri/nokogiri.c

@@ -90,6 +90,16 @@ void Init_nokogiri()
                 NOKOGIRI_STR_NEW2(xmlParserVersion)
               );
 
+#ifdef NOKOGIRI_USE_PACKAGED_LIBRARIES
+  rb_const_set(mNokogiri, rb_intern("NOKOGIRI_USE_PACKAGED_LIBRARIES"), Qtrue);
+  rb_const_set(mNokogiri, rb_intern("NOKOGIRI_LIBXML2_PATH"), NOKOGIRI_STR_NEW2(NOKOGIRI_LIBXML2_PATH));
+  rb_const_set(mNokogiri, rb_intern("NOKOGIRI_LIBXSLT_PATH"), NOKOGIRI_STR_NEW2(NOKOGIRI_LIBXSLT_PATH));
+#else
+  rb_const_set(mNokogiri, rb_intern("NOKOGIRI_USE_PACKAGED_LIBRARIES"), Qfalse);
+  rb_const_set(mNokogiri, rb_intern("NOKOGIRI_LIBXML2_PATH"), Qnil);
+  rb_const_set(mNokogiri, rb_intern("NOKOGIRI_LIBXSLT_PATH"), Qnil);
+#endif
+
 #ifdef LIBXML_ICONV_ENABLED
   rb_const_set(mNokogiri, rb_intern("LIBXML_ICONV_ENABLED"), Qtrue);
 #else

data/ext/nokogiri/xslt_stylesheet.c

@@ -209,7 +209,7 @@ static void * initFunc(xsltTransformContextPtr ctxt, const xmlChar *uri)
     int i;
 
     for(i = 0; i < RARRAY_LEN(methods); i++) {
-	VALUE method_name = rb_obj_as_string(rb_ary_entry(methods, i));
+	VALUE method_name = rb_obj_as_string(RARRAY_PTR(methods)[i]);
 	xsltRegisterExtFunction(ctxt,
           (unsigned char *)StringValuePtr(method_name), uri, method_caller);
     }

data/lib/nokogiri/version.rb

@@ -1,6 +1,6 @@
 module Nokogiri
   # The version of Nokogiri you are using
-  VERSION = '1.5.11'
+  VERSION = '1.6.0.rc1'
 
   class VersionInfo # :nodoc:
     def jruby?
@@ -12,7 +12,7 @@ module Nokogiri
     end
 
     def loaded_parser_version
-      LIBXML_PARSER_VERSION.scan(/^(\d+)(\d\d)(\d\d)(?!\d)/).first.collect{ |j|
+      LIBXML_PARSER_VERSION.scan(/^(.*)(..)(..)$/).first.collect{ |j|
         j.to_i
       }.join(".")
     end
@@ -25,6 +25,14 @@ module Nokogiri
       defined?(LIBXML_VERSION)
     end
 
+    def libxml2_using_system?
+      ! libxml2_using_packaged?
+    end
+
+    def libxml2_using_packaged?
+      NOKOGIRI_USE_PACKAGED_LIBRARIES
+    end
+
     def warnings
       return [] unless libxml2?
 
@@ -49,6 +57,13 @@ module Nokogiri
       if libxml2?
         hash_info['libxml']              = {}
         hash_info['libxml']['binding']   = 'extension'
+        if libxml2_using_packaged?
+          hash_info['libxml']['source']  = "packaged"
+          hash_info['libxml']['libxml2_path'] = NOKOGIRI_LIBXML2_PATH
+          hash_info['libxml']['libxslt_path'] = NOKOGIRI_LIBXSLT_PATH
+        else
+          hash_info['libxml']['source']  = "system"
+        end
         hash_info['libxml']['compiled']  = compiled_parser_version
         hash_info['libxml']['loaded']    = loaded_parser_version
         hash_info['warnings']            = warnings

data/lib/nokogiri/xml/document.rb

@@ -45,13 +45,12 @@ module Nokogiri
         # Give the options to the user
         yield options if block_given?
 
-        return new if empty_doc?(string_or_io)
-
         doc = if string_or_io.respond_to?(:read)
           url ||= string_or_io.respond_to?(:path) ? string_or_io.path : nil
           read_io(string_or_io, url, encoding, options.to_i)
         else
           # read_memory pukes on empty docs
+          return new if string_or_io.nil? or string_or_io.empty?
           read_memory(string_or_io, url, encoding, options.to_i)
         end
 
@@ -261,12 +260,6 @@ module Nokogiri
       end
 
       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
-
       def implied_xpath_context
         "/"
       end

data/lib/nokogiri/xml/node.rb

@@ -917,23 +917,18 @@ module Nokogiri
       end
 
       def coerce data # :nodoc:
-        case data
-        when XML::NodeSet
-          return data
-        when XML::DocumentFragment
-          return data.children
-        when String
-          return fragment(data).children
-        when Document, XML::Attr
-          # unacceptable
-        when XML::Node
-          return data
-        end
+        return data                    if data.is_a?(XML::NodeSet)
+        return data.children           if data.is_a?(XML::DocumentFragment)
+        return fragment(data).children if data.is_a?(String)
 
-        raise ArgumentError, <<-EOERR
+        if data.is_a?(Document) || data.is_a?(XML::Attr) || !data.is_a?(XML::Node)
+          raise ArgumentError, <<-EOERR
 Requires a Node, NodeSet or String argument, and cannot accept a #{data.class}.
 (You probably want to select a node from the Document with at() or search(), or create a new Node via Node.new().)
-        EOERR
+          EOERR
+        end
+
+        data
       end
 
       def implied_xpath_context