nokogiri 1.10.8-java → 1.11.0.rc3-java

data/ext/nokogiri/extconf.rb

@@ -24,6 +24,10 @@ def openbsd?
   RbConfig::CONFIG['target_os'] =~ /openbsd/
 end
 
+def aix?
+  RbConfig::CONFIG["target_os"] =~ /aix/
+end
+
 def nix?
   ! (windows? || solaris? || darwin?)
 end
@@ -419,7 +423,7 @@ if windows?
   $CFLAGS << " -DXP_WIN -DXP_WIN32 -DUSE_INCLUDED_VASPRINTF"
 end
 
-if solaris?
+if solaris? || aix?
   $CFLAGS << " -DUSE_INCLUDED_VASPRINTF"
 end
 
@@ -466,7 +470,7 @@ else
   # The gem version constraint in the Rakefile is not respected at install time.
   # Keep this version in sync with the one in the Rakefile !
   require 'rubygems'
-  gem 'mini_portile2', '~> 2.4.0'
+  gem 'mini_portile2', '~> 2.5.0'
   require 'mini_portile2'
   message "Using mini_portile version #{MiniPortile::VERSION}\n"
 
@@ -486,50 +490,60 @@ else
           url: "http://zlib.net/fossils/#{recipe.name}-#{recipe.version}.tar.gz",
           sha256: dependencies["zlib"]["sha256"]
         }]
-      class << recipe
-        attr_accessor :cross_build_p
-
-        def configure
-          Dir.chdir work_path do
-            mk = File.read 'win32/Makefile.gcc'
-            File.open 'win32/Makefile.gcc', 'wb' do |f|
-              f.puts "BINARY_PATH = #{path}/bin"
-              f.puts "LIBRARY_PATH = #{path}/lib"
-              f.puts "INCLUDE_PATH = #{path}/include"
-              mk.sub!(/^PREFIX\s*=\s*$/, "PREFIX = #{host}-") if cross_build_p
-              f.puts mk
+      if windows?
+        class << recipe
+          attr_accessor :cross_build_p
+
+          def configure
+            Dir.chdir work_path do
+              mk = File.read 'win32/Makefile.gcc'
+              File.open 'win32/Makefile.gcc', 'wb' do |f|
+                f.puts "BINARY_PATH = #{path}/bin"
+                f.puts "LIBRARY_PATH = #{path}/lib"
+                f.puts "INCLUDE_PATH = #{path}/include"
+                mk.sub!(/^PREFIX\s*=\s*$/, "PREFIX = #{host}-") if cross_build_p
+                f.puts mk
+              end
             end
           end
-        end
 
-        def configured?
-          Dir.chdir work_path do
-            !! (File.read('win32/Makefile.gcc') =~ /^BINARY_PATH/)
+          def configured?
+            Dir.chdir work_path do
+              !! (File.read('win32/Makefile.gcc') =~ /^BINARY_PATH/)
+            end
           end
-        end
 
-        def compile
-          execute "compile", "make -f win32/Makefile.gcc"
-        end
+          def compile
+            execute "compile", "make -f win32/Makefile.gcc"
+          end
 
-        def install
-          execute "install", "make -f win32/Makefile.gcc install"
+          def install
+            execute "install", "make -f win32/Makefile.gcc install"
+          end
+        end
+        recipe.cross_build_p = cross_build_p
+      else
+        class << recipe
+          def configure
+            execute "configure", ["env", "CHOST=#{host}", "CFLAGS=-fPIC #{ENV['CFLAGS']}", "./configure", "--static", configure_prefix]
+          end
         end
       end
-      recipe.cross_build_p = cross_build_p
     end
 
-    libiconv_recipe = process_recipe("libiconv", dependencies["libiconv"]["version"], static_p, cross_build_p) do |recipe|
-      recipe.files = [{
-          url: "http://ftp.gnu.org/pub/gnu/libiconv/#{recipe.name}-#{recipe.version}.tar.gz",
-          sha256: dependencies["libiconv"]["sha256"]
-        }]
-      recipe.configure_options += [
-        "CPPFLAGS=-Wall",
-        "CFLAGS=-O2 -g",
-        "CXXFLAGS=-O2 -g",
-        "LDFLAGS="
-      ]
+    unless nix?
+      libiconv_recipe = process_recipe("libiconv", dependencies["libiconv"]["version"], static_p, cross_build_p) do |recipe|
+        recipe.files = [{
+            url: "http://ftp.gnu.org/pub/gnu/libiconv/#{recipe.name}-#{recipe.version}.tar.gz",
+            sha256: dependencies["libiconv"]["sha256"]
+          }]
+        recipe.configure_options += [
+          "CPPFLAGS=-Wall",
+          "CFLAGS=-O2 -g",
+          "CXXFLAGS=-O2 -g",
+          "LDFLAGS="
+        ]
+      end
     end
   else
     if darwin? && !have_header('iconv.h')
@@ -617,7 +631,6 @@ EOM
       end
 
       # Defining a macro that expands to a C string; double quotes are significant.
-      $CPPFLAGS << ' ' << "-DNOKOGIRI_#{recipe.name.upcase}_PATH=\"#{recipe.path}\"".inspect
       $CPPFLAGS << ' ' << "-DNOKOGIRI_#{recipe.name.upcase}_PATCHES=\"#{recipe.patch_files.map { |path| File.basename(path) }.join(' ')}\"".inspect
 
       case libname

data/ext/nokogiri/nokogiri.c

@@ -76,24 +76,30 @@ void Init_nokogiri()
   mNokogiriHtmlSax  = rb_define_module_under(mNokogiriHtml, "SAX");
 
   rb_const_set( mNokogiri,
-                rb_intern("LIBXML_VERSION"),
+                rb_intern("LIBXML_COMPILED_VERSION"),
                 NOKOGIRI_STR_NEW2(LIBXML_DOTTED_VERSION)
               );
   rb_const_set( mNokogiri,
-                rb_intern("LIBXML_PARSER_VERSION"),
+                rb_intern("LIBXML_LOADED_VERSION"),
                 NOKOGIRI_STR_NEW2(xmlParserVersion)
               );
 
+
+  rb_const_set( mNokogiri,
+                rb_intern("LIBXSLT_COMPILED_VERSION"),
+                NOKOGIRI_STR_NEW2(LIBXSLT_DOTTED_VERSION)
+              );
+  rb_const_set( mNokogiri,
+                rb_intern("LIBXSLT_LOADED_VERSION"),
+                NOKOGIRI_STR_NEW2(xsltEngineVersion)
+              );
+
 #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));
   rb_const_set(mNokogiri, rb_intern("NOKOGIRI_LIBXML2_PATCHES"), rb_str_split(NOKOGIRI_STR_NEW2(NOKOGIRI_LIBXML2_PATCHES), " "));
   rb_const_set(mNokogiri, rb_intern("NOKOGIRI_LIBXSLT_PATCHES"), rb_str_split(NOKOGIRI_STR_NEW2(NOKOGIRI_LIBXSLT_PATCHES), " "));
 #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);
   rb_const_set(mNokogiri, rb_intern("NOKOGIRI_LIBXML2_PATCHES"), Qnil);
   rb_const_set(mNokogiri, rb_intern("NOKOGIRI_LIBXSLT_PATCHES"), Qnil);
 #endif

data/ext/nokogiri/nokogiri.h

@@ -1,6 +1,18 @@
 #ifndef NOKOGIRI_NATIVE
 #define NOKOGIRI_NATIVE
 
+#if _MSC_VER
+#ifndef WIN32_LEAN_AND_MEAN
+#define WIN32_LEAN_AND_MEAN
+#endif /* WIN32_LEAN_AND_MEAN */
+#ifndef WIN32
+#define WIN32
+#endif /* WIN32 */
+#include <winsock2.h>
+#include <ws2tcpip.h>
+#include <windows.h>
+#endif
+
 #include <stdlib.h>
 #include <string.h>
 #include <assert.h>
@@ -29,6 +41,7 @@ int vasprintf (char **strp, const char *fmt, va_list ap);
 #include <libxml/relaxng.h>
 #include <libxml/xinclude.h>
 #include <libxslt/extensions.h>
+#include <libxslt/xsltconfig.h>
 #include <libxml/c14n.h>
 #include <ruby.h>
 #include <ruby/st.h>

data/ext/nokogiri/xml_document.c

@@ -1,6 +1,6 @@
 #include <xml_document.h>
 
-static int dealloc_node_i(xmlNodePtr key, xmlNodePtr node, xmlDocPtr doc)
+static int dealloc_node_i2(xmlNodePtr key, xmlNodePtr node, xmlDocPtr doc)
 {
   switch(node->type) {
   case XML_ATTRIBUTE_NODE:
@@ -20,6 +20,11 @@ static int dealloc_node_i(xmlNodePtr key, xmlNodePtr node, xmlDocPtr doc)
   return ST_CONTINUE;
 }
 
+static int dealloc_node_i(st_data_t key, st_data_t node, st_data_t doc)
+{
+  return dealloc_node_i2((xmlNodePtr)key, (xmlNodePtr)node, (xmlDocPtr)doc);
+}
+
 static void remove_private(xmlNodePtr node)
 {
   xmlNodePtr child;
@@ -38,6 +43,15 @@ static void remove_private(xmlNodePtr node)
   node->_private = NULL;
 }
 
+static void mark(xmlDocPtr doc)
+{
+  nokogiriTuplePtr tuple = (nokogiriTuplePtr)doc->_private;
+  if(tuple) {
+      rb_gc_mark(tuple->doc);
+      rb_gc_mark(tuple->node_cache);
+  }
+}
+
 static void dealloc(xmlDocPtr doc)
 {
   st_table *node_hash;
@@ -588,7 +602,7 @@ VALUE Nokogiri_wrap_xml_document(VALUE klass, xmlDocPtr doc)
 
   VALUE rb_doc = Data_Wrap_Struct(
       klass ? klass : cNokogiriXmlDocument,
-      0,
+      mark,
       dealloc,
       doc
   );

data/ext/nokogiri/xml_io.c

@@ -2,12 +2,13 @@
 
 static ID id_read, id_write;
 
-VALUE read_check(VALUE *args) {
+VALUE read_check(VALUE val) {
+  VALUE *args = (VALUE *)val;
   return rb_funcall(args[0], id_read, 1, args[1]);
 }
 
-VALUE read_failed(void) {
-	return Qundef;
+VALUE read_failed(VALUE arg, VALUE exc) {
+  return Qundef;
 }
 
 int io_read_callback(void * ctx, char * buffer, int len) {
@@ -30,12 +31,13 @@ int io_read_callback(void * ctx, char * buffer, int len) {
   return (int)safe_len;
 }
 
-VALUE write_check(VALUE *args) {
+VALUE write_check(VALUE val) {
+  VALUE *args = (VALUE *)val;
   return rb_funcall(args[0], id_write, 1, args[1]);
 }
 
-VALUE write_failed(void) {
-	return Qundef;
+VALUE write_failed(VALUE arg, VALUE exc) {
+  return Qundef;
 }
 
 int io_write_callback(void * ctx, char * buffer, int len) {

data/ext/nokogiri/xml_node.c

@@ -1332,6 +1332,25 @@ static VALUE line(VALUE self)
   return INT2NUM(xmlGetLineNo(node));
 }
 
+/*
+ * call-seq:
+ *  line=(num)
+ *
+ * Sets the line for this Node. num must be less than 65535.
+ */
+static VALUE set_line(VALUE self, VALUE num)
+{
+  xmlNodePtr node;
+  Data_Get_Struct(self, xmlNode, node);
+
+  int value = NUM2INT(num);
+  if (value < 65535) {
+    node->line = value;
+  }
+
+  return num;
+}
+
 /*
  * call-seq:
  *  add_namespace_definition(prefix, href)
@@ -1728,6 +1747,7 @@ void init_xml_node()
   rb_define_method(klass, "create_external_subset", create_external_subset, 3);
   rb_define_method(klass, "pointer_id", pointer_id, 0);
   rb_define_method(klass, "line", line, 0);
+  rb_define_method(klass, "line=", set_line, 1);
   rb_define_method(klass, "content", get_native_content, 0);
   rb_define_method(klass, "native_content=", set_native_content, 1);
   rb_define_method(klass, "lang", get_lang, 0);

data/ext/nokogiri/xml_reader.c

@@ -28,35 +28,24 @@ static int has_attributes(xmlTextReaderPtr reader)
 static void Nokogiri_xml_node_namespaces(xmlNodePtr node, VALUE attr_hash)
 {
   xmlNsPtr ns;
-  static char buffer[XMLNS_BUFFER_LEN] ;
-  char *key ;
-  size_t keylen ;
+  VALUE key;
 
   if (node->type != XML_ELEMENT_NODE) return ;
 
   ns = node->nsDef;
   while (ns != NULL) {
 
-    keylen = XMLNS_PREFIX_LEN + (ns->prefix ? (strlen((const char*)ns->prefix) + 1) : 0) ;
-    if (keylen > XMLNS_BUFFER_LEN) {
-      key = (char*)malloc(keylen) ;
-    } else {
-      key = buffer ;
-    }
-
+    key = rb_enc_str_new_cstr(XMLNS_PREFIX, rb_utf8_encoding());
     if (ns->prefix) {
-      sprintf(key, "%s:%s", XMLNS_PREFIX, ns->prefix);
-    } else {
-      sprintf(key, "%s", XMLNS_PREFIX);
+      rb_str_cat_cstr(key, ":");
+      rb_str_cat_cstr(key, (const char*)ns->prefix);
     }
 
+    key = rb_str_conv_enc(key, rb_utf8_encoding(), rb_default_internal_encoding());
     rb_hash_aset(attr_hash,
-        NOKOGIRI_STR_NEW2(key),
+        key,
         (ns->href ? NOKOGIRI_STR_NEW2(ns->href) : Qnil)
     );
-    if (key != buffer) {
-      free(key);
-    }
     ns = ns->next ;
   }
 }

data/ext/nokogiri/xml_schema.c

@@ -133,6 +133,31 @@ static VALUE read_memory(VALUE klass, VALUE content)
   return rb_schema;
 }
 
+/* Schema creation will remove and deallocate "blank" nodes.
+ * If those blank nodes have been exposed to Ruby, they could get freed
+ * out from under the VALUE pointer.  This function checks to see if any of
+ * those nodes have been exposed to Ruby, and if so we should raise an exception.
+ */
+static int has_blank_nodes_p(VALUE cache)
+{
+    long i;
+
+    if (NIL_P(cache)) {
+        return 0;
+    }
+
+    for (i = 0; i < RARRAY_LEN(cache); i++) {
+        xmlNodePtr node;
+        VALUE element = rb_ary_entry(cache, i);
+        Data_Get_Struct(element, xmlNode, node);
+        if (xmlIsBlankNode(node)) {
+            return 1;
+        }
+    }
+
+    return 0;
+}
+
 /*
  * call-seq:
  *  from_document(doc)
@@ -152,6 +177,10 @@ static VALUE from_document(VALUE klass, VALUE document)
   /* In case someone passes us a node. ugh. */
   doc = doc->doc;
 
+  if (has_blank_nodes_p(DOC_NODE_CACHE(doc))) {
+    rb_raise(rb_eArgError, "Creating a schema from a document that has blank nodes exposed to Ruby is dangerous");
+  }
+
   ctx = xmlSchemaNewDocParserCtxt(doc);
 
   errors = rb_ary_new();

data/ext/nokogiri/xslt_stylesheet.c

@@ -109,10 +109,6 @@ static VALUE serialize(VALUE self, VALUE xmlobj)
     return rval ;
 }
 
-static void swallow_superfluous_xml_errors(void * userdata, xmlErrorPtr error, ...)
-{
-}
-
 /*
  *  call-seq:
  *    transform(document, params = [])

data/lib/nokogiri.rb

@@ -1,28 +1,11 @@
 # -*- coding: utf-8 -*-
+# frozen_string_literal: true
 # Modify the PATH on windows so that the external DLLs will get loaded.
 
 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 'nokogiri/jruby/dependencies'
 end
 
 begin
@@ -54,7 +37,7 @@ require 'nokogiri/html/builder'
 #
 #   # Get a Nokogiri::HTML:Document for the page we’re interested in...
 #
-#   doc = Nokogiri::HTML(open('http://www.google.com/search?q=tenderlove'))
+#   doc = Nokogiri::HTML(URI.open('http://www.google.com/search?q=tenderlove'))
 #
 #   # Do funky things with it using Nokogiri::XML::Node methods...
 #

data/lib/nokogiri/css.rb

@@ -1,3 +1,4 @@
+# frozen_string_literal: true
 require 'nokogiri/css/node'
 require 'nokogiri/css/xpath_visitor'
 x = $-w