nokogiri 1.12.3-java → 1.13.1-java

data/ext/java/nokogiri/internals/XmlDomParserContext.java

@@ -1,30 +1,24 @@
 package nokogiri.internals;
 
-import static nokogiri.internals.NokogiriHelpers.getNokogiriClass;
-import static nokogiri.internals.NokogiriHelpers.isBlank;
-
-import java.io.IOException;
-import java.util.ArrayList;
-import java.util.List;
-
+import nokogiri.XmlDocument;
+import nokogiri.XmlDtd;
+import nokogiri.XmlSyntaxError;
 import org.apache.xerces.parsers.DOMParser;
-import org.jruby.Ruby;
-import org.jruby.RubyArray;
-import org.jruby.RubyClass;
-import org.jruby.RubyFixnum;
+import org.jruby.*;
 import org.jruby.exceptions.RaiseException;
-import org.jruby.runtime.ThreadContext;
 import org.jruby.runtime.Helpers;
+import org.jruby.runtime.ThreadContext;
 import org.jruby.runtime.builtin.IRubyObject;
 import org.w3c.dom.Document;
 import org.w3c.dom.Node;
 import org.w3c.dom.NodeList;
 import org.xml.sax.SAXException;
 
-import nokogiri.NokogiriService;
-import nokogiri.XmlDocument;
-import nokogiri.XmlDtd;
-import nokogiri.XmlSyntaxError;
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.List;
+
+import static nokogiri.internals.NokogiriHelpers.isBlank;
 
 /**
  * Parser class for XML DOM processing. This class actually parses XML document
@@ -48,7 +42,6 @@ public class XmlDomParserContext extends ParserContext
   protected static final String FEATURE_NOT_EXPAND_ENTITY =
     "http://apache.org/xml/features/dom/create-entity-ref-nodes";
   protected static final String FEATURE_VALIDATION = "http://xml.org/sax/features/validation";
-  private static final String XINCLUDE_FEATURE_ID = "http://apache.org/xml/features/xinclude";
   private static final String SECURITY_MANAGER = "http://apache.org/xml/properties/security-manager";
 
   protected ParserContext.Options options;
@@ -69,17 +62,17 @@ public class XmlDomParserContext extends ParserContext
     this.options = new ParserContext.Options(RubyFixnum.fix2long(options));
     java_encoding = NokogiriHelpers.getValidEncodingOrNull(encoding);
     ruby_encoding = encoding;
-    initErrorHandler();
+    initErrorHandler(runtime);
     initParser(runtime);
   }
 
   protected void
-  initErrorHandler()
+  initErrorHandler(Ruby runtime)
   {
     if (options.recover) {
-      errorHandler = new NokogiriNonStrictErrorHandler(options.noError, options.noWarning);
+      errorHandler = new NokogiriNonStrictErrorHandler(runtime, options.noError, options.noWarning);
     } else {
-      errorHandler = new NokogiriStrictErrorHandler(options.noError, options.noWarning);
+      errorHandler = new NokogiriStrictErrorHandler(runtime, options.noError, options.noWarning);
     }
   }
 
@@ -161,12 +154,10 @@ public class XmlDomParserContext extends ParserContext
   mapErrors(ThreadContext context, NokogiriErrorHandler errorHandler)
   {
     final Ruby runtime = context.runtime;
-    final List<Exception> errors = errorHandler.getErrors();
+    final List<RubyException> errors = errorHandler.getErrors();
     final IRubyObject[] errorsAry = new IRubyObject[errors.size()];
     for (int i = 0; i < errors.size(); i++) {
-      XmlSyntaxError xmlSyntaxError = XmlSyntaxError.createXMLSyntaxError(runtime);
-      xmlSyntaxError.setException(errors.get(i));
-      errorsAry[i] = xmlSyntaxError;
+      errorsAry[i] = errors.get(i);
     }
     return runtime.newArrayNoCopy(errorsAry);
   }

data/ext/nokogiri/extconf.rb

@@ -1,5 +1,8 @@
 # frozen_string_literal: true
-ENV["RC_ARCHS"] = "" if RUBY_PLATFORM =~ /darwin/
+
+# rubocop:disable Style/GlobalVars
+
+ENV["RC_ARCHS"] = "" if RUBY_PLATFORM.include?("darwin")
 
 require "mkmf"
 require "rbconfig"
@@ -12,16 +15,14 @@ PACKAGE_ROOT_DIR = File.expand_path(File.join(File.dirname(__FILE__), "..", ".."
 REQUIRED_LIBXML_VERSION = "2.6.21"
 RECOMMENDED_LIBXML_VERSION = "2.9.3"
 
-# The gem version constraint in the Rakefile is not respected at install time.
-# Keep this version in sync with the one in the Rakefile !
-REQUIRED_MINI_PORTILE_VERSION = "~> 2.6.1"
+REQUIRED_MINI_PORTILE_VERSION = "~> 2.7.0" # keep this version in sync with the one in the gemspec
 REQUIRED_PKG_CONFIG_VERSION = "~> 1.1"
 
 # Keep track of what versions of what libraries we build against
 OTHER_LIBRARY_VERSIONS = {}
 
 NOKOGIRI_HELP_MESSAGE = <<~HELP
-  USAGE: ruby #{$0} [options]
+  USAGE: ruby #{$PROGRAM_NAME} [options]
 
     Flags that are always valid:
 
@@ -175,23 +176,23 @@ def config_system_libraries?
 end
 
 def windows?
-  RbConfig::CONFIG["target_os"] =~ /mingw32|mswin/
+  RbConfig::CONFIG["target_os"].match?(/mingw|mswin/)
 end
 
 def solaris?
-  RbConfig::CONFIG["target_os"] =~ /solaris/
+  RbConfig::CONFIG["target_os"].include?("solaris")
 end
 
 def darwin?
-  RbConfig::CONFIG["target_os"] =~ /darwin/
+  RbConfig::CONFIG["target_os"].include?("darwin")
 end
 
 def openbsd?
-  RbConfig::CONFIG["target_os"] =~ /openbsd/
+  RbConfig::CONFIG["target_os"].include?("openbsd")
 end
 
 def aix?
-  RbConfig::CONFIG["target_os"] =~ /aix/
+  RbConfig::CONFIG["target_os"].include?("aix")
 end
 
 def nix?
@@ -284,17 +285,16 @@ ensure
 end
 
 def abort_could_not_find_library(lib)
-  abort("-----\n#{caller[0]}\n#{lib} is missing. Please locate mkmf.log to investigate how it is failing.\n-----")
+  callers = caller(1..2).join("\n")
+  abort("-----\n#{callers}\n#{lib} is missing. Please locate mkmf.log to investigate how it is failing.\n-----")
 end
 
-def chdir_for_build
+def chdir_for_build(&block)
   # When using rake-compiler-dock on Windows, the underlying Virtualbox shared
   # folders don't support symlinks, but libiconv expects it for a build on
   # Linux. We work around this limitation by using the temp dir for cooking.
-  build_dir = ENV["RCD_HOST_RUBY_PLATFORM"].to_s =~ /mingw|mswin|cygwin/ ? "/tmp" : "."
-  Dir.chdir(build_dir) do
-    yield
-  end
+  build_dir = /mingw|mswin|cygwin/.match?(ENV["RCD_HOST_RUBY_PLATFORM"].to_s) ? "/tmp" : "."
+  Dir.chdir(build_dir, &block)
 end
 
 def sh_export_path(path)
@@ -402,9 +402,9 @@ def iconv_configure_flags
   abort_could_not_find_library("libiconv")
 end
 
-def process_recipe(name, version, static_p, cross_p, cacheable_p=true)
+def process_recipe(name, version, static_p, cross_p, cacheable_p = true)
   require "rubygems"
-  gem("mini_portile2", REQUIRED_MINI_PORTILE_VERSION)
+  gem("mini_portile2", REQUIRED_MINI_PORTILE_VERSION) # gemspec is not respected at install time
   require "mini_portile2"
   message("Using mini_portile version #{MiniPortile::VERSION}\n")
 
@@ -413,9 +413,13 @@ def process_recipe(name, version, static_p, cross_p, cacheable_p=true)
   end
 
   MiniPortile.new(name, version).tap do |recipe|
+    def recipe.port_path
+      "#{@target}/#{RUBY_PLATFORM}/#{@name}/#{@version}"
+    end
+
     recipe.target = File.join(PACKAGE_ROOT_DIR, "ports") if cacheable_p
-    # Prefer host_alias over host in order to use i586-mingw32msvc as
-    # correct compiler prefix for cross build, but use host if not set.
+    # Prefer host_alias over host in order to use the correct compiler prefix for cross build, but
+    # use host if not set.
     recipe.host = RbConfig::CONFIG["host_alias"].empty? ? RbConfig::CONFIG["host"] : RbConfig::CONFIG["host_alias"]
     recipe.configure_options << "--libdir=#{File.join(recipe.path, "lib")}"
 
@@ -462,7 +466,7 @@ def process_recipe(name, version, static_p, cross_p, cacheable_p=true)
     end
 
     if RbConfig::CONFIG["target_cpu"] == "universal"
-      %w[CFLAGS LDFLAGS].each do |key|
+      ["CFLAGS", "LDFLAGS"].each do |key|
         unless env[key].include?("-arch")
           env[key] = concat_flags(env[key], RbConfig::CONFIG["ARCH_FLAG"])
         end
@@ -473,7 +477,7 @@ def process_recipe(name, version, static_p, cross_p, cacheable_p=true)
       "#{key}=#{value.strip}"
     end
 
-    checkpoint = "#{recipe.target}/#{recipe.name}-#{recipe.version}-#{recipe.host}.installed"
+    checkpoint = "#{recipe.target}/#{recipe.name}-#{recipe.version}-#{RUBY_PLATFORM}.installed"
     if File.exist?(checkpoint) && !recipe.source_directory
       message("Building Nokogiri with a packaged version of #{name}-#{version}.\n")
     else
@@ -487,7 +491,7 @@ def process_recipe(name, version, static_p, cross_p, cacheable_p=true)
         message("The following patches are being applied:\n")
 
         recipe.patch_files.each do |patch|
-          message("  - %s\n" % File.basename(patch))
+          message(format("  - %s\n", File.basename(patch)))
         end
       end
 
@@ -656,7 +660,7 @@ else
   if cross_build_p || windows?
     zlib_recipe = process_recipe("zlib", dependencies["zlib"]["version"], static_p, cross_build_p) do |recipe|
       recipe.files = [{
-        url: "http://zlib.net/fossils/#{recipe.name}-#{recipe.version}.tar.gz",
+        url: "https://zlib.net/fossils/#{recipe.name}-#{recipe.version}.tar.gz",
         sha256: dependencies["zlib"]["sha256"],
       }]
       if windows?
@@ -700,7 +704,7 @@ else
           end
 
           def compile
-            if host =~ /darwin/
+            if /darwin/.match?(host)
               execute("compile", "make AR=#{host}-libtool")
             else
               super
@@ -714,10 +718,13 @@ else
       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",
+          url: "https://ftp.gnu.org/pub/gnu/libiconv/#{recipe.name}-#{recipe.version}.tar.gz",
           sha256: dependencies["libiconv"]["sha256"],
         }]
 
+        # The libiconv configure script doesn't accept "arm64" host string but "aarch64"
+        recipe.host = recipe.host.gsub("arm64-apple-darwin", "aarch64-apple-darwin")
+
         cflags = concat_flags(ENV["CFLAGS"], "-O2", "-U_FORTIFY_SOURCE", "-g")
 
         recipe.configure_options += [
@@ -741,12 +748,21 @@ else
       Tools" to open the developer site, download the installer for your OS
       version and run it.
       -----
-      EOM
+    EOM
   end
 
-  unless windows?
-    preserving_globals { local_have_library("z", "gzdopen", "zlib.h") } ||
-      abort("zlib is missing; necessary for building libxml2")
+  if zlib_recipe
+    append_cppflags("-I#{zlib_recipe.path}/include")
+    $LIBPATH = ["#{zlib_recipe.path}/lib"] | $LIBPATH
+    ensure_package_configuration(opt: "zlib", pc: "zlib", lib: "z",
+      headers: "zlib.h", func: "gzdopen")
+  end
+
+  if libiconv_recipe
+    append_cppflags("-I#{libiconv_recipe.path}/include")
+    $LIBPATH = ["#{libiconv_recipe.path}/lib"] | $LIBPATH
+    ensure_package_configuration(opt: "iconv", pc: "iconv", lib: "iconv",
+      headers: "iconv.h", func: "iconv_open")
   end
 
   libxml2_recipe = process_recipe("libxml2", dependencies["libxml2"]["version"], static_p, cross_build_p) do |recipe|
@@ -765,7 +781,6 @@ else
 
     if zlib_recipe
       recipe.configure_options << "--with-zlib=#{zlib_recipe.path}"
-      cflags = concat_flags(cflags, "-I#{zlib_recipe.path}/include")
     end
 
     if libiconv_recipe
@@ -834,9 +849,6 @@ else
   append_cppflags("-DNOKOGIRI_PACKAGED_LIBRARIES")
   append_cppflags("-DNOKOGIRI_PRECOMPILED_LIBRARIES") if cross_build_p
 
-  $LIBPATH = ["#{zlib_recipe.path}/lib"] | $LIBPATH if zlib_recipe
-  $LIBPATH = ["#{libiconv_recipe.path}/lib"] | $LIBPATH if libiconv_recipe
-
   $libs = $libs.shellsplit.tap do |libs|
     [libxml2_recipe, libxslt_recipe].each do |recipe|
       libname = recipe.name[/\Alib(.+)\z/, 1]
@@ -929,7 +941,7 @@ libgumbo_recipe = process_recipe("libgumbo", "1.0.0-nokogiri", static_p, cross_b
 
       env = { "CC" => gcc_cmd, "CFLAGS" => cflags }
       if config_cross_build?
-        if host =~ /darwin/
+        if /darwin/.match?(host)
           env["AR"] = "#{host}-libtool"
           env["ARFLAGS"] = "-o"
         else

data/ext/nokogiri/xml_document.c

@@ -533,59 +533,59 @@ block_caller(void *ctx, xmlNodePtr c_node, xmlNodePtr c_parent_node)
 static VALUE
 rb_xml_document_canonicalize(int argc, VALUE *argv, VALUE self)
 {
-  VALUE mode;
-  VALUE incl_ns;
-  VALUE with_comments;
-  xmlChar **ns;
-  long ns_len, i;
+  VALUE rb_mode;
+  VALUE rb_namespaces;
+  VALUE rb_comments_p;
+  xmlChar **c_namespaces;
 
-  xmlDocPtr doc;
-  xmlOutputBufferPtr buf;
-  xmlC14NIsVisibleCallback cb = NULL;
-  void *ctx = NULL;
+  xmlDocPtr c_doc;
+  xmlOutputBufferPtr c_obuf;
+  xmlC14NIsVisibleCallback c_callback_wrapper = NULL;
+  void *rb_callback = NULL;
 
   VALUE rb_cStringIO;
-  VALUE io;
+  VALUE rb_io;
 
-  rb_scan_args(argc, argv, "03", &mode, &incl_ns, &with_comments);
+  rb_scan_args(argc, argv, "03", &rb_mode, &rb_namespaces, &rb_comments_p);
+  if (!NIL_P(rb_mode)) { Check_Type(rb_mode, T_FIXNUM); }
+  if (!NIL_P(rb_namespaces)) { Check_Type(rb_namespaces, T_ARRAY); }
 
-  Data_Get_Struct(self, xmlDoc, doc);
+  Data_Get_Struct(self, xmlDoc, c_doc);
 
   rb_cStringIO = rb_const_get_at(rb_cObject, rb_intern("StringIO"));
-  io           = rb_class_new_instance(0, 0, rb_cStringIO);
-  buf          = xmlAllocOutputBuffer(NULL);
+  rb_io = rb_class_new_instance(0, 0, rb_cStringIO);
+  c_obuf = xmlAllocOutputBuffer(NULL);
 
-  buf->writecallback = (xmlOutputWriteCallback)noko_io_write;
-  buf->closecallback = (xmlOutputCloseCallback)noko_io_close;
-  buf->context       = (void *)io;
+  c_obuf->writecallback = (xmlOutputWriteCallback)noko_io_write;
+  c_obuf->closecallback = (xmlOutputCloseCallback)noko_io_close;
+  c_obuf->context = (void *)rb_io;
 
   if (rb_block_given_p()) {
-    cb = block_caller;
-    ctx = (void *)rb_block_proc();
+    c_callback_wrapper = block_caller;
+    rb_callback = (void *)rb_block_proc();
   }
 
-  if (NIL_P(incl_ns)) {
-    ns = NULL;
+  if (NIL_P(rb_namespaces)) {
+    c_namespaces = NULL;
   } else {
-    Check_Type(incl_ns, T_ARRAY);
-    ns_len = RARRAY_LEN(incl_ns);
-    ns = calloc((size_t)ns_len + 1, sizeof(xmlChar *));
-    for (i = 0 ; i < ns_len ; i++) {
-      VALUE entry = rb_ary_entry(incl_ns, i);
-      ns[i] = (xmlChar *)StringValueCStr(entry);
+    long ns_len = RARRAY_LEN(rb_namespaces);
+    c_namespaces = calloc((size_t)ns_len + 1, sizeof(xmlChar *));
+    for (int j = 0 ; j < ns_len ; j++) {
+      VALUE entry = rb_ary_entry(rb_namespaces, j);
+      c_namespaces[j] = (xmlChar *)StringValueCStr(entry);
     }
   }
 
+  xmlC14NExecute(c_doc, c_callback_wrapper, rb_callback,
+                 (int)(NIL_P(rb_mode) ? 0 : NUM2INT(rb_mode)),
+                 c_namespaces,
+                 (int)RTEST(rb_comments_p),
+                 c_obuf);
 
-  xmlC14NExecute(doc, cb, ctx,
-                 (int)(NIL_P(mode)        ? 0 : NUM2INT(mode)),
-                 ns,
-                 (int)      RTEST(with_comments),
-                 buf);
-
-  xmlOutputBufferClose(buf);
+  free(c_namespaces);
+  xmlOutputBufferClose(c_obuf);
 
-  return rb_funcall(io, rb_intern("string"), 0);
+  return rb_funcall(rb_io, rb_intern("string"), 0);
 }
 
 VALUE

data/ext/nokogiri/xml_document_fragment.c

@@ -28,8 +28,6 @@ new (int argc, VALUE *argv, VALUE klass)
   rb_node = noko_xml_node_wrap(klass, node);
   rb_obj_call_init(rb_node, argc, argv);
 
-  if (rb_block_given_p()) { rb_yield(rb_node); }
-
   return rb_node;
 }
 

data/ext/nokogiri/xml_dtd.c

@@ -57,9 +57,9 @@ entities(VALUE self)
 
 /*
  * call-seq:
- *   notations
+ *   notations() → Hash<name(String)⇒Notation>
  *
- * Get a hash of the notations for this DTD.
+ * [Returns] All the notations for this DTD in a Hash of Notation +name+ to Notation.
  */
 static VALUE
 notations(VALUE self)

data/ext/nokogiri/xml_encoding_handler.c

@@ -3,70 +3,82 @@
 VALUE cNokogiriEncodingHandler;
 
 
+static void
+_xml_encoding_handler_dealloc(xmlCharEncodingHandlerPtr c_handler)
+{
+  /* make sure iconv handlers are cleaned up and freed */
+  xmlCharEncCloseFunc(c_handler);
+}
+
+
 /*
  * call-seq: Nokogiri::EncodingHandler.[](name)
  *
  * Get the encoding handler for +name+
  */
 static VALUE
-get(VALUE klass, VALUE key)
+rb_xml_encoding_handler_s_get(VALUE klass, VALUE key)
 {
   xmlCharEncodingHandlerPtr handler;
 
   handler = xmlFindCharEncodingHandler(StringValueCStr(key));
   if (handler) {
-    return Data_Wrap_Struct(klass, NULL, NULL, handler);
+    return Data_Wrap_Struct(klass, NULL, _xml_encoding_handler_dealloc, handler);
   }
 
   return Qnil;
 }
 
+
 /*
  * call-seq: Nokogiri::EncodingHandler.delete(name)
  *
  * Delete the encoding alias named +name+
  */
 static VALUE
-delete (VALUE klass, VALUE name)
+rb_xml_encoding_handler_s_delete(VALUE klass, VALUE name)
 {
   if (xmlDelEncodingAlias(StringValueCStr(name))) { return Qnil; }
 
   return Qtrue;
 }
 
+
 /*
  * call-seq: Nokogiri::EncodingHandler.alias(from, to)
  *
  * Alias encoding handler with name +from+ to name +to+
  */
 static VALUE
-alias(VALUE klass, VALUE from, VALUE to)
+rb_xml_encoding_handler_s_alias(VALUE klass, VALUE from, VALUE to)
 {
   xmlAddEncodingAlias(StringValueCStr(from), StringValueCStr(to));
 
   return to;
 }
 
+
 /*
  * call-seq: Nokogiri::EncodingHandler.clear_aliases!
  *
  * Remove all encoding aliases.
  */
 static VALUE
-clear_aliases(VALUE klass)
+rb_xml_encoding_handler_s_clear_aliases(VALUE klass)
 {
   xmlCleanupEncodingAliases();
 
   return klass;
 }
 
+
 /*
  * call-seq: name
  *
  * Get the name of this EncodingHandler
  */
 static VALUE
-name(VALUE self)
+rb_xml_encoding_handler_name(VALUE self)
 {
   xmlCharEncodingHandlerPtr handler;
 
@@ -75,6 +87,7 @@ name(VALUE self)
   return NOKOGIRI_STR_NEW2(handler->name);
 }
 
+
 void
 noko_init_xml_encoding_handler()
 {
@@ -82,9 +95,10 @@ noko_init_xml_encoding_handler()
 
   rb_undef_alloc_func(cNokogiriEncodingHandler);
 
-  rb_define_singleton_method(cNokogiriEncodingHandler, "[]", get, 1);
-  rb_define_singleton_method(cNokogiriEncodingHandler, "delete", delete, 1);
-  rb_define_singleton_method(cNokogiriEncodingHandler, "alias", alias, 2);
-  rb_define_singleton_method(cNokogiriEncodingHandler, "clear_aliases!", clear_aliases, 0);
-  rb_define_method(cNokogiriEncodingHandler, "name", name, 0);
+  rb_define_singleton_method(cNokogiriEncodingHandler, "[]", rb_xml_encoding_handler_s_get, 1);
+  rb_define_singleton_method(cNokogiriEncodingHandler, "delete", rb_xml_encoding_handler_s_delete, 1);
+  rb_define_singleton_method(cNokogiriEncodingHandler, "alias", rb_xml_encoding_handler_s_alias, 2);
+  rb_define_singleton_method(cNokogiriEncodingHandler, "clear_aliases!", rb_xml_encoding_handler_s_clear_aliases, 0);
+
+  rb_define_method(cNokogiriEncodingHandler, "name", rb_xml_encoding_handler_name, 0);
 }