nokogiri 1.15.4 → 1.16.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 244e673a87f9559a0aa597c75f7c8848f96f67db9d6fa8b9b21fecc466c472d2
-  data.tar.gz: f046f870e270f8daa484bb16fc8915a7228dfee3d6c61220e78fea753a2250ee
+  metadata.gz: eff8843eb712576ae46f11395a29e0daf5b2b99fbec124a604c784281fab2f69
+  data.tar.gz: c4aa6b876e82af3b9d6620cda1040130e17e493d80e7dc84c6c7715d7783d1a1
 SHA512:
-  metadata.gz: bce186e13d536128d231590d143bd40db88cf1c3008e8d0775b411b1ea7e26eb27ff81af47695b0b073a603bdc111361e56626dbdc7ae43f0f419f06b7128f26
-  data.tar.gz: 252a25c25d530d20012094fa6ba73c87823ca40dd73ec42cde8f579ff14a5c8bc8a3e799e37c9e5848bcc2c4d4ae3d94b88f8b0e6d0771136549bbfd5bd4b7c8
+  metadata.gz: 350715c0687e8d463bc56b666f2113f21ec778998d4862ae88c13a6dbe5ab471ffff1c123870a5ceb9934b0e645ce9e01e25e9242c140244e234076a47cd1922
+  data.tar.gz: 5f617afb7c937a1f321d2f11c13995bd26a7344c9137a63198c88e56512f9cbb8d86d1cb4184cd67dbca70fa380fc2e35cbcd49e487bf73bde323312db547cff

data/Gemfile

@@ -5,35 +5,38 @@ source "https://rubygems.org"
 gemspec
 
 group :development do
+  # ruby 3.4.0-dev removed some gems from the default set
+  #
+  # TODO: we should be able to remove these as our gem dependencies sort it out and we pull them in
+  # transitively.
+  gem "mutex_m"
+
   # bootstrapping
   gem "bundler", "~> 2.3"
-  gem "rake", "= 13.0.6"
+  gem "rake", "13.1.0"
 
   # building extensions
-  gem "rake-compiler", "= 1.2.1"
-  gem "rake-compiler-dock", "= 1.3.0"
-
-  # documentation
-  gem "hoe-markdown", "= 1.4.0"
+  gem "rake-compiler", "1.2.6"
+  gem "rake-compiler-dock", "1.4.0"
 
   # parser generator
   gem "rexical", "= 1.0.7"
 
   # tests
-  gem "minitest", "5.18.0"
-  gem "minitest-reporters", "1.6.0"
-  gem "ruby_memcheck", "1.3.2"
+  gem "minitest", "5.21.2"
+  gem "minitest-parallel_fork", "2.0.0"
+  gem "ruby_memcheck", "2.3.0"
   gem "rubyzip", "~> 2.3.2"
   gem "simplecov", "= 0.21.2"
 
   # rubocop
   if Gem::Requirement.new("~> 3.0").satisfied_by?(Gem::Version.new(RUBY_VERSION))
-    gem "rubocop", "1.51.0"
-    gem "rubocop-minitest", "0.31.0"
+    gem "rubocop", "1.60.2"
+    gem "rubocop-minitest", "0.34.5"
     gem "rubocop-packaging", "0.5.2"
-    gem "rubocop-performance", "1.18.0"
+    gem "rubocop-performance", "1.20.2"
     gem "rubocop-rake", "= 0.6.0"
-    gem "rubocop-shopify", "2.13.0"
+    gem "rubocop-shopify", "2.14.0"
   end
 end
 
@@ -41,5 +44,5 @@ end
 # `bundle config set --local without rdoc`
 # Then re-run `bundle install`.
 group :rdoc do
-  gem "rdoc", "6.5.0"
+  gem "rdoc", "6.6.2"
 end

data/README.md

@@ -55,6 +55,7 @@ There are a few ways to ask exploratory questions:
 
 - The Nokogiri mailing list is active at https://groups.google.com/group/nokogiri-talk
 - Open an issue using the "Help Request" template at https://github.com/sparklemotion/nokogiri/issues
+- Open a discussion at https://github.com/sparklemotion/nokogiri/discussions
 
 Please do not mail the maintainers at their personal addresses.
 
@@ -90,6 +91,8 @@ We bump `Major.Minor.Patch` versions following this guidance:
 - Updating packaged libraries for non-security-related reasons.
 - Dropping support for EOLed Ruby versions. [Some folks find this objectionable](https://github.com/sparklemotion/nokogiri/issues/1568), but [SemVer says this is OK if the public API hasn't changed](https://semver.org/#what-should-i-do-if-i-update-my-own-dependencies-without-changing-the-public-api).
 - Backwards-incompatible changes to internal or private methods and constants. These are detailed in the "Changes" section of each changelog entry.
+- Removal of deprecated methods or parameters, after a generous transition period; usually when those methods or parameters are rarely-used or dangerous to the user. Essentially, removals that do not justify a major version bump.
+
 
 `Patch`:
 
@@ -111,7 +114,7 @@ You can help sponsor the maintainers of this software through one of these organ
 
 Requirements:
 
-- Ruby >= 2.7
+- Ruby >= 3.0
 - JRuby >= 9.4.0.0
 
 

data/dependencies.yml

@@ -1,16 +1,17 @@
+
 libxml2:
-  version: "2.11.5"
-  sha256: "3727b078c360ec69fa869de14bd6f75d7ee8d36987b071e6928d4720a28df3a6"
-  # sha-256 hash provided in https://download.gnome.org/sources/libxml2/2.11/libxml2-2.11.5.sha256sum
+  version: "2.12.5"
+  sha256: "a972796696afd38073e0f59c283c3a2f5a560b5268b4babc391b286166526b21"
+  # sha-256 hash provided in https://download.gnome.org/sources/libxml2/2.12/libxml2-2.12.5.sha256sum
 
 libxslt:
-  version: "1.1.38"
-  sha256: "1f32450425819a09acaff2ab7a5a7f8a2ec7956e505d7beeb45e843d0e1ecab1"
-  # sha-256 hash provided in https://download.gnome.org/sources/libxslt/1.1/libxslt-1.1.38.sha256sum
+  version: "1.1.39"
+  sha256: "2a20ad621148339b0759c4d4e96719362dee64c9a096dbba625ba053846349f0"
+  # sha-256 hash provided in https://download.gnome.org/sources/libxslt/1.1/libxslt-1.1.39.sha256sum
 
 zlib:
-  version: "1.2.13"
-  sha256: "b3a24de97a8fdbc835b9833169501030b8977031bcb54b3b3ac13740f846ab30"
+  version: "1.3"
+  sha256: "ff0ba4c292013dbc27530b3a81e1f9a813cd39de01ca5e0f8bf355702efa593e"
   # SHA-256 hash provided on http://zlib.net/
 
 libiconv:

data/ext/nokogiri/extconf.rb

@@ -443,7 +443,7 @@ def process_recipe(name, version, static_p, cross_p, cacheable_p = true)
     # We use 'host' to set compiler prefix for cross-compiling. Prefer host_alias over host. And
     # prefer i686 (what external dev tools use) to i386 (what ruby's configure.ac emits).
     recipe.host = RbConfig::CONFIG["host_alias"].empty? ? RbConfig::CONFIG["host"] : RbConfig::CONFIG["host_alias"]
-    recipe.host = recipe.host.gsub(/i386/, "i686")
+    recipe.host = recipe.host.gsub("i386", "i686")
 
     recipe.target = File.join(PACKAGE_ROOT_DIR, "ports") if cacheable_p
     recipe.configure_options << "--libdir=#{File.join(recipe.path, "lib")}"
@@ -651,8 +651,7 @@ append_cppflags(ENV["CPPFLAGS"].split) unless ENV["CPPFLAGS"].nil?
 append_ldflags(ENV["LDFLAGS"].split) unless ENV["LDFLAGS"].nil?
 $LIBS = concat_flags($LIBS, ENV["LIBS"])
 
-# nokogumbo code uses C90/C99 features, let's make sure older compilers won't give
-# errors/warnings. see #2302
+# libgumbo uses C90/C99 features, see #2302
 append_cflags(["-std=c99", "-Wno-declaration-after-statement"])
 
 # gumbo html5 serialization is slower with O3, let's make sure we use O2
@@ -919,6 +918,7 @@ else
       "--with-c14n",
       "--with-debug",
       "--with-threads",
+      "--without-tls", # see https://github.com/sparklemotion/nokogiri/issues/3031
       "CPPFLAGS=#{cppflags}",
       "CFLAGS=#{cflags}",
     ]
@@ -1087,7 +1087,6 @@ have_func("xmlRelaxNGSetParserStructuredErrors") # introduced in libxml 2.6.24
 have_func("xmlRelaxNGSetValidStructuredErrors") # introduced in libxml 2.6.21
 have_func("xmlSchemaSetValidStructuredErrors") # introduced in libxml 2.6.23
 have_func("xmlSchemaSetParserStructuredErrors") # introduced in libxml 2.6.23
-have_func("rb_gc_location") # introduced in Ruby 2.7
 have_func("rb_category_warning") # introduced in Ruby 3.0
 
 other_library_versions_string = OTHER_LIBRARY_VERSIONS.map { |k, v| [k, v].join(":") }.join(",")

data/ext/nokogiri/html4_sax_push_parser.c

@@ -32,7 +32,7 @@ native_write(VALUE self, VALUE _chunk, VALUE _last_chunk)
 
   if ((status != 0) && !(ctx->options & XML_PARSE_RECOVER)) {
     // TODO: there appear to be no tests for this block
-    xmlErrorPtr e = xmlCtxtGetLastError(ctx);
+    xmlErrorConstPtr e = xmlCtxtGetLastError(ctx);
     Nokogiri_error_raise(NULL, e);
   }
 

data/ext/nokogiri/nokogiri.h

@@ -66,6 +66,13 @@ xmlNodePtr xmlLastElementChild(xmlNodePtr parent);
 #define XMLNS_PREFIX "xmlns"
 #define XMLNS_PREFIX_LEN 6 /* including either colon or \0 */
 
+#ifndef xmlErrorConstPtr
+#  if LIBXML_VERSION >= 21200
+#    define xmlErrorConstPtr const xmlError *
+#  else
+#    define xmlErrorConstPtr xmlError *
+#  endif
+#endif
 
 #include <ruby.h>
 #include <ruby/st.h>
@@ -227,9 +234,9 @@ void Nokogiri_structured_error_func_save(libxmlStructuredErrorHandlerState *hand
 void Nokogiri_structured_error_func_save_and_set(libxmlStructuredErrorHandlerState *handler_state, void *user_data,
     xmlStructuredErrorFunc handler);
 void Nokogiri_structured_error_func_restore(libxmlStructuredErrorHandlerState *handler_state);
-VALUE Nokogiri_wrap_xml_syntax_error(xmlErrorPtr error);
-void Nokogiri_error_array_pusher(void *ctx, xmlErrorPtr error);
-NORETURN_DECL void Nokogiri_error_raise(void *ctx, xmlErrorPtr error);
+VALUE Nokogiri_wrap_xml_syntax_error(xmlErrorConstPtr error);
+void Nokogiri_error_array_pusher(void *ctx, xmlErrorConstPtr error);
+NORETURN_DECL void Nokogiri_error_raise(void *ctx, xmlErrorConstPtr error);
 void Nokogiri_marshal_xpath_funcall_and_return_values(xmlXPathParserContextPtr ctx, int nargs, VALUE handler,
     const char *function_name) ;
 

data/ext/nokogiri/test_global_handlers.c

@@ -3,7 +3,7 @@
 static VALUE foreign_error_handler_block = Qnil;
 
 static void
-foreign_error_handler(void *user_data, xmlErrorPtr c_error)
+foreign_error_handler(void *user_data, xmlErrorConstPtr c_error)
 {
   rb_funcall(foreign_error_handler_block, rb_intern("call"), 0);
 }

data/ext/nokogiri/xml_cdata.c

@@ -33,8 +33,7 @@ rb_xml_cdata_s_new(int argc, VALUE *argv, VALUE klass)
 
   if (!rb_obj_is_kind_of(rb_document, cNokogiriXmlDocument)) {
     xmlNodePtr deprecated_node_type_arg;
-    // TODO: deprecate allowing Node
-    NOKO_WARN_DEPRECATION("Passing a Node as the first parameter to CDATA.new is deprecated. Please pass a Document instead. This will become an error in a future release of Nokogiri.");
+    NOKO_WARN_DEPRECATION("Passing a Node as the first parameter to CDATA.new is deprecated. Please pass a Document instead. This will become an error in Nokogiri v1.17.0."); // TODO: deprecated in v1.15.3, remove in v1.17.0
     Noko_Node_Get_Struct(rb_document, xmlNode, deprecated_node_type_arg);
     c_document = deprecated_node_type_arg->doc;
   } else {

data/ext/nokogiri/xml_document.c

@@ -74,8 +74,10 @@ dealloc(void *data)
 
   ruby_xfree(doc->_private);
 
+#if defined(__GNUC__) && __GNUC__ >= 5
 #pragma GCC diagnostic push
 #pragma GCC diagnostic ignored "-Wdeprecated-declarations" // xmlDeregisterNodeDefault is deprecated as of libxml2 2.11.0
+#endif
   /*
    * libxml-ruby < 3.0.0 uses xmlDeregisterNodeDefault. If the user is using one of those older
    * versions, the registered callback from libxml-ruby will access the _private pointers set by
@@ -90,7 +92,9 @@ dealloc(void *data)
   if (xmlDeregisterNodeDefaultValue) {
     remove_private((xmlNodePtr)doc);
   }
+#if defined(__GNUC__) && __GNUC__ >= 5
 #pragma GCC diagnostic pop
+#endif
 
   xmlFreeDoc(doc);
 }
@@ -337,7 +341,7 @@ read_io(VALUE klass,
   xmlSetStructuredErrorFunc(NULL, NULL);
 
   if (doc == NULL) {
-    xmlErrorPtr error;
+    xmlErrorConstPtr error;
 
     xmlFreeDoc(doc);
 
@@ -383,7 +387,7 @@ read_memory(VALUE klass,
   xmlSetStructuredErrorFunc(NULL, NULL);
 
   if (doc == NULL) {
-    xmlErrorPtr error;
+    xmlErrorConstPtr error;
 
     xmlFreeDoc(doc);
 
@@ -537,7 +541,7 @@ create_entity(int argc, VALUE *argv, VALUE self)
         );
 
   if (NULL == ptr) {
-    xmlErrorPtr error = xmlGetLastError();
+    xmlErrorConstPtr error = xmlGetLastError();
     if (error) {
       rb_exc_raise(Nokogiri_wrap_xml_syntax_error(error));
     } else {

data/ext/nokogiri/xml_namespace.c

@@ -42,7 +42,6 @@ _xml_namespace_dealloc(void *ptr)
   xmlFree(ns);
 }
 
-#ifdef HAVE_RB_GC_LOCATION
 static void
 _xml_namespace_update_references(void *ptr)
 {
@@ -51,9 +50,6 @@ _xml_namespace_update_references(void *ptr)
     ns->_private = (void *)rb_gc_location((VALUE)ns->_private);
   }
 }
-#else
-#  define _xml_namespace_update_references 0
-#endif
 
 static const rb_data_type_t nokogiri_xml_namespace_type_with_dealloc = {
   .wrap_struct_name = "Nokogiri::XML::Namespace#with_dealloc",

data/ext/nokogiri/xml_node.c

@@ -28,7 +28,6 @@ _xml_node_mark(void *ptr)
   }
 }
 
-#ifdef HAVE_RB_GC_LOCATION
 static void
 _xml_node_update_references(void *ptr)
 {
@@ -38,9 +37,6 @@ _xml_node_update_references(void *ptr)
     node->_private = (void *)rb_gc_location((VALUE)node->_private);
   }
 }
-#else
-#  define _xml_node_update_references 0
-#endif
 
 static const rb_data_type_t nokogiri_node_type = {
   .wrap_struct_name = "Nokogiri::XML::Node",
@@ -961,7 +957,7 @@ internal_subset(VALUE self)
  * - +new_parent_doc+
  *   The new node's parent Document. Defaults to the this node's document.
  *
- * [Returns] The new Nokgiri::XML::Node
+ * [Returns] The new Nokogiri::XML::Node
  */
 static VALUE
 duplicate_node(int argc, VALUE *argv, VALUE self)
@@ -2058,8 +2054,7 @@ rb_xml_node_new(int argc, VALUE *argv, VALUE klass)
     rb_raise(rb_eArgError, "document must be a Nokogiri::XML::Node");
   }
   if (!rb_obj_is_kind_of(rb_document_node, cNokogiriXmlDocument)) {
-    // TODO: deprecate allowing Node
-    NOKO_WARN_DEPRECATION("Passing a Node as the second parameter to Node.new is deprecated. Please pass a Document instead, or prefer an alternative constructor like Node#add_child. This will become an error in a future release of Nokogiri.");
+    NOKO_WARN_DEPRECATION("Passing a Node as the second parameter to Node.new is deprecated. Please pass a Document instead, or prefer an alternative constructor like Node#add_child. This will become an error in Nokogiri v1.17.0."); // TODO: deprecated in v1.13.0, remove in v1.17.0
   }
   Noko_Node_Get_Struct(rb_document_node, xmlNode, c_document_node);
 
@@ -2138,7 +2133,7 @@ process_xincludes(VALUE self, VALUE options)
   xmlSetStructuredErrorFunc(NULL, NULL);
 
   if (rcode < 0) {
-    xmlErrorPtr error;
+    xmlErrorConstPtr error;
 
     error = xmlGetLastError();
     if (error) {
@@ -2174,6 +2169,8 @@ in_context(VALUE self, VALUE _str, VALUE _options)
 
   /* Twiddle global variable because of a bug in libxml2.
    * http://git.gnome.org/browse/libxml2/commit/?id=e20fb5a72c83cbfc8e4a8aa3943c6be8febadab7
+   *
+   * TODO: this is fixed, and HTML_PARSE_NOIMPLIED is defined, in libxml2 2.7.7
    */
 #ifndef HTML_PARSE_NOIMPLIED
   htmlHandleOmittedElem(0);

data/ext/nokogiri/xml_reader.c

@@ -5,8 +5,14 @@ VALUE cNokogiriXmlReader;
 static void
 xml_reader_deallocate(void *data)
 {
+  // free the document separately because we _may_ have triggered preservation by calling
+  // xmlTextReaderCurrentDoc during a read_more.
   xmlTextReaderPtr reader = data;
+  xmlDocPtr doc = xmlTextReaderCurrentDoc(reader);
   xmlFreeTextReader(reader);
+  if (doc) {
+    xmlFreeDoc(doc);
+  }
 }
 
 static const rb_data_type_t xml_reader_type = {
@@ -166,48 +172,6 @@ rb_xml_reader_namespaces(VALUE rb_reader)
   return rb_namespaces ;
 }
 
-/*
-  :call-seq: attribute_nodes() → Array<Nokogiri::XML::Attr>
-
-  Get the attributes of the current node as an Array of XML:Attr
-
-  ⚠ This method is deprecated and unsafe to use. It will be removed in a future version of Nokogiri.
-
-  See related: #attribute_hash, #attributes
- */
-static VALUE
-rb_xml_reader_attribute_nodes(VALUE rb_reader)
-{
-  xmlTextReaderPtr c_reader;
-  xmlNodePtr c_node;
-  VALUE attr_nodes;
-  int j;
-
-  // TODO: deprecated, remove in Nokogiri v1.15, see https://github.com/sparklemotion/nokogiri/issues/2598
-  // After removal, we can also remove all the "node_has_a_document" special handling from xml_node.c
-  NOKO_WARN_DEPRECATION("Reader#attribute_nodes is deprecated and will be removed in a future version of Nokogiri. Please use Reader#attribute_hash instead.");
-
-  TypedData_Get_Struct(rb_reader, xmlTextReader, &xml_reader_type, c_reader);
-
-  if (! has_attributes(c_reader)) {
-    return rb_ary_new() ;
-  }
-
-  c_node = xmlTextReaderExpand(c_reader);
-  if (c_node == NULL) {
-    return Qnil;
-  }
-
-  attr_nodes = noko_xml_node_attrs(c_node);
-
-  /* ensure that the Reader won't be GCed as long as a node is referenced */
-  for (j = 0 ; j < RARRAY_LEN(attr_nodes) ; j++) {
-    rb_iv_set(rb_ary_entry(attr_nodes, j), "@reader", rb_reader);
-  }
-
-  return attr_nodes;
-}
-
 /*
   :call-seq: attribute_hash() → Hash<String ⇒ String>
 
@@ -554,9 +518,10 @@ static VALUE
 read_more(VALUE self)
 {
   xmlTextReaderPtr reader;
-  xmlErrorPtr error;
+  xmlErrorConstPtr error;
   VALUE error_list;
   int ret;
+  xmlDocPtr c_document;
 
   TypedData_Get_Struct(self, xmlTextReader, &xml_reader_type, reader);
 
@@ -566,6 +531,16 @@ read_more(VALUE self)
   ret = xmlTextReaderRead(reader);
   xmlSetStructuredErrorFunc(NULL, NULL);
 
+  c_document = xmlTextReaderCurrentDoc(reader);
+  if (c_document && c_document->encoding == NULL) {
+    VALUE constructor_encoding = rb_iv_get(self, "@encoding");
+    if (RTEST(constructor_encoding)) {
+      c_document->encoding = xmlStrdup(BAD_CAST StringValueCStr(constructor_encoding));
+    } else {
+      c_document->encoding = xmlStrdup(BAD_CAST "UTF-8");
+    }
+  }
+
   if (ret == 1) { return self; }
   if (ret == 0) { return Qnil; }
 
@@ -749,15 +724,18 @@ rb_xml_reader_encoding(VALUE rb_reader)
   const char *parser_encoding;
   VALUE constructor_encoding;
 
+  TypedData_Get_Struct(rb_reader, xmlTextReader, &xml_reader_type, c_reader);
+  parser_encoding = (const char *)xmlTextReaderConstEncoding(c_reader);
+  if (parser_encoding) {
+    return NOKOGIRI_STR_NEW2(parser_encoding);
+  }
+
   constructor_encoding = rb_iv_get(rb_reader, "@encoding");
   if (RTEST(constructor_encoding)) {
     return constructor_encoding;
   }
 
-  TypedData_Get_Struct(rb_reader, xmlTextReader, &xml_reader_type, c_reader);
-  parser_encoding = (const char *)xmlTextReaderConstEncoding(c_reader);
-  if (parser_encoding == NULL) { return Qnil; }
-  return NOKOGIRI_STR_NEW2(parser_encoding);
+  return Qnil;
 }
 
 void
@@ -778,7 +756,6 @@ noko_init_xml_reader(void)
   rb_define_method(cNokogiriXmlReader, "attribute", reader_attribute, 1);
   rb_define_method(cNokogiriXmlReader, "attribute_at", attribute_at, 1);
   rb_define_method(cNokogiriXmlReader, "attribute_count", attribute_count, 0);
-  rb_define_method(cNokogiriXmlReader, "attribute_nodes", rb_xml_reader_attribute_nodes, 0);
   rb_define_method(cNokogiriXmlReader, "attribute_hash", rb_xml_reader_attribute_hash, 0);
   rb_define_method(cNokogiriXmlReader, "attributes?", attributes_eh, 0);
   rb_define_method(cNokogiriXmlReader, "base_uri", rb_xml_reader_base_uri, 0);

data/ext/nokogiri/xml_relax_ng.c

@@ -93,7 +93,7 @@ xml_relax_ng_parse_schema(
   xmlRelaxNGFreeParserCtxt(c_parser_context);
 
   if (NULL == c_schema) {
-    xmlErrorPtr error = xmlGetLastError();
+    xmlErrorConstPtr error = xmlGetLastError();
     if (error) {
       Nokogiri_error_raise(NULL, error);
     } else {

data/ext/nokogiri/xml_sax_parser_context.c

@@ -59,6 +59,10 @@ parse_io(VALUE klass, VALUE io, VALUE encoding)
                                (xmlInputReadCallback)noko_io_read,
                                (xmlInputCloseCallback)noko_io_close,
                                (void *)io, enc);
+  if (!ctxt) {
+    rb_raise(rb_eRuntimeError, "failed to create xml sax parser context");
+  }
+
   if (ctxt->sax) {
     xmlFree(ctxt->sax);
     ctxt->sax = NULL;

data/ext/nokogiri/xml_sax_push_parser.c

@@ -59,7 +59,7 @@ native_write(VALUE self, VALUE _chunk, VALUE _last_chunk)
 
   if (xmlParseChunk(ctx, chunk, size, Qtrue == _last_chunk ? 1 : 0)) {
     if (!(ctx->options & XML_PARSE_RECOVER)) {
-      xmlErrorPtr e = xmlCtxtGetLastError(ctx);
+      xmlErrorConstPtr e = xmlCtxtGetLastError(ctx);
       Nokogiri_error_raise(NULL, e);
     }
   }

data/ext/nokogiri/xml_schema.c

@@ -146,7 +146,7 @@ xml_schema_parse_schema(
   xmlSchemaFreeParserCtxt(c_parser_context);
 
   if (NULL == c_schema) {
-    xmlErrorPtr error = xmlGetLastError();
+    xmlErrorConstPtr error = xmlGetLastError();
     if (error) {
       Nokogiri_error_raise(NULL, error);
     } else {
@@ -222,8 +222,7 @@ rb_xml_schema_s_from_document(int argc, VALUE *argv, VALUE klass)
 
   if (!rb_obj_is_kind_of(rb_document, cNokogiriXmlDocument)) {
     xmlNodePtr deprecated_node_type_arg;
-    // TODO: deprecate allowing Node
-    NOKO_WARN_DEPRECATION("Passing a Node as the first parameter to Schema.from_document is deprecated. Please pass a Document instead. This will become an error in a future release of Nokogiri.");
+    NOKO_WARN_DEPRECATION("Passing a Node as the first parameter to Schema.from_document is deprecated. Please pass a Document instead. This will become an error in Nokogiri v1.17.0."); // TODO: deprecated in v1.15.3, remove in v1.17.0
     Noko_Node_Get_Struct(rb_document, xmlNode, deprecated_node_type_arg);
     c_document = deprecated_node_type_arg->doc;
   } else {

data/ext/nokogiri/xml_syntax_error.c

@@ -26,7 +26,7 @@ Nokogiri_structured_error_func_restore(libxmlStructuredErrorHandlerState *handle
 }
 
 void
-Nokogiri_error_array_pusher(void *ctx, xmlErrorPtr error)
+Nokogiri_error_array_pusher(void *ctx, xmlErrorConstPtr error)
 {
   VALUE list = (VALUE)ctx;
   Check_Type(list, T_ARRAY);
@@ -34,13 +34,13 @@ Nokogiri_error_array_pusher(void *ctx, xmlErrorPtr error)
 }
 
 void
-Nokogiri_error_raise(void *ctx, xmlErrorPtr error)
+Nokogiri_error_raise(void *ctx, xmlErrorConstPtr error)
 {
   rb_exc_raise(Nokogiri_wrap_xml_syntax_error(error));
 }
 
 VALUE
-Nokogiri_wrap_xml_syntax_error(xmlErrorPtr error)
+Nokogiri_wrap_xml_syntax_error(xmlErrorConstPtr error)
 {
   VALUE msg, e, klass;
 

data/ext/nokogiri/xml_text.c

@@ -28,8 +28,7 @@ rb_xml_text_s_new(int argc, VALUE *argv, VALUE klass)
 
   if (!rb_obj_is_kind_of(rb_document, cNokogiriXmlDocument)) {
     xmlNodePtr deprecated_node_type_arg;
-    // TODO: deprecate allowing Node
-    NOKO_WARN_DEPRECATION("Passing a Node as the second parameter to Text.new is deprecated. Please pass a Document instead. This will become an error in a future release of Nokogiri.");
+    NOKO_WARN_DEPRECATION("Passing a Node as the second parameter to Text.new is deprecated. Please pass a Document instead. This will become an error in Nokogiri v1.17.0."); // TODO: deprecated in v1.15.3, remove in v1.17.0
     Noko_Node_Get_Struct(rb_document, xmlNode, deprecated_node_type_arg);
     c_document = deprecated_node_type_arg->doc;
   } else {

data/ext/nokogiri/xml_xpath_context.c

@@ -321,11 +321,8 @@ handler_lookup(void *data, const xmlChar *c_name, const xmlChar *c_ns_uri)
   VALUE rb_handler = (VALUE)data;
   if (rb_respond_to(rb_handler, rb_intern((const char *)c_name))) {
     if (c_ns_uri == NULL) {
-      NOKO_WARN_DEPRECATION(
-        "A custom XPath or CSS handler function named '%s' is being invoked without a namespace."
-        " Please update your query to reference this function as 'nokogiri:%s'."
-        " Invoking custom handler functions without a namespace is deprecated and support will be removed in a future release of Nokogiri.",
-        c_name, c_name);
+      NOKO_WARN_DEPRECATION("A custom XPath or CSS handler function named '%s' is being invoked without a namespace. Please update your query to reference this function as 'nokogiri:%s'. Invoking custom handler functions without a namespace is deprecated and will become an error in Nokogiri v1.17.0.",
+                            c_name, c_name); // deprecated in v1.15.0, remove in v1.17.0
     }
     return method_caller;
   }

data/ext/nokogiri/xslt_stylesheet.c

@@ -71,7 +71,12 @@ Nokogiri_wrap_xslt_stylesheet(xsltStylesheetPtr ss)
  * call-seq:
  *   parse_stylesheet_doc(document)
  *
- * Parse a stylesheet from +document+.
+ * Parse an XSLT::Stylesheet from +document+.
+ *
+ * [Parameters]
+ * - +document+ (Nokogiri::XML::Document) the document to be parsed.
+ *
+ * [Returns] Nokogiri::XSLT::Stylesheet
  */
 static VALUE
 parse_stylesheet_doc(VALUE klass, VALUE xmldocobj)
@@ -104,7 +109,7 @@ parse_stylesheet_doc(VALUE klass, VALUE xmldocobj)
  * call-seq:
  *   serialize(document)
  *
- * Serialize +document+ to an xml string.
+ * Serialize +document+ to an xml string, as specified by the +method+ parameter in the Stylesheet.
  */
 static VALUE
 rb_xslt_stylesheet_serialize(VALUE self, VALUE xmlobj)
@@ -133,7 +138,7 @@ rb_xslt_stylesheet_serialize(VALUE self, VALUE xmlobj)
  *   transform(document)
  *   transform(document, params = {})
  *
- * Apply an XSLT stylesheet to an XML::Document.
+ * Transform an XML::Document as defined by an XSLT::Stylesheet.
  *
  * [Parameters]
  * - +document+ (Nokogiri::XML::Document) the document to be transformed.

data/gumbo-parser/Makefile

@@ -13,6 +13,23 @@ LDFLAGS := -pthread
 
 all: check
 
+oss-fuzz:
+	./fuzzer/build-ossfuzz.sh
+
+fuzzers: fuzzer-normal fuzzer-asan fuzzer-ubsan fuzzer-msan
+
+fuzzer-normal:
+	./fuzzer/build.sh
+
+fuzzer-asan:
+	SANITIZER=asan ./fuzzer/build.sh
+
+fuzzer-ubsan:
+	SANITIZER=ubsan ./fuzzer/build.sh
+
+fuzzer-msan:
+	SANITIZER=msan ./fuzzer/build.sh
+
 # don't try to regenerate ragel or gperf files in CI, that should be a development-only action and
 # the generated files should be committed to SCM
 ifneq ($(CI),true)
@@ -81,6 +98,7 @@ coverage:
 
 clean:
 	$(RM) -r build
+	$(RM) -r fuzzer/build fuzzer/src-* fuzzer/gumbo_corpus
 
 build/src/flags: | build/src
 	@echo 'old_CC := $(CC)' > $@

data/gumbo-parser/src/parser.c

@@ -4826,14 +4826,17 @@ GumboOutput* gumbo_parse_with_options (
       // to a token.
       if (token.type == GUMBO_TOKEN_END_TAG &&
           token.v.end_tag.tag == GUMBO_TAG_UNKNOWN)
+      {
         gumbo_free(token.v.end_tag.name);
+        token.v.end_tag.name = NULL;
+      }
+      if (unlikely(state->_open_elements.length > max_tree_depth)) {
+        parser._output->status = GUMBO_STATUS_TREE_TOO_DEEP;
+        gumbo_debug("Tree depth limit exceeded.\n");
+        break;
+      }
     }
 
-    if (unlikely(state->_open_elements.length > max_tree_depth)) {
-      parser._output->status = GUMBO_STATUS_TREE_TOO_DEEP;
-      gumbo_debug("Tree depth limit exceeded.\n");
-      break;
-    }
 
     ++loop_count;
     assert(loop_count < 1000000000UL);

data/gumbo-parser/src/tokenizer.c

@@ -506,6 +506,7 @@ static void abandon_current_tag(GumboParser* parser) {
   for (unsigned int i = 0; i < tag_state->_attributes.length; ++i) {
     gumbo_destroy_attribute(tag_state->_attributes.data[i]);
   }
+  gumbo_free(tag_state->_name);
   gumbo_free(tag_state->_attributes.data);
   mark_tag_state_as_empty(tag_state);
   gumbo_string_buffer_destroy(&tag_state->_buffer);