nokogiri 1.12.5 → 1.14.3

data/gumbo-parser/src/parser.c

@@ -24,7 +24,7 @@
 #include "ascii.h"
 #include "attribute.h"
 #include "error.h"
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 #include "insertion_mode.h"
 #include "macros.h"
 #include "parser.h"
@@ -537,7 +537,7 @@ static GumboInsertionMode get_current_template_insertion_mode (
   if (modes->length == 0) {
     return GUMBO_INSERTION_MODE_INITIAL;
   }
-  return (GumboInsertionMode) modes->data[(modes->length - 1)];
+  return (GumboInsertionMode)(intptr_t) modes->data[(modes->length - 1)];
 }
 
 // Returns true if the specified token is either a start or end tag
@@ -2940,7 +2940,7 @@ static void handle_in_body(GumboParser* parser, GumboToken* token) {
       TAG(DETAILS), TAG(DIALOG), TAG(DIR), TAG(DIV), TAG(DL), TAG(FIELDSET),
       TAG(FIGCAPTION), TAG(FIGURE), TAG(FOOTER), TAG(HEADER), TAG(HGROUP),
       TAG(MAIN), TAG(MENU), TAG(NAV), TAG(OL), TAG(P), TAG(SECTION),
-      TAG(SUMMARY), TAG(UL)
+      TAG(SUMMARY), TAG(UL), TAG(SEARCH)
     })
   ) {
     maybe_implicitly_close_p_tag(parser, token);
@@ -3018,7 +3018,7 @@ static void handle_in_body(GumboParser* parser, GumboToken* token) {
       TAG(CENTER), TAG(DETAILS), TAG(DIALOG), TAG(DIR), TAG(DIV), TAG(DL),
       TAG(FIELDSET), TAG(FIGCAPTION), TAG(FIGURE), TAG(FOOTER), TAG(HEADER),
       TAG(HGROUP), TAG(LISTING), TAG(MAIN), TAG(MENU), TAG(NAV), TAG(OL),
-      TAG(PRE), TAG(SECTION), TAG(SUMMARY), TAG(UL)
+      TAG(PRE), TAG(SECTION), TAG(SUMMARY), TAG(UL), TAG(SEARCH)
     })
   ) {
     GumboTag tag = token->v.end_tag.tag;
@@ -3057,6 +3057,9 @@ static void handle_in_body(GumboParser* parser, GumboToken* token) {
         ignore_token(parser);
         return;
       }
+      // Since we remove the form node without popping, we need to make sure
+      // that we flush any text nodes at the end of the form.
+      maybe_flush_text_node_buffer(parser);
       // This differs from implicitly_close_tags because we remove *only* the
       // <form> element; other nodes are left in scope.
       generate_implied_end_tags(parser, GUMBO_TAG_LAST, NULL);
@@ -3434,7 +3437,7 @@ static void handle_in_table(GumboParser* parser, GumboToken* token) {
      || token->type == GUMBO_TOKEN_WHITESPACE
      || token->type == GUMBO_TOKEN_NULL)
     && node_tag_in_set(get_current_node(parser), &(const TagSet) {
-      TAG(TABLE), TAG(TBODY), TAG(TFOOT), TAG(THEAD), TAG(TR)
+      TAG(TABLE), TAG(TBODY), TAG(TEMPLATE), TAG(TFOOT), TAG(THEAD), TAG(TR)
     })
   ) {
     // The "pending table character tokens" list described in the spec is
@@ -4423,11 +4426,6 @@ static void handle_in_foreign_content(GumboParser* parser, GumboToken* token) {
     /* Parse error */
     parser_add_parse_error(parser, token);
 
-    /*
-     * Fragment case: If the parser was originally created for the HTML
-     * fragment parsing algorithm, then act as described in the "any other
-     * start tag" entry below.
-     */
     while (
       !(
         is_mathml_integration_point(get_current_node(parser))
@@ -4437,12 +4435,6 @@ static void handle_in_foreign_content(GumboParser* parser, GumboToken* token) {
     ) {
       pop_current_node(parser);
     }
-    // XXX: The spec currently says to handle this using the in body insertion
-    // mode rules. That seems wrong. See
-    // <https://github.com/whatwg/html/issues/6808>. Instead, use the current
-    // insertion mode which seems like it works.
-    //
-    // handle_in_body(parser, token);
     handle_html_content(parser, token);
     return;
   }

data/gumbo-parser/src/replacement.h

@@ -2,7 +2,7 @@
 #define GUMBO_REPLACEMENT_H_
 
 #include <stddef.h>
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 
 typedef struct {
   const char *const from;

data/gumbo-parser/src/string_buffer.h

@@ -4,7 +4,7 @@
 #include <stdbool.h>
 #include <stddef.h>
 
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 
 #ifdef __cplusplus
 extern "C" {

data/gumbo-parser/src/string_piece.c

@@ -17,7 +17,7 @@
 
 #include <stddef.h>
 #include <string.h>
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 #include "ascii.h"
 
 bool gumbo_string_equals (

data/gumbo-parser/src/svg_attrs.c

@@ -1,7 +1,7 @@
 /* ANSI-C code produced by gperf version 3.1 */
-/* Command-line: gperf -m100 lib/svg_attrs.gperf  */
+/* Command-line: gperf -m100 src/svg_attrs.gperf  */
 /* Computed positions: -k'1,10,$' */
-/* Filtered by: mk/gperf-filter.sed */
+/* Filtered by: gperf-filter.sed */
 
 #include "replacement.h"
 #include "macros.h"

data/gumbo-parser/src/svg_tags.c

@@ -1,7 +1,7 @@
 /* ANSI-C code produced by gperf version 3.1 */
-/* Command-line: gperf -m100 lib/svg_tags.gperf  */
+/* Command-line: gperf -m100 src/svg_tags.gperf  */
 /* Computed positions: -k'3,7' */
-/* Filtered by: mk/gperf-filter.sed */
+/* Filtered by: gperf-filter.sed */
 
 #include "replacement.h"
 #include "macros.h"

data/gumbo-parser/src/tag.c

@@ -14,7 +14,7 @@
  limitations under the License.
 */
 
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 #include "util.h"
 #include "tag_lookup.h"
 
@@ -172,6 +172,7 @@ static const char kGumboTagNames[GUMBO_TAG_LAST+1][15] = {
     [GUMBO_TAG_TT] = "tt",
     [GUMBO_TAG_RTC] = "rtc",
     [GUMBO_TAG_DIALOG] = "dialog",
+    [GUMBO_TAG_SEARCH] = "search",
 
     [GUMBO_TAG_UNKNOWN] = "",
     [GUMBO_TAG_LAST] = "",

data/gumbo-parser/src/tag_lookup.c

@@ -1,14 +1,14 @@
 /* ANSI-C code produced by gperf version 3.1 */
-/* Command-line: gperf -m100 lib/tag_lookup.gperf  */
+/* Command-line: gperf -m100 src/tag_lookup.gperf  */
 /* Computed positions: -k'1-2,$' */
-/* Filtered by: mk/gperf-filter.sed */
+/* Filtered by: gperf-filter.sed */
 
 #include "tag_lookup.h"
 #include "macros.h"
 #include "ascii.h"
 #include <string.h>
 
-#define TOTAL_KEYWORDS 150
+#define TOTAL_KEYWORDS 151
 #define MIN_WORD_LENGTH 1
 #define MAX_WORD_LENGTH 14
 #define MIN_HASH_VALUE 9
@@ -26,7 +26,7 @@ hash (register const char *str, register size_t len)
       272, 272, 272, 272, 272, 272, 272, 272, 272, 272,
       272, 272, 272, 272, 272, 272, 272, 272, 272, 272,
       272, 272, 272, 272, 272, 272, 272, 272, 272, 272,
-      272, 272, 272, 272, 272, 272, 272, 272, 272,   9,
+      272, 272, 272, 272, 272, 272, 272, 272, 272,  11,
         7,   6,   4,   4,   3,   4,   3,   3, 272, 272,
       272, 272, 272, 272, 272,  70,  83, 152,   7,  16,
        61,  98,   5,  76, 102, 126,  12,  19,  54,  54,
@@ -69,7 +69,7 @@ gumbo_tag_lookup (register const char *str, register size_t len)
   static const unsigned char lengthtable[] =
     {
        0,  0,  0,  0,  0,  0,  0,  0,  0,  1,  0,  0,  0,  2,
-       2,  2,  2,  6,  2,  6,  2,  4,  0,  7,  6,  3,  0,  3,
+       2,  2,  2,  6,  2,  6,  6,  4,  2,  7,  6,  3,  0,  3,
        0,  6,  6,  8,  5,  0,  0,  4,  5,  5,  8,  0,  2,  4,
        5,  2,  0,  5,  4,  2,  0,  7,  0,  8,  5,  0,  0,  0,
        0,  0,  0,  5,  3,  4,  5,  1,  4,  0,  4,  1,  2,  8,
@@ -111,9 +111,9 @@ gumbo_tag_lookup (register const char *str, register size_t len)
       {"spacer", GUMBO_TAG_SPACER},
       {"h2", GUMBO_TAG_H2},
       {"header", GUMBO_TAG_HEADER},
-      {"h1", GUMBO_TAG_H1},
+      {"search", GUMBO_TAG_SEARCH},
       {"head", GUMBO_TAG_HEAD},
-      {(char*)0,GUMBO_TAG_UNKNOWN},
+      {"h1", GUMBO_TAG_H1},
       {"details", GUMBO_TAG_DETAILS},
       {"select", GUMBO_TAG_SELECT},
       {"dir", GUMBO_TAG_DIR},

data/gumbo-parser/src/tag_lookup.gperf

@@ -167,3 +167,4 @@ spacer, GUMBO_TAG_SPACER
 tt, GUMBO_TAG_TT
 rtc, GUMBO_TAG_RTC
 dialog, GUMBO_TAG_DIALOG
+search, GUMBO_TAG_SEARCH

data/gumbo-parser/src/tag_lookup.h

@@ -1,7 +1,7 @@
 #ifndef GUMBO_TAG_LOOKUP_H_
 #define GUMBO_TAG_LOOKUP_H_
 
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 
 typedef struct {
     const char *key;

data/gumbo-parser/src/token_buffer.h

@@ -20,7 +20,7 @@
 #include <stdbool.h>
 #include <stddef.h>
 
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 
 #ifdef __cplusplus
 extern "C" {

data/gumbo-parser/src/tokenizer.c

@@ -50,7 +50,7 @@
 #include "attribute.h"
 #include "char_ref.h"
 #include "error.h"
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 #include "parser.h"
 #include "string_buffer.h"
 #include "token_type.h"

data/gumbo-parser/src/tokenizer.h

@@ -7,7 +7,7 @@
 #include <stdbool.h>
 #include <stddef.h>
 
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 #include "token_type.h"
 #include "tokenizer_states.h"
 

data/gumbo-parser/src/utf8.c

@@ -22,7 +22,7 @@
 #include <string.h>
 
 #include "error.h"
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 #include "parser.h"
 #include "ascii.h"
 #include "vector.h"

data/gumbo-parser/src/utf8.h

@@ -19,7 +19,7 @@
 #include <stdbool.h>
 #include <stddef.h>
 
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 #include "macros.h"
 
 #ifdef __cplusplus

data/gumbo-parser/src/util.c

@@ -19,7 +19,7 @@
 #include <stdlib.h>
 #include <string.h>
 #include "util.h"
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 
 void* gumbo_alloc(size_t size) {
   void* ptr = malloc(size);
@@ -63,6 +63,4 @@ void gumbo_debug(const char* format, ...) {
   va_end(args);
   fflush(stdout);
 }
-#else
-void gumbo_debug(const char* UNUSED_ARG(format), ...) {}
 #endif

data/gumbo-parser/src/util.h

@@ -21,7 +21,11 @@ void* gumbo_realloc(void* ptr, size_t size) RETURNS_NONNULL;
 void gumbo_free(void* ptr);
 
 // Debug wrapper for printf
+#ifdef GUMBO_DEBUG
 void gumbo_debug(const char* format, ...) PRINTF(1);
+#else
+static inline void PRINTF(1) gumbo_debug(const char* UNUSED_ARG(format), ...) {};
+#endif
 
 #ifdef __cplusplus
 }

data/gumbo-parser/src/vector.h

@@ -1,7 +1,7 @@
 #ifndef GUMBO_VECTOR_H_
 #define GUMBO_VECTOR_H_
 
-#include "gumbo.h"
+#include "nokogiri_gumbo.h"
 
 #ifdef __cplusplus
 extern "C" {

data/lib/nokogiri/class_resolver.rb

@@ -0,0 +1,67 @@
+# coding: utf-8
+# frozen_string_literal: true
+
+require "set"
+
+module Nokogiri
+  #
+  # Some classes in Nokogiri are namespaced as a group, for example
+  # Document, DocumentFragment, and Builder.
+  #
+  # It's sometimes necessary to look up the related class, e.g.:
+  #
+  #   XML::Builder → XML::Document
+  #   HTML4::Builder → HTML4::Document
+  #   HTML5::Document → HTML5::DocumentFragment
+  #
+  # This module is included into those key classes who need to do this.
+  #
+  module ClassResolver
+    # #related_class restricts matching namespaces to those matching this set.
+    VALID_NAMESPACES = Set.new(["HTML", "HTML4", "HTML5", "XML"])
+
+    # :call-seq:
+    #   related_class(class_name) → Class
+    #
+    # Find a class constant within the
+    #
+    # Some examples:
+    #
+    #   Nokogiri::XML::Document.new.related_class("DocumentFragment")
+    #   # => Nokogiri::XML::DocumentFragment
+    #   Nokogiri::HTML4::Document.new.related_class("DocumentFragment")
+    #   # => Nokogiri::HTML4::DocumentFragment
+    #
+    # Note this will also work for subclasses that follow the same convention, e.g.:
+    #
+    #   Loofah::HTML::Document.new.related_class("DocumentFragment")
+    #   # => Loofah::HTML::DocumentFragment
+    #
+    # And even if it's a subclass, this will iterate through the superclasses:
+    #
+    #   class ThisIsATopLevelClass < Nokogiri::HTML4::Builder ; end
+    #   ThisIsATopLevelClass.new.related_class("Document")
+    #   # => Nokogiri::HTML4::Document
+    #
+    def related_class(class_name)
+      klass = nil
+      inspecting = self.class
+
+      while inspecting
+        namespace_path = inspecting.name.split("::")[0..-2]
+        inspecting = inspecting.superclass
+
+        next unless VALID_NAMESPACES.include?(namespace_path.last)
+
+        related_class_name = (namespace_path << class_name).join("::")
+        klass = begin
+          Object.const_get(related_class_name)
+        rescue NameError
+          nil
+        end
+        break if klass
+      end
+      klass
+    end
+  end
+end

data/lib/nokogiri/css/node.rb

@@ -1,7 +1,8 @@
 # frozen_string_literal: true
+
 module Nokogiri
   module CSS
-    class Node
+    class Node # :nodoc:
       ALLOW_COMBINATOR_ON_SELF = [:DIRECT_ADJACENT_SELECTOR, :FOLLOWING_SELECTOR, :CHILD_SELECTOR]
 
       # Get the type of this node
@@ -10,25 +11,25 @@ module Nokogiri
       attr_accessor :value
 
       # Create a new Node with +type+ and +value+
-      def initialize type, value
+      def initialize(type, value)
         @type = type
         @value = value
       end
 
       # Accept +visitor+
-      def accept visitor
+      def accept(visitor)
         visitor.send(:"visit_#{type.to_s.downcase}", self)
       end
 
       ###
       # Convert this CSS node to xpath with +prefix+ using +visitor+
-      def to_xpath prefix = '//', visitor = XPathVisitor.new
-        prefix = '.' if ALLOW_COMBINATOR_ON_SELF.include?(type) && value.first.nil?
+      def to_xpath(prefix, visitor)
+        prefix = "." if ALLOW_COMBINATOR_ON_SELF.include?(type) && value.first.nil?
         prefix + visitor.accept(self)
       end
 
       # Find a node by type using +types+
-      def find_by_type types
+      def find_by_type(types)
         matches = []
         matches << self if to_type == types
         @value.each do |v|
@@ -39,9 +40,9 @@ module Nokogiri
 
       # Convert to_type
       def to_type
-        [@type] + @value.map { |n|
+        [@type] + @value.filter_map do |n|
           n.to_type if n.respond_to?(:to_type)
-        }.compact
+        end
       end
 
       # Convert to array