nokogumbo 1.1.3 → 1.1.4

checksums.yaml

@@ -1,15 +1,15 @@
 ---
 !binary "U0hBMQ==":
   metadata.gz: !binary |-
-    NTBmOGQzODk2MTllYjczZmQ0NDlmMjFiZTgwNDBlODY0ZGRhYmFhOA==
+    OGU3MmZhYzUyY2IzYTZiN2RhMDk4YWE3NDViMDg5Zjg4NGZkN2YyNA==
   data.tar.gz: !binary |-
-    MDU5ZGZjZjRjNmUyYTBmMGYxOGMxMDgwMDI5YzJmZDFmYmNjZmM3OA==
+    NzRlMGZkOWUwN2I1YmM5ZmY4ZWEyYTZlOTBlMjJhMjM5ZTJiZDJlYQ==
 !binary "U0hBNTEy":
   metadata.gz: !binary |-
-    MDc2MzdiYzdjNmFlMjRmZTdiODFlMzI5MjAxOTdmM2I1ZmZjNGZhMzBmNDYy
-    ZmU1NDM3NDhmMjBlYzQ2MDY2NGFlNzBiNWExNGIzYmExYTc5NGRiMzViYjZh
-    NzQ2NmFhOGY5NzIxMDQ2Nzg2OWViNTFmY2VkYWEzYjI0ZTA4NjE=
+    ZGJiYmM4NGYyY2U5MWFhMjAzMWI4MmE2MjQ3OTQ0OTM0MTMyY2E5MWExNjU0
+    MTJjOTExZTBlZDM3NGNjNjBmNjMwZTA1MjQ2MWFkZjI5ODZkY2Q3NzMzZmFi
+    MWE2ODNkZjA0OGU4M2ZjYjE4NWVkZTdmZjM0MTQ2NGQ3OTQzYjU=
   data.tar.gz: !binary |-
-    MzEwOGJlY2UxOGFhM2M5MzVhNGI3ODJlYWMwNzIzM2U0YTdhY2U4ZDYwNmY4
-    MzAzOGJhY2ZlZGQwYTA0ZWY4NmJkYWE5OWNkYmViNTgwNWU4ZmVjMThkZjVk
-    YWMwMmZjYTA5ZTEyNGMwZjQ3OTAxYTgyMWQyODBmMzY5OTY1OTQ=
+    NTY5NWIwOGNmYjc1ZDcxYzZkMjA0MTUzZTUyYWU2ODIwZTQ0OGE5Mzc0MjUz
+    NTg0Y2Y3ZjRhMjgzYzYzMDE4NmY4NTVkZGQ1ZjMwOTkwM2NiOTcyMTRmOTM0
+    MGNmOWI0OWQ0Y2E4MTdkMmU1MWU0YWM5NTQ2YjU5YzcwYjI2Yzk=

data/gumbo-parser/src/error.c

@@ -135,7 +135,7 @@ static const char* find_next_newline(
 
 GumboError* gumbo_add_error(GumboParser* parser) {
   int max_errors = parser->_options->max_errors;
-  if (max_errors < 0 && parser->_output->errors.length >= max_errors) {
+  if (max_errors >= 0 && parser->_output->errors.length >= max_errors) {
     return NULL;
   }
   GumboError* error = gumbo_parser_allocate(parser, sizeof(GumboError));

data/gumbo-parser/src/gumbo.h

@@ -366,7 +366,7 @@ GumboTag gumbo_tag_enum(const char* tagname);
 /**
  * Attribute namespaces.
  * HTML includes special handling for XLink, XML, and XMLNS namespaces on
- * attributes.  Everything else goes in the generatic "NONE" namespace.
+ * attributes.  Everything else goes in the generic "NONE" namespace.
  */
 typedef enum {
   GUMBO_ATTR_NAMESPACE_NONE,

data/gumbo-parser/src/parser.c

@@ -43,7 +43,7 @@ static void* malloc_wrapper(void* unused, size_t size) {
 }
 
 static void free_wrapper(void* unused, void* ptr) {
-  return free(ptr);
+  free(ptr);
 }
 
 const GumboOptions kGumboDefaultOptions = {
@@ -537,25 +537,6 @@ static bool is_in_static_list(
   return false;
 }
 
-static void push_template_insertion_mode(
-    GumboParser* parser, GumboInsertionMode mode) {
-  gumbo_vector_add(
-      parser, (void*) mode, &parser->_parser_state->_template_insertion_modes);
-}
-
-static void pop_template_insertion_mode(GumboParser* parser) {
-  gumbo_vector_pop(parser, &parser->_parser_state->_template_insertion_modes);
-}
-
-static GumboInsertionMode get_current_template_insertion_mode(
-    GumboParser* parser) {
-  GumboVector* template_insertion_modes =
-      &parser->_parser_state->_template_insertion_modes;
-  assert(template_insertion_modes->length > 0);
-  return (GumboInsertionMode) template_insertion_modes->data[
-      template_insertion_modes->length - 1];
-}
-
 static void set_insertion_mode(GumboParser* parser, GumboInsertionMode mode) {
   parser->_parser_state->_insertion_mode = mode;
 }
@@ -1735,10 +1716,10 @@ static bool maybe_add_doctype_error(
     GumboParser* parser, const GumboToken* token) {
   const GumboTokenDocType* doctype = &token->v.doc_type;
   bool html_doctype = !strcmp(doctype->name, kDoctypeHtml.data);
-  if (!html_doctype ||
-      doctype->has_public_identifier ||
-      (doctype->has_system_identifier && !strcmp(
-          doctype->system_identifier, kSystemIdLegacyCompat.data)) ||
+  if ((!html_doctype ||
+       doctype->has_public_identifier ||
+       (doctype->has_system_identifier && !strcmp(
+          doctype->system_identifier, kSystemIdLegacyCompat.data))) &&
       !(html_doctype && (
           doctype_matches(doctype, &kPublicIdHtml4_0,
                           &kSystemIdRecHtml4_0, true) ||
@@ -2372,6 +2353,13 @@ static bool handle_in_body(GumboParser* parser, GumboToken* token) {
       node = pop_current_node(parser);
     } while (node != state->_open_elements.data[1]);
 
+    // Removing & destroying the body node is going to kill any nodes that have
+    // been added to the list of active formatting elements, and so we should
+    // clear it to prevent a use-after-free if the list of active formatting
+    // elements is reconstructed afterwards.  This may happen if whitespace
+    // follows the </frameset>.
+    clear_active_formatting_elements(parser);
+
     // Remove the body node.  We may want to factor this out into a generic
     // helper, but right now this is the only code that needs to do this.
     GumboVector* children = &parser->_output->root->v.element.children;
@@ -3543,6 +3531,10 @@ static bool handle_after_frameset(GumboParser* parser, GumboToken* token) {
   } else if (tag_is(token, kStartTag, GUMBO_TAG_HTML)) {
     return handle_in_body(parser, token);
   } else if (tag_is(token, kEndTag, GUMBO_TAG_HTML)) {
+    GumboNode* html = parser->_parser_state->_open_elements.data[0];
+    assert(node_tag_is(html, GUMBO_TAG_HTML));
+    record_end_of_element(
+        parser->_parser_state->_current_token, &html->v.element);
     set_insertion_mode(parser, GUMBO_INSERTION_MODE_AFTER_AFTER_FRAMESET);
     return true;
   } else if (tag_is(token, kStartTag, GUMBO_TAG_NOFRAMES)) {

data/gumbo-parser/src/tokenizer.c

@@ -442,7 +442,8 @@ static void finish_token(GumboParser* parser, GumboToken* token) {
   reset_token_start_point(tokenizer);
   token->original_text.length =
       tokenizer->_token_start - token->original_text.data;
-  if (token->original_text.length > 0 && token->original_text.data[token->original_text.length - 1] == '\r') {
+  if (token->original_text.length > 0 &&
+      token->original_text.data[token->original_text.length - 1] == '\r') {
     // The UTF8 iterator will ignore carriage returns in the input stream, which
     // means that the next token may start one past a \r character.  The pointer
     // arithmetic above results in that \r being appended to the original text
@@ -815,6 +816,7 @@ static void finish_attribute_value(GumboParser* parser) {
     // Duplicate attribute name detected in an earlier state, so we have to
     // ignore the value.
     tag_state->_drop_next_attr_value = false;
+    reinitialize_tag_buffer(parser);
     return;
   }
 

data/gumbo-parser/src/utf8.c

@@ -208,6 +208,12 @@ void utf8iterator_init(
 }
 
 void utf8iterator_next(Utf8Iterator* iter) {
+  if (iter->_current == -1) {
+    // If we're already at EOF, bail out before advancing anything to avoid
+    // reading past the end of the buffer.  It's easier to catch this case here
+    // than litter the code with lots of individual checks for EOF.
+    return;
+  }
   iter->_start += iter->_width;
   // We update positions based on the *last* character read, so that the first
   // character following a newline is at column 1 in the next line.

data/gumbo-parser/src/util.c

@@ -36,7 +36,7 @@ void* gumbo_parser_allocate(GumboParser* parser, size_t num_bytes) {
 }
 
 void gumbo_parser_deallocate(GumboParser* parser, void* ptr) {
-  return parser->_options->deallocator(parser->_options->userdata, ptr);
+  parser->_options->deallocator(parser->_options->userdata, ptr);
 }
 
 char* gumbo_copy_stringz(GumboParser* parser, const char* str) {

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: nokogumbo
 version: !ruby/object:Gem::Version
-  version: 1.1.3
+  version: 1.1.4
 platform: ruby
 authors:
 - Sam Ruby
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-11-16 00:00:00.000000000 Z
+date: 2014-03-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri