herb 0.4.2-arm-linux-musl → 0.5.0-arm-linux-musl

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 38baf45d2d2561accd20ac4777e751ae868ea1600d02c48c0c40508a43c660db
-  data.tar.gz: 2b34cb8e1a7ecd60da4a4f681ae214626cfd86ec278d814e012183a042e61d04
+  metadata.gz: d6b246d2644a8da136d49f95d24c0eef630626a29dac5f73dd3fe7e7647eefab
+  data.tar.gz: 1db4506721e5562fc251f4befcb95c8c1c33bed20fd45029f98f93d46878a4a2
 SHA512:
-  metadata.gz: 92db40ae1b357b3059678d19a6f7b57b465b8127f9238b7344b79d726e69f1dcfda972b83bc0ff0a21aebcd342617cd7a7a11f65714260bf5c19873579187d6a
-  data.tar.gz: 463070bea23faaa6545d0f1e77490e555c678a2a1b4826b6defadb2e2e7bbbb494a4f88ae9eea5bd46b797ef410891abb5ec618bd5366c08b83ed9df60bcf33c
+  metadata.gz: a94df98536a8afbb6fa638c0b123b5f326f041116aa8b84d2e7736d05e8445754d86877bb4ac09e336ee034ddb091b8350f8eda4398a53873f15efe4349cc79f
+  data.tar.gz: fdf7c424a37bb7a35a6521ddc3307082cfcd800e6dbbb2383e99358826813275d86acd977e9a7b98cd01e2273abd98664272eea97fac8b66e98d4b919f1a571e

data/README.md

@@ -37,7 +37,7 @@ Herb provides a complete ecosystem of HTML+ERB tooling, designed to simplify and
   Automatic, consistent formatting for HTML+ERB files, reducing manual styling and enforcing a standard across projects. Currently in experimental preview - use with caution on version-controlled files.
 
 - **Herb Linter** ([available now](https://herb-tools.dev/projects/linter)):  
-  Static analysis for your HTML+ERB templates to enforce best practices and quickly identify common mistakes with 17 configurable rules.
+  Static analysis for your HTML+ERB templates to enforce best practices and quickly identify common mistakes with plenty of rules.
 
 You can use Herb programmatically in **Ruby**, as well as in **JavaScript** via Node.js, WebAssembly, or directly in browsers.
 

data/ext/herb/error_helpers.c

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/ext/herb/error_helpers.c.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/ext/herb/error_helpers.c.erb
 
 #include <ruby.h>
 

data/ext/herb/error_helpers.h

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/ext/herb/error_helpers.h.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/ext/herb/error_helpers.h.erb
 
 #ifndef HERB_EXTENSION_ERROR_HELPERS_H
 #define HERB_EXTENSION_ERROR_HELPERS_H

data/ext/herb/nodes.c

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/ext/herb/nodes.c.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/ext/herb/nodes.c.erb
 
 #include <ruby.h>
 
@@ -415,7 +415,7 @@ static VALUE rb_erb_content_node_from_c_struct(AST_ERB_CONTENT_NODE_T* erb_conte
   VALUE erb_content_node_tag_opening = rb_token_from_c_struct(erb_content_node->tag_opening);
   VALUE erb_content_node_content = rb_token_from_c_struct(erb_content_node->content);
   VALUE erb_content_node_tag_closing = rb_token_from_c_struct(erb_content_node->tag_closing);
-  /* #<Herb::Template::AnalyzedRubyField:0x00007ffffed79140 @name="analyzed_ruby", @options={kind: nil}> */
+  /* #<Herb::Template::AnalyzedRubyField:0x00007ffffed60a50 @name="analyzed_ruby", @options={kind: nil}> */
   VALUE erb_content_node_analyzed_ruby = Qnil;
   VALUE erb_content_node_parsed = (erb_content_node->parsed) ? Qtrue : Qfalse;
   VALUE erb_content_node_valid = (erb_content_node->valid) ? Qtrue : Qfalse;

data/ext/herb/nodes.h

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/ext/herb/nodes.h.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/ext/herb/nodes.h.erb
 
 #ifndef HERB_EXTENSION_NODES_H
 #define HERB_EXTENSION_NODES_H

data/lib/herb/ast/nodes.rb

@@ -2,7 +2,7 @@
 # typed: true
 
 # NOTE: This file is generated by the templates/template.rb script and should not be
-# modified manually. See /Users/marcoroth/Development/herb-release-6/templates/lib/herb/ast/nodes.rb.erb
+# modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/lib/herb/ast/nodes.rb.erb
 
 module Herb
   module AST

data/lib/herb/cli.rb

@@ -110,8 +110,8 @@ class Herb::CLI
                   project.no_interactive = no_interactive
                   project.no_log_file = no_log_file
                   project.no_timing = no_timing
-                  project.parse!
-                  exit(0)
+                  has_issues = project.parse!
+                  exit(has_issues ? 1 : 0)
                 when "parse"
                   Herb.parse(file_content)
                 when "lex"

data/lib/herb/errors.rb

@@ -2,7 +2,7 @@
 # typed: true
 
 # NOTE: This file is generated by the templates/template.rb script and should not be
-# modified manually. See /Users/marcoroth/Development/herb-release-6/templates/lib/herb/errors.rb.erb
+# modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/lib/herb/errors.rb.erb
 
 module Herb
   module Errors

data/lib/herb/project.rb

@@ -366,6 +366,8 @@ module Herb
         end
 
         puts "\nResults saved to #{output_file}" unless no_log_file
+
+        problem_files.any?
       ensure
         log.close unless no_log_file
       end

data/lib/herb/version.rb

@@ -2,5 +2,5 @@
 # typed: true
 
 module Herb
-  VERSION = "0.4.2"
+  VERSION = "0.5.0"
 end

data/lib/herb/visitor.rb

@@ -2,7 +2,7 @@
 # typed: true
 
 # NOTE: This file is generated by the templates/template.rb script and should not be
-# modified manually. See /Users/marcoroth/Development/herb-release-6/templates/lib/herb/visitor.rb.erb
+# modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/lib/herb/visitor.rb.erb
 
 module Herb
   class Visitor

data/sig/serialized_ast_errors.rbs

@@ -2,7 +2,7 @@
 # typed: true
 
 # NOTE: This file is generated by the templates/template.rb script and should not be
-# modified manually. See /Users/marcoroth/Development/herb-release-6/templates/sig/serialized_ast_errors.rbs.erb
+# modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/sig/serialized_ast_errors.rbs.erb
 
 module Herb
   type serialized_unexpected_error = serialized_error & {

data/sig/serialized_ast_nodes.rbs

@@ -2,7 +2,7 @@
 # typed: true
 
 # NOTE: This file is generated by the templates/template.rb script and should not be
-# modified manually. See /Users/marcoroth/Development/herb-release-6/templates/sig/serialized_ast_nodes.rbs.erb
+# modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/sig/serialized_ast_nodes.rbs.erb
 
 module Herb
   type serialized_document_node = serialized_node & {

data/src/analyze.c

@@ -50,7 +50,8 @@ static bool analyze_erb_content(const AST_NODE_T* node, void* data) {
     AST_ERB_CONTENT_NODE_T* erb_content_node = (AST_ERB_CONTENT_NODE_T*) node;
 
     const char* opening = erb_content_node->tag_opening->value;
-    if (strcmp(opening, "<%%") != 0 && strcmp(opening, "<%%=") != 0) {
+
+    if (strcmp(opening, "<%%") != 0 && strcmp(opening, "<%%=") != 0 && strcmp(opening, "<%#") != 0) {
       analyzed_ruby_T* analyzed = herb_analyze_ruby(erb_content_node->content->value);
 
       if (false) { pretty_print_analyed_ruby(analyzed, erb_content_node->content->value); }

data/src/ast_nodes.c

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/src/ast_nodes.c.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/src/ast_nodes.c.erb
 
 #include <stdio.h>
 #include <stdbool.h>

data/src/ast_pretty_print.c

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/src/ast_pretty_print.c.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/src/ast_pretty_print.c.erb
 
 #include "include/ast_node.h"
 #include "include/ast_nodes.h"

data/src/errors.c

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/src/errors.c.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/src/errors.c.erb
 
 #include "include/array.h"
 #include "include/errors.h"

data/src/include/ast_nodes.h

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/src/include/ast_nodes.h.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/src/include/ast_nodes.h.erb
 
 #ifndef HERB_AST_NODES_H
 #define HERB_AST_NODES_H

data/src/include/ast_pretty_print.h

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/src/include/ast_pretty_print.h.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/src/include/ast_pretty_print.h.erb
 
 #ifndef HERB_AST_PRETTY_PRINT_H
 #define HERB_AST_PRETTY_PRINT_H

data/src/include/errors.h

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/src/include/errors.h.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/src/include/errors.h.erb
 
 #ifndef HERB_ERRORS_H
 #define HERB_ERRORS_H

data/src/include/parser.h

@@ -5,10 +5,22 @@
 #include "ast_node.h"
 #include "lexer.h"
 
+typedef enum {
+  FOREIGN_CONTENT_UNKNOWN = 0,
+  FOREIGN_CONTENT_SCRIPT,
+  FOREIGN_CONTENT_STYLE,
+  // FOREIGN_CONTENT_RUBY,
+  // FOREIGN_CONTENT_TEMPLATE
+} foreign_content_type_T;
+
+typedef enum { PARSER_STATE_DATA, PARSER_STATE_FOREIGN_CONTENT } parser_state_T;
+
 typedef struct PARSER_STRUCT {
   lexer_T* lexer;
   token_T* current_token;
   array_T* open_tags_stack;
+  parser_state_T state;
+  foreign_content_type_T foreign_content_type;
 } parser_T;
 
 parser_T* parser_init(lexer_T* lexer);

data/src/include/parser_helpers.h

@@ -24,6 +24,15 @@ void parser_append_literal_node_from_buffer(
 
 bool parser_in_svg_context(const parser_T* parser);
 
+foreign_content_type_T parser_get_foreign_content_type(const char* tag_name);
+bool parser_is_foreign_content_tag(const char* tag_name);
+const char* parser_get_foreign_content_closing_tag(foreign_content_type_T type);
+
+void parser_enter_foreign_content(parser_T* parser, foreign_content_type_T type);
+void parser_exit_foreign_content(parser_T* parser);
+
+bool parser_is_expected_closing_tag_name(const char* tag_name, foreign_content_type_T expected_type);
+
 token_T* parser_advance(parser_T* parser);
 token_T* parser_consume_if_present(parser_T* parser, token_type_T type);
 token_T* parser_consume_expected(parser_T* parser, token_type_T type, array_T* array);

data/src/include/token_struct.h

@@ -28,6 +28,7 @@ typedef enum {
   TOKEN_SLASH,       // /
   TOKEN_EQUALS,      // =
   TOKEN_QUOTE,       // ", '
+  TOKEN_BACKTICK,    // `
   TOKEN_DASH,        // -
   TOKEN_UNDERSCORE,  // _
   TOKEN_EXCLAMATION, // !

data/src/include/utf8.h

@@ -0,0 +1,11 @@
+#ifndef HERB_UTF8_H
+#define HERB_UTF8_H
+
+#include <stdbool.h>
+#include <stdlib.h>
+
+int utf8_char_byte_length(unsigned char first_byte);
+int utf8_sequence_length(const char* str, size_t position, size_t max_length);
+bool utf8_is_valid_continuation_byte(unsigned char byte);
+
+#endif

data/src/include/version.h

@@ -1,6 +1,6 @@
 #ifndef HERB_VERSION_H
 #define HERB_VERSION_H
 
-#define HERB_VERSION "0.4.2"
+#define HERB_VERSION "0.5.0"
 
 #endif

data/src/lexer.c

@@ -1,6 +1,7 @@
 #include "include/buffer.h"
 #include "include/lexer_peek_helpers.h"
 #include "include/token.h"
+#include "include/utf8.h"
 #include "include/util.h"
 
 #include <ctype.h>
@@ -84,6 +85,23 @@ static void lexer_advance(lexer_T* lexer) {
   }
 }
 
+static void lexer_advance_utf8_bytes(lexer_T* lexer, int byte_count) {
+  if (byte_count <= 0) { return; }
+
+  if (lexer_has_more_characters(lexer) && !lexer_eof(lexer)) {
+    if (!is_newline(lexer->current_character)) { lexer->current_column++; }
+
+    lexer->current_position += byte_count;
+
+    if (lexer->current_position >= lexer->source_length) {
+      lexer->current_position = lexer->source_length;
+      lexer->current_character = '\0';
+    } else {
+      lexer->current_character = lexer->source[lexer->current_position];
+    }
+  }
+}
+
 static void lexer_advance_by(lexer_T* lexer, const size_t count) {
   for (size_t i = 0; i < count; i++) {
     lexer_advance(lexer);
@@ -116,6 +134,35 @@ static token_T* lexer_advance_current(lexer_T* lexer, const token_type_T type) {
   return lexer_advance_with(lexer, (char[]) { lexer->current_character, '\0' }, type);
 }
 
+static token_T* lexer_advance_utf8_character(lexer_T* lexer, const token_type_T type) {
+  int char_byte_length = utf8_sequence_length(lexer->source, lexer->current_position, lexer->source_length);
+
+  if (char_byte_length <= 1) { return lexer_advance_current(lexer, type); }
+
+  char* utf8_char = malloc(char_byte_length + 1);
+
+  if (!utf8_char) { return lexer_advance_current(lexer, type); }
+
+  for (int i = 0; i < char_byte_length; i++) {
+    if (lexer->current_position + i >= lexer->source_length) {
+      free(utf8_char);
+      return lexer_advance_current(lexer, type);
+    }
+
+    utf8_char[i] = lexer->source[lexer->current_position + i];
+  }
+
+  utf8_char[char_byte_length] = '\0';
+
+  lexer_advance_utf8_bytes(lexer, char_byte_length);
+
+  token_T* token = token_init(utf8_char, type, lexer);
+
+  free(utf8_char);
+
+  return token;
+}
+
 static token_T* lexer_match_and_advance(lexer_T* lexer, const char* value, const token_type_T type) {
   if (strncmp(lexer->source + lexer->current_position, value, strlen(value)) == 0) {
     return lexer_advance_with(lexer, value, type);
@@ -232,7 +279,7 @@ token_T* lexer_next_token(lexer_T* lexer) {
   if (isspace(lexer->current_character)) { return lexer_parse_whitespace(lexer); }
 
   if (lexer->current_character == '\xC2' && lexer_peek(lexer, 1) == '\xA0') {
-    return lexer_advance_with(lexer, "\xC2\xA0", TOKEN_NBSP);
+    return lexer_advance_utf8_character(lexer, TOKEN_NBSP);
   }
 
   switch (lexer->current_character) {
@@ -278,11 +325,12 @@ token_T* lexer_next_token(lexer_T* lexer) {
 
     case '"':
     case '\'': return lexer_advance_current(lexer, TOKEN_QUOTE);
+    case '`': return lexer_advance_current(lexer, TOKEN_BACKTICK);
 
     default: {
       if (isalnum(lexer->current_character)) { return lexer_parse_identifier(lexer); }
 
-      return lexer_advance_current(lexer, TOKEN_CHARACTER);
+      return lexer_advance_utf8_character(lexer, TOKEN_CHARACTER);
     }
   }
 }

data/src/parser.c

@@ -9,6 +9,7 @@
 #include "include/parser_helpers.h"
 #include "include/token.h"
 #include "include/token_matchers.h"
+#include "include/util.h"
 
 #include <stdio.h>
 #include <stdlib.h>
@@ -16,6 +17,7 @@
 #include <strings.h>
 
 static void parser_parse_in_data_state(parser_T* parser, array_T* children, array_T* errors);
+static void parser_parse_foreign_content(parser_T* parser, array_T* children, array_T* errors);
 static AST_ERB_CONTENT_NODE_T* parser_parse_erb_tag(parser_T* parser);
 
 size_t parser_sizeof(void) {
@@ -28,6 +30,8 @@ parser_T* parser_init(lexer_T* lexer) {
   parser->lexer = lexer;
   parser->current_token = lexer_next_token(lexer);
   parser->open_tags_stack = array_init(16);
+  parser->state = PARSER_STATE_DATA;
+  parser->foreign_content_type = FOREIGN_CONTENT_UNKNOWN;
 
   return parser;
 }
@@ -184,14 +188,96 @@ static AST_HTML_TEXT_NODE_T* parser_parse_text_content(parser_T* parser, array_T
 
 static AST_HTML_ATTRIBUTE_NAME_NODE_T* parser_parse_html_attribute_name(parser_T* parser) {
   array_T* errors = array_init(8);
-  token_T* identifier = parser_consume_if_present(parser, TOKEN_IDENTIFIER);
 
-  if (identifier == NULL) { parser_append_unexpected_token_error(parser, TOKEN_IDENTIFIER, errors); }
+  token_T* at_token = parser_consume_if_present(parser, TOKEN_AT);
+  token_T* first_token = NULL;
+
+  if (at_token != NULL) {
+    first_token = parser_consume_if_present(parser, TOKEN_IDENTIFIER);
+
+    if (first_token == NULL) {
+      parser_append_unexpected_token_error(parser, TOKEN_IDENTIFIER, errors);
+
+      AST_HTML_ATTRIBUTE_NAME_NODE_T* attribute_name =
+        ast_html_attribute_name_node_init(at_token, at_token->location->start, at_token->location->end, errors);
+
+      token_free(at_token);
+
+      return attribute_name;
+    }
+  } else {
+    first_token = parser_consume_if_present(parser, TOKEN_IDENTIFIER);
+
+    if (first_token == NULL) {
+      parser_append_unexpected_token_error(parser, TOKEN_IDENTIFIER, errors);
+      AST_HTML_ATTRIBUTE_NAME_NODE_T* attribute_name = ast_html_attribute_name_node_init(NULL, NULL, NULL, errors);
+      return attribute_name;
+    }
+  }
+
+  buffer_T name_buffer = buffer_new();
+
+  position_T* start_position;
+
+  if (at_token != NULL) {
+    buffer_append(&name_buffer, at_token->value);
+    start_position = position_copy(at_token->location->start);
+  } else {
+    start_position = position_copy(first_token->location->start);
+  }
+
+  buffer_append(&name_buffer, first_token->value);
+
+  position_T* end_position = position_copy(first_token->location->end);
+  size_t range_end = first_token->range->to;
+
+  while (parser->current_token->type == TOKEN_CHARACTER && parser->current_token->value
+         && strcmp(parser->current_token->value, ".") == 0) {
+
+    token_T* dot_token = parser_advance(parser);
+
+    buffer_append(&name_buffer, dot_token->value);
+    position_free(end_position);
+
+    end_position = position_copy(dot_token->location->end);
+    range_end = dot_token->range->to;
+
+    token_free(dot_token);
+
+    if (parser->current_token->type == TOKEN_IDENTIFIER) {
+      token_T* next_identifier = parser_advance(parser);
+
+      buffer_append(&name_buffer, next_identifier->value);
+      position_free(end_position);
+
+      end_position = position_copy(next_identifier->location->end);
+      range_end = next_identifier->range->to;
+      token_free(next_identifier);
+    } else {
+      break;
+    }
+  }
+
+  token_T* combined_token = calloc(1, sizeof(token_T));
+  combined_token->value = herb_strdup(name_buffer.value);
+  combined_token->type = TOKEN_IDENTIFIER;
+  combined_token->location =
+    location_from(start_position->line, start_position->column, end_position->line, end_position->column);
+
+  size_t range_start = at_token != NULL ? at_token->range->from : first_token->range->from;
+  combined_token->range = range_init(range_start, range_end);
 
   AST_HTML_ATTRIBUTE_NAME_NODE_T* attribute_name =
-    ast_html_attribute_name_node_init(identifier, identifier->location->start, identifier->location->end, errors);
+    ast_html_attribute_name_node_init(combined_token, start_position, end_position, errors);
+
+  buffer_free(&name_buffer);
+  position_free(start_position);
+  position_free(end_position);
+  token_free(first_token);
 
-  token_free(identifier);
+  if (at_token != NULL) { token_free(at_token); }
+
+  token_free(combined_token);
 
   return attribute_name;
 }
@@ -300,6 +386,30 @@ static AST_HTML_ATTRIBUTE_VALUE_NODE_T* parser_parse_html_attribute_value(parser
   // <div id="home">
   if (token_is(parser, TOKEN_QUOTE)) { return parser_parse_quoted_html_attribute_value(parser, children, errors); }
 
+  if (token_is(parser, TOKEN_BACKTICK)) {
+    token_T* token = parser_advance(parser);
+    position_T* start = position_copy(token->location->start);
+    position_T* end = position_copy(token->location->end);
+
+    append_unexpected_error(
+      "Invalid quote character for HTML attribute",
+      "single quote (') or double quote (\")",
+      "backtick (`)",
+      start,
+      end,
+      errors
+    );
+
+    AST_HTML_ATTRIBUTE_VALUE_NODE_T* value =
+      ast_html_attribute_value_node_init(NULL, children, NULL, false, start, end, errors);
+
+    position_free(start);
+    position_free(end);
+    token_free(token);
+
+    return value;
+  }
+
   token_T* token = parser_advance(parser);
 
   append_unexpected_error(
@@ -329,9 +439,19 @@ static AST_HTML_ATTRIBUTE_VALUE_NODE_T* parser_parse_html_attribute_value(parser
 static AST_HTML_ATTRIBUTE_NODE_T* parser_parse_html_attribute(parser_T* parser) {
   AST_HTML_ATTRIBUTE_NAME_NODE_T* attribute_name = parser_parse_html_attribute_name(parser);
 
+  while (token_is_any_of(parser, TOKEN_WHITESPACE, TOKEN_NEWLINE)) {
+    token_T* whitespace = parser_advance(parser);
+    token_free(whitespace);
+  }
+
   token_T* equals = parser_consume_if_present(parser, TOKEN_EQUALS);
 
   if (equals != NULL) {
+    while (token_is_any_of(parser, TOKEN_WHITESPACE, TOKEN_NEWLINE)) {
+      token_T* whitespace = parser_advance(parser);
+      token_free(whitespace);
+    }
+
     AST_HTML_ATTRIBUTE_VALUE_NODE_T* attribute_value = parser_parse_html_attribute_value(parser);
 
     AST_HTML_ATTRIBUTE_NODE_T* attribute_node = ast_html_attribute_node_init(
@@ -390,10 +510,15 @@ static AST_HTML_OPEN_TAG_NODE_T* parser_parse_html_open_tag(parser_T* parser) {
       continue;
     }
 
+    if (parser->current_token->type == TOKEN_AT) {
+      array_append(children, parser_parse_html_attribute(parser));
+      continue;
+    }
+
     parser_append_unexpected_error(
       parser,
       "Unexpected Token",
-      "TOKEN_IDENTIFIER, TOKEN_ERB_START,TOKEN_WHITESPACE, or TOKEN_NEWLINE",
+      "TOKEN_IDENTIFIER, TOKEN_AT, TOKEN_ERB_START,TOKEN_WHITESPACE, or TOKEN_NEWLINE",
       errors
     );
   }
@@ -441,6 +566,12 @@ static AST_HTML_CLOSE_TAG_NODE_T* parser_parse_html_close_tag(parser_T* parser)
 
   token_T* tag_opening = parser_consume_expected(parser, TOKEN_HTML_TAG_START_CLOSE, errors);
   token_T* tag_name = parser_consume_expected(parser, TOKEN_IDENTIFIER, errors);
+
+  while (token_is_any_of(parser, TOKEN_WHITESPACE, TOKEN_NEWLINE)) {
+    token_T* whitespace = parser_advance(parser);
+    token_free(whitespace);
+  }
+
   token_T* tag_closing = parser_consume_expected(parser, TOKEN_HTML_TAG_END, errors);
 
   if (tag_name != NULL && is_void_element(tag_name->value) && parser_in_svg_context(parser) == false) {
@@ -502,7 +633,13 @@ static AST_HTML_ELEMENT_NODE_T* parser_parse_html_regular_element(
 
   parser_push_open_tag(parser, open_tag->tag_name);
 
-  parser_parse_in_data_state(parser, body, errors);
+  if (open_tag->tag_name->value && parser_is_foreign_content_tag(open_tag->tag_name->value)) {
+    foreign_content_type_T content_type = parser_get_foreign_content_type(open_tag->tag_name->value);
+    parser_enter_foreign_content(parser, content_type);
+    parser_parse_foreign_content(parser, body, errors);
+  } else {
+    parser_parse_in_data_state(parser, body, errors);
+  }
 
   if (!token_is(parser, TOKEN_HTML_TAG_START_CLOSE)) { return parser_handle_missing_close_tag(open_tag, body, errors); }
 
@@ -593,6 +730,83 @@ static AST_ERB_CONTENT_NODE_T* parser_parse_erb_tag(parser_T* parser) {
   return erb_node;
 }
 
+static void parser_parse_foreign_content(parser_T* parser, array_T* children, array_T* errors) {
+  buffer_T content = buffer_new();
+  position_T* start = position_copy(parser->current_token->location->start);
+  const char* expected_closing_tag = parser_get_foreign_content_closing_tag(parser->foreign_content_type);
+
+  if (expected_closing_tag == NULL) {
+    parser_exit_foreign_content(parser);
+    position_free(start);
+    buffer_free(&content);
+
+    return;
+  }
+
+  while (!token_is(parser, TOKEN_EOF)) {
+    if (token_is(parser, TOKEN_ERB_START)) {
+      parser_append_literal_node_from_buffer(parser, &content, children, start);
+
+      AST_ERB_CONTENT_NODE_T* erb_node = parser_parse_erb_tag(parser);
+      array_append(children, erb_node);
+
+      position_free(start);
+      start = position_copy(parser->current_token->location->start);
+
+      continue;
+    }
+
+    if (token_is(parser, TOKEN_HTML_TAG_START_CLOSE)) {
+      size_t saved_position = parser->lexer->current_position;
+      size_t saved_line = parser->lexer->current_line;
+      size_t saved_column = parser->lexer->current_column;
+      size_t saved_previous_position = parser->lexer->previous_position;
+      size_t saved_previous_line = parser->lexer->previous_line;
+      size_t saved_previous_column = parser->lexer->previous_column;
+
+      char saved_char = parser->lexer->current_character;
+      lexer_state_T saved_state = parser->lexer->state;
+
+      token_T* next_token = lexer_next_token(parser->lexer);
+      bool is_potential_match = false;
+
+      if (next_token && next_token->type == TOKEN_IDENTIFIER && next_token->value) {
+        is_potential_match = parser_is_expected_closing_tag_name(next_token->value, parser->foreign_content_type);
+      }
+
+      parser->lexer->current_position = saved_position;
+      parser->lexer->current_line = saved_line;
+      parser->lexer->current_column = saved_column;
+      parser->lexer->previous_position = saved_previous_position;
+      parser->lexer->previous_line = saved_previous_line;
+      parser->lexer->previous_column = saved_previous_column;
+      parser->lexer->current_character = saved_char;
+      parser->lexer->state = saved_state;
+
+      if (next_token) { token_free(next_token); }
+
+      if (is_potential_match) {
+        parser_append_literal_node_from_buffer(parser, &content, children, start);
+        parser_exit_foreign_content(parser);
+
+        position_free(start);
+        buffer_free(&content);
+
+        return;
+      }
+    }
+
+    token_T* token = parser_advance(parser);
+    buffer_append(&content, token->value);
+    token_free(token);
+  }
+
+  parser_append_literal_node_from_buffer(parser, &content, children, start);
+  parser_exit_foreign_content(parser);
+  position_free(start);
+  buffer_free(&content);
+}
+
 static void parser_parse_in_data_state(parser_T* parser, array_T* children, array_T* errors) {
   while (token_is_none_of(parser, TOKEN_HTML_TAG_START_CLOSE, TOKEN_EOF)) {
     if (token_is(parser, TOKEN_ERB_START)) {
@@ -618,12 +832,14 @@ static void parser_parse_in_data_state(parser_T* parser, array_T* children, arra
     if (token_is_any_of(
           parser,
           TOKEN_AMPERSAND,
+          TOKEN_AT,
           TOKEN_CHARACTER,
           TOKEN_COLON,
           TOKEN_DASH,
           TOKEN_EQUALS,
           TOKEN_EXCLAMATION,
           TOKEN_IDENTIFIER,
+          TOKEN_NBSP,
           TOKEN_NEWLINE,
           TOKEN_PERCENT,
           TOKEN_QUOTE,
@@ -639,8 +855,8 @@ static void parser_parse_in_data_state(parser_T* parser, array_T* children, arra
     parser_append_unexpected_error(
       parser,
       "Unexpected token",
-      "TOKEN_ERB_START, TOKEN_HTML_DOCTYPE, TOKEN_HTML_COMMENT_START, TOKEN_IDENTIFIER, TOKEN_WHITESPACE, or "
-      "TOKEN_NEWLINE",
+      "TOKEN_ERB_START, TOKEN_HTML_DOCTYPE, TOKEN_HTML_COMMENT_START, TOKEN_IDENTIFIER, TOKEN_WHITESPACE, "
+      "TOKEN_NBSP, TOKEN_AT, or TOKEN_NEWLINE",
       errors
     );
   }

data/src/parser_helpers.c

@@ -8,6 +8,7 @@
 #include "include/lexer.h"
 #include "include/parser.h"
 #include "include/token.h"
+#include "include/token_matchers.h"
 
 #include <stdio.h>
 #include <strings.h>
@@ -54,6 +55,43 @@ bool parser_in_svg_context(const parser_T* parser) {
   return false;
 }
 
+// ===== Foreign Content Handling =====
+
+foreign_content_type_T parser_get_foreign_content_type(const char* tag_name) {
+  if (tag_name == NULL) { return FOREIGN_CONTENT_UNKNOWN; }
+
+  if (strcasecmp(tag_name, "script") == 0) { return FOREIGN_CONTENT_SCRIPT; }
+  if (strcasecmp(tag_name, "style") == 0) { return FOREIGN_CONTENT_STYLE; }
+
+  return FOREIGN_CONTENT_UNKNOWN;
+}
+
+bool parser_is_foreign_content_tag(const char* tag_name) {
+  return parser_get_foreign_content_type(tag_name) != FOREIGN_CONTENT_UNKNOWN;
+}
+
+const char* parser_get_foreign_content_closing_tag(foreign_content_type_T type) {
+  switch (type) {
+    case FOREIGN_CONTENT_SCRIPT: return "script";
+    case FOREIGN_CONTENT_STYLE: return "style";
+    default: return NULL;
+  }
+}
+
+void parser_enter_foreign_content(parser_T* parser, foreign_content_type_T type) {
+  if (parser == NULL) { return; }
+
+  parser->state = PARSER_STATE_FOREIGN_CONTENT;
+  parser->foreign_content_type = type;
+}
+
+void parser_exit_foreign_content(parser_T* parser) {
+  if (parser == NULL) { return; }
+
+  parser->state = PARSER_STATE_DATA;
+  parser->foreign_content_type = FOREIGN_CONTENT_UNKNOWN;
+}
+
 void parser_append_unexpected_error(parser_T* parser, const char* description, const char* expected, array_T* errors) {
   token_T* token = parser_advance(parser);
 
@@ -166,3 +204,11 @@ void parser_handle_mismatched_tags(
     );
   }
 }
+
+bool parser_is_expected_closing_tag_name(const char* tag_name, foreign_content_type_T expected_type) {
+  const char* expected_tag_name = parser_get_foreign_content_closing_tag(expected_type);
+
+  if (expected_tag_name == NULL || tag_name == NULL) { return false; }
+
+  return strcmp(tag_name, expected_tag_name) == 0;
+}

data/src/token.c

@@ -55,6 +55,7 @@ const char* token_type_to_string(const token_type_T type) {
     case TOKEN_HTML_COMMENT_END: return "TOKEN_HTML_COMMENT_END";
     case TOKEN_EQUALS: return "TOKEN_EQUALS";
     case TOKEN_QUOTE: return "TOKEN_QUOTE";
+    case TOKEN_BACKTICK: return "TOKEN_BACKTICK";
     case TOKEN_DASH: return "TOKEN_DASH";
     case TOKEN_UNDERSCORE: return "TOKEN_UNDERSCORE";
     case TOKEN_EXCLAMATION: return "TOKEN_EXCLAMATION";

data/src/utf8.c

@@ -0,0 +1,46 @@
+#include "include/utf8.h"
+
+// UTF-8 byte patterns:
+//   0xxxxxxx = 1 byte (ASCII)
+//   110xxxxx = 2 bytes
+//   1110xxxx = 3 bytes
+//   11110xxx = 4 bytes
+int utf8_char_byte_length(unsigned char first_byte) {
+  if ((first_byte & 0x80) == 0) {
+    return 1;
+  } else if ((first_byte & 0xE0) == 0xC0) {
+    return 2;
+  } else if ((first_byte & 0xF0) == 0xE0) {
+    return 3;
+  } else if ((first_byte & 0xF8) == 0xF0) {
+    return 4;
+  }
+
+  return 1;
+}
+
+// Continuation bytes have pattern 10xxxxxx
+bool utf8_is_valid_continuation_byte(unsigned char byte) {
+  return (byte & 0xC0) == 0x80;
+}
+
+int utf8_sequence_length(const char* str, size_t position, size_t max_length) {
+  if (position >= max_length) { return 0; }
+
+  unsigned char first_byte = (unsigned char) str[position];
+  int expected_length = utf8_char_byte_length(first_byte);
+
+  if (position + expected_length > max_length) {
+    return 1; // Not enough bytes, treat as single byte
+  }
+
+  if (expected_length > 1) {
+    for (int i = 1; i < expected_length; i++) {
+      if (!utf8_is_valid_continuation_byte((unsigned char) str[position + i])) {
+        return 1; // Invalid continuation byte, treat first byte as single byte
+      }
+    }
+  }
+
+  return expected_length;
+}

data/src/visitor.c

@@ -1,5 +1,5 @@
 // NOTE: This file is generated by the templates/template.rb script and should not
-// be modified manually. See /Users/marcoroth/Development/herb-release-6/templates/src/visitor.c.erb
+// be modified manually. See /Users/marcoroth/Development/herb-release-0.5.0/templates/src/visitor.c.erb
 
 #include <stdio.h>
 

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: herb
 version: !ruby/object:Gem::Version
-  version: 0.4.2
+  version: 0.5.0
 platform: arm-linux-musl
 authors:
 - Marco Roth
 bindir: exe
 cert_chain: []
-date: 2025-07-28 00:00:00.000000000 Z
+date: 2025-08-17 00:00:00.000000000 Z
 dependencies: []
 description: Powerful and seamless HTML-aware ERB parsing and tooling.
 email:
@@ -125,6 +125,7 @@ files:
 - src/include/token.h
 - src/include/token_matchers.h
 - src/include/token_struct.h
+- src/include/utf8.h
 - src/include/util.h
 - src/include/version.h
 - src/include/visitor.h
@@ -144,6 +145,7 @@ files:
 - src/ruby_parser.c
 - src/token.c
 - src/token_matchers.c
+- src/utf8.c
 - src/util.c
 - src/visitor.c
 homepage: https://herb-tools.dev