herb 0.0.1 → 0.1.0

data/src/include/ast_nodes.h

@@ -0,0 +1,303 @@
+// NOTE: This file is generated by the templates/template.rb script and should not
+// be modified manually. See /Users/marcoroth/Development/herb-release/templates/src/include/ast_nodes.h.erb
+
+#ifndef HERB_AST_NODES_H
+#define HERB_AST_NODES_H
+
+#include <stdbool.h>
+#include <prism.h>
+
+#include "array.h"
+#include "buffer.h"
+#include "position.h"
+#include "location.h"
+#include "token_struct.h"
+#include "analyzed_ruby.h"
+
+typedef enum {
+  AST_DOCUMENT_NODE,
+  AST_LITERAL_NODE,
+  AST_HTML_OPEN_TAG_NODE,
+  AST_HTML_CLOSE_TAG_NODE,
+  AST_HTML_SELF_CLOSE_TAG_NODE,
+  AST_HTML_ELEMENT_NODE,
+  AST_HTML_ATTRIBUTE_VALUE_NODE,
+  AST_HTML_ATTRIBUTE_NAME_NODE,
+  AST_HTML_ATTRIBUTE_NODE,
+  AST_HTML_TEXT_NODE,
+  AST_HTML_COMMENT_NODE,
+  AST_HTML_DOCTYPE_NODE,
+  AST_WHITESPACE_NODE,
+  AST_ERB_CONTENT_NODE,
+  AST_ERB_END_NODE,
+  AST_ERB_ELSE_NODE,
+  AST_ERB_IF_NODE,
+  AST_ERB_BLOCK_NODE,
+  AST_ERB_WHEN_NODE,
+  AST_ERB_CASE_NODE,
+  AST_ERB_WHILE_NODE,
+  AST_ERB_UNTIL_NODE,
+  AST_ERB_FOR_NODE,
+  AST_ERB_RESCUE_NODE,
+  AST_ERB_ENSURE_NODE,
+  AST_ERB_BEGIN_NODE,
+  AST_ERB_UNLESS_NODE,
+} ast_node_type_T;
+
+typedef struct AST_NODE_STRUCT {
+  ast_node_type_T type;
+  location_T* location;
+  // maybe a range too?
+  array_T* errors;
+} AST_NODE_T;
+
+
+typedef struct AST_DOCUMENT_NODE_STRUCT {
+  AST_NODE_T base;
+  array_T* children;
+} AST_DOCUMENT_NODE_T;
+
+typedef struct AST_LITERAL_NODE_STRUCT {
+  AST_NODE_T base;
+  const char* content;
+} AST_LITERAL_NODE_T;
+
+typedef struct AST_HTML_OPEN_TAG_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* tag_name;
+  token_T* tag_closing;
+  array_T* children;
+  bool is_void;
+} AST_HTML_OPEN_TAG_NODE_T;
+
+typedef struct AST_HTML_CLOSE_TAG_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* tag_name;
+  token_T* tag_closing;
+} AST_HTML_CLOSE_TAG_NODE_T;
+
+typedef struct AST_HTML_SELF_CLOSE_TAG_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* tag_name;
+  array_T* attributes;
+  token_T* tag_closing;
+  bool is_void;
+} AST_HTML_SELF_CLOSE_TAG_NODE_T;
+
+typedef struct AST_HTML_ELEMENT_NODE_STRUCT {
+  AST_NODE_T base;
+  struct AST_HTML_OPEN_TAG_NODE_STRUCT* open_tag;
+  token_T* tag_name;
+  array_T* body;
+  struct AST_HTML_CLOSE_TAG_NODE_STRUCT* close_tag;
+  bool is_void;
+} AST_HTML_ELEMENT_NODE_T;
+
+typedef struct AST_HTML_ATTRIBUTE_VALUE_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* open_quote;
+  array_T* children;
+  token_T* close_quote;
+  bool quoted;
+} AST_HTML_ATTRIBUTE_VALUE_NODE_T;
+
+typedef struct AST_HTML_ATTRIBUTE_NAME_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* name;
+} AST_HTML_ATTRIBUTE_NAME_NODE_T;
+
+typedef struct AST_HTML_ATTRIBUTE_NODE_STRUCT {
+  AST_NODE_T base;
+  struct AST_HTML_ATTRIBUTE_NAME_NODE_STRUCT* name;
+  token_T* equals;
+  struct AST_HTML_ATTRIBUTE_VALUE_NODE_STRUCT* value;
+} AST_HTML_ATTRIBUTE_NODE_T;
+
+typedef struct AST_HTML_TEXT_NODE_STRUCT {
+  AST_NODE_T base;
+  const char* content;
+} AST_HTML_TEXT_NODE_T;
+
+typedef struct AST_HTML_COMMENT_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* comment_start;
+  array_T* children;
+  token_T* comment_end;
+} AST_HTML_COMMENT_NODE_T;
+
+typedef struct AST_HTML_DOCTYPE_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  array_T* children;
+  token_T* tag_closing;
+} AST_HTML_DOCTYPE_NODE_T;
+
+typedef struct AST_WHITESPACE_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* value;
+} AST_WHITESPACE_NODE_T;
+
+typedef struct AST_ERB_CONTENT_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  analyzed_ruby_T* analyzed_ruby;
+  bool parsed;
+  bool valid;
+} AST_ERB_CONTENT_NODE_T;
+
+typedef struct AST_ERB_END_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+} AST_ERB_END_NODE_T;
+
+typedef struct AST_ERB_ELSE_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* statements;
+} AST_ERB_ELSE_NODE_T;
+
+typedef struct AST_ERB_IF_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* statements;
+  AST_NODE_T* subsequent;
+  struct AST_ERB_END_NODE_STRUCT* end_node;
+} AST_ERB_IF_NODE_T;
+
+typedef struct AST_ERB_BLOCK_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* body;
+  struct AST_ERB_END_NODE_STRUCT* end_node;
+} AST_ERB_BLOCK_NODE_T;
+
+typedef struct AST_ERB_WHEN_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* statements;
+} AST_ERB_WHEN_NODE_T;
+
+typedef struct AST_ERB_CASE_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* children;
+  array_T* conditions;
+  struct AST_ERB_ELSE_NODE_STRUCT* else_clause;
+  struct AST_ERB_END_NODE_STRUCT* end_node;
+} AST_ERB_CASE_NODE_T;
+
+typedef struct AST_ERB_WHILE_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* statements;
+  struct AST_ERB_END_NODE_STRUCT* end_node;
+} AST_ERB_WHILE_NODE_T;
+
+typedef struct AST_ERB_UNTIL_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* statements;
+  struct AST_ERB_END_NODE_STRUCT* end_node;
+} AST_ERB_UNTIL_NODE_T;
+
+typedef struct AST_ERB_FOR_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* statements;
+  struct AST_ERB_END_NODE_STRUCT* end_node;
+} AST_ERB_FOR_NODE_T;
+
+typedef struct AST_ERB_RESCUE_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* statements;
+  struct AST_ERB_RESCUE_NODE_STRUCT* subsequent;
+} AST_ERB_RESCUE_NODE_T;
+
+typedef struct AST_ERB_ENSURE_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* statements;
+} AST_ERB_ENSURE_NODE_T;
+
+typedef struct AST_ERB_BEGIN_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* statements;
+  struct AST_ERB_RESCUE_NODE_STRUCT* rescue_clause;
+  struct AST_ERB_ELSE_NODE_STRUCT* else_clause;
+  struct AST_ERB_ENSURE_NODE_STRUCT* ensure_clause;
+  struct AST_ERB_END_NODE_STRUCT* end_node;
+} AST_ERB_BEGIN_NODE_T;
+
+typedef struct AST_ERB_UNLESS_NODE_STRUCT {
+  AST_NODE_T base;
+  token_T* tag_opening;
+  token_T* content;
+  token_T* tag_closing;
+  array_T* statements;
+  struct AST_ERB_ELSE_NODE_STRUCT* else_clause;
+  struct AST_ERB_END_NODE_STRUCT* end_node;
+} AST_ERB_UNLESS_NODE_T;
+
+AST_DOCUMENT_NODE_T* ast_document_node_init(array_T* children, position_T* start_position, position_T* end_position, array_T* errors);
+AST_LITERAL_NODE_T* ast_literal_node_init(const char* content, position_T* start_position, position_T* end_position, array_T* errors);
+AST_HTML_OPEN_TAG_NODE_T* ast_html_open_tag_node_init(token_T* tag_opening, token_T* tag_name, token_T* tag_closing, array_T* children, bool is_void, position_T* start_position, position_T* end_position, array_T* errors);
+AST_HTML_CLOSE_TAG_NODE_T* ast_html_close_tag_node_init(token_T* tag_opening, token_T* tag_name, token_T* tag_closing, position_T* start_position, position_T* end_position, array_T* errors);
+AST_HTML_SELF_CLOSE_TAG_NODE_T* ast_html_self_close_tag_node_init(token_T* tag_opening, token_T* tag_name, array_T* attributes, token_T* tag_closing, bool is_void, position_T* start_position, position_T* end_position, array_T* errors);
+AST_HTML_ELEMENT_NODE_T* ast_html_element_node_init(struct AST_HTML_OPEN_TAG_NODE_STRUCT* open_tag, token_T* tag_name, array_T* body, struct AST_HTML_CLOSE_TAG_NODE_STRUCT* close_tag, bool is_void, position_T* start_position, position_T* end_position, array_T* errors);
+AST_HTML_ATTRIBUTE_VALUE_NODE_T* ast_html_attribute_value_node_init(token_T* open_quote, array_T* children, token_T* close_quote, bool quoted, position_T* start_position, position_T* end_position, array_T* errors);
+AST_HTML_ATTRIBUTE_NAME_NODE_T* ast_html_attribute_name_node_init(token_T* name, position_T* start_position, position_T* end_position, array_T* errors);
+AST_HTML_ATTRIBUTE_NODE_T* ast_html_attribute_node_init(struct AST_HTML_ATTRIBUTE_NAME_NODE_STRUCT* name, token_T* equals, struct AST_HTML_ATTRIBUTE_VALUE_NODE_STRUCT* value, position_T* start_position, position_T* end_position, array_T* errors);
+AST_HTML_TEXT_NODE_T* ast_html_text_node_init(const char* content, position_T* start_position, position_T* end_position, array_T* errors);
+AST_HTML_COMMENT_NODE_T* ast_html_comment_node_init(token_T* comment_start, array_T* children, token_T* comment_end, position_T* start_position, position_T* end_position, array_T* errors);
+AST_HTML_DOCTYPE_NODE_T* ast_html_doctype_node_init(token_T* tag_opening, array_T* children, token_T* tag_closing, position_T* start_position, position_T* end_position, array_T* errors);
+AST_WHITESPACE_NODE_T* ast_whitespace_node_init(token_T* value, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_CONTENT_NODE_T* ast_erb_content_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, analyzed_ruby_T* analyzed_ruby, bool parsed, bool valid, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_END_NODE_T* ast_erb_end_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_ELSE_NODE_T* ast_erb_else_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* statements, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_IF_NODE_T* ast_erb_if_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* statements, AST_NODE_T* subsequent, struct AST_ERB_END_NODE_STRUCT* end_node, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_BLOCK_NODE_T* ast_erb_block_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* body, struct AST_ERB_END_NODE_STRUCT* end_node, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_WHEN_NODE_T* ast_erb_when_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* statements, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_CASE_NODE_T* ast_erb_case_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* children, array_T* conditions, struct AST_ERB_ELSE_NODE_STRUCT* else_clause, struct AST_ERB_END_NODE_STRUCT* end_node, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_WHILE_NODE_T* ast_erb_while_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* statements, struct AST_ERB_END_NODE_STRUCT* end_node, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_UNTIL_NODE_T* ast_erb_until_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* statements, struct AST_ERB_END_NODE_STRUCT* end_node, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_FOR_NODE_T* ast_erb_for_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* statements, struct AST_ERB_END_NODE_STRUCT* end_node, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_RESCUE_NODE_T* ast_erb_rescue_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* statements, struct AST_ERB_RESCUE_NODE_STRUCT* subsequent, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_ENSURE_NODE_T* ast_erb_ensure_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* statements, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_BEGIN_NODE_T* ast_erb_begin_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* statements, struct AST_ERB_RESCUE_NODE_STRUCT* rescue_clause, struct AST_ERB_ELSE_NODE_STRUCT* else_clause, struct AST_ERB_ENSURE_NODE_STRUCT* ensure_clause, struct AST_ERB_END_NODE_STRUCT* end_node, position_T* start_position, position_T* end_position, array_T* errors);
+AST_ERB_UNLESS_NODE_T* ast_erb_unless_node_init(token_T* tag_opening, token_T* content, token_T* tag_closing, array_T* statements, struct AST_ERB_ELSE_NODE_STRUCT* else_clause, struct AST_ERB_END_NODE_STRUCT* end_node, position_T* start_position, position_T* end_position, array_T* errors);
+
+const char* ast_node_type_to_string(AST_NODE_T* node);
+const char* ast_node_human_type(AST_NODE_T* node);
+
+#endif

data/src/include/ast_pretty_print.h

@@ -0,0 +1,17 @@
+// NOTE: This file is generated by the templates/template.rb script and should not
+// be modified manually. See /Users/marcoroth/Development/herb-release/templates/src/include/ast_pretty_print.h.erb
+
+#ifndef HERB_AST_PRETTY_PRINT_H
+#define HERB_AST_PRETTY_PRINT_H
+
+#include "ast_nodes.h"
+#include "buffer.h"
+
+void ast_pretty_print_node(
+  AST_NODE_T* node,
+  size_t indent,
+  size_t relative_indent,
+  buffer_T* buffer
+);
+
+#endif

data/src/include/buffer.h

@@ -0,0 +1,36 @@
+#ifndef HERB_BUFFER_H
+#define HERB_BUFFER_H
+
+#include <stdbool.h>
+#include <stdlib.h>
+
+typedef struct BUFFER_STRUCT {
+  char* value;
+  size_t length;
+  size_t capacity;
+} buffer_T;
+
+bool buffer_init(buffer_T* buffer);
+buffer_T buffer_new(void);
+
+bool buffer_increase_capacity(buffer_T* buffer, size_t required_length);
+bool buffer_reserve(buffer_T* buffer, size_t min_capacity);
+
+void buffer_append(buffer_T* buffer, const char* text);
+void buffer_append_with_length(buffer_T* buffer, const char* text, size_t length);
+void buffer_append_char(buffer_T* buffer, char character);
+void buffer_append_repeated(buffer_T* buffer, char character, size_t length);
+void buffer_append_whitespace(buffer_T* buffer, size_t length);
+void buffer_prepend(buffer_T* buffer, const char* text);
+void buffer_concat(buffer_T* destination, buffer_T* source);
+
+char* buffer_value(const buffer_T* buffer);
+
+size_t buffer_length(const buffer_T* buffer);
+size_t buffer_capacity(const buffer_T* buffer);
+size_t buffer_sizeof(void);
+
+void buffer_clear(buffer_T* buffer);
+void buffer_free(buffer_T* buffer);
+
+#endif

data/src/include/errors.h

@@ -0,0 +1,125 @@
+// NOTE: This file is generated by the templates/template.rb script and should not
+// be modified manually. See /Users/marcoroth/Development/herb-release/templates/src/include/errors.h.erb
+
+#ifndef HERB_ERRORS_H
+#define HERB_ERRORS_H
+
+#include "array.h"
+#include "buffer.h"
+#include "errors.h"
+#include "location.h"
+#include "position.h"
+#include "token.h"
+
+typedef enum {
+  UNEXPECTED_ERROR,
+  UNEXPECTED_TOKEN_ERROR,
+  MISSING_OPENING_TAG_ERROR,
+  MISSING_CLOSING_TAG_ERROR,
+  TAG_NAMES_MISMATCH_ERROR,
+  QUOTES_MISMATCH_ERROR,
+  VOID_ELEMENT_CLOSING_TAG_ERROR,
+  UNCLOSED_ELEMENT_ERROR,
+  RUBY_PARSE_ERROR,
+} error_type_T;
+
+typedef struct ERROR_STRUCT {
+  error_type_T type;
+  location_T* location;
+  char* message;
+} ERROR_T;
+
+
+typedef struct {
+  ERROR_T base;
+  const char* description;
+  const char* expected;
+  const char* found;
+} UNEXPECTED_ERROR_T;
+
+typedef struct {
+  ERROR_T base;
+  token_type_T expected_type;
+  token_T* found;
+} UNEXPECTED_TOKEN_ERROR_T;
+
+typedef struct {
+  ERROR_T base;
+  token_T* closing_tag;
+} MISSING_OPENING_TAG_ERROR_T;
+
+typedef struct {
+  ERROR_T base;
+  token_T* opening_tag;
+} MISSING_CLOSING_TAG_ERROR_T;
+
+typedef struct {
+  ERROR_T base;
+  token_T* opening_tag;
+  token_T* closing_tag;
+} TAG_NAMES_MISMATCH_ERROR_T;
+
+typedef struct {
+  ERROR_T base;
+  token_T* opening_quote;
+  token_T* closing_quote;
+} QUOTES_MISMATCH_ERROR_T;
+
+typedef struct {
+  ERROR_T base;
+  token_T* tag_name;
+  const char* expected;
+  const char* found;
+} VOID_ELEMENT_CLOSING_TAG_ERROR_T;
+
+typedef struct {
+  ERROR_T base;
+  token_T* opening_tag;
+} UNCLOSED_ELEMENT_ERROR_T;
+
+typedef struct {
+  ERROR_T base;
+  const char* error_message;
+  const char* diagnostic_id;
+  const char* level;
+} RUBY_PARSE_ERROR_T;
+
+UNEXPECTED_ERROR_T* unexpected_error_init(const char* description, const char* expected, const char* found, position_T* start, position_T* end);
+void append_unexpected_error(const char* description, const char* expected, const char* found, position_T* start, position_T* end, array_T* errors);
+UNEXPECTED_TOKEN_ERROR_T* unexpected_token_error_init(token_type_T expected_type, token_T* found, position_T* start, position_T* end);
+void append_unexpected_token_error(token_type_T expected_type, token_T* found, position_T* start, position_T* end, array_T* errors);
+MISSING_OPENING_TAG_ERROR_T* missing_opening_tag_error_init(token_T* closing_tag, position_T* start, position_T* end);
+void append_missing_opening_tag_error(token_T* closing_tag, position_T* start, position_T* end, array_T* errors);
+MISSING_CLOSING_TAG_ERROR_T* missing_closing_tag_error_init(token_T* opening_tag, position_T* start, position_T* end);
+void append_missing_closing_tag_error(token_T* opening_tag, position_T* start, position_T* end, array_T* errors);
+TAG_NAMES_MISMATCH_ERROR_T* tag_names_mismatch_error_init(token_T* opening_tag, token_T* closing_tag, position_T* start, position_T* end);
+void append_tag_names_mismatch_error(token_T* opening_tag, token_T* closing_tag, position_T* start, position_T* end, array_T* errors);
+QUOTES_MISMATCH_ERROR_T* quotes_mismatch_error_init(token_T* opening_quote, token_T* closing_quote, position_T* start, position_T* end);
+void append_quotes_mismatch_error(token_T* opening_quote, token_T* closing_quote, position_T* start, position_T* end, array_T* errors);
+VOID_ELEMENT_CLOSING_TAG_ERROR_T* void_element_closing_tag_error_init(token_T* tag_name, const char* expected, const char* found, position_T* start, position_T* end);
+void append_void_element_closing_tag_error(token_T* tag_name, const char* expected, const char* found, position_T* start, position_T* end, array_T* errors);
+UNCLOSED_ELEMENT_ERROR_T* unclosed_element_error_init(token_T* opening_tag, position_T* start, position_T* end);
+void append_unclosed_element_error(token_T* opening_tag, position_T* start, position_T* end, array_T* errors);
+RUBY_PARSE_ERROR_T* ruby_parse_error_init(const char* error_message, const char* diagnostic_id, const char* level, position_T* start, position_T* end);
+void append_ruby_parse_error(const char* error_message, const char* diagnostic_id, const char* level, position_T* start, position_T* end, array_T* errors);
+
+void error_init(ERROR_T* error, error_type_T type, position_T* start, position_T* end);
+
+size_t error_sizeof(void);
+error_type_T error_type(ERROR_T* error);
+
+char* error_message(ERROR_T* error);
+
+const char* error_type_to_string(ERROR_T* error);
+const char* error_human_type(ERROR_T* error);
+
+void error_free(ERROR_T* error);
+
+void error_pretty_print(ERROR_T* error, size_t indent, size_t relative_indent, buffer_T* buffer);
+
+void error_pretty_print_array(
+  const char* name, array_T* array, size_t indent, size_t relative_indent, bool last_property,
+  buffer_T* buffer
+);
+
+#endif

data/src/include/extract.h

@@ -0,0 +1,20 @@
+#ifndef HERB_EXTRACT_H
+#define HERB_EXTRACT_H
+
+#include "buffer.h"
+
+typedef enum {
+  HERB_EXTRACT_LANGUAGE_RUBY,
+  HERB_EXTRACT_LANGUAGE_HTML,
+} herb_extract_language_T;
+
+void herb_extract_ruby_to_buffer(const char* source, buffer_T* output);
+void herb_extract_html_to_buffer(const char* source, buffer_T* output);
+
+char* herb_extract_ruby_with_semicolons(const char* source);
+void herb_extract_ruby_to_buffer_with_semicolons(const char* source, buffer_T* output);
+
+char* herb_extract(const char* source, herb_extract_language_T language);
+char* herb_extract_from_file(const char* path, herb_extract_language_T language);
+
+#endif

data/src/include/herb.h

@@ -0,0 +1,32 @@
+#ifndef HERB_H
+#define HERB_H
+
+#include "array.h"
+#include "ast_node.h"
+#include "buffer.h"
+#include "extract.h"
+
+#include <stdint.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+void herb_lex_to_buffer(const char* source, buffer_T* output);
+void herb_lex_json_to_buffer(const char* source, buffer_T* output);
+
+array_T* herb_lex(const char* source);
+array_T* herb_lex_file(const char* path);
+
+AST_DOCUMENT_NODE_T* herb_parse(const char* source);
+
+const char* herb_version(void);
+const char* herb_prism_version(void);
+
+void herb_free_tokens(array_T** tokens);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif

data/src/include/html_util.h

@@ -0,0 +1,13 @@
+#ifndef HERB_HTML_UTIL_H
+#define HERB_HTML_UTIL_H
+
+#include <stdbool.h>
+
+bool is_void_element(const char* tag_name);
+bool is_html4_void_element(const char* tag_name);
+
+char* html_opening_tag_string(const char* tag_name);
+char* html_closing_tag_string(const char* tag_name);
+char* html_self_closing_tag_string(const char* tag_name);
+
+#endif

data/src/include/io.h

@@ -0,0 +1,9 @@
+#ifndef HERB_IO_H
+#define HERB_IO_H
+
+#include <stdio.h>
+#include <stdlib.h>
+
+char* herb_read_file(const char* filename);
+
+#endif

data/src/include/json.h

@@ -0,0 +1,28 @@
+#ifndef HERB_JSON_H
+#define HERB_JSON_H
+
+#include "buffer.h"
+
+void json_start_root_object(buffer_T* json);
+void json_start_root_array(buffer_T* json);
+
+void json_escape_string(buffer_T* json, const char* string);
+
+void json_add_string(buffer_T* json, const char* key, const char* value);
+void json_add_int(buffer_T* json, const char* key, int value);
+void json_add_size_t(buffer_T* json, const char* key, size_t value);
+void json_add_double(buffer_T* json, const char* key, double value);
+void json_add_bool(buffer_T* json, const char* key, int value);
+
+void json_add_raw_string(buffer_T* json, const char* string);
+
+void json_start_object(buffer_T* json, const char* key);
+void json_end_object(buffer_T* json);
+
+void json_start_array(buffer_T* json, const char* key);
+void json_end_array(buffer_T* json);
+
+void json_double_to_string(double value, char* buffer);
+void json_int_to_string(int value, char* buffer);
+
+#endif

data/src/include/lexer.h

@@ -0,0 +1,13 @@
+#ifndef HERB_LEXER_H
+#define HERB_LEXER_H
+
+#include "lexer_struct.h"
+#include "token_struct.h"
+
+lexer_T* lexer_init(const char* source);
+token_T* lexer_next_token(lexer_T* lexer);
+token_T* lexer_error(lexer_T* lexer, const char* message);
+
+void lexer_free(lexer_T* lexer);
+
+#endif

data/src/include/lexer_peek_helpers.h

@@ -0,0 +1,23 @@
+#ifndef HERB_LEXER_PEEK_H
+#define HERB_LEXER_PEEK_H
+
+#include "lexer_struct.h"
+
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+char lexer_peek(const lexer_T* lexer, int offset);
+bool lexer_peek_for_doctype(const lexer_T* lexer, int offset);
+
+bool lexer_peek_for_html_comment_start(const lexer_T* lexer, int offset);
+bool lexer_peek_for_html_comment_end(const lexer_T* lexer, int offset);
+
+bool lexer_peek_erb_close_tag(const lexer_T* lexer, int offset);
+bool lexer_peek_erb_dash_close_tag(const lexer_T* lexer, int offset);
+bool lexer_peek_erb_percent_close_tag(const lexer_T* lexer, int offset);
+bool lexer_peek_erb_end(const lexer_T* lexer, int offset);
+
+char lexer_backtrack(const lexer_T* lexer, int offset);
+
+#endif

data/src/include/lexer_struct.h

@@ -0,0 +1,32 @@
+#ifndef HERB_LEXER_STRUCT_H
+#define HERB_LEXER_STRUCT_H
+
+#include <stdbool.h>
+#include <stdlib.h>
+
+typedef enum {
+  STATE_DATA,
+  STATE_ERB_CONTENT,
+  STATE_ERB_CLOSE,
+} lexer_state_T;
+
+typedef struct LEXER_STRUCT {
+  const char* source;
+  size_t source_length;
+
+  size_t current_line;
+  size_t current_column;
+  size_t current_position;
+
+  size_t previous_line;
+  size_t previous_column;
+  size_t previous_position;
+
+  char current_character;
+  lexer_state_T state;
+  size_t stall_counter;
+  size_t last_position;
+  bool stalled;
+} lexer_T;
+
+#endif

data/src/include/location.h

@@ -0,0 +1,25 @@
+#ifndef HERB_LOCATION_H
+#define HERB_LOCATION_H
+
+#include <stdlib.h>
+
+#include "position.h"
+
+typedef struct LOCATION_STRUCT {
+  position_T* start;
+  position_T* end;
+} location_T;
+
+location_T* location_init(position_T* start, position_T* end);
+location_T* location_from(size_t start_line, size_t start_column, size_t end_line, size_t end_column);
+
+position_T* location_start(location_T* location);
+position_T* location_end_(location_T* location);
+
+size_t location_sizeof(void);
+
+location_T* location_copy(location_T* location);
+
+void location_free(location_T* location);
+
+#endif