propane/assets/parser.h.erb

/**
 * @file
 *
 * This file is generated by Propane.
 */

#pragma once

#include <stdint.h>
#include <stddef.h>

/**************************************************************************
 * Public types
 *************************************************************************/

/* Result codes. */
#define <%= @grammar.prefix.upcase %>SUCCESS 0u
#define <%= @grammar.prefix.upcase %>DECODE_ERROR 1u
#define <%= @grammar.prefix.upcase %>UNEXPECTED_INPUT 2u
#define <%= @grammar.prefix.upcase %>UNEXPECTED_TOKEN 3u
#define <%= @grammar.prefix.upcase %>DROP 4u
#define <%= @grammar.prefix.upcase %>EOF 5u
#define <%= @grammar.prefix.upcase %>USER_TERMINATED 6u

/** Token type. */
typedef <%= get_type_for(@grammar.terminate_token_id) %> <%= @grammar.prefix %>token_t;

/** Token IDs. */
<% @grammar.tokens.each_with_index do |token, index| %>
#define TOKEN_<%= token.code_name %> <%= index %>u
<%   unless token.id == index %>
<%     raise "Token ID (#{token.id}) does not match index (#{index}) for token #{token.name}!" %>
<%   end %>
<% end %>
#define INVALID_TOKEN_ID <%= @grammar.invalid_token_id %>u
#define TERMINATE_TOKEN_ID <%= @grammar.terminate_token_id %>u

/** Code point type. */
typedef uint32_t <%= @grammar.prefix %>code_point_t;

/** User header code blocks. */
<%= @grammar.code_blocks.fetch("header", "") %>

<% if @grammar.ast %>
/** Parser values type. */
typedef <%= @grammar.ptype %> <%= @grammar.prefix %>value_t;
<% else %>
/** Parser values type(s). */
typedef union
{
<%   @grammar.ptypes.each do |name, typestring| %>
    <%= typestring %> v_<%= name %>;
<%   end %>
} <%= @grammar.prefix %>value_t;
<% end %>

<% if @grammar.ast %>
/** AST node types. @{ */
typedef struct <%= @grammar.ast_prefix %>Token<%= @grammar.ast_suffix %>
{
    <%= @grammar.prefix %>token_t token;
    <%= @grammar.prefix %>value_t pvalue;
} <%= @grammar.ast_prefix %>Token<%= @grammar.ast_suffix %>;

<%   @parser.rule_sets.each do |name, rule_set| %>
<%     next if name.start_with?("$") %>
<%     next if rule_set.optional? %>
struct <%= name %>;
<%   end %>

<%   @parser.rule_sets.each do |name, rule_set| %>
<%     next if name.start_with?("$") %>
<%     next if rule_set.optional? %>
typedef struct <%= @grammar.ast_prefix %><%= name %><%= @grammar.ast_suffix %>
{
<%     rule_set.ast_fields.each do |fields| %>
    union
    {
<%       fields.each do |field_name, type| %>
        struct <%= type %> * <%= field_name %>;
<%       end %>
    };
<%     end %>
} <%= @grammar.ast_prefix %><%= name %><%= @grammar.ast_suffix %>;

<%   end %>
/** @} */
<% end %>

/**
 * A structure to keep track of parser position.
 *
 * This is useful for reporting errors, etc...
 */
typedef struct
{
    /** Input text row (0-based). */
    uint32_t row;

    /** Input text column (0-based). */
    uint32_t col;
} <%= @grammar.prefix %>position_t;

/** Lexed token information. */
typedef struct
{
    /** Text position where the token was found. */
    <%= @grammar.prefix %>position_t position;

    /** Number of input bytes used by the token. */
    size_t length;

    /** Token that was lexed. */
    <%= @grammar.prefix %>token_t token;

    /** Parser value associated with the token. */
    <%= @grammar.prefix %>value_t pvalue;
} <%= @grammar.prefix %>token_info_t;

/**
 * Lexer and parser context.
 *
 * The user must allocate an instance of this structure and pass it to any
 * public API function.
 */
typedef struct
{
    /* Lexer context data. */

    /** Input text. */
    uint8_t const * input;

    /** Input text length. */
    size_t input_length;

    /** Input text index (byte offset). */
    size_t input_index;

    /** Input text position (row/column). */
    <%= @grammar.prefix %>position_t text_position;

    /** Current lexer mode. */
    size_t mode;

    /* Parser context data. */

    /** Parse result value. */
<% if @grammar.ast %>
    <%= @grammar.ast_prefix %><%= @grammar.start_rule %><%= @grammar.ast_suffix %> * parse_result;
<% else %>
    <%= @grammar.prefix %>value_t parse_result;
<% end %>

    /** Unexpected token received. */
    <%= @grammar.prefix %>token_t token;

    /** User terminate code. */
    size_t user_terminate_code;
} <%= @grammar.prefix %>context_t;

/**************************************************************************
 * Public data
 *************************************************************************/

/** Token names. */
extern const char * <%= @grammar.prefix %>token_names[];

void <%= @grammar.prefix %>context_init(<%= @grammar.prefix %>context_t * context, uint8_t const * input, size_t input_length);

size_t <%= @grammar.prefix %>decode_code_point(uint8_t const * input, size_t input_length,
        <%= @grammar.prefix %>code_point_t * out_code_point, uint8_t * out_code_point_length);

size_t <%= @grammar.prefix %>lex(<%= @grammar.prefix %>context_t * context, <%= @grammar.prefix %>token_info_t * out_token_info);

size_t <%= @grammar.prefix %>parse(<%= @grammar.prefix %>context_t * context);

<% if @grammar.ast %>
<%= @grammar.ast_prefix %><%= @grammar.start_rule %><%= @grammar.ast_suffix %> * <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context);
<% else %>
<%= start_rule_type[1] %> <%= @grammar.prefix %>result(<%= @grammar.prefix %>context_t * context);
<% end %>

<%= @grammar.prefix %>position_t <%= @grammar.prefix %>position(<%= @grammar.prefix %>context_t * context);

size_t <%= @grammar.prefix %>user_terminate_code(<%= @grammar.prefix %>context_t * context);

<%= @grammar.prefix %>token_t <%= @grammar.prefix %>token(<%= @grammar.prefix %>context_t * context);