Data Types
type	mod_parser::t_parser
	Parser. More...

Functions/Subroutines
subroutine	mod_parser::parser_initialize (parser, filename, keywords)
	Initialize the parser. More...

subroutine	mod_parser::parser_finalize (parser)
	Finalize the parser. More...

subroutine	mod_parser::parser_new_scope (parser, tok, kind)
	Start a new scope. More...

subroutine	mod_parser::parser_pop_scope (parser, tok)
	Remove the scope at the top of the scope stack. More...

subroutine	mod_parser::parser_reset_scope (parser)
	Reset the parser to the beginning of the scope. More...

subroutine	mod_parser::parser_get_identifier_type (parser, tok, id_kind)
	Find the type of an identifier. More...

logical function	mod_parser::parser_is_identifier_defined (parser, tok)
	Check if a given identifier is defined. More...

subroutine	mod_parser::parser_add_identifier (parser, tok, intgr, dbl, bool, string)
	Add an identifier to the current scope. More...

integer function	mod_parser::parser_get_integer_identifier (parser, tok)
	Get the value of an integer identifier. More...

double precision function	mod_parser::parser_get_double_identifier (parser, tok)
	Get the value of a double identifier. More...

logical function	mod_parser::parser_get_boolean_identifier (parser, tok)
	Get the value of a boolean identifier. More...

character(len=:) function, allocatable	mod_parser::parser_get_string_identifier (parser, tok)
	Get the value of a string identifier. More...

subroutine	mod_parser::parser_get (parser, tok)
	Get the next token. More...

subroutine	mod_parser::parser_peek (parser, tok)
	Get the next token but not consume it. More...

subroutine	mod_parser::parser_unget (parser, tok)
	Unget a token. More...

logical function	mod_parser::parser_next_token (parser, kind, peek)
	Check if the next token is of a specific kind. More...

logical function	mod_parser::parser_next_keyword (parser, keyword_id)
	Check if the next token is a keyword of a specific kind. More...

subroutine	mod_parser::parser_expect (parser, kind, has_leading_space, message)
	Ensure that the next token is of a specific kind. More...

subroutine	mod_parser::parser_throw_error (parser, tok, message)
	Display an error message. More...

Detailed Description

Brief description

Let us recall the lexer/parser organization with the 3 main units:

                                keyword list
                                      v
        ┌──────────┐            ┌───────────┐          ┏━━━━━━━━━━━━┓
 files  │          │ characters │           │  tokens  ┃            ┃ ───────>
 ─────> │   File   │ ──────────>│   Lexer   │ ───────> ┃   Parser   ┃  tokens / actions
        │          │            │           │          ┃            ┃ <───────
        └──────────┘            └───────────┘          ┗━━━━━━━━━━━━┛
       mod_source_file            mod_lexer             mod_parser
                                                        mod_identifier
                                                        mod_scope

Usage

A complete usage of the parser can be found in the OBJ Wavefront reader code.

Let us consider the source code example of the lexer section with the file example.txt:

# Print a message to screen
print "I love Notus <3";
 
# Compute the square of 42 an print the result to screen
square 42;

Specifications:

A comment line starts with a '#'
An instruction line must ends with ';' (tk_semicolon)
The language has two keywords with the following syntax:
- 'print': print tk_string tk_semicolon
- 'square': square tk_integer tk_semicolon

The following code read the source file and execute the instructions.

! Define an enumerator to identify keywords
!enum, bind(c)
!   enumerator :: kw_print = 1
!   enumerator :: kw_square
!end enum
 
type(t_parser) :: parser
type(t_keyword_name), dimension(2) :: keywords
type(t_token) :: tok, next_tok
 
! Create a list of keywords
keywords(kw_print )%name = "print"
keywords(kw_square)%name = "square"
 
! Initialize the parser with the file name and the keyword list
call parser_initialize(parser, "example.txt", keywords)
 
do
   ! Get the next token
   call parser%get(tok)
 
   select case(tok%kind)
   case(tk_eof)
      write(*,'("EOF")')
      exit
 
   case(tk_keyword)
      select case(tok%keyword_id)
      case(kw_print)
         ! Get the next token
         call parser%get(next_tok)
 
         ! Throw an error if no string is found after the keyword
         if (next_tok%kind /= tk_string) call parser%throw_error(next_tok, "'print' keyword requires a string")
 
         ! Ensure the presence of a semicolon
         call parser%expect(tk_semicolon)
 
         ! Print the message to standard output
         write(*,'(a)') next_tok%string_value
 
      case(kw_square)
         ! Get the next token
         call parser%get(next_tok)
 
         ! Throw an error if no integer is found after the keyword
         if (next_tok%kind /= tk_integer) call parser%throw_error(next_tok, "'square' keyword requires an integer")
 
         ! Ensure the presence of a semicolon
         call parser%expect(tk_semicolon)
 
         ! Print the square of the integer to standard output
         write(*,'(g0)') next_tok%integer_value**2
 
      end select
 
   case default
      call parser%throw_error("Forbidden token '"//trim(token_id_to_string(tok%kind))//"'")
   end select
end do

Expected output:

I love Notus <3
1764
EOF

Function/Subroutine Documentation

◆ parser_add_identifier()

subroutine mod_parser::parser_add_identifier	(	class(t_parser)	parser,
		type(t_token)	tok,
		integer, intent(in), optional	intgr,
		double precision, intent(in), optional	dbl,
		logical, intent(in), optional	bool,
		character(len=*), intent(in), optional	string
	)

Parameters

[in,out]	parser	Parser
[in]	tok	Identifier token
[in]	intgr	If present, the identifier is an integer (id_integer)
[in]	dbl	If present, the identifier is a double (id_double)
[in]	bool	If present, the identifier is a boolean (id_boolean)
[in]	string	If present, the identifier is a string (id_string)

◆ parser_expect()

subroutine mod_parser::parser_expect	(	class(t_parser)	parser,
		integer, intent(in)	kind,
		logical, intent(in), optional	has_leading_space,
		character(len=*), intent(in), optional	message
	)

If the kind of the next token is the same as the one provided, the next token is consumed. Otherwise, a parser error is thrown.

Typical usage to ensure that the next token is a semicolon ';':

call parser%expect(tk_semicolon)

Parameters

[in,out]	parser	Parser
[in]	kind	required kind of the next token
[in]	has_leading_space	optional argument that check if the token has a leading space
[in]	message	optional message to display the expected token is not found

◆ parser_finalize()

subroutine mod_parser::parser_finalize ( type(t_parser) parser )

Parameters

[in,out] parser Parser to finalize

◆ parser_get()

subroutine mod_parser::parser_get	(	class(t_parser)	parser,
		type(t_token), intent(out)	tok
	)

Act as a wrapper for lexer_read_token.

Throw an error if the lexer encounters an error.

Parameters

[in,out]	parser	Parser
[out]	tok	Token

◆ parser_get_boolean_identifier()

logical function mod_parser::parser_get_boolean_identifier	(	class(t_parser)	parser,
		type(t_token)	tok
	)

Throw an error when if the identifier is not found.

Parameters

[in,out]	parser	Parser
[in]	tok	Identifier token (tokkink == tk_identifier)

Returns: Boolean value of the identifier

◆ parser_get_double_identifier()

double precision function mod_parser::parser_get_double_identifier	(	class(t_parser)	parser,
		type(t_token)	tok
	)

Throw an error when if the identifier is not found.

Parameters

[in,out]	parser	Parser
[in]	tok	Identifier token (tokkink == tk_identifier)

Returns: Double value of the identifier

◆ parser_get_identifier_type()

subroutine mod_parser::parser_get_identifier_type	(	class(t_parser)	parser,
		type(t_token), intent(in)	tok,
		integer, intent(out)	id_kind
	)

Behavior:

Find an identifier starting with the current scope.
If an identifier is not found on the current scope, search deeper in the scope stack.
Throw an error when the identifier is not found.

Parameters

[in,out]	parser	Parser
[in]	tok	Identifier token (tokkind == tk_identifier)
[in]	id_kind	Type of identifier (id_integer, id_double, id_boolean, id_string)

◆ parser_get_integer_identifier()

integer function mod_parser::parser_get_integer_identifier	(	class(t_parser)	parser,
		type(t_token)	tok
	)

Throw an error when if the identifier is not found.

Parameters

[in,out]	parser	Parser
[in]	tok	Identifier token (tokkink == tk_identifier)

Returns: Integer value of the identifier

◆ parser_get_string_identifier()

character(len=:) function, allocatable mod_parser::parser_get_string_identifier	(	class(t_parser)	parser,
		type(t_token)	tok
	)

Throw an error when if the identifier is not found.

Parameters

[in,out]	parser	Parser
[in]	tok	Identifier token (tokkink == tk_identifier)

Returns: String value of the identifier

◆ parser_initialize()

subroutine mod_parser::parser_initialize	(	type(t_parser), intent(inout)	parser,
		character(len=*), intent(in)	filename,
		type(t_keyword_name), dimension(:), intent(in)	keywords
	)

The parser initializes a lexer with filename and keywords.

Parameters

[in,out]	parser	Parser to initialize
[in]	filename	Source file name
[in]	keywords	List of keywords

◆ parser_is_identifier_defined()

logical function mod_parser::parser_is_identifier_defined	(	class(t_parser), intent(inout)	parser,
		type(t_token), intent(in)	tok
	)

Parameters

[in,out]	parser	Parser
[in]	tok	Identifier token

Returns: Return .true. if the given identifier is defined

◆ parser_new_scope()

subroutine mod_parser::parser_new_scope	(	class(t_parser)	parser,
		type(t_token), intent(in)	tok,
		integer, intent(in), optional	kind
	)

Start a new scope beginning at the position of the token tok. It is possible to attribute a flag to identify the scope with the kind argument.

Parameters

[in,out]	parser	Parser
[in]	tok	Token where the scope starts
[in]	kind	Flag to identify the scope

◆ parser_next_keyword()

logical function mod_parser::parser_next_keyword	(	class(t_parser)	parser,
		integer, intent(in)	keyword_id
	)

The function returns false if the next token is not a keyword or the keyword has the wrong id.

Parameters

[in,out]	parser	Parser
[in]	keyword_id	wished id of the next keyword

◆ parser_next_token()

logical function mod_parser::parser_next_token	(	class(t_parser)	parser,
		integer, intent(in)	kind,
		logical, intent(in), optional	peek
	)

If the kind of the next token is the same as the one provided, the next token is consumed (unless peek is set to true) and the function returns .true.. Otherwise, the next token is not consumed and the function returns .false..

Act as a wrapper for lexer_next_token.

Parameters

[in,out]	parser	Parser
[in]	kind	wished kind of the next token
[in]	peek	peek token instead of consumming it

◆ parser_peek()

subroutine mod_parser::parser_peek	(	class(t_parser)	parser,
		type(t_token), intent(out)	tok
	)

Act as a wrapper for lexer_peek_token.

Parameters

[in,out]	parser	Parser
[out]	tok	Contains the next token

◆ parser_pop_scope()

subroutine mod_parser::parser_pop_scope	(	class(t_parser)	parser,
		type(t_token), intent(in)	tok
	)

If an error occurs, a message is thrown with a position determined by the token arguement tok.

Parameters

[in,out]	parser	Parser
[in]	tok	Token to tell where a potential error occurs.

◆ parser_reset_scope()

subroutine mod_parser::parser_reset_scope ( class(t_parser) parser )

Parameters

[in,out] parser Parser

◆ parser_throw_error()

subroutine mod_parser::parser_throw_error	(	class(t_parser)	parser,
		type(t_token), intent(in)	tok,
		character(len=*), intent(in)	message
	)

Throw an error message to the user with the line, the column and the file where the provided token was read.

Parameters

[in,out]	parser	current parser
[in]	tok	token where the error occurs
[in]	message	error message

◆ parser_unget()

subroutine mod_parser::parser_unget	(	class(t_parser)	parser,
		type(t_token), intent(in)	tok
	)

Act as a wrapper for lexer_unget_token.

Parameters

[in,out]	parser	Parser
[in]	tok	Token to unget

Data Types

Functions/Subroutines

Detailed Description

Brief description

Usage

Function/Subroutine Documentation

◆ parser_add_identifier()

◆ parser_expect()

◆ parser_finalize()

◆ parser_get()

◆ parser_get_boolean_identifier()

◆ parser_get_double_identifier()

◆ parser_get_identifier_type()

◆ parser_get_integer_identifier()

◆ parser_get_string_identifier()

◆ parser_initialize()

◆ parser_is_identifier_defined()

◆ parser_new_scope()

◆ parser_next_keyword()

◆ parser_next_token()

◆ parser_peek()

◆ parser_pop_scope()

◆ parser_reset_scope()

◆ parser_throw_error()

◆ parser_unget()