parser.h File Reference

The parser used to parse Ruby source. More...

#include "prism/defines.h"
#include "prism/ast.h"
#include "prism/encoding.h"
#include "prism/options.h"
#include "prism/static_literals.h"
#include "prism/util/pm_constant_pool.h"
#include "prism/util/pm_list.h"
#include "prism/util/pm_newline_list.h"
#include "prism/util/pm_string.h"
#include <stdbool.h>

Include dependency graph for parser.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Data Structures
struct	pm_heredoc_lex_mode_t
	All of the information necessary to store to lexing a heredoc. More...
struct	pm_lex_mode
	When lexing Ruby source, the lexer has a small amount of state to tell which kind of token it is currently lexing. More...
struct	pm_context_node
	This is a node in a linked list of contexts. More...
struct	pm_comment
	This is a node in the linked list of comments that we've found while parsing. More...
struct	pm_magic_comment_t
	This is a node in the linked list of magic comments that we've found while parsing. More...
struct	pm_lex_callback_t
	When you are lexing through a file, the lexer needs all of the information that the parser additionally provides (for example, the local table). More...
struct	pm_local_t
	This tracks an individual local variable in a certain lexical context, as well as the number of times is it read. More...
struct	pm_locals
	This is a set of local variables in a certain lexical context (method, class, module, etc.). More...
struct	pm_scope
	This struct represents a node in a linked list of scopes. More...
struct	pm_parser
	This struct represents the overall parser. More...

Macros
#define	PM_LEX_STACK_SIZE   4
	We pre-allocate a certain number of lex states in order to avoid having to call malloc too many times while parsing.

Typedefs
typedef struct pm_lex_mode	pm_lex_mode_t
	When lexing Ruby source, the lexer has a small amount of state to tell which kind of token it is currently lexing.
typedef struct pm_parser	pm_parser_t
	The parser used to parse Ruby source.
typedef struct pm_context_node	pm_context_node_t
	This is a node in a linked list of contexts.
typedef struct pm_comment	pm_comment_t
	This is a node in the linked list of comments that we've found while parsing.
typedef void(*	pm_encoding_changed_callback_t) (pm_parser_t *parser)
	When the encoding that is being used to parse the source is changed by prism, we provide the ability here to call out to a user-defined function.
typedef uint8_t	pm_shareable_constant_value_t
	The type of shareable constant value that can be set.
typedef struct pm_locals	pm_locals_t
	This is a set of local variables in a certain lexical context (method, class, module, etc.).
typedef uint8_t	pm_scope_parameters_t
	The flags about scope parameters that can be set.
typedef struct pm_scope	pm_scope_t
	This struct represents a node in a linked list of scopes.
typedef uint32_t	pm_state_stack_t
	A struct that represents a stack of boolean values.

Enumerations
enum	pm_lex_state_bit_t {   PM_LEX_STATE_BIT_BEG , PM_LEX_STATE_BIT_END , PM_LEX_STATE_BIT_ENDARG , PM_LEX_STATE_BIT_ENDFN ,   PM_LEX_STATE_BIT_ARG , PM_LEX_STATE_BIT_CMDARG , PM_LEX_STATE_BIT_MID , PM_LEX_STATE_BIT_FNAME ,   PM_LEX_STATE_BIT_DOT , PM_LEX_STATE_BIT_CLASS , PM_LEX_STATE_BIT_LABEL , PM_LEX_STATE_BIT_LABELED ,   PM_LEX_STATE_BIT_FITEM }
	This enum provides various bits that represent different kinds of states that the lexer can track. More...
enum	pm_lex_state_t {   PM_LEX_STATE_NONE = 0 , PM_LEX_STATE_BEG = (1 << PM_LEX_STATE_BIT_BEG) , PM_LEX_STATE_END = (1 << PM_LEX_STATE_BIT_END) , PM_LEX_STATE_ENDARG = (1 << PM_LEX_STATE_BIT_ENDARG) ,   PM_LEX_STATE_ENDFN = (1 << PM_LEX_STATE_BIT_ENDFN) , PM_LEX_STATE_ARG = (1 << PM_LEX_STATE_BIT_ARG) , PM_LEX_STATE_CMDARG = (1 << PM_LEX_STATE_BIT_CMDARG) , PM_LEX_STATE_MID = (1 << PM_LEX_STATE_BIT_MID) ,   PM_LEX_STATE_FNAME = (1 << PM_LEX_STATE_BIT_FNAME) , PM_LEX_STATE_DOT = (1 << PM_LEX_STATE_BIT_DOT) , PM_LEX_STATE_CLASS = (1 << PM_LEX_STATE_BIT_CLASS) , PM_LEX_STATE_LABEL = (1 << PM_LEX_STATE_BIT_LABEL) ,   PM_LEX_STATE_LABELED = (1 << PM_LEX_STATE_BIT_LABELED) , PM_LEX_STATE_FITEM = (1 << PM_LEX_STATE_BIT_FITEM) , PM_LEX_STATE_BEG_ANY = PM_LEX_STATE_BEG | PM_LEX_STATE_MID | PM_LEX_STATE_CLASS , PM_LEX_STATE_ARG_ANY = PM_LEX_STATE_ARG | PM_LEX_STATE_CMDARG ,   PM_LEX_STATE_END_ANY = PM_LEX_STATE_END | PM_LEX_STATE_ENDARG | PM_LEX_STATE_ENDFN }
	This enum combines the various bits from the above enum into individual values that represent the various states of the lexer.
enum	pm_heredoc_quote_t { PM_HEREDOC_QUOTE_NONE , PM_HEREDOC_QUOTE_SINGLE = '\'' , PM_HEREDOC_QUOTE_DOUBLE = '"' , PM_HEREDOC_QUOTE_BACKTICK = '`' }
	The type of quote that a heredoc uses.
enum	pm_heredoc_indent_t { PM_HEREDOC_INDENT_NONE , PM_HEREDOC_INDENT_DASH , PM_HEREDOC_INDENT_TILDE }
	The type of indentation that a heredoc uses.
enum	pm_context_t {   PM_CONTEXT_NONE = 0 , PM_CONTEXT_BEGIN , PM_CONTEXT_BEGIN_ENSURE , PM_CONTEXT_BEGIN_ELSE ,   PM_CONTEXT_BEGIN_RESCUE , PM_CONTEXT_BLOCK_BRACES , PM_CONTEXT_BLOCK_KEYWORDS , PM_CONTEXT_BLOCK_ENSURE ,   PM_CONTEXT_BLOCK_ELSE , PM_CONTEXT_BLOCK_RESCUE , PM_CONTEXT_CASE_WHEN , PM_CONTEXT_CASE_IN ,   PM_CONTEXT_CLASS , PM_CONTEXT_CLASS_ENSURE , PM_CONTEXT_CLASS_ELSE , PM_CONTEXT_CLASS_RESCUE ,   PM_CONTEXT_DEF , PM_CONTEXT_DEF_ENSURE , PM_CONTEXT_DEF_ELSE , PM_CONTEXT_DEF_RESCUE ,   PM_CONTEXT_DEF_PARAMS , PM_CONTEXT_DEFINED , PM_CONTEXT_DEFAULT_PARAMS , PM_CONTEXT_ELSE ,   PM_CONTEXT_ELSIF , PM_CONTEXT_EMBEXPR , PM_CONTEXT_FOR , PM_CONTEXT_FOR_INDEX ,   PM_CONTEXT_IF , PM_CONTEXT_LAMBDA_BRACES , PM_CONTEXT_LAMBDA_DO_END , PM_CONTEXT_LAMBDA_ENSURE ,   PM_CONTEXT_LAMBDA_ELSE , PM_CONTEXT_LAMBDA_RESCUE , PM_CONTEXT_LOOP_PREDICATE , PM_CONTEXT_MAIN ,   PM_CONTEXT_MODULE , PM_CONTEXT_MODULE_ENSURE , PM_CONTEXT_MODULE_ELSE , PM_CONTEXT_MODULE_RESCUE ,   PM_CONTEXT_MULTI_TARGET , PM_CONTEXT_PARENS , PM_CONTEXT_POSTEXE , PM_CONTEXT_PREDICATE ,   PM_CONTEXT_PREEXE , PM_CONTEXT_RESCUE_MODIFIER , PM_CONTEXT_SCLASS , PM_CONTEXT_SCLASS_ENSURE ,   PM_CONTEXT_SCLASS_ELSE , PM_CONTEXT_SCLASS_RESCUE , PM_CONTEXT_TERNARY , PM_CONTEXT_UNLESS ,   PM_CONTEXT_UNTIL , PM_CONTEXT_WHILE }
	While parsing, we keep track of a stack of contexts. More...
enum	pm_comment_type_t { PM_COMMENT_INLINE , PM_COMMENT_EMBDOC }
	This is the type of a comment that we've found while parsing.

Detailed Description

The parser used to parse Ruby source.

Macro Definition Documentation

PM_LEX_STACK_SIZE

#define PM_LEX_STACK_SIZE   4

We pre-allocate a certain number of lex states in order to avoid having to call malloc too many times while parsing.

You really shouldn't need more than this because you only really nest deeply when doing string interpolation.

Typedef Documentation

pm_lex_mode_t

typedef struct pm_lex_mode pm_lex_mode_t

When lexing Ruby source, the lexer has a small amount of state to tell which kind of token it is currently lexing.

For example, when we find the start of a string, the first token that we return is a TOKEN_STRING_BEGIN token. After that the lexer is now in the PM_LEX_STRING mode, and will return tokens that are found as part of a string.

pm_locals_t

typedef struct pm_locals pm_locals_t

This is a set of local variables in a certain lexical context (method, class, module, etc.).

We need to track how many times these variables are read in order to warn if they only get written.

pm_scope_t

typedef struct pm_scope pm_scope_t

This struct represents a node in a linked list of scopes.

Some scopes can see into their parent scopes, while others cannot.

Enumeration Type Documentation

pm_lex_state_bit_t

enum pm_lex_state_bit_t

This enum provides various bits that represent different kinds of states that the lexer can track.

This is used to determine which kind of token to return based on the context of the parser.

pm_context_t

enum pm_context_t

While parsing, we keep track of a stack of contexts.

This is helpful for error recovery so that we can pop back to a previous context when we hit a token that is understood by a parent context but not by the current context.

Enumerator
PM_CONTEXT_NONE	a null context, used for returning a value from a function
PM_CONTEXT_BEGIN	a begin statement
PM_CONTEXT_BEGIN_ENSURE	an ensure statement with an explicit begin
PM_CONTEXT_BEGIN_ELSE	a rescue else statement with an explicit begin
PM_CONTEXT_BEGIN_RESCUE	a rescue statement with an explicit begin
PM_CONTEXT_BLOCK_BRACES	expressions in block arguments using braces
PM_CONTEXT_BLOCK_KEYWORDS	expressions in block arguments using do..end
PM_CONTEXT_BLOCK_ENSURE	an ensure statement within a do..end block
PM_CONTEXT_BLOCK_ELSE	a rescue else statement within a do..end block
PM_CONTEXT_BLOCK_RESCUE	a rescue statement within a do..end block
PM_CONTEXT_CASE_WHEN	a case when statements
PM_CONTEXT_CASE_IN	a case in statements
PM_CONTEXT_CLASS	a class declaration
PM_CONTEXT_CLASS_ENSURE	an ensure statement within a class statement
PM_CONTEXT_CLASS_ELSE	a rescue else statement within a class statement
PM_CONTEXT_CLASS_RESCUE	a rescue statement within a class statement
PM_CONTEXT_DEF	a method definition
PM_CONTEXT_DEF_ENSURE	an ensure statement within a method definition
PM_CONTEXT_DEF_ELSE	a rescue else statement within a method definition
PM_CONTEXT_DEF_RESCUE	a rescue statement within a method definition
PM_CONTEXT_DEF_PARAMS	a method definition's parameters
PM_CONTEXT_DEFINED	a defined? expression
PM_CONTEXT_DEFAULT_PARAMS	a method definition's default parameter
PM_CONTEXT_ELSE	an else clause
PM_CONTEXT_ELSIF	an elsif clause
PM_CONTEXT_EMBEXPR	an interpolated expression
PM_CONTEXT_FOR	a for loop
PM_CONTEXT_FOR_INDEX	a for loop's index
PM_CONTEXT_IF	an if statement
PM_CONTEXT_LAMBDA_BRACES	a lambda expression with braces
PM_CONTEXT_LAMBDA_DO_END	a lambda expression with do..end
PM_CONTEXT_LAMBDA_ENSURE	an ensure statement within a lambda expression
PM_CONTEXT_LAMBDA_ELSE	a rescue else statement within a lambda expression
PM_CONTEXT_LAMBDA_RESCUE	a rescue statement within a lambda expression
PM_CONTEXT_LOOP_PREDICATE	the predicate clause of a loop statement
PM_CONTEXT_MAIN	the top level context
PM_CONTEXT_MODULE	a module declaration
PM_CONTEXT_MODULE_ENSURE	an ensure statement within a module statement
PM_CONTEXT_MODULE_ELSE	a rescue else statement within a module statement
PM_CONTEXT_MODULE_RESCUE	a rescue statement within a module statement
PM_CONTEXT_MULTI_TARGET	a multiple target expression
PM_CONTEXT_PARENS	a parenthesized expression
PM_CONTEXT_POSTEXE	an END block
PM_CONTEXT_PREDICATE	a predicate inside an if/elsif/unless statement
PM_CONTEXT_PREEXE	a BEGIN block
PM_CONTEXT_RESCUE_MODIFIER	a modifier rescue clause
PM_CONTEXT_SCLASS	a singleton class definition
PM_CONTEXT_SCLASS_ENSURE	an ensure statement with a singleton class
PM_CONTEXT_SCLASS_ELSE	a rescue else statement with a singleton class
PM_CONTEXT_SCLASS_RESCUE	a rescue statement with a singleton class
PM_CONTEXT_TERNARY	a ternary expression
PM_CONTEXT_UNLESS	an unless statement
PM_CONTEXT_UNTIL	an until statement
PM_CONTEXT_WHILE	a while statement