Module Mopsa_c_parser.C_ASTSource

C_AST - Simpler C AST

Provides a simpler version of the AST converted from Clang.

The goal of this AST is to:

Unique identifiers (for variables and functions).

Sourcemodule UidMap : sig ... end
Sourcemodule UidSet : sig ... end
Sourcemodule StringMap : sig ... end
module RangeMap : sig ... end

Source locations.

Comments in file.

preprocessor macros

Sourcetype character_kind = Clang_AST.character_kind
Sourcetype target_info = Clang_AST.target_info

Operators

Sourcetype binary_arithmetic =
  1. | ADD
    (*
    *)
  2. | SUB
    (*
    *)
  3. | MUL
    (*

    *

    *)
  4. | DIV
    (*

    /

    *)
  5. | MOD
    (*

    %

    *)
  6. | LEFT_SHIFT
    (*

    <<

    *)
  7. | RIGHT_SHIFT
    (*

    >>

    *)
  8. | BIT_AND
    (*

    &

    *)
  9. | BIT_OR
    (*

    |

    *)
  10. | BIT_XOR
    (*

    ^

    *)
Sourcetype binary_logical =
  1. | LESS
    (*

    <

    *)
  2. | LESS_EQUAL
    (*

    <=

    *)
  3. | GREATER
    (*

    >

    *)
  4. | GREATER_EQUAL
    (*

    >=

    *)
  5. | EQUAL
    (*

    ==

    *)
  6. | NOT_EQUAL
    (*

    !=

    *)
  7. | LOGICAL_AND
    (*

    &&

    *)
  8. | LOGICAL_OR
    (*

    ||

    *)
Sourcetype binary_operator =
  1. | O_arithmetic of binary_arithmetic
  2. | O_logical of binary_logical
Sourcetype unary_operator =
  1. | NEG
    (*
    *)
  2. | BIT_NOT
    (*

    ~

    *)
  3. | LOGICAL_NOT
    (*

    !

    *)
Sourcetype explicitness =
  1. | EXPLICIT
  2. | IMPLICIT
  3. | BITCAST
    (*

    Whether a cast is explicit (in the source) or implicit (added for implicit conversions

    *)
Sourcetype inc_location =
  1. | PRE
  2. | POST
    (*

    Whether an incrementation is performed before (PRE) or after (POST) the expression value is used

    *)
Sourcetype inc_direction =
  1. | INC
  2. | DEC
    (*

    Whether an incrementation is ++ or --

    *)

Types

Sourcetype signedness =
  1. | SIGNED
  2. | UNSIGNED
    (*

    Whether an integer type is signed or not.

    *)
Sourcetype integer_type =
  1. | Char of signedness
    (*

    plain 'char', where the signeness is defined by the platform

    *)
  2. | SIGNED_CHAR
  3. | UNSIGNED_CHAR
  4. | SIGNED_SHORT
  5. | UNSIGNED_SHORT
  6. | SIGNED_INT
  7. | UNSIGNED_INT
  8. | SIGNED_LONG
  9. | UNSIGNED_LONG
  10. | SIGNED_LONG_LONG
  11. | UNSIGNED_LONG_LONG
  12. | SIGNED_INT128
  13. | UNSIGNED_INT128
    (*

    Integer types.

    *)
Sourcetype float_type =
  1. | FLOAT
  2. | DOUBLE
  3. | LONG_DOUBLE
  4. | FLOAT128
  5. | FLOAT16
    (*

    Floating-point types.

    *)
Sourcetype record_kind =
  1. | STRUCT
  2. | UNION
    (*

    Whether a record type is a struct or a union.

    *)
Sourcetype qualifier = {
  1. qual_is_const : bool;
}

Type qualifiers. For now, only 'const' is useful.

Sourcetype typ =
  1. | T_void
    (*

    Void type.

    *)
  2. | T_bool
  3. | T_integer of integer_type
  4. | T_float of float_type
  5. | T_complex of float_type
  6. | T_pointer of type_qual
    (*

    Scalar types.

    *)
  7. | T_array of type_qual * array_length
    (*

    Arrays.

    *)
  8. | T_bitfield of typ * int
    (*

    Bitfields, with bit-width, only used in struct.

    *)
  9. | T_function of function_type option
    (*

    Function, with or without a prototype

    *)
  10. | T_builtin_fn
    (*

    Built-in functions

    *)
  11. | T_typedef of typedef
    (*

    Typedefs

    *)
  12. | T_record of record_type
    (*

    struct and union

    *)
  13. | T_enum of enum_type
    (*

    enums

    *)
  14. | T_vector of vector_type
    (*

    GCC vectors

    *)
  15. | T_attributed of typ * attr
  16. | T_unknown_builtin of string
    (*

    unknown builtin

    *)

Types.

Sourceand type_qual = typ * qualifier

Type with qualifier.

Sourceand typedef = {
  1. typedef_org_name : string;
    (*

    name as in source

    *)
  2. mutable typedef_uid : uid;
  3. mutable typedef_unique_name : string;
    (*

    unique name

    *)
  4. mutable typedef_def : type_qual;
    (*

    declaration

    *)
  5. mutable typedef_range : range;
    (*

    declaration location

    *)
  6. mutable typedef_com : comment list;
    (*

    comments associated to the declaration

    *)
  7. mutable typedef_attrs : attr list;
    (*

    attributes

    *)
}
Sourceand record_type = {
  1. record_kind : record_kind;
  2. record_org_name : string;
    (*

    name as in source, may be empty

    *)
  3. mutable record_uid : uid;
    (*

    unique identifier

    *)
  4. mutable record_unique_name : string;
    (*

    unique, non-empty name

    *)
  5. mutable record_defined : bool;
    (*

    false if only declared

    *)
  6. mutable record_sizeof : Z.t;
    (*

    size of record, in bytes

    *)
  7. mutable record_alignof : Z.t;
    (*

    alignment, in bytes

    *)
  8. mutable record_fields : record_field array;
  9. mutable record_range : range;
    (*

    declaration location

    *)
  10. mutable record_com : comment list;
    (*

    comments associated to the declaration

    *)
  11. mutable record_attrs : attr list;
    (*

    attributes

    *)
}

Struct or union type.

Sourceand record_field = {
  1. field_uid : uid;
    (*

    unique identifier

    *)
  2. field_org_name : string;
    (*

    may be empty for anonymous or padding fields

    *)
  3. field_name : string;
    (*

    non-empty name

    *)
  4. field_offset : int;
  5. field_bit_offset : int;
  6. mutable field_type : type_qual;
  7. field_range : range;
    (*

    declaration location

    *)
  8. field_record : record_type;
  9. field_index : int;
  10. mutable field_com : comment list;
    (*

    comments associated to the declaration

    *)
  11. mutable field_attrs : attr list;
    (*

    attributes

    *)
}

Struct or union field.

Sourceand enum_type = {
  1. enum_org_name : string;
    (*

    name as in source, may be empty

    *)
  2. mutable enum_uid : uid;
    (*

    unoque identifier

    *)
  3. mutable enum_unique_name : string;
    (*

    unique, non-empty name

    *)
  4. mutable enum_defined : bool;
    (*

    false if only declared

    *)
  5. mutable enum_values : enum_value list;
  6. mutable enum_integer_type : integer_type option;
  7. mutable enum_range : range;
    (*

    declaration location

    *)
  8. mutable enum_com : comment list;
    (*

    comments associated to the declaration

    *)
  9. mutable enum_attrs : attr list;
    (*

    attributes

    *)
}

Enumerated type.

Sourceand enum_value = {
  1. enum_val_uid : uid;
    (*

    unique identifier

    *)
  2. enum_val_org_name : string;
    (*

    name as in source

    *)
  3. enum_val_unique_name : string;
    (*

    unique name

    *)
  4. enum_val_value : Z.t;
  5. enum_val_enum : enum_type;
  6. enum_val_range : range;
  7. mutable enum_val_com : comment list;
    (*

    comments associated to the declaration

    *)
  8. mutable enum_val_attrs : attr list;
    (*

    attributes

    *)
}

A possible value in an enumerated type.

Sourceand array_length =
  1. | No_length
  2. | Length_cst of Z.t
  3. | Length_expr of expr
    (*

    Length of an array.

    *)
Sourceand function_type = {
  1. mutable ftype_return : type_qual;
  2. mutable ftype_params : type_qual list;
  3. ftype_variadic : bool;
}

Type of functions and prototypes.

Sourceand vector_type = {
  1. vector_type : type_qual;
    (*

    element type

    *)
  2. vector_size : int;
    (*

    size, in elements

    *)
  3. vector_kind : int;
  4. vector_sizeof : Z.t;
    (*

    size, in bytes

    *)
}

GCC vector types.

Variables and functions

Sourceand variable = {
  1. var_uid : uid;
    (*

    unique identifier

    *)
  2. var_org_name : string;
    (*

    original name

    *)
  3. var_unique_name : string;
    (*

    unique name for globals and statics

    *)
  4. mutable var_kind : variable_kind;
    (*

    variable kind and life-time

    *)
  5. mutable var_type : type_qual;
  6. mutable var_init : init option;
  7. mutable var_before_stmts : statement list;
    (*

    statements executed before the declaration of the variable

    *)
  8. mutable var_after_stmts : statement list;
    (*

    statements executed after the declaration of the variable

    *)
  9. mutable var_range : range;
  10. mutable var_com : comment list;
    (*

    comments associated to the declaration

    *)
  11. mutable var_init_org : init option;
    (*

    initialisation _before_ simplification, if different from var_init; used for printing

    *)
  12. mutable var_attrs : attr list;
    (*

    attributes

    *)
}
Sourceand variable_kind =
  1. | Variable_global
    (*

    global shared among translation units

    *)
  2. | Variable_extern
    (*

    declared but not defined

    *)
  3. | Variable_local of func
    (*

    local to a function

    *)
  4. | Variable_parameter of func
    (*

    formal argument

    *)
  5. | Variable_file_static of translation_unit
    (*

    restricted to a translation unit

    *)
  6. | Variable_func_static of func
    (*

    restricted to a function

    *)
Sourceand func = {
  1. func_uid : uid;
    (*

    unique identifier

    *)
  2. func_org_name : string;
    (*

    original name

    *)
  3. func_unique_name : string;
    (*

    unique name for globals and statics

    *)
  4. func_is_static : bool;
  5. mutable func_return : type_qual;
    (*

    type of returned value

    *)
  6. mutable func_parameters : variable array;
    (*

    function parameters

    *)
  7. mutable func_body : block option;
    (*

    function body

    *)
  8. mutable func_static_vars : variable list;
    (*

    static variables declared in the function

    *)
  9. mutable func_local_vars : variable list;
    (*

    local variables declared in the function (excluding parameters)

    *)
  10. mutable func_variadic : bool;
    (*

    whether the has a variable number of arguments

    *)
  11. mutable func_range : range;
    (*

    range of the full declaration

    *)
  12. mutable func_name_range : range;
    (*

    range of the name part of the declaration

    *)
  13. mutable func_com : (comment * macro StringMap.t) list;
    (*

    comments associated to the declaration

    *)
  14. mutable func_attrs : attr list;
    (*

    function attributes

    *)
}

Statements and expressions

Sourceand statement = statement_kind * range
Sourceand block = {
  1. blk_stmts : statement list;
    (*

    statements in the block

    *)
  2. blk_local_vars : variable list;
    (*

    local variables declared within the block (excluding sub-blocks); these should be deallocated when exiting the block

    *)
}
Sourceand statement_kind =
  1. | S_local_declaration of variable
    (*

    local variable declaration

    *)
  2. | S_expression of expr
    (*

    expression statement

    *)
  3. | S_block of block
    (*

    statement block

    *)
  4. | S_if of expr * block * block
    (*

    else branch

    *)
  5. | S_while of expr * block
    (*

    body

    *)
  6. | S_do_while of block * expr
    (*

    condition

    *)
  7. | S_for of block * expr option * expr option * block
    (*

    body

    *)
  8. | S_jump of jump_kind
    (*

    jump instruction

    *)
  9. | S_target of target_kind
    (*

    target of a jump

    *)
  10. | S_asm of asm_kind
    (*

    asm statement

    *)
  11. | S_attributed of statement * attr list
    (*

    statement with attributes

    *)
Sourceand jump_kind =
  1. | S_goto of string * scope_update
  2. | S_break of scope_update
  3. | S_continue of scope_update
  4. | S_return of expr option * scope_update
  5. | S_switch of expr * block
    (*

    various ways to jump

    *)
Sourceand target_kind =
  1. | S_label of string
  2. | S_case of expr list * scope_update
  3. | S_default of scope_update
    (*

    various targets of jumps

    *)
Sourceand scope_update = {
  1. mutable scope_var_added : variable list;
  2. mutable scope_var_removed : variable list;
}

variables to remove and to add (uninitialized) when jumping into another scope; may be attached to a jump source (break, continue, return, goto) or a jump target (switch case and default), depending on what's more convinient

Sourceand asm_kind = {
  1. asm_style : asm_style;
  2. asm_is_simple : bool;
  3. asm_is_volatile : bool;
  4. asm_body : string;
  5. asm_outputs : asm_output array;
  6. asm_inputs : asm_input array;
  7. asm_clobbers : string array;
  8. asm_labels : string array;
}

and asm statement

Sourceand asm_output = {
  1. asm_output_string : string;
  2. asm_output_expr : expr;
  3. asm_output_constraint : asm_output_constraint;
}
Sourceand asm_input = {
  1. asm_input_string : string;
  2. asm_input_expr : expr;
}
Sourceand asm_output_constraint = Clang_AST.asm_output_constraint
Sourceand expr_kind =
  1. | E_conditional of expr * expr * expr
    (*

    else

    *)
  2. | E_binary_conditional of expr * expr
    (*

    else

    *)
  3. | E_array_subscript of expr * expr
    (*

    index

    *)
  4. | E_member_access of expr * int * string
    (*

    field

    *)
  5. | E_arrow_access of expr * int * string
    (*

    field

    *)
  6. | E_compound_assign of expr * type_qual * binary_arithmetic * expr * type_qual
    (*

    type of the result, before converting back to lvalue type

    *)
  7. | E_binary of binary_operator * expr * expr
    (*

    right argument

    *)
  8. | E_assign of expr * expr
    (*

    rvalue

    *)
  9. | E_comma of expr * expr
    (*

    , operator

    *)
  10. | E_unary of unary_operator * expr
  11. | E_increment of inc_direction * inc_location * expr
  12. | E_address_of of expr
    (*

    & operator (address of lvalue)

    *)
  13. | E_deref of expr
    (*

    * operator (pointer dereference)

    *)
  14. | E_cast of expr * explicitness
  15. | E_call of expr * expr array
    (*

    actual arguments

    *)
  16. | E_character_literal of Z.t * character_kind
    (*

    literal character

    *)
  17. | E_integer_literal of Z.t
    (*

    literal integer

    *)
  18. | E_float_literal of string
    (*

    literal float, in binary string notation

    *)
  19. | E_string_literal of string * character_kind
  20. | E_null_ptr
  21. | E_compound_literal of init
  22. | E_variable of variable
  23. | E_function of func
  24. | E_predefined of string
    (*

    predefined identifier

    *)
  25. | E_statement of block
    (*

    a block of statements viewed as an expression (GCC extension)

    *)
  26. | E_var_args of expr
    (*

    __builtin_va_arg

    *)
  27. | E_atomic of int * expr * expr
  28. | E_convert_vector of expr
  29. | E_vector_element of expr * string
    (*

    accessor

    *)
  30. | E_shuffle_vector of expr array
Sourceand init =
  1. | I_init_expr of expr
  2. | I_init_list of init list * init option
    (*

    filler

    *)
  3. | I_init_implicit of type_qual
Sourceand translation_unit = {
  1. tu_uid : int;
  2. tu_name : string;
  3. tu_range : range;
}
Sourceand project = {
  1. proj_name : string;
    (*

    project name

    *)
  2. proj_tu : translation_unit list;
    (*

    translation units composing the project

    *)
  3. proj_target : target_info;
  4. proj_typedefs : typedef StringMap.t;
    (*

    typedefs, by unique name

    *)
  5. proj_enums : enum_type StringMap.t;
    (*

    enums, by unique name

    *)
  6. proj_records : record_type StringMap.t;
    (*

    records, by unique name

    *)
  7. proj_vars : variable StringMap.t;
    (*

    variables with global lifetime, by unique name

    *)
  8. proj_funcs : func StringMap.t;
    (*

    functions, by unique name

    *)
  9. proj_file_scope_asm : string RangeMap.t;
    (*

    file-scope assembly directives

    *)
  10. proj_files : string list;
    (*

    list of parsed files

    *)
  11. proj_comments : (comment * macro StringMap.t) list RangeMap.t;
    (*

    all comments

    *)
  12. proj_macros : macro StringMap.t;
    (*

    macros, by name

    *)
  13. proj_opts : string list;
    (*

    selected options pased to the compiler we'd like to keep

    *)
}

A project is a set of translation units linked together.

Useful definitions

Sourceval no_qual : qualifier

No qualifier.

Sourceval int_type : typ * qualifier
Sourceval bool_type : typ * qualifier
Sourceval void_ptr_type : typ * qualifier
Sourceval uchar_ptr_type : typ * qualifier
Sourceval merge_qualifiers : qualifier -> qualifier -> qualifier
Sourceval resolve_typedef : type_qual -> type_qual

Replace toplevel occurences of typedefs with their definition.

Sourceval as_int_type : type_qual -> integer_type

Interprets the type as an integer type, if possible

Sourceval as_float_type : type_qual -> float_type

Interprets the type as an integer type, if possible

Sourceval type_is_scalar : typ -> bool

Whether a type yields a scalar value.

Variable utilities

Sourceval variable_is_global : variable_kind -> bool

Whether variables of this kind have global lifetime or not.

Sourceval variable_is_static : variable_kind -> bool

Whether variables of this kind are static or not.

Sourceval variable_is_file_static : variable_kind -> bool

Whether variables of this kind are file-level static.