Type information

│ English (en) │ français (fr) │

Back to contents FPC internals

Type information

Architecture

(last updated for fpc version 1.0.x)

A type declaration , which is the basis for the symbol table, since inherently everything comes down to a type after parsing is a special structure with two principal fields, which point to a symbol table entry which is the type name, and the actual definition which gives the information on other symbols in the type, the size of the type and other such information.

type
  TType = object
    Sym: PSym;      // Points to the symbol table of this type
    Def: PDef;      // Points to the actual definition of this type
  end;

Definition types

(last updated for fpc version 1.0.x)

Definitions represent the type information for all possible symbols which can be encountered by the parser. The definition types are associated with symbols in the symbol table, and are used by the parsing process (among other things) to perform type checking.

The current possible definition types are enumerated in TDefType and can have one of the following symbolic values:

http://www.pjh2.de/fpc/CompilerInternalsFigure07.png

deftype of TDef object	Description
AbstractDef
ArrayDef	array type definition
RecordDef	record type definition
PointerDef	pointer type definition
OrdDef	ordinal (numeric value) type definition
StringDef	string type definition
EnumDef	enumeration type definition
ProcDef	procedure type definition
ObjectDef	object or class type definition
ErrorDef	error definition (empty, used for error recovery)
FileDef	file type definition
FormalDef
SetDef	set type definition
ProcVarDef	procedure variable type definition
FloatDef	floating point type definition
ClassrefDef
ForwardDef

base definition (TDef)

All type definitions are based on this object. Therefore all derived object all posess the fields in this object in addition to their own private fields.

type
  PDef = ^TDef;
  TDef = object(TSymTableEntry)
    TypeSym: PTypeSym;                    // Pointer to symbol table entry for this type
                                          // definition
    InitTable_Label: PAsmLabel;           // Label to initialization information (required
                                          // for some complex types)
    Rtti_Label: PAsmLabel;                // Label to the runtime type information.
    NextGlobal: PDef;
    PreviousGlobal: PDef;
    SaveSize: Longint;                    // Size in bytes of the data definition
    DefType: TDefType;                    // Indicates the definition type (see table ??).
    Has_InitTable: Boolean;
    Has_Rtti: Boolean;
    Is_Def_Stab_Written: TDefStabStatus;  // Can be one of the following states :
                                          // (Not_Written, written, Being_Written)
                                          // which indicates if the debug information
                                          // for this type has been defined or not.
     GlobalNb: Longint;                   // Internal stabs debug information type signature
                                          // (each type definition has a numeric
                                          // signature).
  end;

file definition (TFileDef)

The file definition can occur in only some rare instances, when a file of type is parsed, a file definition of that type will be created. Furthermore, internally, a definition for a Text file type and untyped File type are created when the system unit is loaded. These types are always defined when compiling any unit or program.

type
  PFileDef = ^TFileDef;
  TFileDef = object(TDef)
    FileTyp: TFileTyp;        // Indicates what type of file definition it is 
                              // (text, untyped or typed).
    TypedFileType: TType;     // In the case of a typed file definition, 
                              // definition of the type of the file
  end;

forward definition (TForwardDef)

The forward definition is created, when a type is declared before an actual definition exists. This is the case, when, for example type PMyObject = TMyObject, while TMyObject has yet to be defined.

type
  PForwardDef = ^TForwardDef;
  TForwardDef = object(TDef)
    toSymName: String;         // The symbol name for this forward declaration
                               // (the actual real definition does not exist yet)
    ForwardPos: TFilePosInfo;  // Indicates file position where this forward
                               // definition was declared.
  end;

error definition (TErrorDef)

This definition is actually an empty definition entry. When the parser encounters an error when parsing a definition instead of putting nothing in the type for a symbol, it puts this entry. This avoids illegal memory accesses later in parsing.

pointer definition (TPointerDef)

The pointer definition is used for distinguishing between different types of pointers in the compiler, and are created at each typename parsing construct found.

type
  PPointerDef = ^TPointerDef;
  TPointerDef = object(TDef)
    Is_Far: Boolean;         // Used to indicate if this is a far pointer or not 
                             // (this flag is cpu-specific)
    PointerType: TType;      // This indicates to what type definition this pointer
                             // points to.
  end;

object definition (TObjectDef)

The object definition is created each time an object declaration is found in the type declaration section.

type
  PObjectDef = ^TObjectDef;
  TObjectDef = object(TDef)
    ChildOf: PObjectDef;                 // This is a pointer to the parent object
                                         // definition. It is set to nil, if
                                         // this object definition has no parent.
    ObjName: PString;                    // This is the object name
    SymTable: PSymTable;                 // This is a pointer to the symbol
                                         // table entries within this object.
    ObjectOptions: TObjectOptions;       // The options for this object, see
                                         // the following table for the possible
                                         // options for the object.
    VMT_Offset: Longint;                 // This is the offset from the start
                                         // of the object image in memory
                                         // where the virtual method table is
                                         // located.
    Writing_Class_Record_Stab: Boolean;
  end;

Object Options(TObjectOptions)	Description
oo_is_class	This is a delphi styled class declaration, and not a Turbo Pascal object.
oo_is_forward	This flag is set to indicate that the object has been declared in a type section, but there is no implementation yet.
oo_has_virtual	This object / class contains virtual methods
oo_has_private	This object / class contains private fields or methods
oo_has_protected	This object / class contains protected fields or methods
oo_has_constructor	This object / class has a constructor method
oo_has_destructor	This object / class has a destructor method
oo_has_vmt	This object / class has a virtual method table
oo_has_msgstr	This object / class contains one or more message handlers
oo_has_msgint	This object / class contains one or more message handlers
oo_has_abstract	This object / class contains one or more abstract methods
oo_can_have_published	the class has runtime type information, i.e. you can publish properties
oo_cpp_class	the object/class uses an C++ compatible class layout
oo_interface	this class is a delphi styled interface

array definition (TArrayDef)

This definition is created when an array type declaration is parsed. It contains all the information necessary for array type checking and code generation.

type
  PArrayDef = ^TArrayDef;
  TArrayDef = object(TDef)
    IsVariant: Boolean;
    IsConstructor : Boolean;
    RangeNr: Longint;          // Label number associated with the index values
                               // when range checking is on
    LowRange : Longint;        // The lower index range of the array definition
    HighRange : Longint;       // The higher index range of the array definition
    ElementType : TType;       // The type information for the elements of the array
    RangeType : TType;         // The type information for the index ranges of the array
    IsArrayofConst : Boolean;
  end;

record definition (TRecordDef)

The record definition entry is created each time a record type declaration is parsed. It contains the symbol table to the elements in the record.

type
  PRecordDef = ^TRecordDef;
  TRecordDef = object(TDef)
    SymTable: PSymTable;    // This is a pointer to the symbol table entries within
                            // this record.
  end;

ordinal definition (TOrdDef)

This type definition is the one used for all ordinal values such as char, bytes and other numeric integer type values. Some of the predefined type definitions are automatically created and loaded when the compiler starts. Others are created at compile time, when declared.

type
  POrdDef = ^TOrdDef;
  TOrdDef = object(TDef)
    Low: Longint;        // The minimum value of this ordinal type
    High: Longint;       // The maximum value of this ordinal type
    Typ: TBaseType;      // The type of ordinal value
  end;

Base ordinal type (TBaseType)	Description
uauto	user defined ordinal type definition
uvoid	Represents a void return value or node
uchar	ASCII character (1 byte)
u8bit	unsigned 8-bit value
u16bit	unsigned 16-bit value
u32bit	unsigned 32-bit value
s16bit	signed 16-bit value
s32bit	signed 32-bit value
bool8bit	boolean 8-bit value
bool16bit	boolean 16-bit value
bool32bit	boolean 32-bit value
u64bit	unsigned 64-bit value (not fully supported/tested)
s64bit	signed 64-bit value
uwidechar	Currently not supported and unused

float definition (TFloatDef)

This type definition is the one used for all floating point values such as SINGLE, DOUBLE. Some of the predefined type definitions are automatically created and loaded when the compiler starts.

type
  PFloatDef = ^TFloatDef;
  TFloatDef = object(TDef)
    Typ: TFloatType;       // The type of floating point value.
  end;

Base floating point type (TFloatType)	Description
s32real	IEEE Single precision floating point value
s64real	IEEE Double precision floating point value
s80real	Extended precision floating point value (cpu-specific, usually maps to double)
s64comp	63-bit signed value, using 1 bit for sign indication
f16bit	Unsupported
f32bit	Unsupported

abstract procedure definition (tabstractprocdef)

This is the base of all routine type definitions. This object is abstract, and is not directly used in a useful way. The derived object of this object are used for the actual parsing process.

type
  PAbstractProcDef = ^TAbstractProcDef;
  TAbstractProcDef = object(TDef)
    SymtableLevel: Byte;
    Fpu_Used: Byte;                     // Number of floating point registers
                                        // used in this routine
    RetType: TType;                     // Type information for the return value
                                        // (uvoid if it returns nothing)
    ProcTypeOption: TProcTypeOption;    // Indicates the type of routine it is.
    ProcCallOptions: TProcCallOptions;  // Indicates the calling convention
                                        // of the routine.
    ProcOptions: TProcOptions;          // Indicates general procedure options.
    Para: PLinkedList;                  // This is a linked list of parameters
                                        // (pparaitem list)
  end;

Procedure options (TProcTypeOption)	Description
poType_ProgInit	Routine is the program entry point (defined as 'main' in the compiler).
poType_UnitInit	Routine is the unit initialization code (defined as unitname_init in the compiler
poType_UnitFinalize	Routine is the unit exit code (defined as unitname_finalize in the compiler)
poType_Constructor	Routine is an object or class constructor
poType_Destructor	Routine is an object or class destructor
poType_Operator	Procedure is an operator

call options (TProcCallOptions)	Description
pocall_clearstack	The routine caller clears the stack upon return
pocall_leftright	Send parameters to routine from left to right
pocall_cdecl	Passing parameters is done using the GCC alignment scheme, passing parameter values is directly copied into the stack space
pocall_register	unused (Send parameters via registers)
pocall_stdcall	Passing parameters is done using GCC alignment scheme, standard GCC registers are saved
pocall_safecall	Standard GCC registers are saved
pocall_palmsssyscall	This is a special syscall macro for embedded system
pocall_system	unused
pocall_inline	Routine is an inline assembler macro (not a true call)
pocall_internproc	System unit code generator helper routine
pocall_internconst	System unit code generator helper macro routine

routine options (TProcOptions)	Description
po_classmethod	This is a class method
po_virtualmethod	This is a virtual method
po_abstractmethod	This is an abstract method
po_staticmethod	This is a static method
po_overridingmethod	This is an overriden method (with po_virtual flag usually)
po_methodpointer	This is a method pointer (not a normal routine pointer)
po_containsself	self is passed explicitly as a parameter to the method
po_interrupt	This routine is an interrupt handler
po_iocheck	IO checking should be done after a call to the procedure
po_assembler	The routine is in assembler
po_msgstr	method for string message handling
po_msgint	method for int message handling
po_exports	Routine has export directive
po_external	Routine is external (in other object or lib)
po_savestdregs	Routine entry should save all registers used by GCC
po_saveregisters	Routine entry should save all registers
po_overload	Routine is declared as being overloaded

procedural variable definition (TProcVarDef)

This definition is created when a procedure variable type is declared. It gives information on the type of a procedure, and is used when assigning and directly calling a routine through a pointer.

type
  PProcVarDef = ^TProcVarDef;
  TProcVarDef = object(TAbstractProcDef)
  end;

procedure definition (TProcDef)

When a procedure head is parsed, the definition of the routine is created. Thereafter, other fields containing information on the definition of the routine are populated as required.

type
  PProcDef = ^TProcDef;
  TProcDef = object(TAbstractProcDef)
    ForwardDef: Boolean;               // TRUE if this is a forward definition
    InterfaceDef: Boolean;
    ExtNumber: Longint;
    MessageInf: TMessageInf;
    NextOverloaded: PProcDef;
    FileInfo: TFilePosInfo;            // Position in source code for the declaration of
                                       // this routine. Used for error management.
    Localst: PSymTable;                // The local variables symbol table
    Parast: PSymTable;                 // The parameter symbol table
    ProcSym: PProcSym;                 // Points to owner of this definition
    LastRef: PRef;
    DefRef: PRef;
    CrossRef: PRef;
    LastWritten: PRef;
    RefCount: Longint;
    _Class: ProbjectDef;
    Code: Pointer;                     // The actual code for the routine 
                                       // (only for inlined routines)
    UsedRegisters: TRegisterSet;       // The set of registers used in this routine
    HasForward: Boolean;
    Count: Boolean;
    Is_Used: Boolean;
  end;

string definition (TStringDef)

This definition represents all string types as well as derived types. Some of the default string type definitions are loaded when the compiler starts up. Others are created at compile time as they are declared with a specific length type.

type
  PStringDef = ^TStringDef;
  TStringDef = object(TDef)
    String_Typ: TStringType;   // Indicates the string type definition
    Len: Longint;              // This is the maximum length which can have the string
  end;

String type (TStringType)	Description
st_default	Depends on current compiler switches, can either be a st_ShortString or st_AnsiString
st_shortstring	short string (length byte followed by actual ASCII characters (1 byte/char))
st_longstring	long string (length longint followed by actual ASCII characters (1 byte/char))
st_ansistring	long string garbage collected (pointer to a length, reference count followed by actual ASCII characters (1 byte/char))
st_widestring	long string garbage collected (pointer to a length, reference count followed by actual unicode characters (1 word/char (utf16)))

enumeration definition (TEnumDef)

An enumeration definition is created each time an enumeration is declared and parsed. Each element in the enumeration will be added to the enumeration symtable and if $SCOPEDENUMS directive is off then to the unit symtable also.

type
  tenumdef = class(tstoreddef)
    has_jumps: Boolean;   // True if enum has jumps between elements
    minval: aint;         // Value of the first element in the enumeration
    maxval: aint;         // Value of the last element in the enumeration
    BaseDef: tenumdef;    // In the case where the enumeration is a 
                          // subrange of another enumeration or a unique 
                          // enumeration copied from the another, this gives 
                          // information on the base range of the elements
   symtable: TSymtable;   // Enumeration symbol table - stores enumeration elements
  end;

set definition (TSetDef)

This definition is created when a set type construct is parsed (set of declaration).

type
  PSetDef = ^TSetDef;
  TSetDef = object(TDef)
    SetType: TSetType; Indicates the storage type of the set.
    ElementType: TType; Points the type definition and symbol
      table to the elements in the set.
  end;

set type (TSetType)	Description
NormSet	Normal set of up to 256 elements (32 byte storage space required)
SmallSet	Small set of up to 32 elements (4 byte storage space)
VarSet	Variable number of element set (storage size is dependent on number of elements) (currently unused and unsupported)

Definition interface

(last updated for fpc version 1.0.x)

routines

TDef.Size

Declaration:	function TDef.Size: Longint;
Description:	This method returns the true size of the memory space required in bytes for this type definition (after alignment considerations).

TDef.Alignment

Declaration:	function TDef.Alignment: Longint;
Description:	This method returns the alignment of the data for complex types such as records and objects, otherwise returns 0 or 1 (no alignment).

Next chapter: The parser