elna/source/Lexer.elna

727 lines
35 KiB
Plaintext

implementation module Lexer;
from FIO import ReadNBytes;
from SYSTEM import ADR;
from Storage import DEALLOCATE, ALLOCATE;
from Strings import Length;
from MemUtils import MemZero;
from StrCase import Lower;
const
ChunkSize = 65536;
type
(*
* Classification table assigns each possible character to a group (class). All
* characters of the same group a handled equivalently.
*
* Classification:
*)
TransitionClass = (
transitionClassInvalid,
transitionClassDigit,
transitionClassAlpha,
transitionClassSpace,
transitionClassColon,
transitionClassEquals,
transitionClassLeftParen,
transitionClassRightParen,
transitionClassAsterisk,
transitionClassUnderscore,
transitionClassSingle,
transitionClassHex,
transitionClassZero,
transitionClassX,
transitionClassEof,
transitionClassDot,
transitionClassMinus,
transitionClassSingleQuote,
transitionClassDoubleQuote,
transitionClassGreater,
transitionClassLess,
transitionClassOther
);
TransitionState = (
transitionStateStart,
transitionStateColon,
transitionStateIdentifier,
transitionStateDecimal,
transitionStateGreater,
transitionStateMinus,
transitionStateLeftParen,
transitionStateLess,
transitionStateDot,
transitionStateComment,
transitionStateClosingComment,
transitionStateCharacter,
transitionStateString,
transitionStateLeadingZero,
transitionStateDecimalSuffix,
transitionStateEnd
);
TransitionAction = procedure(PLexer, PLexerToken);
Transition = record
Action: TransitionAction;
NextState: TransitionState
end;
VAR
Classification: ARRAY[1..128] OF TransitionClass;
Transitions: ARRAY[0..MAX(TransitionState)] OF ARRAY[0..MAX(TransitionClass)] OF Transition;
PROCEDURE InitializeClassification();
BEGIN
Classification[1] := transitionClassEof; (* NUL *)
Classification[2] := transitionClassInvalid; (* SOH *)
Classification[3] := transitionClassInvalid; (* STX *)
Classification[4] := transitionClassInvalid; (* ETX *)
Classification[5] := transitionClassInvalid; (* EOT *)
Classification[6] := transitionClassInvalid; (* EMQ *)
Classification[7] := transitionClassInvalid; (* ACK *)
Classification[8] := transitionClassInvalid; (* BEL *)
Classification[9] := transitionClassInvalid; (* BS *)
Classification[10] := transitionClassSpace; (* HT *)
Classification[11] := transitionClassSpace; (* LF *)
Classification[12] := transitionClassInvalid; (* VT *)
Classification[13] := transitionClassInvalid; (* FF *)
Classification[14] := transitionClassSpace; (* CR *)
Classification[15] := transitionClassInvalid; (* SO *)
Classification[16] := transitionClassInvalid; (* SI *)
Classification[17] := transitionClassInvalid; (* DLE *)
Classification[18] := transitionClassInvalid; (* DC1 *)
Classification[19] := transitionClassInvalid; (* DC2 *)
Classification[20] := transitionClassInvalid; (* DC3 *)
Classification[21] := transitionClassInvalid; (* DC4 *)
Classification[22] := transitionClassInvalid; (* NAK *)
Classification[23] := transitionClassInvalid; (* SYN *)
Classification[24] := transitionClassInvalid; (* ETB *)
Classification[25] := transitionClassInvalid; (* CAN *)
Classification[26] := transitionClassInvalid; (* EM *)
Classification[27] := transitionClassInvalid; (* SUB *)
Classification[28] := transitionClassInvalid; (* ESC *)
Classification[29] := transitionClassInvalid; (* FS *)
Classification[30] := transitionClassInvalid; (* GS *)
Classification[31] := transitionClassInvalid; (* RS *)
Classification[32] := transitionClassInvalid; (* US *)
Classification[33] := transitionClassSpace; (* Space *)
Classification[34] := transitionClassSingle; (* ! *)
Classification[35] := transitionClassDoubleQuote; (* " *)
Classification[36] := transitionClassOther; (* # *)
Classification[37] := transitionClassOther; (* $ *)
Classification[38] := transitionClassSingle; (* % *)
Classification[39] := transitionClassSingle; (* & *)
Classification[40] := transitionClassSingleQuote; (* ' *)
Classification[41] := transitionClassLeftParen; (* ( *)
Classification[42] := transitionClassRightParen; (* ) *)
Classification[43] := transitionClassAsterisk; (* * *)
Classification[44] := transitionClassSingle; (* + *)
Classification[45] := transitionClassSingle; (* , *)
Classification[46] := transitionClassMinus; (* - *)
Classification[47] := transitionClassDot; (* . *)
Classification[48] := transitionClassSingle; (* / *)
Classification[49] := transitionClassZero; (* 0 *)
Classification[50] := transitionClassDigit; (* 1 *)
Classification[51] := transitionClassDigit; (* 2 *)
Classification[52] := transitionClassDigit; (* 3 *)
Classification[53] := transitionClassDigit; (* 4 *)
Classification[54] := transitionClassDigit; (* 5 *)
Classification[55] := transitionClassDigit; (* 6 *)
Classification[56] := transitionClassDigit; (* 7 *)
Classification[57] := transitionClassDigit; (* 8 *)
Classification[58] := transitionClassDigit; (* 9 *)
Classification[59] := transitionClassColon; (* : *)
Classification[60] := transitionClassSingle; (* ; *)
Classification[61] := transitionClassLess; (* < *)
Classification[62] := transitionClassEquals; (* = *)
Classification[63] := transitionClassGreater; (* > *)
Classification[64] := transitionClassOther; (* ? *)
Classification[65] := transitionClassSingle; (* @ *)
Classification[66] := transitionClassAlpha; (* A *)
Classification[67] := transitionClassAlpha; (* B *)
Classification[68] := transitionClassAlpha; (* C *)
Classification[69] := transitionClassAlpha; (* D *)
Classification[70] := transitionClassAlpha; (* E *)
Classification[71] := transitionClassAlpha; (* F *)
Classification[72] := transitionClassAlpha; (* G *)
Classification[73] := transitionClassAlpha; (* H *)
Classification[74] := transitionClassAlpha; (* I *)
Classification[75] := transitionClassAlpha; (* J *)
Classification[76] := transitionClassAlpha; (* K *)
Classification[77] := transitionClassAlpha; (* L *)
Classification[78] := transitionClassAlpha; (* M *)
Classification[79] := transitionClassAlpha; (* N *)
Classification[80] := transitionClassAlpha; (* O *)
Classification[81] := transitionClassAlpha; (* P *)
Classification[82] := transitionClassAlpha; (* Q *)
Classification[83] := transitionClassAlpha; (* R *)
Classification[84] := transitionClassAlpha; (* S *)
Classification[85] := transitionClassAlpha; (* T *)
Classification[86] := transitionClassAlpha; (* U *)
Classification[87] := transitionClassAlpha; (* V *)
Classification[88] := transitionClassAlpha; (* W *)
Classification[89] := transitionClassAlpha; (* X *)
Classification[90] := transitionClassAlpha; (* Y *)
Classification[91] := transitionClassAlpha; (* Z *)
Classification[92] := transitionClassSingle; (* [ *)
Classification[93] := transitionClassOther; (* \ *)
Classification[94] := transitionClassSingle; (* ] *)
Classification[95] := transitionClassSingle; (* ^ *)
Classification[96] := transitionClassUnderscore; (* _ *)
Classification[97] := transitionClassOther; (* ` *)
Classification[98] := transitionClassHex; (* a *)
Classification[99] := transitionClassHex; (* b *)
Classification[100] := transitionClassHex; (* c *)
Classification[101] := transitionClassHex; (* d *)
Classification[102] := transitionClassHex; (* e *)
Classification[103] := transitionClassHex; (* f *)
Classification[104] := transitionClassAlpha; (* g *)
Classification[105] := transitionClassAlpha; (* h *)
Classification[106] := transitionClassAlpha; (* i *)
Classification[107] := transitionClassAlpha; (* j *)
Classification[108] := transitionClassAlpha; (* k *)
Classification[109] := transitionClassAlpha; (* l *)
Classification[110] := transitionClassAlpha; (* m *)
Classification[111] := transitionClassAlpha; (* n *)
Classification[112] := transitionClassAlpha; (* o *)
Classification[113] := transitionClassAlpha; (* p *)
Classification[114] := transitionClassAlpha; (* q *)
Classification[115] := transitionClassAlpha; (* r *)
Classification[116] := transitionClassAlpha; (* s *)
Classification[117] := transitionClassAlpha; (* t *)
Classification[118] := transitionClassAlpha; (* u *)
Classification[119] := transitionClassAlpha; (* v *)
Classification[120] := transitionClassAlpha; (* w *)
Classification[121] := transitionClassX; (* x *)
Classification[122] := transitionClassAlpha; (* y *)
Classification[123] := transitionClassAlpha; (* z *)
Classification[124] := transitionClassOther; (* { *)
Classification[125] := transitionClassSingle; (* | *)
Classification[126] := transitionClassOther; (* } *)
Classification[127] := transitionClassSingle; (* ~ *)
Classification[128] := transitionClassInvalid (* DEL *)
END InitializeClassification;
PROCEDURE CompareKeyword(Keyword: ARRAY OF CHAR; TokenStart: PLexerBuffer; TokenEnd: PLexerBuffer): BOOLEAN;
VAR
Result: BOOLEAN;
Index: CARDINAL;
BEGIN
Index := 0;
Result := TRUE;
WHILE (Index < Length(Keyword)) AND (TokenStart <> TokenEnd) AND Result DO
Result := (Keyword[Index] = TokenStart^) OR (Lower(Keyword[Index]) = TokenStart^);
INC(TokenStart);
INC(Index)
END;
RETURN (Index = Length(Keyword)) AND (TokenStart = TokenEnd) AND Result
END CompareKeyword;
(* Reached the end of file. *)
PROCEDURE TransitionActionEof(ALexer: PLexer; AToken: PLexerToken);
BEGIN
AToken^.Kind := lexerKindEof
END TransitionActionEof;
(* Add the character to the token currently read and advance to the next character. *)
PROCEDURE TransitionActionAccumulate(ALexer: PLexer; AToken: PLexerToken);
BEGIN
INC(ALexer^.Current)
END TransitionActionAccumulate;
(* The current character is not a part of the token. Finish the token already
* read. Don't advance to the next character. *)
PROCEDURE TransitionActionFinalize(ALexer: PLexer; AToken: PLexerToken);
BEGIN
IF ALexer^.Start^ = ':' THEN
AToken^.Kind := lexerKindColon
ELSIF ALexer^.Start^ = '>' THEN
AToken^.Kind := lexerKindGreaterThan
ELSIF ALexer^.Start^ = '<' THEN
AToken^.Kind := lexerKindLessThan
ELSIF ALexer^.Start^ = '(' THEN
AToken^.Kind := lexerKindLeftParen
ELSIF ALexer^.Start^ = '-' THEN
AToken^.Kind := lexerKindLeftParen
ELSIF ALexer^.Start^ = '.' THEN
AToken^.Kind := lexerKindDot
END
END TransitionActionFinalize;
(* An action for tokens containing multiple characters. *)
PROCEDURE TransitionActionComposite(ALexer: PLexer; AToken: PLexerToken);
BEGIN
IF ALexer^.Start^ = '<' THEN
IF ALexer^.Current^ = '>' THEN
AToken^.Kind := lexerKindNotEqual
ELSIF ALexer^.Current^ = '=' THEN
AToken^.Kind := lexerKindLessEqual
END
ELSIF (ALexer^.Start^ = '>') AND (ALexer^.Current^ = '=') THEN
AToken^.Kind := lexerKindGreaterEqual
ELSIF (ALexer^.Start^ = '.') AND (ALexer^.Current^ = '.') THEN
AToken^.Kind := lexerKindRange
ELSIF (ALexer^.Start^ = ':') AND (ALexer^.Current^ = '=') THEN
AToken^.Kind := lexerKindAssignment
END;
INC(ALexer^.Current)
END TransitionActionComposite;
(* Skip a space. *)
PROCEDURE TransitionActionSkip(ALexer: PLexer; AToken: PLexerToken);
BEGIN
INC(ALexer^.Current);
INC(ALexer^.Start)
END TransitionActionSkip;
(* 0x04. Delimited string action. *)
PROCEDURE TransitionActionDelimited(ALexer: PLexer; AToken: PLexerToken);
BEGIN
IF ALexer^.Start^ = '(' THEN
AToken^.Kind := lexerKindComment
ELSIF ALexer^.Start^ = '"' THEN
AToken^.Kind := lexerKindCharacter
ELSIF ALexer^.Start^ = "'" THEN
AToken^.Kind := lexerKindString
END;
INC(ALexer^.Current)
END TransitionActionDelimited;
(* Finalize keyword or identifier. *)
PROCEDURE TransitionActionKeyId(ALexer: PLexer; AToken: PLexerToken);
BEGIN
IF CompareKeyword('PROGRAM', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindProgram
ELSIF CompareKeyword('IMPORT', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindImport
ELSIF CompareKeyword('CONST', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindConst
ELSIF CompareKeyword('VAR', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindVar
ELSIF CompareKeyword('IF', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindIf
ELSIF CompareKeyword('THEN', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindThen
ELSIF CompareKeyword('ELSIF', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindElsif
ELSIF CompareKeyword('ELSE', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindElse
ELSIF CompareKeyword('WHILE', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindWhile
ELSIF CompareKeyword('DO', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindDo
ELSIF CompareKeyword('PROCEDURE', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindProc
ELSIF CompareKeyword('BEGIN', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindBegin
ELSIF CompareKeyword('END', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindEnd
ELSIF CompareKeyword('TYPE', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindType
ELSIF CompareKeyword('RECORD', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindRecord
ELSIF CompareKeyword('UNION', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindUnion
ELSIF CompareKeyword('NIL', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindNull
ELSIF CompareKeyword('AND', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindAnd
ELSIF CompareKeyword('OR', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindOr
ELSIF CompareKeyword('RETURN', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindReturn
ELSIF CompareKeyword('DEFINITION', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindDefinition
ELSIF CompareKeyword('TO', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindTo
ELSIF CompareKeyword('CASE', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindCase
ELSIF CompareKeyword('OF', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindOf
ELSIF CompareKeyword('FROM', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindFrom
ELSIF CompareKeyword('MODULE', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindModule
ELSIF CompareKeyword('IMPLEMENTATION', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindImplementation
ELSIF CompareKeyword('POINTER', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindPointer
ELSIF CompareKeyword('ARRAY', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindArray
ELSIF CompareKeyword('TRUE', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindBoolean;
AToken^.booleanKind := TRUE
ELSIF CompareKeyword('FALSE', ALexer^.Start, ALexer^.Current) THEN
AToken^.Kind := lexerKindBoolean;
AToken^.booleanKind := FALSE
ELSE
AToken^.Kind := lexerKindIdentifier
END;
END TransitionActionKeyId;
(* Action for tokens containing only one character. The character cannot be
* followed by other characters forming a composite token. *)
PROCEDURE TransitionActionSingle(ALexer: PLexer; AToken: PLexerToken);
BEGIN
CASE ALexer^.Current^ OF
'&': AToken^.Kind := lexerKindAnd |
';': AToken^.Kind := lexerKindSemicolon |
',': AToken^.Kind := lexerKindComma |
')': AToken^.Kind := lexerKindRightParen |
'[': AToken^.Kind := lexerKindLeftSquare |
']': AToken^.Kind := lexerKindRightSquare |
'^': AToken^.Kind := lexerKindHat |
'=': AToken^.Kind := lexerKindEqual |
'+': AToken^.Kind := lexerKindPlus |
'/': AToken^.Kind := lexerKindDivision |
'%': AToken^.Kind := lexerKindRemainder |
'@': AToken^.Kind := lexerKindAt |
'|': AToken^.Kind := lexerKindPipe
END;
INC(ALexer^.Current)
END TransitionActionSingle;
(* Handle an integer literal. *)
PROCEDURE TransitionActionInteger(ALexer: PLexer; AToken: PLexerToken);
BEGIN
AToken^.Kind := lexerKindInteger
END TransitionActionInteger;
PROCEDURE SetDefaultTransition(CurrentState: TransitionState; DefaultAction: TransitionAction; NextState: TransitionState);
VAR DefaultTransition: Transition;
BEGIN
DefaultTransition.Action := DefaultAction;
DefaultTransition.NextState := NextState;
Transitions[ORD(CurrentState)][ORD(transitionClassInvalid)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassDigit)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassAlpha)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassSpace)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassColon)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassEquals)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassLeftParen)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassRightParen)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassAsterisk)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassUnderscore)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassSingle)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassHex)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassZero)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassX)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassEof)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassDot)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassMinus)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassSingleQuote)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassDoubleQuote)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassGreater)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassLess)] := DefaultTransition;
Transitions[ORD(CurrentState)][ORD(transitionClassOther)] := DefaultTransition;
END SetDefaultTransition;
(*
* The transition table describes transitions from one state to another, given
* a symbol (character class).
*
* The table has m rows and n columns, where m is the amount of states and n is
* the amount of classes. So given the current state and a classified character
* the table can be used to look up the next state.
*
* Each cell is a word long.
* - The least significant byte of the word is a row number (beginning with 0).
* It specifies the target state. "ff" means that this is an end state and no
* transition is possible.
* - The next byte is the action that should be performed when transitioning.
* For the meaning of actions see labels in the lex_next function, which
* handles each action.
*)
PROCEDURE InitializeTransitions();
BEGIN
(* Start state. *)
Transitions[ORD(transitionStateStart)][ORD(transitionClassInvalid)].Action := NIL;
Transitions[ORD(transitionStateStart)][ORD(transitionClassInvalid)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateStart)][ORD(transitionClassDigit)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassDigit)].NextState := transitionStateDecimal;
Transitions[ORD(transitionStateStart)][ORD(transitionClassAlpha)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassAlpha)].NextState := transitionStateIdentifier;
Transitions[ORD(transitionStateStart)][ORD(transitionClassSpace)].Action := TransitionActionSkip;
Transitions[ORD(transitionStateStart)][ORD(transitionClassSpace)].NextState := transitionStateStart;
Transitions[ORD(transitionStateStart)][ORD(transitionClassColon)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassColon)].NextState := transitionStateColon;
Transitions[ORD(transitionStateStart)][ORD(transitionClassEquals)].Action := TransitionActionSingle;
Transitions[ORD(transitionStateStart)][ORD(transitionClassEquals)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateStart)][ORD(transitionClassLeftParen)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassLeftParen)].NextState := transitionStateLeftParen;
Transitions[ORD(transitionStateStart)][ORD(transitionClassRightParen)].Action := TransitionActionSingle;
Transitions[ORD(transitionStateStart)][ORD(transitionClassRightParen)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateStart)][ORD(transitionClassAsterisk)].Action := TransitionActionSingle;
Transitions[ORD(transitionStateStart)][ORD(transitionClassAsterisk)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateStart)][ORD(transitionClassUnderscore)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassUnderscore)].NextState := transitionStateIdentifier;
Transitions[ORD(transitionStateStart)][ORD(transitionClassSingle)].Action := TransitionActionSingle;
Transitions[ORD(transitionStateStart)][ORD(transitionClassSingle)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateStart)][ORD(transitionClassHex)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassHex)].NextState := transitionStateIdentifier;
Transitions[ORD(transitionStateStart)][ORD(transitionClassZero)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassZero)].NextState := transitionStateLeadingZero;
Transitions[ORD(transitionStateStart)][ORD(transitionClassX)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassX)].NextState := transitionStateIdentifier;
Transitions[ORD(transitionStateStart)][ORD(transitionClassEof)].Action := TransitionActionEof;
Transitions[ORD(transitionStateStart)][ORD(transitionClassEof)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateStart)][ORD(transitionClassDot)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassDot)].NextState := transitionStateDot;
Transitions[ORD(transitionStateStart)][ORD(transitionClassMinus)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassMinus)].NextState := transitionStateMinus;
Transitions[ORD(transitionStateStart)][ORD(transitionClassSingleQuote)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassSingleQuote)].NextState := transitionStateCharacter;
Transitions[ORD(transitionStateStart)][ORD(transitionClassDoubleQuote)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassDoubleQuote)].NextState := transitionStateString;
Transitions[ORD(transitionStateStart)][ORD(transitionClassGreater)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassGreater)].NextState := transitionStateGreater;
Transitions[ORD(transitionStateStart)][ORD(transitionClassLess)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateStart)][ORD(transitionClassLess)].NextState := transitionStateLess;
Transitions[ORD(transitionStateStart)][ORD(transitionClassOther)].Action := NIL;
Transitions[ORD(transitionStateStart)][ORD(transitionClassOther)].NextState := transitionStateEnd;
(* Colon state. *)
SetDefaultTransition(transitionStateColon, TransitionActionFinalize, transitionStateEnd);
Transitions[ORD(transitionStateColon)][ORD(transitionClassEquals)].Action := TransitionActionComposite;
Transitions[ORD(transitionStateColon)][ORD(transitionClassEquals)].NextState := transitionStateEnd;
(* Identifier state. *)
SetDefaultTransition(transitionStateIdentifier, TransitionActionKeyId, transitionStateEnd);
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassDigit)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassDigit)].NextState := transitionStateIdentifier;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassAlpha)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassAlpha)].NextState := transitionStateIdentifier;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassUnderscore)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassUnderscore)].NextState := transitionStateIdentifier;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassHex)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassHex)].NextState := transitionStateIdentifier;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassZero)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassZero)].NextState := transitionStateIdentifier;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassX)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateIdentifier)][ORD(transitionClassX)].NextState := transitionStateIdentifier;
(* Decimal state. *)
SetDefaultTransition(transitionStateDecimal, TransitionActionInteger, transitionStateEnd);
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassDigit)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassDigit)].NextState := transitionStateDecimal;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassAlpha)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassAlpha)].NextState := transitionStateDecimalSuffix;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassUnderscore)].Action := NIL;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassUnderscore)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassHex)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassHex)].NextState := transitionStateDecimalSuffix;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassZero)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassZero)].NextState := transitionStateDecimal;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassX)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateDecimal)][ORD(transitionClassX)].NextState := transitionStateDecimalSuffix;
(* Greater state. *)
SetDefaultTransition(transitionStateGreater, TransitionActionFinalize, transitionStateEnd);
Transitions[ORD(transitionStateGreater)][ORD(transitionClassEquals)].Action := TransitionActionComposite;
Transitions[ORD(transitionStateGreater)][ORD(transitionClassEquals)].NextState := transitionStateEnd;
(* Minus state. *)
SetDefaultTransition(transitionStateMinus, TransitionActionFinalize, transitionStateEnd);
Transitions[ORD(transitionStateMinus)][ORD(transitionClassGreater)].Action := TransitionActionComposite;
Transitions[ORD(transitionStateMinus)][ORD(transitionClassGreater)].NextState := transitionStateEnd;
(* Left paren state. *)
SetDefaultTransition(transitionStateLeftParen, TransitionActionFinalize, transitionStateEnd);
Transitions[ORD(transitionStateLeftParen)][ORD(transitionClassAsterisk)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateLeftParen)][ORD(transitionClassAsterisk)].NextState := transitionStateComment;
(* Less state. *)
SetDefaultTransition(transitionStateLess, TransitionActionFinalize, transitionStateEnd);
Transitions[ORD(transitionStateLess)][ORD(transitionClassEquals)].Action := TransitionActionComposite;
Transitions[ORD(transitionStateLess)][ORD(transitionClassEquals)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateLess)][ORD(transitionClassGreater)].Action := TransitionActionComposite;
Transitions[ORD(transitionStateLess)][ORD(transitionClassGreater)].NextState := transitionStateEnd;
(* Hexadecimal after 0x. *)
SetDefaultTransition(transitionStateDot, TransitionActionFinalize, transitionStateEnd);
Transitions[ORD(transitionStateDot)][ORD(transitionClassDot)].Action := TransitionActionComposite;
Transitions[ORD(transitionStateDot)][ORD(transitionClassDot)].NextState := transitionStateEnd;
(* Comment. *)
SetDefaultTransition(transitionStateComment, TransitionActionAccumulate, transitionStateComment);
Transitions[ORD(transitionStateComment)][ORD(transitionClassAsterisk)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateComment)][ORD(transitionClassAsterisk)].NextState := transitionStateClosingComment;
Transitions[ORD(transitionStateComment)][ORD(transitionClassEof)].Action := NIL;
Transitions[ORD(transitionStateComment)][ORD(transitionClassEof)].NextState := transitionStateEnd;
(* Closing comment. *)
SetDefaultTransition(transitionStateClosingComment, TransitionActionAccumulate, transitionStateComment);
Transitions[ORD(transitionStateClosingComment)][ORD(transitionClassInvalid)].Action := NIL;
Transitions[ORD(transitionStateClosingComment)][ORD(transitionClassInvalid)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateClosingComment)][ORD(transitionClassRightParen)].Action := TransitionActionDelimited;
Transitions[ORD(transitionStateClosingComment)][ORD(transitionClassRightParen)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateClosingComment)][ORD(transitionClassAsterisk)].Action := TransitionActionAccumulate;
Transitions[ORD(transitionStateClosingComment)][ORD(transitionClassAsterisk)].NextState := transitionStateClosingComment;
Transitions[ORD(transitionStateClosingComment)][ORD(transitionClassEof)].Action := NIL;
Transitions[ORD(transitionStateClosingComment)][ORD(transitionClassEof)].NextState := transitionStateEnd;
(* Character. *)
SetDefaultTransition(transitionStateCharacter, TransitionActionAccumulate, transitionStateCharacter);
Transitions[ORD(transitionStateCharacter)][ORD(transitionClassInvalid)].Action := NIL;
Transitions[ORD(transitionStateCharacter)][ORD(transitionClassInvalid)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateCharacter)][ORD(transitionClassEof)].Action := NIL;
Transitions[ORD(transitionStateCharacter)][ORD(transitionClassEof)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateCharacter)][ORD(transitionClassSingleQuote)].Action := TransitionActionDelimited;
Transitions[ORD(transitionStateCharacter)][ORD(transitionClassSingleQuote)].NextState := transitionStateEnd;
(* String. *)
SetDefaultTransition(transitionStateString, TransitionActionAccumulate, transitionStateString);
Transitions[ORD(transitionStateString)][ORD(transitionClassInvalid)].Action := NIL;
Transitions[ORD(transitionStateString)][ORD(transitionClassInvalid)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateString)][ORD(transitionClassEof)].Action := NIL;
Transitions[ORD(transitionStateString)][ORD(transitionClassEof)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateString)][ORD(transitionClassDoubleQuote)].Action := TransitionActionDelimited;
Transitions[ORD(transitionStateString)][ORD(transitionClassDoubleQuote)].NextState := transitionStateEnd;
(* Leading zero. *)
SetDefaultTransition(transitionStateLeadingZero, TransitionActionInteger, transitionStateEnd);
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassDigit)].Action := NIL;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassDigit)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassAlpha)].Action := NIL;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassAlpha)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassUnderscore)].Action := NIL;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassUnderscore)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassHex)].Action := NIL;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassHex)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassZero)].Action := NIL;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassZero)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassX)].Action := NIL;
Transitions[ORD(transitionStateLeadingZero)][ORD(transitionClassX)].NextState := transitionStateEnd;
(* Digit with a character suffix. *)
SetDefaultTransition(transitionStateDecimalSuffix, TransitionActionInteger, transitionStateEnd);
Transitions[ORD(transitionStateDecimalSuffix)][ORD(transitionClassAlpha)].Action := NIL;
Transitions[ORD(transitionStateDecimalSuffix)][ORD(transitionClassAlpha)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateDecimalSuffix)][ORD(transitionClassDigit)].Action := NIL;
Transitions[ORD(transitionStateDecimalSuffix)][ORD(transitionClassDigit)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateDecimalSuffix)][ORD(transitionClassHex)].Action := NIL;
Transitions[ORD(transitionStateDecimalSuffix)][ORD(transitionClassHex)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateDecimalSuffix)][ORD(transitionClassZero)].Action := NIL;
Transitions[ORD(transitionStateDecimalSuffix)][ORD(transitionClassZero)].NextState := transitionStateEnd;
Transitions[ORD(transitionStateDecimalSuffix)][ORD(transitionClassX)].Action := NIL;
Transitions[ORD(transitionStateDecimalSuffix)][ORD(transitionClassX)].NextState := transitionStateEnd
END InitializeTransitions;
PROCEDURE LexerInitialize(ALexer: PLexer; Input: File);
BEGIN
ALexer^.Input := Input;
ALexer^.Length := 0;
ALLOCATE(ALexer^.Buffer, ChunkSize);
MemZero(ALexer^.Buffer, ChunkSize);
ALexer^.Size := ChunkSize
END LexerInitialize;
PROCEDURE LexerCurrent(ALexer: PLexer): LexerToken;
VAR
CurrentClass: TransitionClass;
CurrentState: TransitionState;
CurrentTransition: Transition;
Result: LexerToken;
BEGIN
ALexer^.Current := ALexer^.Start;
CurrentState := transitionStateStart;
WHILE CurrentState <> transitionStateEnd DO
CurrentClass := Classification[ORD(ALexer^.Current^) + 1];
CurrentTransition := Transitions[ORD(CurrentState)][ORD(CurrentClass)];
IF CurrentTransition.Action <> NIL THEN
CurrentTransition.Action(ALexer, ADR(Result))
END;
CurrentState := CurrentTransition.NextState
END;
RETURN Result
END LexerCurrent;
PROCEDURE LexerLex(ALexer: PLexer): LexerToken;
BEGIN
IF ALexer^.Length = 0 THEN
ALexer^.Length := ReadNBytes(ALexer^.Input, ChunkSize, ALexer^.Buffer);
ALexer^.Current := ALexer^.Buffer
END;
ALexer^.Start := ALexer^.Current;
RETURN LexerCurrent(ALexer)
END LexerLex;
PROCEDURE LexerDestroy(ALexer: PLexer);
BEGIN
DEALLOCATE(ALexer^.Buffer, ALexer^.Size)
END LexerDestroy;
BEGIN
InitializeClassification();
InitializeTransitions()
END Lexer.