394 lines
9.3 KiB
Haxe
394 lines
9.3 KiB
Haxe
/*
|
|
* Copyright (c) 2008, Nicolas Cannasse
|
|
* All rights reserved.
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions are met:
|
|
*
|
|
* - Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
* - Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE HAXE PROJECT CONTRIBUTORS "AS IS" AND ANY
|
|
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
|
|
* WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
|
|
* DISCLAIMED. IN NO EVENT SHALL THE HAXE PROJECT CONTRIBUTORS BE LIABLE FOR
|
|
* ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
|
|
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
|
|
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
|
|
* DAMAGE.
|
|
*/
|
|
package hscript;
|
|
import hscript.Expr;
|
|
|
|
enum Token {
|
|
TEof;
|
|
TConst( c : Const );
|
|
TId( s : String );
|
|
TOp( s : String );
|
|
TPOpen;
|
|
TPClose;
|
|
TBrOpen;
|
|
TBrClose;
|
|
TDot;
|
|
TComma;
|
|
TSemicolon;
|
|
}
|
|
|
|
class Parser {
|
|
|
|
// config / variables
|
|
public var line : Int;
|
|
public var opChars : String;
|
|
public var identChars : String;
|
|
public var opPriority : Array<String>;
|
|
public var unopsPrefix : Array<String>;
|
|
public var unopsSuffix : Array<String>;
|
|
|
|
// implementation
|
|
var char : Null<Int>;
|
|
var ops : Array<Bool>;
|
|
var idents : Array<Bool>;
|
|
var tokens : haxe.FastList<Token>;
|
|
|
|
public function new() {
|
|
line = 1;
|
|
opChars = "+*/-=!><&|^%";
|
|
identChars = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789_";
|
|
opPriority = [
|
|
"=",
|
|
"||","&&",
|
|
"==","!=",">","<",">=","<=",
|
|
"|","&","^",
|
|
"<<",">>",">>>",
|
|
"+","-",
|
|
"*","/",
|
|
"%"
|
|
];
|
|
unopsPrefix = ["!","++","--","-"];
|
|
unopsSuffix = ["++","--"];
|
|
}
|
|
|
|
public function parseString( s : String ) {
|
|
return parse( new haxe.io.StringInput(s) );
|
|
}
|
|
|
|
public function parse( s : haxe.io.Input ) {
|
|
char = null;
|
|
ops = new Array();
|
|
idents = new Array();
|
|
tokens = new haxe.FastList<Token>();
|
|
for( i in 0...opChars.length )
|
|
ops[opChars.charCodeAt(i)] = true;
|
|
for( i in 0...identChars.length )
|
|
idents[identChars.charCodeAt(i)] = true;
|
|
var a = new Array();
|
|
while( true ) {
|
|
var tk = token(s);
|
|
if( tk == TEof ) break;
|
|
tokens.add(tk);
|
|
a.push(parseFullExpr(s));
|
|
}
|
|
return a;
|
|
}
|
|
|
|
function unexpected( tk ) : Dynamic {
|
|
throw Error.EUnexpected(tokenString(tk));
|
|
return null;
|
|
}
|
|
|
|
function parseFullExpr(s) {
|
|
var e = parseExpr(s);
|
|
var tk = token(s);
|
|
if( tk != TSemicolon && tk != TEof )
|
|
switch( e ) {
|
|
case EBlock(_): tokens.add(tk);
|
|
default: unexpected(tk);
|
|
}
|
|
return e;
|
|
}
|
|
|
|
function parseExpr( s : haxe.io.Input ) {
|
|
var tk = token(s);
|
|
switch( tk ) {
|
|
case TId(id):
|
|
var e = parseStructure(s,id);
|
|
if( e == null )
|
|
return parseExprNext(s,EIdent(id));
|
|
return e;
|
|
case TConst(c):
|
|
return parseExprNext(s,EConst(c));
|
|
case TPOpen:
|
|
var e = parseExpr(s);
|
|
tk = token(s);
|
|
if( tk != TPClose ) unexpected(tk);
|
|
return parseExprNext(s,EParent(e));
|
|
case TBrOpen:
|
|
var a = new Array();
|
|
while( true ) {
|
|
tk = token(s);
|
|
if( tk == TBrClose )
|
|
break;
|
|
tokens.add(tk);
|
|
a.push(parseFullExpr(s));
|
|
}
|
|
return EBlock(a);
|
|
case TOp(op):
|
|
var found;
|
|
for( x in unopsPrefix )
|
|
if( x == op )
|
|
return EUnop(op,true,parseExpr(s));
|
|
return unexpected(tk);
|
|
default:
|
|
return unexpected(tk);
|
|
}
|
|
}
|
|
|
|
function parseStructure( s, id ) {
|
|
return switch( id ) {
|
|
case "if":
|
|
var cond = parseExpr(s);
|
|
var e1 = parseExpr(s);
|
|
var e2 = null;
|
|
var semic = false;
|
|
var tk = token(s);
|
|
if( tk == TSemicolon ) {
|
|
semic = true;
|
|
tk = token(s);
|
|
}
|
|
if( Type.enumEq(tk,TId("else")) )
|
|
e2 = parseExpr(s);
|
|
else {
|
|
tokens.add(tk);
|
|
if( semic ) tokens.add(TSemicolon);
|
|
}
|
|
EIf(cond,e1,e2);
|
|
case "var":
|
|
var tk = token(s);
|
|
var ident;
|
|
switch(tk) {
|
|
case TId(id): ident = id;
|
|
default: unexpected(tk);
|
|
}
|
|
tk = token(s);
|
|
var e;
|
|
if( Type.enumEq(tk,TOp("=")) )
|
|
e = parseExpr(s);
|
|
else
|
|
tokens.add(tk);
|
|
EVar(ident,e);
|
|
case "while":
|
|
var econd = parseExpr(s);
|
|
var e = parseExpr(s);
|
|
EWhile(econd,e);
|
|
case "for":
|
|
var tk = token(s);
|
|
if( tk != TPOpen ) unexpected(tk);
|
|
tk = token(s);
|
|
var vname;
|
|
switch( tk ) {
|
|
case TId(id): vname = id;
|
|
default: unexpected(tk);
|
|
}
|
|
tk = token(s);
|
|
if( !Type.enumEq(tk,TId("in")) ) unexpected(tk);
|
|
var eiter = parseExpr(s);
|
|
tk = token(s);
|
|
if( tk != TPClose ) unexpected(tk);
|
|
EFor(vname,eiter,parseExpr(s));
|
|
case "break": EBreak;
|
|
case "continue": EContinue;
|
|
case "else": unexpected(TId(id));
|
|
default: null;
|
|
}
|
|
}
|
|
|
|
function priority(op) {
|
|
for( i in 0...opPriority.length )
|
|
if( opPriority[i] == op )
|
|
return i;
|
|
return -1;
|
|
}
|
|
|
|
function parseExprNext( s : haxe.io.Input, e1 : Expr ) {
|
|
var tk = token(s);
|
|
switch( tk ) {
|
|
case TOp(op):
|
|
for( x in unopsSuffix )
|
|
if( x == op )
|
|
return EUnop(op,false,e1);
|
|
var e2 = parseExpr(s);
|
|
switch( e2 ) {
|
|
case EBinop(op2,e2,e3):
|
|
if( priority(op) > priority(op2) )
|
|
return EBinop(op2,EBinop(op,e1,e2),e3);
|
|
default:
|
|
}
|
|
return EBinop(op,e1,e2);
|
|
case TDot:
|
|
tk = token(s);
|
|
var field;
|
|
switch(tk) {
|
|
case TId(id): field = id;
|
|
default: unexpected(tk);
|
|
}
|
|
return parseExprNext(s,EField(e1,field));
|
|
case TPOpen:
|
|
var args = new Array();
|
|
tk = token(s);
|
|
if( tk != TPClose ) {
|
|
tokens.add(tk);
|
|
while( true ) {
|
|
args.push(parseExpr(s));
|
|
tk = token(s);
|
|
switch( tk ) {
|
|
case TComma:
|
|
case TPClose: break;
|
|
default: unexpected(tk);
|
|
}
|
|
}
|
|
}
|
|
return ECall(e1,args);
|
|
default:
|
|
tokens.add(tk);
|
|
return e1;
|
|
}
|
|
}
|
|
|
|
function readChar( s : haxe.io.Input ) {
|
|
return try s.readByte() catch( e : Dynamic ) 0;
|
|
}
|
|
|
|
function readString( s : haxe.io.Input, until ) {
|
|
var c;
|
|
var b = new StringBuf();
|
|
var esc = false;
|
|
while( true ) {
|
|
try {
|
|
c = s.readByte();
|
|
} catch( e : Dynamic ) {
|
|
throw Error.EUnterminatedString;
|
|
}
|
|
if( esc ) {
|
|
esc = false;
|
|
switch( c ) {
|
|
case 110: b.addChar(10); // \n
|
|
case 114: b.addChar(13); // \r
|
|
case 116: b.addChar(9); // \t
|
|
case 39: b.addChar(39); // \'
|
|
case 34: b.addChar(34); // \"
|
|
case 92: b.addChar(92); // \\
|
|
default: throw Error.EInvalidChar(c);
|
|
}
|
|
} else if( c == 92 )
|
|
esc = true;
|
|
else if( c == until )
|
|
break;
|
|
else
|
|
b.addChar(c);
|
|
}
|
|
return b.toString();
|
|
}
|
|
|
|
function token( s : haxe.io.Input ) {
|
|
if( !tokens.isEmpty() )
|
|
return tokens.pop();
|
|
var char;
|
|
if( this.char == null )
|
|
char = readChar(s);
|
|
else {
|
|
char = this.char;
|
|
this.char = null;
|
|
}
|
|
while( true ) {
|
|
switch( char ) {
|
|
case 0: return TEof;
|
|
case 32,9,13: // space, tab, CR
|
|
case 10: line++; // LF
|
|
case 48,49,50,51,52,53,54,55,56,57: // 0...9
|
|
var n = char - 48;
|
|
var exp = 0;
|
|
while( true ) {
|
|
char = readChar(s);
|
|
exp *= 10;
|
|
switch( char ) {
|
|
case 48,49,50,51,52,53,54,55,56,57:
|
|
n = n * 10 + (char - 48);
|
|
case 46:
|
|
if( exp > 0 )
|
|
throw Error.EInvalidChar(char);
|
|
exp = 1;
|
|
default:
|
|
this.char = char;
|
|
return TConst( (exp > 0) ? CFloat(n * 10 / exp) : CInt(n) );
|
|
}
|
|
}
|
|
case 59: return TSemicolon;
|
|
case 40: return TPOpen;
|
|
case 41: return TPClose;
|
|
case 44: return TComma;
|
|
case 46: return TDot;
|
|
case 123: return TBrOpen;
|
|
case 125: return TBrClose;
|
|
case 39: return TConst( CString(readString(s,39)) );
|
|
case 34: return TConst( CString(readString(s,34)) );
|
|
default:
|
|
if( ops[char] ) {
|
|
var op = Std.chr(char);
|
|
while( true ) {
|
|
char = readChar(s);
|
|
if( !ops[char] ) {
|
|
this.char = char;
|
|
return TOp(op);
|
|
}
|
|
op += Std.chr(char);
|
|
}
|
|
}
|
|
if( idents[char] ) {
|
|
var id = Std.chr(char);
|
|
while( true ) {
|
|
char = readChar(s);
|
|
if( !idents[char] ) {
|
|
this.char = char;
|
|
return TId(id);
|
|
}
|
|
id += Std.chr(char);
|
|
}
|
|
}
|
|
throw Error.EInvalidChar(char);
|
|
}
|
|
char = readChar(s);
|
|
}
|
|
return null;
|
|
}
|
|
|
|
function constString( c ) {
|
|
return switch(c) {
|
|
case CInt(v): Std.string(v);
|
|
case CFloat(f): Std.string(f);
|
|
case CString(s): s; // TODO : escape + quote
|
|
}
|
|
}
|
|
|
|
function tokenString( t ) {
|
|
return switch( t ) {
|
|
case TEof: "<eof>";
|
|
case TConst(c): constString(c);
|
|
case TId(s): s;
|
|
case TOp(s): s;
|
|
case TPOpen: "(";
|
|
case TPClose: ")";
|
|
case TBrOpen: "{";
|
|
case TBrClose: "}";
|
|
case TDot: ".";
|
|
case TComma: ",";
|
|
case TSemicolon: ";";
|
|
}
|
|
}
|
|
|
|
}
|