change readme and add new idea for lexer

This commit is contained in:
Aleksey Chichenkov 2019-01-25 14:25:21 +03:00
parent 04420a5f9f
commit 2374e2b5d1
4 changed files with 268 additions and 203 deletions

View File

@ -1,6 +1,7 @@
# re2-js-generator
### Install guide
You need install npm and node v.10;
npm install
@ -12,8 +13,10 @@ Read re2c manual: http://re2c.org/manual/manual.html
or
node parse_source_lexeme.js
node main.js
Result fill write in <lexer.js> file.
Result fill write in out.js file.
Report for bugs: rolahd@yandex.ru

440
lexer.l
View File

@ -1,190 +1,255 @@
var lex = [
"ERR",
"DELIM",
var types = [
"LSB",
"RSB",
"LCB",
"RCB",
"INTEGER",
"COLON",
"COMMA",
"DOT",
"REM",
"GT",
"GTE",
"LT",
"LTE",
"EQ",
"NEQ",
"LIKE",
"NLIKE",
"AND",
"OR",
"NOT",
"ADDRESS",
"TIME",
"TIMEDIFF",
"INTEGER_LITERAL",
"FLOAT_LITERAL",
"BOOL_LITERAL",
"INTEGER_LITERAL",
"ID"
];
var searchString = function (_str, _quote, _yylexstart, _yycursor) {
var found_back_slash = false;
_yycursor++;
while(_yycursor < _str.length){
var char = _str[_yycursor];
if(_quote == '"') {
switch (char) {
case "\\":
found_back_slash = true;
break;
case '"':
if(!found_back_slash) {
return { success: true, pos: _yycursor + 1 }
}
found_back_slash = false;
break;
}
} else if(_quote == "'") {
switch (char) {
case "\\":
found_back_slash = true;
break;
case "'":
if(!found_back_slash) {
return { success: true, pos: _yycursor + 1 }
}
found_back_slash = false;
break;
}
}
_yycursor++;
}
return {success: false, pos: _yycursor + 1 }
var errors = {
"-2": "not found close quote or singleQuote",
"-1": "not found any lexemes or errors or anything else",
"0": "success",
"1": "found unknown symbol"
};
var addLexeme = function(_str, _yylexstart, _yycursor, _lexeme) {
if(_lexeme !== "ERR"){
console.log(print_f("found lex: %s; start: %s; end: %s; result => %s", _lexeme, _yylexstart, _yycursor, _str.substring(_yylexstart, _yycursor)));
} else {
console.log(print_f("search end\n"));
return true;
}
var Lexer = function(_string){
this._last_found_lexeme = {error: -1};
this._end = false;
this._error = false;
this._string = _string;
this._state = 1;
this._yy_char = null;
this._yy_lex_start = 0;
this._yy_cursor = 0;
this._yy_marker = 0;
this._yy_accept = 0;
};
var unknownSymbol = function(_str, _yylexstart, _yycursor){
throw print_f("Found unknown symbol on position: %s", _yycursor)
};
Lexer.prototype = {
types: types,
errors: errors,
_notFoundCloseQuote: function() {
this._error = true;
this._last_found_lexeme = {
error: 2,
start: this._yy_lex_start,
end: this._yy_cursor
};
var notFoundCloseQuote = function(_str, _yylexstart, _yycursor) {
console.log( print_f("Not found close quote start: %s", _yycursor));
throw print_f("Not found close quote start: %s", _yycursor);
};
var start_search = function(_str) {
console.log("start search", _str);
var id = 1;
var yych = null;
var YYLEXSTART = 0;
var yyaccept = 0;
var YYCURSOR = 0;
var YYMARKER = 0;
var str = _str;
var reset = function(){
yyaccept = 0;
id = 1;
YYLEXSTART = YYCURSOR;
YYMARKER = YYCURSOR;
console.log( print_f("Not found close quote start: %s", this._yy_cursor));
},
_unknownSymbol: function(){
this._error = true;
this._last_found_lexeme = {
error: 1,
start: this._yy_lex_start,
end: this._yy_cursor
};
while(true)
{
switch(id) /*!re2c
re2c:define:YYCTYPE = _r2c_var_;
re2c:yyfill:enable = 0;
console.log( print_f("Found unknown symbol on position: %s", this._yy_cursor));
},
_foundLexeme: function(_lexeme) {
console.log(print_f("found lex: %s; start: %s; end: %s; result => %s", _lexeme, this._yy_lex_start, this._yy_cursor, this._string.substring(this._yy_lex_start, this._yy_cursor)));
this._last_found_lexeme = {
error: 0,
lexeme: _lexeme,
start: this._yy_lex_start,
end: this._yy_cursor
};
},
_endOfString: function(){
console.log(print_f("search end\n"));
this._end = true;
this._last_found_lexeme = {
error: -2
};
},
_searchString: function () {
var _quote = this._string[this._yy_cursor - 1];
var found_back_slash = false;
while(this._yy_cursor < this._string.length){
this._yy_char = this._string[this._yy_cursor];
if(_quote == '"') {
switch (this._yy_char) {
case "\\":
found_back_slash = true;
break;
case '"':
if(!found_back_slash) {
this._yy_cursor++;
this._foundLexeme("STRING_LITERAL");
return;
}
found_back_slash = false;
break;
}
} else if(_quote == "'") {
switch (this._yy_char) {
case "\\":
found_back_slash = true;
break;
case "'":
if(!found_back_slash) {
this._yy_cursor++;
this._foundLexeme("STRING_LITERAL");
return;
}
found_back_slash = false;
break;
}
}
this._yy_cursor++;
}
D = [0-9];
end = "\x00";
L = [A-Za-z_];
RL = [\U00000400-\U00000451];
this._notFoundCloseQuote();
},
_set_next: function(){
this._yy_accept = 0;
this._state = 1;
this._yy_lex_start = this._yy_cursor;
this._yy_marker = this._yy_cursor;
},
next: function(){
if(this._end || this._error) return null;
this.search();
return this.token();
},
token: function(){
return this._last_found_lexeme;
},
search: function(){
if(this._end) return false;
CR = "\r";
LF = "\n";
CRLF = CR?LF;
INTEGER = "-"?D+;
SP = " ";
TAB = "\t";
while(true){
switch(id) /*!re2c
re2c:define:YYCTYPE = _r2c_var_;
re2c:define:YYCURSOR = this._yy_cursor;
re2c:define:YYMARKER = this._yy_marker;
re2c:yyfill:enable = 0;
DELIM = SP|TAB|CR|LF;
D = [0-9];
end = "\x00";
L = [A-Za-z_];
RL = [\U00000400-\U00000451];
LSB = "[";
RSB = "]";
LCB = "(";
RCB = ")";
COLON = ":";
COMMA = ",";
DOT = ".";
REM = "%";
GT = ">";
GTE = ">=";
LT = "<";
LTE = "<=";
EQ = "==";
NEQ = "!=";
CR = "\r";
LF = "\n";
CRLF = CR?LF;
INTEGER = "-"?D+;
SP = " ";
TAB = "\t";
AND = 'AND';
OR = 'OR';
NOT = 'NOT';
LIKE = 'LIKE';
NLIKE = 'NLIKE';
DELIM = SP|TAB|CR|LF;
ADDRESS = "Address";
TIME = "Time";
TIMEDIFF = "TimeDiff";
LSB = "[";
RSB = "]";
LCB = "(";
RCB = ")";
COLON = ":";
COMMA = ",";
DOT = ".";
REM = "%";
GT = ">";
GTE = ">=";
LT = "<";
LTE = "<=";
EQ = "==";
NEQ = "!=";
BOOL_LITERAL = 'true'|'false';
FLOAT_LITERAL = "-"? D* "." D+ ("e" "-"? D+)?;
INTEGER_LITERAL = INTEGER;
ID = L(L|D)*;
AND = 'AND';
OR = 'OR';
NOT = 'NOT';
LIKE = 'LIKE';
NLIKE = 'NLIKE';
QU = "\"";
SQU = "'";
ADDRESS = "Address";
TIME = "Time";
TIMEDIFF = "TimeDiff";
BOOL_LITERAL = 'true'|'false';
FLOAT_LITERAL = "-"? D* "." D+ ("e" "-"? D+)?;
INTEGER_LITERAL = INTEGER;
ID = L(L|D)*;
QU = "\"";
SQU = "'";
end { if(addLexeme(str, YYLEXSTART, YYCURSOR, "ERR")) return; reset(); break; }
end { this._endOfString(); return; }
LSB { if(addLexeme(str, YYLEXSTART, YYCURSOR, "LSB")) return; reset(); break; }
RSB { if(addLexeme(str, YYLEXSTART, YYCURSOR, "RSB")) return; reset(); break; }
LCB { if(addLexeme(str, YYLEXSTART, YYCURSOR, "LCB")) return; reset(); break; }
RCB { if(addLexeme(str, YYLEXSTART, YYCURSOR, "RCB")) return; reset(); break; }
COLON { if(addLexeme(str, YYLEXSTART, YYCURSOR, "COLON")) return; reset(); break; }
COMMA { if(addLexeme(str, YYLEXSTART, YYCURSOR, "COMMA")) return; reset(); break; }
DOT { if(addLexeme(str, YYLEXSTART, YYCURSOR, "DOT")) return; reset(); break; }
REM { if(addLexeme(str, YYLEXSTART, YYCURSOR, "REM")) return; reset(); break; }
LSB { this._foundLexeme("LSB"); this._set_next(); return; }
RSB { this._foundLexeme("RSB"); this._set_next(); return; }
LCB { this._foundLexeme("LCB"); this._set_next(); return; }
RCB { this._foundLexeme("RCB"); this._set_next(); return; }
COLON { this._foundLexeme("COLON"); this._set_next(); return; }
COMMA { this._foundLexeme("COMMA"); this._set_next(); return; }
DOT { this._foundLexeme("DOT"); this._set_next(); return; }
REM { this._foundLexeme("REM"); this._set_next(); return; }
GT { if(addLexeme(str, YYLEXSTART, YYCURSOR, "GT")) return; reset(); break; }
GTE { if(addLexeme(str, YYLEXSTART, YYCURSOR, "GTE")) return; reset(); break; }
LT { if(addLexeme(str, YYLEXSTART, YYCURSOR, "LT")) return; reset(); break; }
LTE { if(addLexeme(str, YYLEXSTART, YYCURSOR, "LTE")) return; reset(); break; }
EQ { if(addLexeme(str, YYLEXSTART, YYCURSOR, "EQ")) return; reset(); break; }
NEQ { if(addLexeme(str, YYLEXSTART, YYCURSOR, "NEQ")) return; reset(); break; }
LIKE { if(addLexeme(str, YYLEXSTART, YYCURSOR, "LIKE")) return; reset(); break; }
NLIKE { if(addLexeme(str, YYLEXSTART, YYCURSOR, "NLIKE")) return; reset(); break; }
GT { this._foundLexeme("GT"); this._set_next(); return; }
GTE { this._foundLexeme("GTE"); this._set_next(); return; }
LT { this._foundLexeme("LT"); this._set_next(); return; }
LTE { this._foundLexeme("LTE"); this._set_next(); return; }
EQ { this._foundLexeme("EQ"); this._set_next(); return; }
NEQ { this._foundLexeme("NEQ"); this._set_next(); return; }
LIKE { this._foundLexeme("LIKE"); this._set_next(); return; }
NLIKE { this._foundLexeme("NLIKE"); this._set_next(); return; }
AND { if(addLexeme(str, YYLEXSTART, YYCURSOR, "AND")) return; reset(); break; }
OR { if(addLexeme(str, YYLEXSTART, YYCURSOR, "OR")) return; reset(); break; }
NOT { if(addLexeme(str, YYLEXSTART, YYCURSOR, "NOT")) return; reset(); break; }
AND { this._foundLexeme("AND"); this._set_next(); return; }
OR { this._foundLexeme("OR"); this._set_next(); return; }
NOT { this._foundLexeme("NOT"); this._set_next(); return; }
ADDRESS { if(addLexeme(str, YYLEXSTART, YYCURSOR, "ADDRESS")) return; reset(); break; }
TIME { if(addLexeme(str, YYLEXSTART, YYCURSOR, "TIME")) return; reset(); break; }
TIMEDIFF { if(addLexeme(str, YYLEXSTART, YYCURSOR, "TIMEDIFF")) return; reset(); break; }
ADDRESS { this._foundLexeme("ADDRESS"); this._set_next(); return; }
TIME { this._foundLexeme("TIME"); this._set_next(); return; }
TIMEDIFF { this._foundLexeme("TIMEDIFF"); this._set_next(); return; }
INTEGER_LITERAL { if(addLexeme(str, YYLEXSTART, YYCURSOR, "INTEGER_LITERAL")) return; reset(); break; }
FLOAT_LITERAL { if(addLexeme(str, YYLEXSTART, YYCURSOR, "FLOAT_LITERAL")) return; reset(); break; }
BOOL_LITERAL { if(addLexeme(str, YYLEXSTART, YYCURSOR, "BOOL_LITERAL")) return; reset(); break; }
INTEGER_LITERAL { this._foundLexeme("INTEGER_LITERAL"); this._set_next(); return; }
FLOAT_LITERAL { this._foundLexeme("FLOAT_LITERAL"); this._set_next(); return; }
BOOL_LITERAL { this._foundLexeme("BOOL_LITERAL"); this._set_next(); return; }
ID { if(addLexeme(str, YYLEXSTART, YYCURSOR, "ID")) return; reset(); break; }
DELIM { reset(); break; }
ID { this._foundLexeme("ID"); this._set_next(); return; }
DELIM { this._set_next(); break; }
QU|SQU { id = 100000000; break;}
QU|SQU { id = 100000000; break;}
[^] { if(unknownSymbol(str, YYLEXSTART, YYCURSOR)) return; reset(); break; }
*/ENDER}
[^] { this._unknownSymbol(); this._set_next(); return; }
*/ENDER}
}
};
var print_f = function () {
var print_f = function() {
var r_str = "";
var next = arguments[0];
var rx = /(%[a-zA-Z]{1})/;
var a = 1, match;
var a = 1,
match;
while (match = rx.exec(next)) {
var prev = next.substring(0, match.index);
var macro = next.substring(match.index + 1, match.index + 2);
@ -193,17 +258,16 @@ var print_f = function () {
var arg = arguments[a];
if(arg !== undefined) {
if (arg !== undefined) {
switch (macro) {
case "s":
if(arg.to_string && !arg.toString) r_str += arg.to_string();
if(arg.toString && !arg.to_string) r_str += arg.toString();
r_str += arg.toString();
break;
case "i":
r_str += (arg.to_number && arg.to_number()) || parseInt(arg);
r_str += parseInt(arg);
break;
case "f":
r_str += (arg.to_number && arg.to_number()) || parseFloat(arg);
r_str += parseFloat(arg);
break;
}
} else {
@ -217,41 +281,41 @@ var print_f = function () {
return r_str;
};
console.log("TEST SINGLE")
start_search("[");
start_search("]");
start_search("(");
start_search(")");
start_search(":");
start_search(",");
start_search(".");
start_search("%");
start_search(">");
start_search(">=");
start_search("<");
start_search("<=");
start_search("==");
start_search("!=");
start_search("AND");
start_search("and");
start_search("OR");
start_search("or");
start_search("NOT");
start_search("not");
start_search("LIKE");
start_search("like");
start_search("NLIKE");
start_search("nlike");
start_search("Address");
start_search("Time");
start_search("TimeDiff");
console.log("TEST SINGLE");
(new Lexer("[")).search();
(new Lexer("]")).search();
(new Lexer("(")).search();
(new Lexer(")").search());
(new Lexer(":")).search();
(new Lexer(",")).search();
(new Lexer(".")).search();
(new Lexer("%")).search();
(new Lexer(">")).search();
(new Lexer(">=")).search();
(new Lexer("<")).search();
(new Lexer("<=")).search();
(new Lexer("==")).search();
(new Lexer("!=")).search();
(new Lexer("AND")).search();
(new Lexer("and")).search();
(new Lexer("OR")).search();
(new Lexer("or")).search();
(new Lexer("NOT")).search();
(new Lexer("not")).search();
(new Lexer("LIKE")).search();
(new Lexer("like")).search();
(new Lexer("NLIKE")).search();
(new Lexer("nlike")).search();
(new Lexer("Address")).search();
(new Lexer("Time")).search();
(new Lexer("TimeDiff")).search();
console.log("TEST ALL")
start_search("[ ] ( ) : , . % > >= < <= == != AND and OR or NOT not LIKE like NLIKE nlike Address Time TimeDiff ");
var lex_test_all = new Lexer("[ ] ( ) : , . % > >= < <= == != AND and OR or NOT not LIKE like NLIKE nlike Address Time TimeDiff 'sdfadfasdf' \"asdfasfd\" ")
var _lex;
while(_lex = lex_test_all.next()){
console.log("IN while:", _lex.lexeme);
}
console.log("TEST STRING LITERAL");
start_search(' "111\\\"11\\\"1" "222222" ');
start_search(" '111\\\'11\\\'1' '222222' ");
console.log("TEST FAILS");
start_search(' sdfasdfasdfsdf "fasdf');
(new Lexer(' "111\\\"11\\\"1" "222222" ')).search();
(new Lexer(" '111\\\'11\\\'1' '222222' ")).search();

17
main.js
View File

@ -14,13 +14,12 @@ exec("re2c -i lexer.l", function(err, stdout, stderr) {
var post_process_lexer = function (_string) {
var search_string = fs.readFileSync("search_string.js", "utf8");
// insert last case for string detect
_string = _string.replace(/\}\nENDER}/gm, "yy100000000: { " + search_string + " reset(); break; }}}");
_string = _string.replace(/\}\nENDER}/gm, "yy100000000: { this._searchString(); this._set_next(); return; }}}");
_string = _string.replace(/^.*(_r2c_var_.*;|unsigned int yyaccept = 0;)\n/gm, ""); // replace var yych;
_string = _string.replace(/(yych = \*YYCURSOR);\n/gm, "\tcase 1:\n yych = str[YYCURSOR];\n"); // insert "case 1:" before;
_string = _string.replace(/\*(.*?);/gm, "str[$1];"); // замена разыменовываний
_string = _string.replace(/(yych = \*this._yy_cursor);\n/gm, "\tcase 1:\n yych = this._string[this._yy_cursor];\n"); // insert "case 1:" before;
_string = _string.replace(/\*(.*?);/gm, "this._string[$1];"); // замена разыменовываний
_string = _string.replace(/^yy(\d*?):/gm, "case $1:"); // replace goto marker onto case
_string = _string.replace(/\) goto yy(\d*?);/gm, ") { id = $1; break; }"); // replace goto inside if
_string = _string.replace(/goto yy(\d*?);/gm, "id = $1; break;"); // replace goto outside if
@ -29,10 +28,16 @@ var post_process_lexer = function (_string) {
_string = _string.replace(/0x00/gm, 'undefined'); // replace 0x00
// black magic
_string = _string.replace(/(switch \(yych\) \{[\s\S]*?})/gm, "(function(){$1})(); break;"); // добавим замыкание что бы обработать свиче в свиче
_string = _string.replace(/(switch \(yych\) \{[\s\S]*?})/gm, "(function(){$1}.bind(this))(); break;"); // добавим замыкание что бы обработать свиче в свиче
_string = _string.replace(/switch\((id)\)/gm, "switch(this._state)"); // replace id to this._state
_string = _string.replace(/id = (\d.*?);/gm, "this._state = $1;"); // replace id = n to this._state = n
_string = _string.replace(/yyaccept/gm, "this._yy_accept"); // replace yyaccept to this._yy_accept
_string = _string.replace(/yych/gm, "this._yy_char"); // replace yych to this._yy_char
_string = js_beautify(_string, {indent_size: 4, space_in_empty_paren: true});
fs.writeFileSync("out.js", _string);
fs.writeFileSync("lexer.js", _string);
};

View File

@ -1,7 +0,0 @@
var info = searchString(str, yych, YYLEXSTART, YYCURSOR);
if(info.success) {
YYCURSOR = info.pos;
addLexeme(str, YYLEXSTART, YYCURSOR, "STRING");
} else {
notFoundCloseQuote(str, YYLEXSTART, YYCURSOR);
}