void error(string msg, int line = -1, int col = -1, Token tok = null)
{
if (tok == null)
tok = reader.Peek();
if (line == -1)
line = tok.Line;
if (col == -1)
col = tok.Column;
msg += ", got '" + tok.Data + "'";
LuaSourceException ex = new LuaSourceException(line, col, msg);
Errors.Add(ex);
//Console.WriteLine(ex.GenerateMessage("sd"));
if (ThrowParsingErrors)
throw ex;
}
public TokenReader Lex(string s)
{
List<Token> tokens = new List<Token>();
src = s;
p = 0;
ln = 1;
col = 1;
while (true)
{
List<Token> leading = new List<Token>();
// eat whitespace/comments
while (true)
{
char c_ = peek();
if (c_ == '#' && peek(1) == '!' && ln == 1)
{
// linux shebang
string sh = "";
while (peek() != '\n' && peek() != '\r' && peek() != '\0')
{
sh += read();
}
//readnl();
leading.Add(new Token { Type = TokenType.Shebang, Data = sh });
}
else if (c_ == ' ' || c_ == '\t')
{
leading.Add(new Token { Type = c_ == ' ' ? TokenType.WhitespaceSpace : TokenType.WhitespaceTab, Data = c_.ToString() });
read();
}
else if (c_ == '-' && peek(1) == '-')
{
read();
read();
char openParen = '\0';
if (peek() == '[')
openParen = read();
string comment = tryReadLongStr();
if (comment == null)
{
comment = "--" + (openParen != '\0' ? openParen.ToString() : "");
while (peek() != '\n' && peek() != '\r' && peek() != '\0')
comment += read();
//read();
//if (peek() == '\n') // e.g. \r\n
// read();
if (comment.Length >= 3 && comment.Substring(0, 3) == "---")
leading.Add(new Token { Type = TokenType.DocumentationComment, Data = comment });
else
leading.Add(new Token { Type = TokenType.ShortComment, Data = comment });
}
else
{
comment = "--" + comment;
leading.Add(new Token { Type = TokenType.LongComment, Data = comment });
}
}
else if (c_ == '\n' || c_ == '\r')
{
leading.Add(new Token { Type = c_ == '\n' ? TokenType.WhitespaceN : TokenType.WhitespaceR, Data = c_.ToString() });
// read handles line changing...
read();
}
else
break;
}
Token t = new Token();
t.Leading = leading;
t.Line = ln;
t.Column = col;
char c = read();
if (c == '\0')
t.Type = TokenType.EndOfStream;
else if (char.IsLetter(c) || c == '_')
{
// ident / keyword
string s4 = c.ToString();
while (char.IsLetter(peek()) ||
peek() == '_' ||
char.IsDigit(peek()) &&
peek() != '\0')
{
s4 += read();
}
t.Data = s4;
if (IsKeyword(s4))
t.Type = TokenType.Keyword;
else
t.Type = TokenType.Ident;
}
else if (char.IsDigit(c) ||
(c == '.' && char.IsDigit(peek())))
{ // negative numbers are handled in unary minus collection
string num = "";
if (c == '0' && matchpeek("xX"))
{
//read(); -> already done
num = "0" + read(); // 'xX'
while (IsHexDigit(peek()))
num += read();
}
#if !VANILLA_LUA
else if (c == '0' && matchpeek("bB"))
{
num = "0" + read(); // read 'bB'
while (char.IsDigit(peek()) || peek() == '_')
num += read();
}
else if (c == '0' && matchpeek("oO"))
{
num = "0" + read(); // read 'oO'
while (char.IsDigit(peek()) || peek() == '_')
num += read();
}
#endif
else
{
num = c.ToString();
bool dec = false;
while (char.IsDigit(peek())
|| peek() == '.'
#if !VANILLA_LUA
|| peek() == '_')
#else
)
#endif
{
num += read();
if (peek() == '.')
{
if (dec)
error("Number has more than one decimal point");
dec = true;
num += read();
}
}
}
#if !VANILLA_LUA
if (matchpeek("PpEe")) // exponent
{
num += read();
if (matchpeek("+-"))
num += read();
while (char.IsDigit(peek()))
num += read();
}
#endif
t.Data = num;
t.Type = TokenType.Number;
}
else if (c == '\'' || c == '"')
{
char delim = c;
string str = "";
while (true)
{
char c2 = read();
if (c2 == '\\')
{
str += "\\";
str += read(); // we won't parse \0xFF, \000, \n, etc here
}
else if (c2 == delim)
break;
else if (c2 == '\0')
error("expected '" + delim + "', not <eof>");
else
str += c2;
}
t.Data = str;
t.Type = delim == '"' ? TokenType.DoubleQuoteString : TokenType.SingleQuoteString;
}
else if (c == '[')
{
string s3 = tryReadLongStr();
if (s3 == null)
{
t.Type = TokenType.Symbol;
t.Data = "[";
}
else
{
t.Type = TokenType.LongString;
t.Data = s3;
}
}
else if (c == '<' || c == '>' || c == '=')
{
t.Type = TokenType.Symbol;
if (peek() == '=' ||
(c == '<' && peek() == '<') ||
(c == '>' && peek() == '>'))
{
t.Data = c.ToString() + read().ToString();
#if !VANILLA_LUA
if (peek() == '=' && (c == '<' || c == '>'))
t.Data += read(); // augmented, e.g. >>=, but not ===
#endif
}
else
t.Data = c.ToString();
}
else if (c == '~')
{
if (peek() == '=')
{
read();
t.Type = TokenType.Symbol;
t.Data = "~=";
}
else
{
t.Type = TokenType.Symbol;
t.Data = "~";
}
}
else if (c == '.')
{
t.Type = TokenType.Symbol;
if (peek() == '.')
{
read(); // read second '.
if (peek() == '.')
{
t.Data = "...";
read(); // read third '.'
}
#if !VANILLA_LUA
else if (peek() == '=') // ..=
{
t.Data = "..=";
read(); // read '='
}
#endif
else
t.Data = "..";
}
else
{
t.Data = ".";
}
}
else if (c == ':')
{
t.Type = TokenType.Symbol;
#if !VANILLA_LUA
if (peek() == ':')
{
read();
t.Data = "::";
}
else
#endif
t.Data = ":";
}
else if (c == '-' && peek() == '>')
{
read(); // read the '>'
t.Data = "->";
t.Type = TokenType.Symbol;
}
else if (c == '^')
{
t.Type = TokenType.Symbol;
if (peek() == '^')
{
read();
#if !VANILLA_LUA
if (peek() == '=')
{
read();
t.Data = "^^=";
}
else
#endif
t.Data = "^^";
}
#if !VANILLA_LUA
else if (peek() == '=')
{
read();
t.Data = "^=";
}
#endif
else
t.Data = "^";
}
#if !VANILLA_LUA
else if (c == '!')
{
t.Type = TokenType.Symbol;
if (peek() == '=')
{
read();
t.Data = "!=";
}
else t.Data = "!";
}
#endif
else if (IsSymbol(c))
{
t.Type = TokenType.Symbol;
t.Data = c.ToString();
#if !VANILLA_LUA
if (peek() == '=')
{
char c2 = peek();
if (c == '+' ||
c == '-' ||
c == '/' ||
c == '*' ||
c == '^' ||
c == '%' ||
c == '&' ||
c == '|')
{
t.Data += "=";
read();
}
}
#endif
}
else
{
p--; // un-read token
col--;
error("Unexpected Symbol '" + c + "'");
read();
}
tokens.Add(t);
if (peek() == '\0')
break;
}
if (tokens.Count > 0 && tokens[tokens.Count - 1].Type != TokenType.EndOfStream)
tokens.Add(new Token { Type = TokenType.EndOfStream });
if (tokens.Count > 1) // 2+
tokens[tokens.Count - 2].FollowingEoSToken = tokens[tokens.Count - 1];
return new TokenReader(tokens);
}
public Token(Token copy)
{
this.token = copy.token;
this.seminfo = new SemInfo(copy.seminfo);
}
bool isAugmentedAssignment(Token token)
{
switch (token.Data)
{
case "..=":
case ">>=":
case "<<=":
case "+=":
case "-=":
case "/=":
case "*=":
case "&=":
case "|=":
case "^=":
case "%=":
case "^^=":
return true;
default:
return false;
}
}