nimp/compiler/source/nimpc/parser.d

83 lines
1.8 KiB
D
Raw Permalink Normal View History

2023-06-27 04:45:28 +00:00
module nimpc.parser;
import streams;
import std.stdio;
import std.regex;
import std.array;
import std.typecons;
import std.uni;
import std.string;
enum TokenType {
KEYWORD,
DATA_TYPE,
SENTENCE_END,
LITERAL_INTEGER,
LITERAL_FLOAT,
LITERAL_BOOLEAN,
LITERAL_STRING,
COMMENT_INLINE,
SYMBOL
}
immutable struct Token {
TokenType type;
string content;
uint line;
uint column;
}
class LexerException : Exception {
const uint sourceLine;
const uint sourceColumn;
this(string msg, uint sourceLine, uint sourceColumn) {
super(msg);
this.sourceLine = sourceLine;
this.sourceColumn = sourceColumn;
}
}
class StreamEndException : Exception {
this() {
super("Stream ended.");
}
}
/**
* Parses a list of tokens from an input stream of lines of code.
* Params:
* input = The lines of input to parse.
* Returns: A list of tokens.
*/
Token[] tokenize(string input) {
Appender!(Token[]) tokenApp;
bool streamEnded = false;
uint line = 0;
uint col = 0;
while (input.length > 0) {
input = stripLeft(input);
if (input.length == 0) break;
if (input[0] == '#') {
writeln("Parsing inline comment.");
// parse inline comment.
size_t endIdx = 0;
while (!(endIdx >= input.length || input[endIdx] == '\n')) {
endIdx++;
}
tokenApp ~= Token(TokenType.COMMENT_INLINE, input[0 .. endIdx], line, col);
input = input[endIdx .. $];
} else if (input.length >= 7 && input[0 .. 7] == "define ") {
tokenApp ~= Token(TokenType.KEYWORD, input[0 .. 6], line, col);
input = input[7 .. $];
} else {
input = input[1 .. $];
}
}
return tokenApp[];
}