Last active
February 7, 2020 11:12
-
-
Save katsaii/5eaaec820ea8f5515896365be05cac63 to your computer and use it in GitHub Desktop.
Tokenises a string into an array of strings and numbers, where anything between double quotes is considered a string.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
/// @desc Tokenises a string into a list of values. | |
/// @param str {String} The string to tokenise. | |
/// @author Kat @katsaii | |
var str = argument0; | |
var tokens = []; | |
var len = string_length(str); | |
for (var i = 1; i <= len; i++) { | |
var prefix = string_char_at(str, i); | |
if (char_is_whitespace(prefix)) then continue; | |
var value = undefined; | |
switch (prefix) { | |
case "\"": | |
// tokenise string | |
var start = i + 1; | |
var count = 0; | |
while (start + count <= len) { | |
var char = string_char_at(str, start + count); | |
if (char == "\"") then break; | |
if (char == "\\") then count++; | |
count++; | |
} | |
i = start + count; | |
value = string_copy(str, start, count); | |
break; | |
default: | |
// tokenise number or word | |
var validNumber = true; | |
var decimal = false; | |
var start = i; | |
var count = 0; | |
while (start + count <= len) { | |
var char = string_char_at(str, start + count); | |
if (char_is_whitespace(char)) then break; | |
switch (char) { | |
case "0": | |
case "1": | |
case "2": | |
case "3": | |
case "4": | |
case "5": | |
case "6": | |
case "7": | |
case "8": | |
case "9": | |
break; | |
case "-": | |
validNumber = count == 0; | |
break; | |
case ".": | |
if (decimal) { | |
validNumber = false; | |
} else { | |
decimal = true; | |
} | |
break; | |
default: | |
validNumber = false; | |
break; | |
} | |
count++; | |
} | |
i = start + count; | |
value = string_copy(str, start, count); | |
if (validNumber) { | |
value = real(value); | |
} else { | |
// keywords | |
switch (value) { | |
case "null": | |
case "void": | |
case "empty": | |
case "()": | |
case "undefined": | |
value = undefined; | |
break; | |
case "noone": | |
value = noone; | |
break; | |
case "global": | |
value = global; | |
break; | |
case "self": | |
value = self; | |
break; | |
case "other": | |
value = other; | |
break; | |
case "all": | |
value = all; | |
break; | |
case "inf": | |
case "infinity": | |
value = infinity; | |
break; | |
case "NaN": | |
value = NaN; | |
break; | |
case "yes": | |
case "enable": | |
case "true": | |
value = true; | |
break; | |
case "no": | |
case "disable": | |
case "false": | |
value = false; | |
break; | |
default: | |
// check for an asset, otherwise just accept the word itself | |
var index = asset_get_index(value); | |
if (index != -1) then value = index; | |
break; | |
} | |
} | |
break; | |
} | |
tokens[@ array_length_1d(tokens)] = value; | |
} | |
return tokens; |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment