Last active
February 19, 2020 09:58
-
-
Save vilicvane/1353586a3f6f46102ac3 to your computer and use it in GitHub Desktop.
A lexer (behave like Angular expression lexer) written with the help of regular expression.
This file contains hidden or 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
/** | |
* A test using regular expression for lexing | |
* by vilicvane<https://vilic.github.io/> | |
* | |
* It turns out to be 50% faster comparing to original Angular lexer before JIT Compilation, but 50% slower then (in Chrome). | |
* And only 1/3 of the original lexer counting the core code lines. | |
* Though it's because the regular expressions are doing most of the work. | |
* | |
* Note: The regular expression and related enum declaration are managed by https://github.com/vilic/regex-tools/. | |
*/ | |
interface Token { | |
index: number; | |
text: string; | |
constant?: boolean; | |
identifier?: boolean; | |
operator?: boolean; | |
value?: any; | |
} | |
/* /$expressionLex/ */ | |
const enum ExpressionLexGroup { | |
string = 1, | |
stringUnexpectedToken = 3, | |
number, | |
numberInvalidExponent, | |
identifier, | |
punctuation, | |
whitespace, | |
operator | |
} | |
var expressionLexRegex = /* /$expressionLex/ */ /((["'])(?:(?!\2|[\\\r\n\u2028\u2029])[\s\S]|\\(?:['"\\bfnrtv]|[^'"\\bfnrtv\dxu\r\n\u2028\u2029]|0(?!\d)|x[\da-fA-F]{2}|u[\da-fA-F]{4})|\\(?:\r?\n|\r(?!\n)|[\u2028\u2029]))*(?:\2|(\\?)))|((?:(?:(?:0|[1-9]\d*)(?:\.\d*)?|\.\d+)(?:[eE](?:[+-]?\d+|()))?|0[xX][\da-fA-F]+))|([a-zA-Z$_][\w$]*)|([(){}[\].,;:?])|(\s+)|([-+*\/%<>]|[!=]={0,2}|[<>]=|&&|\|\|)|(?:)/g; // for poor github syntax highlight: ' | |
function parseExpressionLex(source: string): Token[] { | |
var index = 0; | |
var tokens: Token[] = []; | |
while (index < source.length) { | |
let start = expressionLexRegex.lastIndex = index; | |
let groups = expressionLexRegex.exec(source); | |
index = expressionLexRegex.lastIndex; | |
if (groups[ExpressionLexGroup.string]) { | |
let unexpectedToken = groups[ExpressionLexGroup.stringUnexpectedToken]; | |
if (typeof unexpectedToken == 'string') { | |
if (unexpectedToken == '\\') { | |
throwSyntaxError('Invalid escape'); | |
} else { | |
throwSyntaxError('Unterminated quote'); | |
} | |
} else { | |
let text = groups[0]; | |
tokens.push({ | |
index: start, | |
text, | |
constant: true, | |
value: eval(text) | |
}); | |
} | |
} else if (groups[ExpressionLexGroup.number]) { | |
let invalidExponent = groups[ExpressionLexGroup.numberInvalidExponent]; | |
if (invalidExponent == '') { | |
throwSyntaxError('Invalid exponent'); | |
} else { | |
let text = groups[0]; | |
tokens.push({ | |
index: start, | |
text, | |
constant: true, | |
value: eval(text) | |
}); | |
} | |
} else if (groups[ExpressionLexGroup.identifier]) { | |
tokens.push({ | |
index: start, | |
text: groups[0], | |
identifier: true | |
}); | |
} else if (groups[ExpressionLexGroup.punctuation]) { | |
tokens.push({ | |
index: start, | |
text: groups[0] | |
}); | |
} else if (groups[ExpressionLexGroup.whitespace]) { | |
// do nothing... | |
} else if (groups[ExpressionLexGroup.operator]) { | |
tokens.push({ | |
index: start, | |
text: groups[0], | |
operator: true | |
}); | |
} else { | |
throwSyntaxError('Unexpected token'); | |
} | |
} | |
return tokens; | |
function throwSyntaxError(message: string, errorIndex = index, ignoreSpaces = true) { | |
if (ignoreSpaces) { | |
let originalIndex = errorIndex; | |
for (; errorIndex < source.length; errorIndex++) { | |
if (/\S/.test(source[errorIndex])) { | |
break; | |
} | |
} | |
if (errorIndex == source.length) { | |
errorIndex = originalIndex; | |
} | |
} | |
var i = errorIndex; | |
for (; i >= 0; i--) { | |
if (/[\r\n]/.test(source.charAt(i))) { | |
i++; | |
break; | |
} | |
} | |
var lineNumber = (source.substr(0, i + 1).match(/\r?\n/g) || []).length; | |
var columnNumber = errorIndex - i; | |
var line = (source.substr(i).match(/.*/) || [''])[0]; | |
var arrowLine = Array(columnNumber + 1).join(' ') + '^'; | |
message = `${message} at (${lineNumber},${columnNumber}): | |
${line} | |
${arrowLine}`; | |
throw new SyntaxError(message); | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment