Created
August 12, 2013 16:12
-
-
Save creationix/6212379 to your computer and use it in GitHub Desktop.
Simple state-machine that scans javascript code for requires.
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
// Mine a string for require calls and export the module names | |
// Extract all require calls using a proper state-machine parser. | |
function mine(js) { | |
var names = []; | |
var state = 0; | |
var ident; | |
var quote; | |
var name; | |
var isIdent = /[a-z0-9_.]/i; | |
var isWhitespace = /[ \r\n\t]/; | |
function $start(char) { | |
if (char === "/") { | |
return $slash; | |
} | |
if (char === "'" || char === '"') { | |
quote = char; | |
return $string; | |
} | |
if (isIdent.test(char)) { | |
ident = char; | |
return $ident; | |
} | |
return $start; | |
} | |
function $ident(char) { | |
if (isIdent.test(char)) { | |
ident += char; | |
return $ident; | |
} | |
if (char === "(" && ident === "require") { | |
ident = undefined; | |
return $call; | |
} | |
return $start; | |
} | |
function $call(char) { | |
if (isWhitespace.test(char)) return $call; | |
if (char === "'" || char === '"') { | |
quote = char; | |
name = ""; | |
return $name; | |
} | |
return $start; | |
} | |
function $name(char) { | |
if (char === quote) { | |
return $close; | |
} | |
name += char; | |
return $name; | |
} | |
function $close(char) { | |
if (isWhitespace.test(char)) return $close; | |
if (char === ")") { | |
names.push(name); | |
} | |
name = undefined; | |
return $start; | |
} | |
function $string(char) { | |
if (char === "\\") { | |
return $escape; | |
} | |
if (char === quote) { | |
return $start; | |
} | |
return $string; | |
} | |
function $escape(char) { | |
return $string; | |
} | |
function $slash(char) { | |
if (char === "/") return $lineComment; | |
if (char === "*") return $multilineComment; | |
return $start; | |
} | |
function $lineComment(char) { | |
if (char === "\r" || char === "\n") return $start; | |
return $lineComment; | |
} | |
function $multilineComment(char) { | |
if (char === "*") return $multilineEnding; | |
return $multilineComment; | |
} | |
function $multilineEnding(char) { | |
if (char === "/") return $start; | |
if (char === "*") return $multilingEnding; | |
return $multilineComment; | |
} | |
var state = $start; | |
for (var i = 0, l = js.length; i < l; i++) { | |
state = state(js[i]); | |
} | |
return names; | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
great! would be interesting to test how much faster something like browserify would be with a simple parser like this...