esrefactor (BSD licensed) is a little helper library for ECMAScript refactoring.
With Node.js:
npm install esrefactor
In a browser, include first all the dependents:
<!-- esrefactor depends on these libraries -->
<script src="esprima.js"></src>
<script src="estraverse.js"></src>
<script src="escope.js"></src>
<script src="esrefactor.js"></src>
Before using the API, a context needs to be created:
var ctx = new esrefactor.Context(code);
where code
is the source code.
An identifier, whether it is a variable, a function name, or a function
parameter, can be identified using identify()
. Example:
var ctx = new esrefactor.Context('var x = 42; y = x * 2; z = x / 2');
var id = ctx.identify(4);
The only argument to identify
is the zero-based position index.
The return object has 3 (three) properties:
identifier
: the syntax node associated with the positiondeclaration
: the declaration syntax node for the identifierreferences
: an array of all identical references
If there is no declaration for the identifier (e.g. x = 42
, global
leak), then declaration
will be null.
The resolution of the declaration syntax node and the references array take into account the identifier scope as defined in the official ECMAScript 5.1 Specification (ECMA-262).
Note that if there is no identifier in the given position index,
identify()
will return undefined.
An identifier can be renamed using rename()
. All other identical references
associated with that identifier will be renamed as well, again taking into
account the proper identifier scope. Renaming works for variable declaration,
function name, and function parameter.
For rename()
to work, it needs to have the identification result
(via identify
) and the new name for the identifier.
var ctx = new esrefactor.Context('var x; x = 42');
var id = ctx.identify(4);
var code = ctx.rename(id, 'answer');
In the above example, code
is var answer; answer = 42
.