outerIIFEBody does not handle all IIFE cases

What version of ESLint are you using? 3.0.1

What parser (default, Babel-ESLint, etc.) are you using? espree

Please show your full configuration:

module.exports = {
    "env": {
        "browser": true,
        "amd": true
    },
    "extends": "eslint:recommended",
    "rules": {
        "indent": [
            "error",
            "tab",
            { "outerIIFEBody": 0 }
        ],
        "linebreak-style": [
            "error",
            "windows"
        ],
        "semi": [
            "error",
            "always"
        ],
        "no-unused-vars": [
            "error",
            { "vars": "local", "args": "after-used" }
        ],
        "no-empty": [
            "error",
            { "allowEmptyCatch": true }
        ]
    },
    "globals": {
        "dojo": false,
        "ActiveXObject": false
    }
};

What did you do? Please include the actual source code causing the issue.

window.ERMS.FloatingDialog = (function(evts) {
"use strict";
var my;
my.foo = function() {
    return evts;
};

return my;
})(window.ERMS.Events);

What did you expect to happen? No errors

What actually happened? Please include the actual, raw output from ESLint.

  2:1  error  Expected indentation of 1 tab character but found 0  indent
  3:1  error  Expected indentation of 1 tab character but found 0  indent
  4:1  error  Expected indentation of 1 tab character but found 0  indent
  8:1  error  Expected indentation of 1 tab character but found 0  indent

I believe this case would be covered if the false negatives mentioned by @wojdyr in https://github.com/eslint/eslint/issues/6585#issuecomment-230183366 were handled.

Pasting those examples here for convenience:

(function () {
var i;  // ok
})();

~function () {
var i;  // not
}();

!(function () {
var i;  // not
})();

!function () {
var i;  // not
}();

;(function () {
var i;  // ok
})();

var MyClass = (function () {
var i;  // not
})();

[nzakas]

@TheCycoONE we never intended to support assignment as in your example, and I’m not entirely sure it makes sense to do so because the IIFE isn’t really “outer” at that point. I can see an argument for supporting unary operators before the IIFE, but I’m not sure about assignment.

[nzakas]

The original purpose for this option was to support the jQuery style guide: https://contribute.jquery.org/style-guide/js/#full-file-closures

And, once I looked back at that, it seems that they do support assignments so we should as well.

I’ll champion this change.

@TheCycoONE are you willing to implement this once accepted?