Use of assignment equals instead of comparison ==

[guest]

Loaded dojo.js.uncompressed.js in Aptana IDE and discovered that on line 59 = was used in the while loop rather than the comparison operator ==. Here is the code snippet for search purposes:

while(tn=cn[i++]){

if(!console[tn]){

console[tn] = function(){};

}

}

[dante]

it's in dojo/_base/_loader/bootstrap.js is is harmless. cn is an array defined immediately, and all supported browsers will exit the while when tn = undefined, which cn[i++] will always reach (cn.length)... its console.* debugging stuff.

 var cn = [
                "assert", "count", "debug", "dir", "dirxml", "error", "group",
                "groupEnd", "info", "log", "profile", "profileEnd", "time",
                "timeEnd", "trace", "warn"
        ];
        var i=0, tn;
        while(tn=cn[i++]){
                if(!console[tn]){
                        console[tn] = function(){};
                }
        }

[James Burke]

This is actually working as designed. This is a valid JS construct.

I suggest opening a bug with Aptana to find out why this was reported as an error.

[Adam Peller]

It's considered 'bad style' by some. Crockford's jslint, and now Rhino strict mode, require an extra set of parenthesis around the assignment so that it stands out. Although I know some of the jslint stuff is controversial, I think this is good practice. It looks different than a comparison and stands out to the programmer, who might read it wrong.

[James Burke]

peller: I'm OK wrapping the assignment in another set of parentheses if that solves the problem (will it?). I suppose Aptana is using jslint to generate these messages. Although, it seems to bring up a larger question that I think you have asked before: should the Dojo toolkit pass all the jslint checks? In the past, it seems like there was feedback from other contributors who felt like it was not appropriate.

But I'm fine with doing small things like this if it reduces messages when using dojo in an IDE that uses jslint.

[Adam Peller]

Yup. That's a discussion we should eventually have regarding #3970, where I'd like to put such rules on the pre-commit hooks, if everyone is game.... something I plan to raise *after* 1.0 :-) I haven't looked too closely at the new Rhino feature. Not sure it's even released yet.

[James Burke]

(In [10713]) Fixes #4495. The extra set () should avoid the jslint error.