Official `length` workaround doesn't work

[endash]

About to dig deeper into this, but the be.of.length(x) workaround for have.length(x) no longer seems to work.

I just cloned fresh, npm installed and added a single test for

$('<div></div>').should.be.of.length(1);

And get the following failure:

TypeError: Property 'length' of object function (selector) {
    var obj = flag(this, 'object');
    if (obj instanceof $) {
      this.assert(
          obj.is(selector)
        , 'expected #{this} to be #{exp}'
        , 'expected #{this} not to be #{exp}'
        , selector
      );
    } else {
      _super.apply(this, arguments);
    }
  } is not a function
at Context.<anonymous> (file://localhost/Users/chris/dev/chai-jquery/test/chai-jquery-spec.js:604:37)
at Test.Runnable.run (file://localhost/Users/chris/dev/chai-jquery/node_modules/mocha/mocha.js:4039:32)
at Runner.runTest (file://localhost/Users/chris/dev/chai-jquery/node_modules/mocha/mocha.js:4404:10)
at file://localhost/Users/chris/dev/chai-jquery/node_modules/mocha/mocha.js:4450:12
at next (file://localhost/Users/chris/dev/chai-jquery/node_modules/mocha/mocha.js:4330:14)
at file://localhost/Users/chris/dev/chai-jquery/node_modules/mocha/mocha.js:4339:7
at next (file://localhost/Users/chris/dev/chai-jquery/node_modules/mocha/mocha.js:4287:23)
at file://localhost/Users/chris/dev/chai-jquery/node_modules/mocha/mocha.js:4307:5
at timeslice (file://localhost/Users/chris/dev/chai-jquery/node_modules/mocha/mocha.js:5279:29)

I get a different error in Safari:

'1' is not a function (evaluating 'subject.should.be.of.length(1)')

I suppose it's possible there's something wrong with my setup, but I don't see what it could be...

[endash]

New workaround, for the terminal variation of length: should.have.lengthOf(n)

FWIW the function it's complaining about and spitting out in the error message is the be method overwritten by chai-jquery, not a new addition at all. It is perplexing.

[endash]

Rolling back chai to 1.4.0 has no effect... I'm at a loss to explain why the workaround would have ever worked, given that be is also overridden to be a function, just like have.

[jfirebaugh]

It may well be that it never did work. I should've added it to the test suite when it was first proposed, but neglected to.

[nackjicholson]

Only real work around my team has found is to not use length at all. And instead do expect(selection.length).to.equal(number);

[alexstrat]

chai-jquery's have should maybe be renamed in something that does not interfere with chai's original have ?

[fizker]

@nackjicholson I prefer expect(selection).to.have.property('length', number); over expect(selection.length).to.equal(number);.

I find that version way more readable in case of failure (Failure will read expected ... to have a property 'length' of x, but got y instead of expected x to be y.

[jednano]

I'm getting the same error with:

expect($('.foo')).to.have('h1');

TypeError: '[object Object]' is not a function (evaluating 'expect($('.foo')).to.have('h1')')

[mattfysh]

I ended up opting for:

expect($('body')).with.length(1);

[jednano]

My issue was actually an issue with installing Chai Jquery into Karma. Now, my original expression works just fine.

[marcandre]

with works fine.
Please fix the doc!

[gmfx]

@jedmao: what exactly was the issue having the same prob with a mocha setup

[jednano]

@gumaflux my issue was with both syntaxes. The first isn't officially supported, so that shouldn't be a surprise:

expect(selection).to.have.length(2);

The second, however, was explicitly stated in the documentation as a workaround:

expect(selection).to.be.of.length(2);

This also doesn't work. The way I'm doing it now is pretty much as @nackjicholson has stated:

expect(selection.length).to.eq(2);

This is the only way I've gotten it to work. I'm sure the other ways expressed above work also, but I'm not interested in those syntaxes.

BTW @jfirebaugh, I can confirm that this workaround does, in fact, NOT work! I would love it if it did :)

[glittershark]

👍 more descriptive error messages than "expected 2 to equal 1" would be very nice.

[gmfx]

@jedmao great thanks