1093669 - [system] shrink_ui_test assert.deepEqual Nodes fail in newer version chai

Status: RESOLVED FIXED

(Firefox OS Graveyard :: Gaia::System, defect)

Description

[sv99]

User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:36.0) Gecko/20100101 Firefox/36.0
Build ID: 20141104030202

Steps to reproduce:

163: assert.deepEqual(tempTip.parentNode, fakeParent); fail in newer version chai

really tempTip.parentNode and fakeParent not Equal!! 
used version chai have problems in assert.deepEqual - newer version assert.deepEqual checked objects with his prototypes

Comment 1

[sv99]

I try replace assert.deepEqual with 
    function checkNodes (test, expected) {
      for (var key in expected) {
        if ((key !== 'children') && (key !== 'lastElementChild') 
          && (key !== 'firstElementChild') && (key !== 'childNodes') ) {
          assert.deepEqual(test[key], expected[key], 'key: ' + key);
        }
      }
    }

stopped on forth not equal property!

Comment 2

[Julien Wajsberg [:julienw]]

I think in this specific case we need to use assert.equal, it makes no sense to check DOM nodes using deepEqual. But clearly the test is wrong.

I added ids to all elements we used in the test so that the error message makes sense. The main reason the test is wrong is that:

* in shrinkingUI.prototype._setTip: we append to backgroundElement. But backgroundElement is set once, in the constructor. And we never change this then.

So I fixed the test replacing the deepEqual line with:

    assert.equal(tempTip.parentNode, fakeApp.element.parentNode);

But also removing all the fakeParent initialization (2 first lines in the test).

Comment 3

[sv99]

Attachment: Link to PR

Comment 4

[Julien Wajsberg [:julienw]]

Comment on attachment 8517496 [details] [review]
Link to PR

lgtm, let's ask to Arthur who wrote this at first, and alive as peer.

Comment 5

[Arthur Chen [:arthurcc][inactive after 6/30]]

master: 8fc1eeb2d208e766e48cafd5fb0ff3071d03f57a