Author cheryl.sabella
Recipients cheryl.sabella, terry.reedy
Date 2017-12-26.00:02:15
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <>

I'm not sure if I phrased my initial question correctly.  I've made a pull request to show you what I was thinking.

What I had tried to show with the test case change is that, prior to the guarantee of insertion order on the dictionary, the check `eq(tcl, [C0, f0])` would have failed without the sort because `transform_children` would have returned `[f0, C0]`.  With the sort in place or with the guarantee insertion order, this test doesn't fail.

Date User Action Args
2017-12-26 00:02:16cheryl.sabellasetrecipients: + cheryl.sabella, terry.reedy
2017-12-26 00:02:15cheryl.sabellasetmessageid: <>
2017-12-26 00:02:15cheryl.sabellalinkissue32411 messages
2017-12-26 00:02:15cheryl.sabellacreate