See bug 1228827 comment 4 - the tl;dr is that the "pos" value in a bookmark record is expected to be wrong in all sorts of cases, so we should ignore that in our comparisons.

Review commit: https://reviewboard.mozilla.org/r/55826/diff/#index_header See other reviews: https://reviewboard.mozilla.org/r/55826/

I was running into issues with this with the TPS tests as well (hence bug 1274394 comment 11). I had assumed these were real bookmark sync bugs, but it's good to know they are not.

(In reply to Thom Chiovoloni [:tcsc] from comment #2) > I was running into issues with this with the TPS tests as well (hence bug > 1274394 comment 11). Yeah, I was confused by that comment - you didn't say what failures, but you already seemed to be handling some such errors - hence my comment 12 in that bug :) That bug doesn't mention "pos" at all.

Comment on attachment 8757354 [details] MozReview Request: Bug 1276152 - Ignore differences in the pos property when validating bookmark separators. r?markh https://reviewboard.mozilla.org/r/55826/#review52762 Thanks!

Pushed by cbook@mozilla.com: https://hg.mozilla.org/integration/fx-team/rev/6617e62776cd Ignore differences in the pos property when validating bookmark separators. r=markh