<p>The original discussion on this was:</p>
<pre><code><davewood> can the "problem" be simplified to: "if your foreign key column is not nullable you want join_type => left for complex resultsets to work correctly"?
<ribasushi> davewood: itym s/not nullable/nullable/
<ribasushi> and "for relatonship traversal to work consistently in all situations" (nothing complex about the resultset I gave as example)
<ribasushi> davewood: but yes, that'd be much better - convert it to a final patch and show again?
</code></pre>
<p>The current patch however simply rewords two sentences, while keeping the original confusing semantic structure intact. Is there a reason you didn't go for the much simpler explanation as in the IRC log above?<br>
I could apply the patch as-is, but it imho makes no difference whatsoever wrt the problem you were originally trying to solve.</p>
<p style="font-size:small;-webkit-text-size-adjust:none;color:#666;">—<br>Reply to this email directly or <a href='https://github.com/dbsrgits/dbix-class/pull/37#issuecomment-31225950'>view it on GitHub</a>.<img src='https://github.com/notifications/beacon/302594__eyJzY29wZSI6Ik5ld3NpZXM6QmVhY29uIiwiZXhwaXJlcyI6MTcwMzYwNzU0OCwiZGF0YSI6eyJpZCI6MjI1NjUyMzh9fQ==--5996edb6724c274be5943de1416b83cb88fb35d1.gif' height='1' width='1'></p>