Fix for rangepos crasher. #1704

Merged
merged 1 commit into from Dec 5, 2012

Projects

None yet

5 participants

@retronym
Member
retronym commented Dec 4, 2012

wrapClassTagUnapply was generating an unpositioned tree
which would crash under -Yrangepos. See SI-6338 / SI-6754

Review by @adriaanm (I've bumped this to a 2.10.0 blocker
after the reports or problems in the IDE with Akka.)

Cherry-pick from #1447.

@paulp @retronym paulp Fix for rangepos crasher.
wrapClassTagUnapply was generating an unpositioned tree
which would crash under -Yrangepos.  See SI-6338.
eccdc2f
@adriaanm adriaanm was assigned Dec 4, 2012
@xeno-by
Member
xeno-by commented Dec 4, 2012

If it's a cherry-pick, we're going to have problems propagating this commit to master. Maybe we could rather merge it into 2.10.0-wip retaining the SHA?

@retronym
Member
retronym commented Dec 4, 2012

I don't know what you mean. If we merge, the entire history of that branch is now the history of 2.10.0-wip. I thought git was pretty forgiving of cherry picks getting merged back.

@xeno-by
Member
xeno-by commented Dec 4, 2012

Your commit's sha and Paul's commit's sha are different. Hence to the best of my knowledge we'll have merge conflicts, even though patch contents are identical.

@adriaanm
Member
adriaanm commented Dec 4, 2012

That doesn't necessarily cause a merge conflict, I think.

@adriaanm
Member
adriaanm commented Dec 4, 2012

LGTM, thanks for cleaning up after sloppy, sloppy me, @paulp

@paulp
Contributor
paulp commented Dec 4, 2012

While there is no doubt plenty of cleaning-up-after-you credit to be given, I think retronym is the right target in this case. (And yes, git is forgiving of cherry-picks - you lose "fast forward merges", but that's irrelevant when we're talking about branches which have meaningfully diverged, as these have.)

@adriaanm adriaanm merged commit 18481ce into scala:2.10.0-wip Dec 5, 2012
@adriaanm
Member
adriaanm commented Dec 5, 2012

sorry, i haven't been super coherent the last two days.. more affected by that jetlag thingy than i'd care to admit

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment