Bug 60670

Summary: Remove LOOSE_PASS_OWN_ARRAY_PTR
Product: WebKit Reporter: Patrick R. Gansterer <paroga>
Component: Web Template FrameworkAssignee: Darin Adler <darin>
Status: RESOLVED DUPLICATE    
Severity: Normal CC: abarth, darin, mrowe
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
Patch mrowe: review-

Patrick R. Gansterer
Reported 2011-05-11 15:44:09 PDT
Remove LOOSE_PASS_OWN_ARRAY_PTR
Attachments
Patch (3.54 KB, patch)
2011-05-11 15:50 PDT, Patrick R. Gansterer 		mrowe: review-
DetailsFormatted DiffDiff
View All Add attachment proposed patch, testcase, etc.
Patrick R. Gansterer
Comment 1 2011-05-11 15:50:43 PDT
Created attachment 93201 [details] Patch
Patrick R. Gansterer
Comment 2 2011-05-11 15:54:52 PDT
Does Apple have some internal dependencies on OwnArrayPtr?
Adam Barth
Comment 3 2011-05-11 16:17:23 PDT
Comment on attachment 93201 [details] Patch I think it's safe to remove this if everything compiles. If we screw something up internally for Apple, I'm sure someone will tell us and we can revert this patch.
Mark Rowe (bdash)
Comment 4 2011-05-11 16:57:19 PDT
We added this code back in because it is being used.
Adam Barth
Comment 5 2011-05-11 16:58:34 PDT
(In reply to comment #4) > We added this code back in because it is being used. Is there a timetable after which we can remove the code?
Mark Rowe (bdash)
Comment 6 2011-05-11 17:02:01 PDT
(In reply to comment #5) > (In reply to comment #4) > > We added this code back in because it is being used. > > Is there a timetable after which we can remove the code? Is there an urgent need to remove it?
Adam Barth
Comment 7 2011-05-11 17:04:48 PDT
> Is there an urgent need to remove it? Looks like it hasn't been used by WebKit in since <http://trac.webkit.org/changeset/77785>, which was 3 months ago. I'm happy to wait another 3 months if that's convenient for Apple internally, but I would like to remove the code at some point.
Mark Rowe (bdash)
Comment 8 2011-05-11 17:09:21 PDT
The code can safely be removed when it’s no longer used. Given the tiny amount of code involved, and its unchanging nature, it seems absurd to try and force the issue.
Adam Barth
Comment 9 2011-05-11 17:14:30 PDT
My concern is that without a deadline we'll never remove the code. Presumably we'll have the same issue with the other "LOOSE" code for the other smart pointer types. It seems reasonable to want to not carry this dead code forever. If three months is not enough, how about six months?
Mark Rowe (bdash)
Comment 10 2011-05-11 17:19:52 PDT
> My concern is that without a deadline we'll never remove the code. Was I somehow unclear when I stated that this code can be removed when it is unused?
Darin Adler
Comment 11 2011-05-11 17:27:41 PDT
Adam, please cool your jets on this one. This is a very small number of lines of code, and it’s extremely helpful to me and my team at Apple to have this in there for a short while longer. We *will* remove it when we’re not relying on it any more. You’re coming on too strong. You may recall I specifically called this out at the contributors meeting. There must be more-worthwhile things for us to spend our time on. I’ll take this bug and I promise to deal with it.
Patrick R. Gansterer
Comment 12 2011-05-11 17:30:48 PDT
@darin: maybe you want to take bug 59909 too
Adam Barth
Comment 13 2011-05-11 17:35:18 PDT
> You’re coming on too strong. Sorry if I came across too strongly. That wasn't my intent.
Darin Adler
Comment 14 2011-06-15 12:38:43 PDT
Decided to handle all of these at once in bug 59909. *** This bug has been marked as a duplicate of bug 59909 ***
Note You need to log in before you can comment on or make changes to this bug.