Last Comment Bug 776630 - Switch comm-central from using nsnull to nullptr.
: Switch comm-central from using nsnull to nullptr.
Product: MailNews Core
Classification: Components
Component: Backend (show other bugs)
: Trunk
: x86 All
: -- normal (vote)
: Thunderbird 17.0
Assigned To: Mike Conley (:mconley) - (needinfo me!)
Depends on:
Blocks: 626472
  Show dependency treegraph
Reported: 2012-07-23 11:54 PDT by Mike Conley (:mconley) - (needinfo me!)
Modified: 2012-08-02 07:29 PDT (History)
11 users (show)
See Also:
Crash Signature:
QA Whiteboard:
Iteration: ---
Points: ---

comm-central: switch nsnull to nullptr (1.77 MB, patch)
2012-07-23 11:54 PDT, Mike Conley (:mconley) - (needinfo me!)
no flags Details | Diff | Splinter Review
Switch nsnull to nullptr (1.77 MB, patch)
2012-07-30 07:41 PDT, Mike Conley (:mconley) - (needinfo me!)
philipp: review+
Details | Diff | Splinter Review
Unbitrotted patch (checked in) (1.77 MB, patch)
2012-08-02 07:29 PDT, Mike Conley (:mconley) - (needinfo me!)
no flags Details | Diff | Splinter Review

Description Mike Conley (:mconley) - (needinfo me!) 2012-07-23 11:54:23 PDT
Created attachment 645008 [details] [diff] [review]
comm-central: switch nsnull to nullptr

Bug 626472 is dropping nsnull in favour of nullptr.

I talked to Ehsan, and I think the plan is to land the definition of nullptr soon, and once it's available in mozilla-central, push our patch to try, and test building it using both clang and gcc.

I'll probably get our try server to apply the removal of nsnull against our mozilla-central subdirectory, for added assurance that we still build alright.
Comment 1 Mike Conley (:mconley) - (needinfo me!) 2012-07-24 06:40:38 PDT
Part 1 from Bug 626472 has landed in mozilla-central.

As per our plan, I'm going to push the Attachment 645008 [details] [diff] to try to see if we get green builds.
Comment 2 :Ehsan Akhgari 2012-07-24 09:49:40 PDT
FWIW, nullptr is now on mozilla-central, so you guys can proceed with this whenever you want.
Comment 3 Mike Conley (:mconley) - (needinfo me!) 2012-07-24 11:01:31 PDT
Ok, here are our try results:

Mostly green, but not perfect. The build failures on Windows seem to be unrelated. The bloattest crashes on Linux, and the crashes caused by Mozmill are more disturbing.

Cc'ing Irving, who's awesome at diagnosing crashes and stuff.

I'll start by poking at the Mozmill crashes.
Comment 4 Mike Conley (:mconley) - (needinfo me!) 2012-07-24 11:29:59 PDT
So the crashes on mailbloat seem to also be happening on trunk (sigh, they just showed up this morning), so that appears to be unrelated.
Comment 5 :Ehsan Akhgari 2012-07-24 14:13:03 PDT
FWIW it will be really surprising if this results in crashes etc once you get green builds...
Comment 6 Mike Conley (:mconley) - (needinfo me!) 2012-07-25 13:16:22 PDT
Yeah, it turns out that other stuff landed on mozilla-central that broke us in new and wonderful ways.

I think we're getting close to sorting it all out. Anyhow, those reds on the try build can be safely ignored, and I think we're going to survive the landing.
Comment 7 Mike Conley (:mconley) - (needinfo me!) 2012-07-30 07:40:01 PDT
This spills outside of mailnews, and affects everything under comm-central.
Comment 8 Mike Conley (:mconley) - (needinfo me!) 2012-07-30 07:41:10 PDT
Created attachment 647163 [details] [diff] [review]
Switch nsnull to nullptr

Here's an up to date conversion patch
Comment 9 Philipp Kewisch [:Fallen] 2012-07-30 08:39:14 PDT
Comment on attachment 647163 [details] [diff] [review]
Switch nsnull to nullptr

Review of attachment 647163 [details] [diff] [review]:

r=philipp for calendar parts
Comment 10 Mike Conley (:mconley) - (needinfo me!) 2012-07-30 09:38:47 PDT

My try build looks good (well, good enough, considering the current bustedness of the tree):

Fallen has confirmed that the Calendar bits look sane. I've leafed through the diff, and I don't see anything crazy. What's the protocol for a patch this big?

Should we get a swarm of people to ensure that this thing is OK, or should we just land it?

Comment 11 Aryeh Gregor (:ayg) (away until October 25) 2012-08-02 05:39:10 PDT
On m-c, we just did sed -i 's/\bnsnull\b/nullptr/g' and ehsan gave me r+ for whatever that produced without manual review.  It's really unlikely this could cause problems, since nsnull is currently defined to be nullptr in m-c!
Comment 12 Mike Conley (:mconley) - (needinfo me!) 2012-08-02 07:29:14 PDT
Created attachment 648339 [details] [diff] [review]
Unbitrotted patch (checked in)

Re-ran sed script to un-bitrot, and committed to comm-central as

Note You need to log in before you can comment on or make changes to this bug.