Calendar:QA Chat:2007-09-13:Log

From MozillaWiki
Jump to: navigation, search
[18:30] <mschroeder> hi ulf, hi Andreas
[18:30] <Andreas> hi mschroeder
[18:31] <ulf> Hi mschroeder!
[18:31] <mschroeder> Okay, I think we can start the QA chat.
[18:32] <mschroeder> RC1 is planned for next week.
[18:32] <ulf> right and things move into the right direction imo
[18:33] <mschroeder> I hope we'll gather more people for testing when RC1 is out.
[18:33] <ulf> that would be nice
[18:34] <mschroeder> sipaq and ctalbert do a lot of blog posting to inform the users about the approach of RC1 and the release.
[18:35] <ulf> mschroeder: will we have a sort of QA work session prior to RC1
[18:35] <mschroeder> ulf: good question
[18:36] <ulf> from what I#ve heard we'll see some important fixes in the next couple of days
[18:36] <ulf> and I ask myself how to prevent regression for RC1
[18:36] <mschroeder> ctalbert: How was the QA procedure before the 0.5 release?
[18:37] <mschroeder> ulf: I think a working session is a good idea.
[18:38] <mschroeder> We should do it between freeze and release of RC1.
[18:38] <ulf> good, somewhere early next week?
[18:38] <ctalbert> A working session would be a good idea
[18:38] <ctalbert> One thing we did not do prior to 0.5 was much upgrade testing and it did tend to bite us
[18:39] <mschroeder> ulf: maybe tuesday or thursday
[18:40] <ulf> how about next Wednesday? Tuesday turns out to have too much meetings at least for me
[18:40] <ctalbert> This time, with the timezone fix landing, there are two upgrade paths that are a little worrisome: 0.5 to 0.7 should be tested, and I don't expect too many surprises there, but 0.3 to 0.7 might be a bit wierd.  That's because 0.3 did not have the timezone updates that 0.3.1 and 0.5 had, and that upgrade could be a bit awkward.
[18:40] <ulf> ok, thanks for the hint!
[18:40] <ctalbert> Simon and I got some download numbers and twice as many people downloaded 0.3 as 0.5, so that makes me think we still have some 0.3 builds lurking out there
[18:41] <ctalbert> some == 300,000 or so
[18:41] <mschroeder> On Wednesday I'll be on train most of the time.
[18:41] <ulf> :-(
[18:42] <ulf> so Thursday?
[18:42] <mschroeder> That's better for me.
[18:44] <mschroeder> ctalbert: Will you have time for a working session next week?
[18:45] <ctalbert> maybe early in the morning next week, yes.  Thursday would be the best day.  I'll be at calconnect, but I'll be on the east coast, so it will be easier for me to be awake during your timezone
[18:45] <ulf> mschroeder: what do you think about starting the work session earlier than in the last times?
[18:45] <ctalbert> ulf sounds good to me ^ see my previous comment
[18:45] <mschroeder> yep
[18:46] <ctalbert> If we could take advantage of that fact, then I could be more available and get some stuff done before mountain view wakes up :-)
[18:46] <ulf> I mean Europe people could begin and you guys from the other side of the atlantic join us a bit later?
[18:47] <mschroeder> When should we start?
[18:47] <ulf> when you like
[18:48] <ulf> how about 10.00 UTC? is that too early?
[18:48] <mschroeder> perfect
[18:48] <ulf> cool!
[18:48] <mschroeder> Next point: Localization testing
[18:49] <ulf> can we expect to have localized builds around by then?
[18:49] <ctalbert> 10AM UTC works for me
[18:49] * ctalbert makes an event!
[18:49] <ulf> ctalbert: great!
[18:50] <ctalbert> If Ause can spin us an RC1 in the beginning of the week, then we should have L10N builds by then
[18:50] <mschroeder> 14-15 of 24 l10n tinderboxen are green
[18:50] <ssitter> otherwise one could use the l10n nightly builds
[18:51] <ulf> ctalbert: I'll contact Ause, but if I'm not completely wrong that was the plan anyway
[18:51] <ulf> precondition: the blocker list is down to zero
[18:52] <mschroeder> On Monday we'll have a decision about RC1 and the blocking list afaik.
[18:52] <ctalbert> yeah
[18:52] <ulf> fine! :-)
[18:54] <mschroeder> okay
[18:55] * mschroeder has compiled a list with possible blocker bugs again
[18:56] <ctalbert> mschroeder, can you pastebin the numbers to make it easier for us to follow along?
[18:56] <mschroeder> I have a *tada* tinyurl for bugzilla:
[18:57] <mschroeder> These are nominated, recently filed or reopened bugs.
[18:57] <ctalbert> heh, awesome
[18:57] <ulf> nice, thanks for compiling the kist!
[18:58] <ulf> s/kist/list
[18:59] <mschroeder> First one: bug 357112 
[18:59] <firebot> mschroeder: Bug min, --, ---,, NEW, Drag and drop of multiday-event does't drop on days the shadow suggests
[19:00] <mschroeder> It regressed and the bug was reopened by Omar B.
[19:01] <ctalbert> I'm less worried about that one, and more worried about the "criticial" ones in the list, all things considered
[19:02] <ctalbert> And about the "dead ui" ones, so I think omar's regression is not blocking
[19:02] <mschroeder> okay
[19:02] <mschroeder> Next one: bug 383272
[19:02] <firebot> mschroeder: Bug nor, --, 0.7,, ASSI, [Proto] Recurrence dialog: weekly and monthly recurrencepattern is not in sync with minimonths
[19:02] <mschroeder> Berend set it from 0.7+ to 0.7- today without a comment.
[19:03] <ctalbert> Ulf: can you ask Berend what's up with that one?  Maybe he could comment on it and let us know what happened
[19:03] <ulf> ctalbert: BTW the dead views bug is understood and Daniel is about to fix it atm
[19:03] <ctalbert> sweet
[19:03] <ulf> ctalbert: sure!
[19:03] <daniel_> thanks
[19:04] <mschroeder> next: bug 386479
[19:04] <firebot> mschroeder: Bug nor, --, ---,, ASSI, Switch to Calendar mode don't work properly using buttons, Calendar menu or keyboard shortcut
[19:04] <mschroeder> this one has 0.7? and a reviewed patch
[19:05] <ctalbert> sorry, I think I missed something, ulf, daniel_ can you ask (or do you know) whyberend set bug 383272 to 0.7-?
[19:05] <daniel_> will do
[19:05] <ctalbert> thx
[19:06] <ctalbert> mschroeder: looks like this one will make it.  I'll check it in if Philipp doesn't
[19:07] <mschroeder> Next one: bug 390147
[19:07] <firebot> mschroeder: Bug nor, --, 0.7,, ASSI, Export calendar is not wired up correctly
[19:08] <daniel_> ctalbert: I think fallen keeps track of that: Fallen>	daniel_: bug 386479, no I just missed the r+ :) Checking in now
[19:09] <mschroeder> I think 390147 will be fixed for 0.9.
[19:10] <ctalbert> cool.  looks like 390147 is waiting for christian
[19:10] <mschroeder> so, not blocking?
[19:12] <mschroeder> Next one: bug 390842 
[19:12] <firebot> mschroeder: Bug cri, --, ---,, NEW, Error: redeclaration of var Cc in console, kills Lightning
[19:13] <ctalbert> mschroeder: yeah 390147  is not blocking
[19:13] <ctalbert> imo
[19:13] <mschroeder> daniel_: Can you have a look at bug 390842?
[19:14] <daniel_> hmm
[19:16] <daniel_> I would suggest to provide a waiver for someone substituting all places in a single step once we've landed all patches
[19:16] <daniel_> opinions?
[19:16] <daniel_> meaning a review waiver
[19:18] <mschroeder> daniel_: good idea. can you leave a comment in the bug?
[19:18] <daniel_> clint, are you ok with that or have objections?
[19:18] <ctalbert> I don't understand what a review waiver buys us
[19:19] <daniel_> it doesn't make much sense to me reviewing that; it IMO just blocks the tree, because that patch would be large and instantly bitrotten...
[19:20] <mschroeder> We have decision how to solve the problem and it's not so difficult, so maybe someone steps up to produce a patch.
[19:20] <ctalbert> yeah, that makes sense, I'll review it once we get a patch
[19:20] <ctalbert> we just have to do it quickly
[19:20] <ctalbert> has to be unbitrotten to check it inanyway
[19:20] <daniel_> mschroeder: for what reason?
[19:21] <mschroeder> daniel_: ?
[19:22] <daniel_> IMO could be done in a single step; I would pass a waver to someone I trust.
[19:22] <daniel_> no problem, trivial change.
[19:22] <mschroeder> okay
[19:22] * ctalbert wants to prevent the argument about "unreviewed code being checked in" is all.  It's not that it's a difficult chanfge
[19:23] <daniel_> ctalbert: we agreed on that, and it's our project.
[19:23] <ctalbert> maybe an over the shoulder review would work :-)
[19:24] <ctalbert> ok, we're blocking up the qa chat.  mschroeder what's next sorry
[19:24] <mschroeder> lol
[19:24] <mschroeder> Next one: bug 395883
[19:25] <mschroeder> That's a ugly UI problem. But I think mickey only need 5 minutes for a patch. ;)
[19:25] <ctalbert> firebot bug 395883
[19:25] <firebot> ctalbert: Bug nor, --, ---,, NEW, Readonly dialog says: "Repeat: Occurs ???" on yearly and monthly recurring items.
[19:25] <Andreas> Not nice but IMO non blcking.
[19:26] <ctalbert> +1
[19:27] <mschroeder> okay. maybe I can fix it. ;)
[19:27] <ctalbert> go mschroeder go!
[19:27] <ctalbert> :-)
[19:27] <mschroeder> Next one: bug 395658
[19:28] <firebot> mschroeder: Bug nor, --, ---,, NEW, colliding event boxes in rotated calendar view
[19:29] <ulf> Andreas? I think you agreed with Mickeythat this one does not block? right?
[19:30] <Andreas> Yes, more non blocking.
[19:30] <mschroeder> Next one: bug 393307
[19:30] <firebot> mschroeder: Bug min, --, ---,, NEW, summary is not refreshed correctly when mouse over events
[19:30] <ctalbert> can ya'll update the bug with why, Andreas?
[19:30] <ctalbert> (so we have a record)
[19:31] <Andreas> Ok, I will do so.
[19:31] <ulf> Btw: the overlapped boxes are fixed in the nightlies :-) 
[19:31] <ulf> only rotated view is remaining
[19:33] <ulf> meaning war is over regarding waronboxes - peace brothers :-)
[19:33] <ctalbert> peace
[19:33] <ctalbert> ;-)
[19:33] <ctalbert> Andreas, thanks
[19:33] <mschroeder> Any opinion on 393307?
[19:34] <ulf> tiny repaint issue?
[19:35] <ctalbert> sounds like, probably not blocking
[19:35] <mschroeder> okay
[19:35] <ulf> ctalbert: that's what I thought also
[19:35] <mschroeder> bug 395288 has already a patch by ssitter and 0.7?
[19:35] <firebot> mschroeder: Bug nor, --, ---,, ASSI, [Proto] Event dialog: Wrong Reminder entry shown for reminders greater '7 days before the event star
[19:36] <ctalbert> If mickey can get it reviewed, it would be great to get this.  We've been getting a lot of complaints about reminders on the blog and what not
[19:37] <ctalbert> Guys, I have to head into the office now.  I'll be back online in about twenty minutes or so.
[19:38] * mschroeder thinks this should block. (I hate those alarm/reminder bug reports ;))
[19:38] <mschroeder> ulf, Andreas?
[19:38] <ulf> ssitter, ctalbert: review missing? 
[19:39] <ulf> has anyone spoke w. mickey about it?
[19:40] <mschroeder> ulf: I'll give him a hint on IRC later or tomorrow. ;)
[19:40] <ulf> yoou know, mickey is quite busy these days w. reviews, but I don't have a status on this one
[19:40] <ulf> ok, fine; - thanks!
[19:40] <ssitter> yes, the patch is waiting for review. daniel said it's mickeys code therefore I asked hin for review
[19:40] <mschroeder> ulf: mickeys queue is almost empty according to what he reported at yesterday's conf call.
[19:40] <ssitter> but if someone wants to take over ...
[19:41] * mschroeder only reviews patches with max. 3 lines. ;)
[19:41] <ulf> ssitter: mickey is probably perfect
[19:42] <mschroeder> Last one: bug 394010
[19:42] <firebot> mschroeder: Bug nor, --, ---,, NEW, The recurrece dialog has major user experience flaws
[19:42] <ssitter> mschroeder: no problem, I can split up the patch in pieces for you :)
[19:43] <ulf> enhancement?
[19:43] <mschroeder> ssitter: lol
[19:44] <ulf> ssitter: rotfl
[19:44] <ssitter> it's bad UI but not a blocker I think
[19:44] <mschroeder> ulf: If you want to create a custom monthly recurrence, it's really bad UE at the moment.
[19:45] <ulf> am I missing something? can you elaborate in your words what's happening?
[19:47] <ssitter> I think this is the same issue ctalbert mentioned during the proto dialog review in bug 370435 Comment #28
[19:47] <firebot> ssitter: Bug nor, --, 0.7,, RESO FIXED, Review prototype event dialog
[19:47] <mschroeder> ulf: sorry I meant custom annually. It always selects Jan 1.
[19:47] <ssitter>  > I first click on a recurrence pattern of yearly, and then I click on "custom". The recurrence dialog defaults to "Every 1st of January".  It should at least make the assumption that I want to use my date settings from the event dialog and default to "every 5th of September".  The same holds true for month, and week repetition (day of month and day of week should be the defaults there).
[19:49] <ulf> ok, I absolutely agree w you that recurrence-custom is crap! but is that really blocking this late? 
[19:51] <Andreas> Yes, ugly but non blocking.
[19:52] <ulf> mschroeder: also agreed?
[19:52] <mschroeder> no, but you have more experience than me. :)
[19:53] <Andreas> Sorry guys, I have to leave. Bye.
[19:54] <ulf> mschroeder: can you collect the feedback on recurrence-custom and contact chrisJ thisregarding
[19:56] <mschroeder> ulf: okay. i'll do that.
[19:56] <ulf> that's great! thanks!
[19:59] <mschroeder> That's the end of the QA chat. ;)