<ctalbert> Hi everyone. Let's get started with the QA Chat
[11:32] <ctalbert> There is a test day on Tuesday January 16. I'll try to
be online the entire time from 13:00 to 01:00 UTC
[11:34] <ctalbert> That's my major announcement. It looks like Damian has
some ideas for testing. Damian, do you want to summarize them?
[11:36] <ctalbert> Hmmm....they are pretty self explanatory. I think
these are good ideas. I've also found a site for free ICS sharing,
www.icalshare.com
[11:36] -->| ulf_home (ulf@moz-F628D62D.dip.t-dialin.net) has joined
#calendar-qa
[11:37] <ctalbert> Hi Ulf, we just started and were looking at Damian's
ideas that he posted to the wiki:
http://wiki.mozilla.org/Calendar:Current_QA_Chat
[11:37] <ulf_home> Hi Clint!
[11:37] <damian> Hi, there are note notes in wiki, please refer to
them
[11:38] <damian> this is what I mentioned time ago but then we hadn't
got own calendar, right now xFallenAngel provides one which is
really fast (at least for me)
[11:38] <ctalbert> Ok. I think we can make use of these new
calendars as part of the test day on Tuesday because one thing
we need to verify is that the ICS provider still works after mvl's
recent changes.
[11:38] <mschroeder> I like the idea to prepare some test calendars for
timezones, languages and large numbers of events
[11:39] <damian> thanks
[11:39] <ctalbert> Damian, do you want to create a few litmus test
cases using these new calendars?
[11:39] <damian> sure, why not
[11:39] <ctalbert> Then, I'd encourage all of us to upload data to
them before Tuesday. I have the ability to upload Japanese data, so I
will do that.
[11:39] <damian> ctalbert: can you prepare new subgroup for it?
[11:39] <ctalbert> Sure.
[11:40] <ctalbert> Timezones and International?
[11:40] <ctalbert> Actually, I think timezones should be part
of the Event Subgroup.
[11:40] <damian> yes, this is way to test Japanese Russian
and some others languages
[11:40] <ctalbert> But, I'll make one for internationalization
and localisation.
[11:41] <damian> sounds good, but quite long
[11:41] <ctalbert> I'll shorten it when I put it into Litmus.
[11:41] <damian> oh, that makes sense
[11:42] <ctalbert> thanks for the ideas, damian
[11:42] <damian> how about addresses, u guess this http://wiki.mozilla.org/Calendar:QA_Test_Servers is a good place for
it? put there calendars names
[11:43] <ctalbert> Yes, please add them there.
[11:43] <damian> just to keep all those information in one place
[11:43] <damian> ok
[11:44] <damian> how about the third one with many events?
any idea how to do it well?
[11:45] <damian> I guess creating the same event just like
a copy is not best way?
[11:45] <ctalbert> mvl had a test calendar with many events
on it. I can upload that one to this calendar.
[11:46] <ctalbert> That will give us a good start.
[11:46] <damian> we will see - we can always improve it
before next testcase day which is in two and half week as I remember
[11:47] <ctalbert> Ideally, we need to work on creating a large
calendar of just recurring events, a large one of non-recurring events
and a few with different mixtures of the two (40% non-recurring, 60%
recurring) etc.
[11:47] <ctalbert> But, that won't happen before Tuesday.
[11:47] <ctalbert> I'll add that to the TODO list, actually.
[11:48] <damian> this is good idea, another one with alarms -
I guess we can have one with all but then we need to specify what user
expect when all events are loaded on start up - how many of them are
visible how many alarms must fire and so on
[11:49] <ctalbert> yes.
[11:49] <damian> I guess we will have to specify it after next testday
[11:50] <ctalbert> Yes, let's hold off on this until after the
16th, but it would be nice to have these large calendars for testing
on the January 30th test day.
[11:50] <ctalbert> Let's go on now to the QA Discussion Bugs.
[11:50] <damian> what is going to be then?
[11:51] <ctalbert> The 30th will be a big test day. We should have
performance fixes and timezone stuff by then
[11:52] <ctalbert> ** QA Discussion Bugs ** : http://tinyurl.com/yhwge4
[11:53] <ctalbert> First one: bug 273645: what needs to be discussed?
[11:53] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=273645 nor, --, ---,
nobody@mozilla.org, RESO WORKSFORME, slow reload of remote calendar
from apache webdav server
[11:54] <mschroeder> damian marked it
[11:54] <ctalbert> Damian: in 273645, I agree that a performance
test would be a good thing to have. Should we put that in Litmus, or
were you thinking of something else?
[11:54] <damian> it looks like we've just discussed it
[11:55] <ctalbert> ok. cool
[11:55] <ctalbert> Next up: 300470
[11:56] <damian> this one we can also verify when I prepare calendars ;)
[11:56] <ctalbert> yes, you and I can try this on the test day.
[11:56] <damian> true
[11:57] <ctalbert> I'll make a task to remind me about it, and
clear the whiteboard flag.
[11:57] <ctalbert> Next: bug 31101
[11:57] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=31101 cri, P3, ---,
serhunt@flash.net, VERI WORKSFORME, [SHOCKWAVE] crash on opening Discovery
Channel Mammoth page that uses Shockwave and detects plug-ins
[11:57] <ctalbert> uh...no. bug 311301
[11:57] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=311301 min, --, ---,
nobody@mozilla.org, UNCO, End Date is set back one day if start time is set
[11:58] <ctalbert> What did we want to discuss here? Does this still
happen?
[11:59] <mschroeder> I wondered, if this is really an issue or if we
can mark it invalid?
[11:59] <ctalbert> good question
[12:00] <mschroeder> the bug describes a behavior of lightning and
sunbird, most of us are used to
[12:00] <mschroeder> i don't know what the prototype event dialog
does
[12:01] <ctalbert> Nor do I. The issue still happens in the
non-prototype new event dialog.
[12:01] <ctalbert> I would say it's really an issue because it
is confusing for the user if they get into this state.
[12:02] <ctalbert> jminta: Do you have a recent opinion about
311301?
[12:04] <ctalbert> Perhaps we should detect that the user is
creating an "invalid" situation here and somehow prevent them from
getting there. I think we should go ahead and confirm this one since
the issue still exists and has a potential (though perhaps small)
to confuse people.
[12:04] <ctalbert> Thoughts?
[12:04] <mschroeder> okay.
[12:04] <ulf_home> the prototype UI always sets an offset of
one hour for the end date
[12:05] <ctalbert> That sounds like a better solution than
trying to maintain a duration setting.
[12:05] <ctalbert> which is how this bug happens.
[12:06] <ctalbert> When confirming it, note that it might cease
to be an issue if we decide to go with the prototype event dialog.
[12:06] <ssitter> but this should only happen if the end date
is invalid. otherwise it should keep the duration as is
[12:07] <ctalbert> ssitter, I think that's a good idea.
[12:07] <ctalbert> Best of both worlds.
[12:08] <ctalbert> Shall we go on to QA Wanted?
[12:08] <ctalbert> ** QA Wanted Bugs ** :
https://bugzilla.mozilla.org/buglist.cgi?query_format=advanced&product=Calendar&keywords=qawanted
[12:09] <ctalbert> Bug 340488 is a migration issue. Sebo fixed
some of the migration wizard issues on Windows, so this Tuesday will
be a good time to try this one out and do some QA on it (and we can
test Sebo's fix at the same time).
[12:09] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=340488 nor, --, ---,
nobody@mozilla.org, UNCO, birthday's listed double after upgrade to a
new sunbird version
[12:11] <ctalbert> Bug 277941 is next.
[12:11] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=277941 nor, --, ---,
nobody@mozilla.org, UNCO, FTP password pops up every time I make an
addition to my calendar.
[12:11] <ctalbert> Any takers on that?
[12:11] <ctalbert> I think this is fixed.
[12:12] <ctalbert> I might be able to get to it on Tuesday, but
probably not before. Next up: 285779
[12:14] <mschroeder> does somebody have a tomcat webdav installation?
[12:15] <ctalbert> I thought that was what xFallenAngel's
installation was. Apache + Tomcat
[12:16] <ctalbert> I could be wrong. I'll ask him again next
time I see him.
[12:17] <ctalbert> dmose: Can you help us look at bug 308569?
[12:17] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=308569 nor, --, ---,
dmose@mozilla.org, NEW, CalDAV should specify HTTP charset as UTF8
[12:18] <ctalbert> Ok. I'll ask dmose about that one in a bit.
[12:18] <bbbrowning> yes, I think at this point that's mostly an 'ask dmose'
[12:19] <bbbrowning> at some point someone should probably look at
the on-wire traffic, but I'm not terribly worried that that's wrong
[12:19] <ctalbert> it would be good to do. I haven't had time
to set up everything to do it.
[12:19] <ctalbert> I'd like to do it for both ICS and Caldav
providers.
[12:19] <ctalbert> I'll add that to our TODO list. Perhaps someone
else will have the time.
[12:20] <bbbrowning> I can look at the ICS traffic by or on
Tuesday; I shouldn't verify the CalDAV since it was my patch
[12:21] <ctalbert> Thanks, that'd be a big help. I'll see if
I can get someone to do CalDav, if not, I'll do it later next week
sometime.
[12:22] <ctalbert> I must be going crazy. It looks like the
query just re-sorted itself. I've lost my place. Hmmm...Has anyone
tried bug 339960?
[12:22] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=339960 nor, --, ---,
nobody@mozilla.org, UNCO, event editor adds 2 hours to time when
saving to webdav
[12:24] <damian> I hope more bugs can be testerd and verified
when we will have our own server and it will be easier to encourage
people
[12:24] <ctalbert> Ok. We should be able to confirm that one on
tuesday.
[12:24] <mschroeder> I wanted to test it, but at the moment I have
no Linux installation
[12:24] <ssitter> maybe dupe of the foreign timezone bug?
[12:24] <ctalbert> ssitter, I was wondering that too.
[12:24] <mschroeder> that's possible
[12:25] <ssitter> tz europe/berlin is exactly +2 during DST
[12:26] <damian> Berlin is equal to Poland and I'm UTC+1
[12:26] <ssitter> damian: this was filed 2006-06-01
[12:27] <damian> so?
[12:27] <ssitter> during daylight saving time we have utc+2,
during winter we have utc+1 in germany
[12:28] <damian> oh, I understand
[12:29] <ctalbert> Ok. We still need to check that one out with
a linux installation, either way.
[12:29] <ctalbert> Thanks to xFallenAngel and Jef Driesen, bug
347128 has been extremely well QA'd.
[12:29] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=347128 nor, --, ---,
nobody@mozilla.org, UNCO, Changes to a remote calendar using ftp
(without providing the username/password in the URL) are not
[12:29] <ctalbert> I see mschroeder's post about anonymous
FTP. Should this be a duplicate or is there more going on here?
[12:31] <mschroeder> I don't know, I think it may be somehow connected
[12:34] <ctalbert> From comparing xFallenAngel's output with
Jef's output,
it looks like the issue is that he can never create the test.ics
file on
his FTP server.
[12:35] <ctalbert> In the create event section of Jef's
log: this
part is what bugs me:
ftp> PUT event.ics test.ics
200 PORT command successful
550 test.ics: Permission denied
[12:36] <ctalbert> I will add a comment to this, and remove the
QA
Wanted flag. I think at this point we have all the information that
we need.
[12:36] <mschroeder> okay
[12:37] <ctalbert> We're just about out of time, but we have
three more bugs. Let's do these quickly.
[12:37] <ctalbert> Bug 349330
[12:37] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=349330 nor, --, ---,
nobody@mozilla.org, UNCO, Show country in Holiday Files
[12:37] <ctalbert> First off, this ought to be an enhancement
[12:38] <ctalbert> And it seems like it ought to reside with the
person creating the holiday calendar. I don't think we want to program
in all the country holidays into sunbird.
[12:38] <ctalbert> Should we WONTFIX this?
[12:39] <mschroeder> yes
[12:39] <ssitter> yeah, if you want to show the country add
it to title
[12:39] <ctalbert> ok. I'll do that.
[12:39] <ctalbert> bug 356987
[12:39] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=356987 nor, --, ---,
nobody@mozilla.org, UNCO, calIIcalComponent.startTime error when loading
calendar from Evolution
[12:40] <mschroeder> I think we could mark this invalid.
[12:40] <mschroeder> as I already commented in the bug
[12:41] <ctalbert> You don't have to have VTIMEZONE in the ICS. You
only have to have it if you refer to a VTIMEZONE in DTSTART or DTEND
[12:41] <ssitter> yes. maybe we should file a new bug to handle
this type of error (missing timezone definition) more gracefully in future
[12:42] <ctalbert> Let's do that, close this one as invalid
(since we can't open the ICS file he uploaded).
[12:42] <mschroeder> but we marked some bugs invalid because the
ICS didn't have a VTIMEZONE
[12:43] <ctalbert> Those probably referenced one in DTSTART and
DTEND without specifying it.
[12:43] <mschroeder> oh, i see
[12:44] <ctalbert> Cool. I can go over the three types of time specifications in the date/time fields of RFC 2445 if you want after
the meeting.
[12:44] <ctalbert> Bug 361800 -- unhappily I am doing a windows
restore on my parent's computer this weekend. I'll test this then. :-)
[12:44] <firebot> ctalbert: Bug
https://bugzilla.mozilla.org/show_bug.cgi?id=361800 nor, --, ---,
nobody@mozilla.org, UNCO, Windows Restore causes events to disappear
[12:44] <mschroeder> good ;)
[12:45] <ctalbert> I apologize for going over. Does anyone have
anything else to bring up?
[12:46] <ctalbert> Sounds like no. Thanks again for your help
and patience. I hope to see ya'll on Tuesday at the test day. Have
a good weekend.
[12:46] <ctalbert> I'll be travelling the rest of today and
tomorrow and will be mostly offline.
[12:46] <ctalbert> *** End of QA Chat ***