Calendar:QA Chat:2006-02-22:Log

From MozillaWiki
Jump to: navigation, search
[11:33]	<ctalbert>	===== Chat Starting Header =======
[11:33]	<ctalbert>	Congratulations and Thanks to everone on shipping 0.3.1!!!!!
[11:33]	<ctalbert>	I know we're all glad to have that effort behind us.
[11:34]	<ctalbert>	Thanks to mschroeder, we have an agenda today, and it is here:
[11:34]	=-=	eor|brb is now known as eor
[11:34]	<eor>	i m back
[11:35]	<ctalbert>	One thing I wanted to do before diving into the bugs was to go 
over what went well for the 0.3.1 release. And also to talk about what we can do to
 improve for the 0.5 release
[11:36]	<Archaeopteryx>	one point: i hope will have launched v3 
(remora) and i will talk to the admins to make calendar a trusted add-on so that if 
you send it to amo, it will be automatically published. anyone against this? we have 
the qa testdays
[11:36]	<ctalbert> thing that went very well was that we had a lot of people
 join us and help out with testing on timezones and locales. I thought that was very positive.
[11:37]	<ctalbert>	Archaeopteryx: I'm not specifically against that, I would just 
recommend that you check with lilmatt first to keep him in the loop on that.
[11:38]	<Archaeopteryx>	ok, he can do this if he wants, there has to be probably someone
[11:38]	<ctalbert>	This would only auto-publish the release builds of Lightning right?
 Not the nightlies?
[11:38]	<Archaeopteryx>	yes
[11:38]	<ctalbert>	Then I'd be ok with it.
[11:39]	<ctalbert>	One of the big problems we had during the release were the test 
cases -- I think that they could have been better and easier to run. I'm going to 
expend some effort on those before we get into the 0.5 release cycle.
[11:39]	<ctalbert>	eor: Did you have any feedback from the students?
[11:40]	<ctalbert>	Everyone should feel free to jump in...
[11:40]	<eor>	yes i do
[11:40]	<ctalbert>	cool
[11:40]	<eor>	Timezone test cases are too long for student to start with
[11:41]	<eor>	last time when we were running the testday with them, I was with 
them and it took them more than an hour to get one test case done
[11:41]	<eor>	even though they were happy that they find a bug and were able to 
file the bug, they seems to be a little discourage or in another word bored by the long test case
[11:42]	<ctalbert>	That's good feedback. Damian has a rule that it should take no 
longer than 1 minute to read a litmus test case and no more than 2-5 to run it.
[11:43]	<ctalbert>	The timezone tests were definitely hideous.
[11:43]	<eor>	yeah, the timezone test cases should really be broken down much further
[11:43]	<eor>	i was surprise that we didnt scared them away and they have the patient
 keep going with testing
[11:43]	<eor>	:D
[11:44]	<ctalbert>	That's good. Was there more?
[11:44]	<eor>	yup, we are hoping to encourage more student to get involve, it is
 hard during school year, because they have other courses to focus on
[11:45]	<eor>	however, i m thinking that student in 2nd year should be familiar
 with computer envionrment already. They should be able to run thru simple test cases on litmus
[11:45]	<ctalbert>	I understand. Did anyone else have feedback on the release 
process? Do you see things that we could do differently/better?
[11:46]	<eor>	it is good to start them in 2nd year, because they could being with 
doing user testing to learn more about OSS, and then they could have further development 
for OSS in their 3rd year and etc
[11:48]	<ctalbert>	We are planning (tentatively) to release 0.5 at the end of 
March/beginning of April. Keep an eye on the newsgroup for a post about that schedule. 
So, we'll be adjusting the QA cycle to that schedule soon. In the meantime, we'll 
start the "every two week" test days on March 6.
[11:49]	<mschroeder>	Sounds good.
[11:49]	<ctalbert>	If anyone thinks up something that we should change or do to 
speed up the QA process on the release then let me know, either in here or via email: 
ctalbert.moz at
[11:49]	<ctalbert>	One thing I want to do before then is look at code coverage w.r.t. 
our testcases. That's an undertaking in itself, so if anyone wants to volunteer to help, let me know.
[11:50]	<ctalbert>	Ok. Let's go on to QA Discussion bugs:
[11:51]	<mschroeder>	i added some today
[11:51]	<ctalbert>	Yeah, I saw that in my bugmail. The First bug 362918
[11:51]	<firebot>	ctalbert: Bug 
nor, --, ---,, UNCO, Installer reports insufficient disk space
[11:51]	<mschroeder>	I think, this is in the wrong component.
[11:52]	<ctalbert>	+1
[11:53]	<ctalbert>	we'll move it.
[11:53]	<ctalbert>	368269...
[11:53]	<mschroeder>	I'll do that?
[11:53]	<ctalbert>	ok sounds good
[11:56]	<eor>	dave, my teacher is running a mozilla crash course next week, so 
hopefully we got more people to get involve
[11:56]	<mschroeder>	368292: First the reporter thought one event is duplicated, 
but the event began befor midnight and ended after.
[11:56]	<ctalbert>	eor: That's awesome.
[11:56]	<ctalbert>	mschroeder: I think we can mark it as a dupe, like Stefan mentioned.
[11:56]	<eor>	next week is study week, will see if we can remind the student in to
 club to get more involement too
[11:57]	<firebot> set the Resolution field on bug 368269 
to DUPLICATE of bug 337191.
[11:57]	<firebot>	Bug maj, 
--, ---,, RESO DUPLICATE, When I create a non-recurring event 
it duplicates to the immediate next day
[11:57]	<firebot>	Bug nor, 
--, ---,, NEW, Event/Task Dialog always shows times in OS timezone 
regardless of timezone preferences
[11:57]	<mschroeder>	okay
[11:57]	<ctalbert>	bug 369962 is next
[11:57]	<firebot>	ctalbert: Bug 
min, --, ---,, UNCO, sunbird : changing alarm sound doesn't show
 full path only file name
[11:58]	<ctalbert>	Oh this is that alarm path issue.
[11:58]	<ctalbert>	What it shows you is the chrome path to the file. I think for 
this we should just change the test case. I don't think it is important enough to 
bother with changing the application.
[11:58]	<ctalbert>	What do others think?
[12:00]	<ssitter>	iirc Thunderbird also only shows the filename for the new mail 
[12:00]	<eor>	sorry, just have a question that not related in here, but hoping 
that you can give me feedback
[12:00]	<ctalbert>	ssitter: that's good enough for me. Let's change Litmus.
[12:00]	<eor>	Is "Vista locking out security vendor" a good white paper topic?
[12:01]	<ctalbert>	eor: sure. I can talk to you more about that after the meeting, 
if you want.
[12:01]	<ctalbert>	bug 370059 s next
[12:01]	<firebot>	ctalbert: Bug
 nor, --, ---,, UNCO, Printing color in Lightning gives B/W
[12:01]	<eor>	thanks ctalbert!
[12:03]	<eor>	you want to re-work on some test cases on litmus b4 next release right? 
I can help out with that next week. what about i try to re work some of the timezone
 test cases
[12:03]	<mschroeder>	We have some similar bugs and you (ctalbert) investigated the 
behavior with printing in b/w and color.
[12:04]	<ctalbert>	mschroeder: yes, I remember. I don't know if the printing componentry 
in Lightning is similiar enough to the printing componentry in Sunbird to mark this
 as a DUPE or not.
[12:04]	* ctalbert	imagines it is the same
[12:04]	<ctalbert>	Let's add lilmatt and daniel to the CC list and ask them.
[12:05]	<ssitter>	basically the problem is that we don't have a real print preview
[12:05]	<ctalbert>	Is that what causes all these headaches?
[12:05]	* ctalbert	looks at 350099
[12:05]	<ssitter>	that shows a preview based on the print/page options
[12:06]	<ssitter>	like with/without colors
[12:07]	<ctalbert>	Ah. I get you. Do you think we should dupe 364842, 370059, and 
367898 all against that bug?
[12:07]	<ctalbert>	eor: I would really appreciate help with reworking the testcases! 
[12:07]	<mschroeder>	How can I change product _and_ component for a bug in one
[12:08]	<ctalbert>	mschroeder: Try changing product then clicking on component link?
[12:09]	<ctalbert>	lilmatt: Do you have any advice for what to do with this slew 
of printing bugs?
[12:09]	<lilmatt>	wait until cairo
[12:09]	<ctalbert>	Uh...I meant should we dupe them against the print preview bug 
or not.
[12:09]	<ctalbert>	Do you feel they are all the same issue?
[12:10]	=-=	Fallen|away is now known as xFallenAngel
[12:12]	<mschroeder>	ctalbert: clicking on component link doesn't work... strange
[12:12]	<lilmatt>	not entirely.
[12:12]	<ssitter>	mschroeder: i think if you change the product and than you comitt 
the changes you'll be asked for component
[12:12]	<lilmatt>	I'd make them depend on it
[12:12]	<mschroeder>	ssitter: thanks
[12:12]	<ctalbert>	Okay. Thanks lilmatt, we'll do that.
[12:13]	<ctalbert>	Bug 370803 is mine (Litmus tests -- specifically the timezone 
[12:13]	<firebot>	ctalbert: Bug
 nor, --, ---,, UNCO, litmus-tests lacking information
[12:13]	<firebot> changed the Product on bug 362918
 from Calendar to Toolkit.
[12:13]	<firebot>	Bug nor, --,
 ---,, UNCO, Installer should check the free space for the unc path
[12:14]	<ctalbert>	Bug 371133 is next
[12:14]	<firebot>	ctalbert: Bug 
maj, --, ---,, UNCO, Extra linebreak in ics/webdav publishing 
removes SUMMARY in Apple iCal
[12:16]	<ssitter>	can someone with iCal confirm this issue?
[12:16]	<ctalbert>	sure give me a moment
[12:16]	<ssitter>	otherwise i think it's a duplicate/depends on bug 324198
[12:16]	<firebot>	ssitter: Bug
 nor, --, ---,, NEW, Bad line ends when exporting event with long
[12:19]	<ctalbert>	ssitter: Using the sample from 324198, I can recreate the issue
 in iCAL. I think 371133 is a dupe of 324198
[12:20]	<ssitter>	this answers Comment #7 from bug 324198
[12:21]	<ssitter>	but in that case i maybe mark it dependend
[12:21]	<ctalbert>	too late
[12:21]	<firebot> set the Resolution field on bug 371133 to
 DUPLICATE of bug 324198.
[12:21]	<ctalbert>	yeah, I guess depend is a better relationship there.
[12:23]	<ctalbert>	ssitter: Should I reopen it and mark it as a dupe?
[12:23]	<ctalbert>	sorry, should I reopen it and mark it as *depends*?
[12:24]	<mschroeder>	ctalbert: don't forget to remove the qa flag :)
[12:25]	<ctalbert>	I also forgot to remove the qa flag, so I have to edit it 
[12:25]	<ctalbert>	heh, thanks
[12:25]	<ssitter>	or we just add the info to the summary and proceed as proposed 
by dmose in comment 7
[12:26]	<ctalbert>	ok, let's do that.
[12:27]	=-=	eor is now known as eor|afk
[12:28]	<firebot> requested blocking-calendar0.5 from the 
wind on bug 324198.
[12:28]	<firebot>	Bug nor, --,
 ---,, NEW, Incorrect line endings with long UID loses SUMMARY 
 information in Apple iCal
[12:30]	<ctalbert>	That's the QA Discussion list. I want to deviate from the agenda
[12:31]	<ctalbert>	xFallenAngel: Earlier today, Archaeopteryx and I were discussing 
the FTP authentication issues in bug 347128 and bug 273478
[12:31]	<firebot>	ctalbert: Bug
 nor, --, ---,, RESO WORKSFORME, Changes to a remote calendar using 
 ftp (without providing the username/password in the URL) are not
[12:31]	<firebot>	ctalbert: Bug
 nor, --, ---,, NEW, Anonymous FTP confuses Firefox calendar 
 logging into remote server
[12:32]	<ctalbert>	Can you find a way to summarize the remaining issues with FTP 
authentication and file any needed bugs against Product: "Core" and Component: "Network:FTP"?
[12:33]	<ctalbert>	My hunch is that we are running into problems because we expect
 Sunbird (a calendaring app) to use FTP differently than Firefox (a browser) uses FTP.
[12:33]	<ctalbert>	And the FTP componentry probably defaults to Firefox acceptable 
[12:35]	<xFallenAngel>	I haven't read both bugs in detail yet, but it seems the issue
 described is that anonymous login confuses some ftp servers. As stated in the first 
 bug, the easy workaround for such servers is using ftp://user@host/, I personally 
 think that is enough to WONTFIX all upcoming bugs on that
[12:35]	<xFallenAngel>	those users will have the same probems in firefox
[12:35]	<xFallenAngel>	and probably use the same workaround
[12:36]	<xFallenAngel>	The only other ftp issue I remember is the two-channels bug,
 but I think we have that described somewhere
[12:37]	<Archaeopteryx>	issues: 1. so the user has to write his loginname into his 
profile 2. the prompt is talking about entering username and password for ""
[12:37]	<ctalbert>	Archaeopteryx: Isn't there another issue where the user has to 
put his password into the URL also? Or was that only a configuration issue for that
 particular FTP server?
[12:38]	<Archaeopteryx>	for me worked ftp://user@host to download and edit
[12:38]	<mschroeder>	On the QA TODO list we have: "Need a follow on FTP bug 
regarding the fact that separate channels are used for reading and writing."
[12:40]	<xFallenAngel>	I'll try to whip something up, giving you guys a preview 
before I post, in case I misunderstood of forgot something
[12:40]	<ctalbert>	Archaeopteryx: I think that the prompt business is a known issue
 on the FTP bugs...there are several about prmpt not being completed: 139851 for 
[12:40]	<Archaeopteryx>	thx
[12:41]	<ctalbert>	xFallenAngel: Ok. So do you think we can WONTFIX 273478 and suggest 
the username workaround instead?
[12:42]	=-=	eor|afk is now known as eor
[12:42]	<xFallenAngel>	Well, if the core people have a bright idea how to manage 
this, then its not WONTFIX.
[12:42]	<xFallenAngel>	I'd say when the core bug is filed, dupe them against it.
[12:43]	<ctalbert>	Ok. That makes sense. Then let us know what you come up with 
for a Core bug, and we'll take it from there. Thanks.
[12:43]	<xFallenAngel>	sure thing :)
[12:43]	<ctalbert>	Archaeopteryx, xFallenAngel FTP bugs with "prompt" in them:
[12:44]	<ctalbert>	Okay. That's the end of our time today. I'm going to try to 
carve out some time to work on QA Wanted bugs next week, and I will post something 
up onto the channel topic or the TODO list or both (and the google calendar) to 
let people know about it.
[12:45]	|<--	Andreas has left (Quit: Chatzilla 0.9.67+ [SeaMonkey 
[12:45]	<ctalbert>	Thanks again for all your help in releasing 0.3.1 and every day. 
We couldn't do it without you!