/irc-logs / mozilla / #accessibility / 2012-01-30 / end
Options:
- # Session Start: Mon Jan 30 00:00:00 2012
- # Session Ident: #accessibility
- # [00:15] * Joins: Jamie (jamie@moz-CA26021.jantrid.net)
- # [00:54] * Joins: ehsan (ehsan@moz-A6FE435.build.sjc1.mozilla.com)
- # [01:32] * Quits: @firebot (firebot@moz-67BF9FE6.carolina.res.rr.com) (Ping timeout)
- # [01:32] * Joins: firebot (firebot@moz-67BF9FE6.carolina.res.rr.com)
- # [01:32] * ChanServ sets mode: +o firebot
- # [01:34] * Quits: @firebot (firebot@moz-67BF9FE6.carolina.res.rr.com) (Ping timeout)
- # [01:35] * Joins: firebot (firebot@moz-67BF9FE6.carolina.res.rr.com)
- # [01:35] * ChanServ sets mode: +o firebot
- # [01:37] * Quits: @firebot (firebot@moz-67BF9FE6.carolina.res.rr.com) (Ping timeout)
- # [01:38] * Joins: firebot (firebot@moz-67BF9FE6.carolina.res.rr.com)
- # [01:38] * ChanServ sets mode: +o firebot
- # [02:10] * Quits: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp) (Quit: mike5w3c)
- # [02:26] * Joins: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp)
- # [02:31] * Quits: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp) (Quit: mike5w3c)
- # [02:42] * Joins: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp)
- # [02:44] * Quits: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp) (Ping timeout)
- # [02:50] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
- # [03:31] * Quits: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com) (Ping timeout)
- # [03:32] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
- # [03:53] * khuey is now known as khuey|away
- # [03:54] * khuey|away is now known as khuey
- # [04:08] * khuey is now known as khuey|away
- # [04:18] * Joins: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com)
- # [04:18] * ChanServ sets mode: +o jprmc
- # [04:19] * Quits: @jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com) (Input/output error)
- # [04:35] * Joins: surkov (surkov@DD563F5.F78D7EEB.34044A7F.IP)
- # [04:35] * ChanServ sets mode: +o surkov
- # [04:58] * Joins: hub (hub@moz-E2FCA694.figuiere.net)
- # [04:58] * ChanServ sets mode: +o hub
- # [05:08] * Joins: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp)
- # [05:15] <Jamie> hi everyone
- # [05:15] <Jamie> surkov: quick question if I may? :)
- # [05:15] <@surkov> Jamie: hi, of course
- # [05:16] <Jamie> surkov: Some time ago (I'm not sure when, but my hunch is somewhere in the Firefox 11 timeline), windowed plugins started getting exposed with an extra HWND wrapper
- # [05:16] <Jamie> surkov: previously (firefox 4 and later), you had the single Firefox HWND and the plugin was in a child window of that with class GeckoPluginWindow
- # [05:16] <@surkov> didn't hear about that, that's a problem right?
- # [05:17] <Jamie> now, we have Firefox HWNd -> random MozillaWindowClass wrapper _. GeckoPluginWindow
- # [05:17] <Jamie> well, if it's unintentional, it's a problem
- # [05:17] <Jamie> if not, it's something we need to work around on our side, though I still think we need a better solution for all of this plugin dodginess
- # [05:18] <@surkov> Jamie: that's new for me so can you file a bug please, and we'll figure out what we can do
- # [05:18] <Jamie> sure. just checking whether you knew what was going on off the top of your head
- # [05:18] <Jamie> I have a few bugs to file today :)
- # [05:18] <@surkov> sure, thank you for doing this
- # [05:18] <@surkov> sounds scary :)
- # [05:18] * Jamie grins
- # [05:18] <Jamie> I'm catching up on Mozilla work
- # [05:18] <@surkov> great!
- # [05:24] * Quits: cbrodigan (christineb@moz-FAB212FE.hsd1.ca.comcast.net) (Quit: cbrodigan)
- # [06:41] <Jamie> surkov: hmm. seems i was wrong about the wrapping hwnd. firefox 4 has it too. it just... isn't a problem there.
- # [06:41] <@surkov> hm, ok
- # [06:41] <Jamie> surkov: I seem to recall that a while ago, there was a change to WM_GETOBJECT to not return the top level client accessible for certain HWNDs
- # [06:41] <Jamie> any idea what bug that was for?
- # [06:42] * Jamie remembers the change, but not why
- # [06:42] <@surkov> hm, I don't recall any changes here, at least they weren't done by a11y guys
- # [06:42] <Jamie> hmm
- # [06:43] <Jamie> I seem to remember me being the one that identified that it needed to be done
- # [06:43] <Jamie> I just can't remember why I said that
- # [06:43] * Jamie grumbles
- # [06:58] * Jamie wades through bugzilla
- # [07:11] <@surkov> Jamie: how do you rely on embeds relation on top level document accessible?
- # [07:12] <@surkov> Jamie, I fixed bug 707654 but the patch was backed out because it regressed (bug 720393). The interesting thing running NVDA never hits breakpoint for that relation for me
- # [07:12] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=707654 nor, --, ---, surkov.alexander, REOP, embeds relation on root accessible can return not content document
- # [07:13] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=720393 maj, --, ---, nobody, VERI FIXED, NVDA's virtual buffer often breaks with iFrames, started January 23, 2012 build.
- # [08:14] <Jamie> surkov: we use it for plugins to get back to the document
- # [08:15] <Jamie> see bug 646360 for details
- # [08:15] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=646360 nor, --, ---, nobody, NEW, Plugins should be beneath the document in the accessibility hierarchy
- # [08:20] <Jamie> surkov: we've noticed NAVRELATION_EMBEDS seems ot be a bit unreliable in firefox 9 (as compared with firefox 4), but not sure whether we just didn't notice problems before
- # [08:37] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [08:40] <Jamie> urg. is there anywhere i cna find a .zip build of firefox 10?
- # [09:03] * Quits: ehsan (ehsan@moz-A6FE435.build.sjc1.mozilla.com) (Input/output error)
- # [09:05] * Quits: Jamie (jamie@moz-CA26021.jantrid.net) (Quit: leaving)
- # [09:07] * Quits: @hub (hub@moz-E2FCA694.figuiere.net) (Ping timeout)
- # [09:30] * Joins: victorporof (victorporo@C092FEB2.1C233438.79933D60.IP)
- # [09:42] * Joins: sgimeno (santiago.g@A922400D.48362F18.BCC03455.IP)
- # [10:06] * Joins: MarcoZ (marco.zehe@moz-1B071F06.dip.t-dialin.net)
- # [10:06] * ChanServ sets mode: +o MarcoZ
- # [10:07] <@MarcoZ> Good day all!
- # [10:37] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [11:08] * Quits: @surkov (surkov@DD563F5.F78D7EEB.34044A7F.IP) (Quit: surkov)
- # [11:18] <@firebot> New Core - Disability Access APIs bug 722248 filed by jamie@nvaccess.org.
- # [11:18] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=722248 nor, --, ---, nobody, NEW, Some chrome accessibles report incorrect visibility states
- # [11:26] * Joins: silvia (Adium@moz-7033BBBB.dyn.iinet.net.au)
- # [11:59] * khuey|away is now known as khuey
- # [12:03] * Joins: askalski (akuda@moz-828F8EEB.neoplus.adsl.tpnet.pl)
- # [12:03] * ChanServ sets mode: +o askalski
- # [12:04] <@askalski> morning everyone!
- # [12:05] <@tbsaunde> morning askalski
- # [12:06] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [12:07] <@MarcoZ> Good morning askalski!
- # [12:07] <@MarcoZ> tbsaunde: Any thoughts on my latest comment in bug 591363?
- # [12:07] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=591363 nor, --, mozilla12, bolterbugz, RESO FIXED, (in)visible state is not always correct?
- # [12:10] <@tbsaunde> MarcoZ: the we should it out one?
- # [12:10] <@tbsaunde> MarcoZ: I think we may well want to back it out, but I don't really see any hurry
- # [12:11] <@tbsaunde> I can't see any trouble getting approval for a small backout
- # [12:11] <@tbsaunde> to a well known state
- # [12:11] <@tbsaunde> and who knows how big the fix to Jamies issues will be
- # [12:12] <@tbsaunde> so I'd probably sit tight till we know that
- # [12:12] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [12:16] <@tbsaunde> MarcoZ: that said I don't feel particularly strongly about it
- # [12:19] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Quit: ChatZilla 0.9.88 [Firefox 11.0a2/20120126042008])
- # [12:20] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [12:25] <@MarcoZ> tbsaunde: Thanks, I'll talk to davidb about it when he comes in. But I feel, with the other crash fix that this seems to have introduced, and the lots of comments that followed even after the landing, that this is not finished yet, so I personally feel more comfortable backing this out either now or after the merge from the Aurora channel so it doesn't get carried forward just yet.
- # [12:25] <@MarcoZ> s/crash fix/crash bug/
- # [13:20] <@askalski> MarcoZ, I am doing tests with "python /home/akuda/Moz/obj-ff-dbg/_tests/testing/mochitest/runtests.py --a11y --close-when-done --log-file="/home/Moz/mochitest1.log" --autorun --console-level=ERROR"
- # [13:21] <@askalski> and it hangs
- # [13:21] <@askalski> (firefox open, zero tests, nothing happens)
- # [13:21] <@askalski> (I mean it hangs before tests starts)
- # [13:21] <@askalski> while doing make mochitests-a11y processes
- # [13:21] <@askalski> any clues?
- # [13:44] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [13:45] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [13:46] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Quit: ChatZilla 0.9.88 [Firefox 11.0a2/20120126042008])
- # [13:47] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [13:53] <@MarcoZ> askalski: No idea. It runs OK for me... Hmm....
- # [13:54] <@askalski> MarcoZ, ok, I'll just stike to "make"
- # [13:54] <@askalski> now I have another issue, for a clear repo no tests are found. I removed entire obj dir and rebuild now
- # [13:54] <@askalski> I think my machine is test-cursed
- # [13:54] <@MarcoZ> askalski: Did you put ac_add_option --enable-tests in your .mozconfig? I think tests are not built by default.
- # [13:55] <@askalski> nope
- # [13:55] <@askalski> in either build
- # [13:55] <@askalski> this might be the case...
- # [13:57] <@askalski> yay, rebuilding again!
- # [13:57] <@askalski> feels like gentoo
- # [13:58] <@askalski> I think I start to do 50 pushups each time I rebuild. I'll be in olympic form in no-time!
- # [13:58] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [14:00] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [14:00] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [14:00] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [14:05] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Connection reset by peer)
- # [14:06] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [14:07] <@tbsaunde> askalski: --test-level= DEBUG or INFO may be more useful
- # [14:07] * @tbsaunde has always had tests without explicitly enabling them
- # [14:08] <@tbsaunde> but I don't really know wht to tell you I don't have a scientific method for making them work when I need to run them locally I hit them with a hammer until they do what I need
- # [14:09] <@askalski> tbsaunde, so far I had tests by default too, today it was the first time it changed. can I pass this --test-level option to make instead of python?
- # [14:09] <@askalski> I mean, in the bug I fight description is "make sure mochitest passes" it's like "and make sure to fix European economy"
- # [14:11] <@tbsaunde> askalski: maybe, but I'm not sure how
- # [14:12] <@tbsaunde> askalski: I believe make runs the python script with the highest log level so I'm not sure there's much else you can turn on other than logging we do in the individual tests
- # [14:13] <@tbsaunde> askalski: it might be easier to just run the patch through try to find out which tests are broken
- # [14:13] <@tbsaunde> did we ever give you access btw
- # [14:13] <@askalski> tbsaunde, through try?
- # [14:14] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Quit: ChatZilla 0.9.88 [Firefox 11.0a2/20120126042008])
- # [14:14] <@askalski> tbsaunde, nope, I still work as a outside contributor everywhere else than bugzilla
- # [14:15] <@askalski> at bugzilla I received some privileges
- # [14:16] <@tbsaunde> askalski: wiki.mozilla.org/ReleaseEngineering/TryServer
- # [14:17] <@tbsaunde> askalski: nad please follow wiki.mozilla.org/hacking/committer/
- # [14:18] <@tbsaunde> please tell me what the bug number is when you file :)
- # [14:19] <@askalski> tbsaunde, ok, how to receive an account?
- # [14:19] <@askalski> O, I check my mail if I have one
- # [14:20] <@askalski> tbsaunde, no such page (the second address)
- # [14:22] <@tbsaunde> askalski: oh, weird its www.mozilla.org/hacking/committer/
- # [14:25] <@askalski> tbsaunde, ok
- # [14:28] <@tbsaunde> askalski: best if you can deal with the bug in the next few minutes I need to head out soon
- # [14:29] <@askalski> tbsaunde, which bug?
- # [14:29] <@tbsaunde> askalski: filing the bug in the instructions
- # [14:30] <@askalski> tbsaunde, I still don't understand
- # [14:30] <@askalski> ah, repository access bug
- # [14:31] <@askalski> tbsaunde, ok, I'm still not familiar with the policy that everything (including hardware request) is a bug in mozilla :)
- # [14:31] * Quits: silvia (Adium@moz-7033BBBB.dyn.iinet.net.au) (Quit: Leaving.)
- # [14:33] <@tbsaunde> askalski: heh
- # [14:33] <@askalski> tbsaunde, level 1 is OK?
- # [14:34] <@tbsaunde> askalski: yes
- # [14:35] <@askalski> tbsaunde, ok, I need to generate key. are there some limits in USA? I heard that Patriot Act made some key-lengths illegal
- # [14:36] <@tbsaunde> askalski: I'm not aware of any
- # [14:36] * @tbsaunde usually uses 2048 or 4096 bit rsa
- # [14:36] <@tbsaunde> which afaik is the default now, and I live in the US so shrug
- # [14:36] <@tbsaunde> I can't imagne you reall need more than 4096
- # [14:37] <@tbsaunde> askalski: but you can generatee it after filing the bug :)
- # [14:38] <@askalski> tbsaunde, I got the private, now I need to find out how to extract a public from it :D
- # [14:39] <@tbsaunde> askalski: ssh-keygen creates them at the same time
- # [14:39] <@askalski> I used something different
- # [14:39] <@askalski> but OK
- # [14:39] <@tbsaunde> and names the public key .pub
- # [14:39] <@tbsaunde> what?
- # [14:43] * @tbsaunde has to run
- # [14:43] * tbsaunde is now known as tbsaunde|afk
- # [14:45] <@askalski> tbsaunde|afk, https://bugzilla.mozilla.org/show_bug.cgi?id=722292
- # [14:45] <@firebot> Bug 722292 nor, --, ---, mozillamarcia.knous, NEW, Commit Access (Level 1) for Andrzej Skalski
- # [14:57] <@firebot> marco.zehe@googlemail.com changed the Component on bug 369710 from Widget: Cocoa to Disability Access APIs.
- # [14:57] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=369710 enh, --, mozilla12, hub, RESO FIXED, No talking alerts
- # [15:06] <@firebot> pallavikumarijha@gmail.com requested review from bolterbugz@gmail.co m for attachment 592681 on bug 589784.
- # [15:06] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=589784 nor, --, ---, markcapella, ASSI, Remove role="label" aria processing.
- # [15:10] * Joins: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP)
- # [15:10] * ChanServ sets mode: +qo davidb davidb
- # [15:12] <@davidb> MarcoZ: saw your comment about backing out. I don't mind that happening at all. What are the issues?
- # [15:13] <@askalski> hi David!
- # [15:15] <@MarcoZ> davidb: Hi! The main issues are some issues Jamie brings up in bug 722248.
- # [15:15] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=722248 nor, --, ---, nobody, NEW, Some chrome accessibles report incorrect visibility states
- # [15:16] <@MarcoZ> davidb: But I think there was one crash bug, but I'm not sure what the status of that is. This may be fixed.
- # [15:16] <@MarcoZ> davidb: And what also worries me is the string of comments that followed the last patch and which seem to point at various problems within the patch that landed.
- # [15:16] <@askalski> davidb, I requested for repo access, https://bugzilla.mozilla.org/show_bug.cgi?id=722292
- # [15:16] <@firebot> Bug 722292 nor, --, ---, mozillamarcia.knous, NEW, Commit Access (Level 1) for Andrzej Skalski
- # [15:16] <@davidb> askalski: great
- # [15:17] <@davidb> MarcoZ: Timothy is working to remove views from gecko. I'm not sure why he is using that bug to have the conversation about finding alternate API.
- # [15:17] <@askalski> davidb, still having problems with tests, I did some checkouts and cleaning, rebuild (twice, a stupid mistake) and now I am redoing tests
- # [15:18] <@davidb> ok
- # [15:18] <@MarcoZ> davidb: Oh! And all the time I thought he was proposing that the patch that landed be changed/fixed to use a different API.
- # [15:18] <@davidb> MarcoZ: can you do a/b testing on bug 722248, before and after the invi fix?
- # [15:19] <@MarcoZ> davidb: Can do, but I'm pretty sure it's that bug.
- # [15:19] <@davidb> MarcoZ: well in a sense Timothy is suggesting that, but I'm not sure he knows we already used the views here.
- # [15:19] * Quits: Hixie (ianh@moz-E9FB2A6A.no) (Ping timeout)
- # [15:19] <@davidb> MarcoZ: probably is, but would be good to know.
- # [15:19] <@MarcoZ> Yup will do that in a bit, when I've finished my blog post about the progress on Mac.
- # [15:22] <@davidb> thanks
- # [15:22] * @davidb compiles a contributor patch
- # [15:22] * Joins: Hixie (ianh@moz-E9FB2A6A.no)
- # [15:28] * Joins: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
- # [15:34] <@MarcoZ> davidb: Nice!
- # [15:35] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [15:36] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [15:42] <@davidb> hi jhk
- # [15:42] <@firebot> bolterbugz@gmail.com cancelled review?(bolterbugz@gmail.co m) for attachment 592681 on bug 589784.
- # [15:42] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=589784 nor, --, ---, markcapella, ASSI, Remove role="label" aria processing.
- # [15:43] <jhk> hi davidb
- # [15:43] <@davidb> jhk: what are you working on these days?
- # [15:43] <jhk> https://bugzilla.mozilla.org/show_bug.cgi?id=700538
- # [15:43] <@firebot> Bug 700538 nor, --, ---, nobody, NEW, nsHTMLEditor should override IsRootNode
- # [15:43] <@davidb> oho!
- # [15:43] <@davidb> good good
- # [15:44] <jhk> :)
- # [15:44] <@davidb> jhk: in this channel you can just type bug 700538, and firebot will auto link
- # [15:44] <@davidb> e.g. bug 123456
- # [15:44] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=123456 is not accessible
- # [15:44] <@davidb> bah
- # [15:44] <@davidb> bug 654321
- # [15:44] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=654321 is not accessible
- # [15:44] <@davidb> heheh
- # [15:44] <@davidb> bug 432345
- # [15:45] <jhk> firebot works:)
- # [15:45] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=432345 nor, --, 3.0, steven, VERI FIXED, fix the Thunderbird/Releases page to match the new Mozilla.com look & feel
- # [15:45] <@davidb> :)
- # [15:45] <@firebot> jhk: Sorry, I've no idea what 'works:)' might be.
- # [15:49] * Joins: jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP)
- # [15:49] * ChanServ sets mode: +o jprmc
- # [15:50] <@MarcoZ> LOL
- # [15:50] <@MarcoZ> davidb: Info added. There were a few more a11y changesets that day, but bug 591363 is still the most likely. And all of the problems Jamie describes weren't present in the Jan 10 build that still worked, and all of them appear in the January 11 build.
- # [15:50] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=591363 nor, --, mozilla12, bolterbugz, RESO FIXED, (in)visible state is not always correct?
- # [15:51] <@MarcoZ> davidb: Unless if you find out what the issues are and can fix them, I'd be also willing to keep it and request approval on aurora for the fix once we have it.
- # [15:52] <@MarcoZ> s/unless//
- # [15:52] <@davidb> MarcoZ: new, I'll comment that I agree to back out, on the bug.
- # [15:52] <@davidb> new/naw
- # [15:52] * @askalski going for lunch
- # [15:52] <@davidb> askalski: bon appetite
- # [15:53] <@davidb> MarcoZ: there was a follow up crash fix as well I think
- # [15:54] <@MarcoZ> davidb: Yes, just found it, bug 717505.
- # [15:54] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=717505 cri, --, mozilla12, bolterbugz, RESO FIXED, Crash [@ nsAccessible::VisibilityState] when closing a tab, depending on when accessibility was enab
- # [15:55] <@davidb> ok
- # [15:55] <@davidb> MarcoZ: thanks for keeping on top of this.
- # [15:55] <@MarcoZ> davidb: YW!
- # [16:00] * Quits: @askalski (akuda@moz-828F8EEB.neoplus.adsl.tpnet.pl) (Ping timeout)
- # [16:01] * Joins: askalski (akuda@moz-A5373222.neoplus.adsl.tpnet.pl)
- # [16:01] * ChanServ sets mode: +o askalski
- # [16:01] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [16:01] <@askalski> davidb, can we talk today on skype? I have some serious issues with test, I think I have exceeded all options that comes to my mind, I need to know what to do
- # [16:02] <@askalski> davidb, I'll be back in 20 mins
- # [16:02] <@davidb> yep i have a meeting in 65 minutes
- # [16:02] <@davidb> so before that
- # [16:02] <@davidb> askalski: ^
- # [16:05] * clown is now known as clown_mtg
- # [16:06] * Quits: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com) (Ping timeout)
- # [16:10] * Joins: ehsan (ehsan@F0B20A8D.8458880F.57F33CED.IP)
- # [16:13] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Quit: ChatZilla 0.9.88 [Firefox 12.0a1/20120130061320])
- # [16:13] * @davidb rights a wrong
- # [16:19] <@MarcoZ> Bah, merge issues while trying to back out! :( And my resolver doesn't work.
- # [16:22] <@davidb> ycuk
- # [16:22] <@davidb> yuck
- # [16:23] <@askalski> ok, let me go somewhere noone interupts me and I call you in 5, ok?
- # [16:25] <@askalski> davidb, ok, anytime you're ready
- # [16:25] * @davidb looks for a room
- # [16:29] * Quits: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp) (Quit: mike5w3c)
- # [16:30] <@davidb> MarcoZ: was Trev active earlier?
- # [16:31] <@MarcoZ> davidb: He was, discussing some stuff with askalski.
- # [16:35] * Joins: hub (hub@moz-E2FCA694.figuiere.net)
- # [16:35] * ChanServ sets mode: +o hub
- # [16:36] * Joins: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp)
- # [16:43] <@davidb> ok so I'm guessing sleep will override 1:1
- # [16:43] <@davidb> which is fine
- # [16:44] <@firebot> markcapella@twcny.rr.com requested review from bolterbugz@gmail.co m for attachment 592711 on bug 589784.
- # [16:44] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=589784 nor, --, ---, markcapella, ASSI, Remove role="label" aria processing.
- # [16:45] * tbsaunde|afk is now known as tbsaunde
- # [16:45] * Quits: ehsan (ehsan@F0B20A8D.8458880F.57F33CED.IP) (Input/output error)
- # [16:51] <@davidb> aha tbsaunde is back
- # [16:52] <@tbsaunde> davidb: yaa, was off at class
- # [16:52] <@davidb> ah
- # [16:53] * Joins: shorlander-away (shorlander@moz-853043D6.dhcp.insightbb.com)
- # [16:58] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
- # [16:59] <@MarcoZ> davidb: Yeah that backout is difficult somehow. Seems something else made a change afterwards to that code that we're not aware of, and which we may not want to back out in principal.
- # [16:59] <@davidb> hmm
- # [17:01] <@davidb> MarcoZ: I think it was Trev's roles:: cleanup patch
- # [17:01] <@MarcoZ> davidb: Ouch :(
- # [17:01] <@MarcoZ> davidb: Like I said, I'll review the rollback patch.
- # [17:01] <@MarcoZ> s/rollback/backout
- # [17:01] * @MarcoZ needs to completely re-clone his repo, busted it.
- # [17:02] <@davidb> ok
- # [17:03] * shorlander-away is now known as shorlander
- # [17:06] * Joins: richardschwerdtfeger (richardsch@moz-130F1494.lightspeed.austtx.sbcglobal.net)
- # [17:08] * Quits: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com) (Ping timeout)
- # [17:13] <@davidb> MarcoZ: how long are you around?
- # [17:14] <@davidb> how much longer
- # [17:14] <@davidb> today
- # [17:14] * Joins: ehsan (ehsan@moz-A6FE435.build.sjc1.mozilla.com)
- # [17:18] <@MarcoZ> davidb: GF is out of the house today, so can be until about 2:30 or 2:45 PM your time.
- # [17:18] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [17:18] <@davidb> MarcoZ: I'm tempted to try and fix bug 722248 today
- # [17:19] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=722248 nor, --, ---, nobody, NEW, Some chrome accessibles report incorrect visibility states
- # [17:19] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [17:20] <@tbsaunde> davidb: does it look that simple?
- # [17:21] <@davidb> tbsaunde: possibly.
- # [17:21] <@tbsaunde> (I wouldn't really object to it getting fixed soon and getting approval for aurora)
- # [17:21] <@MarcoZ> davidb: Fine with me, too!
- # [17:25] <@askalski> I am finishing for today
- # [17:25] <@askalski> bye
- # [17:26] * Quits: @askalski (akuda@moz-A5373222.neoplus.adsl.tpnet.pl) (Quit: Wychodzi)
- # [17:27] * @davidb tries something
- # [17:31] <@davidb> didn't work
- # [17:47] <@davidb> oh!
- # [17:47] <@davidb> MarcoZ: I was wrong. I introduce use of the view here.
- # [17:48] <@MarcoZ> davidb: Ah!
- # [17:48] <@MarcoZ> davidb: So?
- # [17:48] <@davidb> so I'm tinkering. removing it. adding back bits of previous
- # [17:51] <@davidb> ok heading for a quick lunch. i'll work up a fix patch after fuel.
- # [18:00] * clown_mtg is now known as clown
- # [18:00] * Quits: clown (clown@67828CC7.C1A51174.9D42CF23.IP) (Quit: Leaving.)
- # [18:00] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
- # [18:02] * Joins: clown (clown@6EDD3027.A6295926.9D42CF23.IP)
- # [18:08] * @davidb brings back
- # [18:08] * Quits: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com) (Ping timeout)
- # [18:10] * Quits: @MarcoZ (marco.zehe@moz-1B071F06.dip.t-dialin.net) (Quit: Reboot. brb!)
- # [18:12] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
- # [18:16] * Joins: MarcoZ (marco.zehe@moz-1B071F06.dip.t-dialin.net)
- # [18:16] * ChanServ sets mode: +o MarcoZ
- # [18:18] <@davidb> ok i think i got it
- # [18:18] * @davidb tidies and runs tests
- # [18:20] <@MarcoZ> davidb: Curious to see whatit'll be!
- # [18:20] <@davidb> the view was lying
- # [18:21] <@davidb> or more likely, i don't understand the expectations of the view methods
- # [18:23] <@davidb> woot! tests pass
- # [18:24] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
- # [18:25] <@davidb> this probably over-corrects though. *thinking*
- # [18:33] <@firebot> markcapella@twcny.rr.com requested review from bolterbugz@gmail.co m for attachment 592751 on bug 589784.
- # [18:33] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=589784 nor, --, ---, markcapella, ASSI, Remove role="label" aria processing.
- # [18:55] * Joins: askalski (akuda@moz-1276DE7A.neoplus.adsl.tpnet.pl)
- # [18:55] * ChanServ sets mode: +o askalski
- # [18:55] <@askalski> davidb, I got the script! it's working and useful!
- # [18:56] <@askalski> davidb, where to post it?
- # [18:56] <@tbsaunde> askalski: interesting what script did davidb know about?
- # [18:56] <@davidb> tbsaunde: github is fine
- # [18:57] <@davidb> oops
- # [18:57] <@davidb> askalski: ^
- # [18:57] <@askalski> tbsaunde, I created a smart diff to mochitest results. it ignores addresses and provides a diff-style output what changed
- # [18:58] <@tbsaunde> oic
- # [18:58] <@askalski> one command to find them all!
- # [19:00] <@davidb> :)
- # [19:05] <@firebot> trev.saunders@gmail.com granted review for attachment 592150 on bug 721772.
- # [19:05] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=721772 nor, --, ---, nobody, NEW, Only accumulate ISimpleDOM usage bool for Telemetry once.
- # [19:07] * Joins: cbrodigan (christineb@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [19:09] <@davidb> MarcoZ: patch coming in about 10 mins
- # [19:09] <@davidb> ditto Trev
- # [19:10] <@tbsaunde> davidb: ack
- # [19:12] <@askalski> https://github.com/askalski/moz-related
- # [19:12] <@davidb> MarcoZ: if you build a patch locally can you tell relatively if there is perf impact compared to trunk?
- # [19:12] <@MarcoZ> davidb: Not sure.
- # [19:12] <@MarcoZ> davidb: I can try.
- # [19:12] <@davidb> ok
- # [19:13] <@davidb> MarcoZ: when does uplift happen?
- # [19:13] <@MarcoZ> davidb: Tomorrow morning PST, probably between 6 and 8 some time.
- # [19:13] <@davidb> ok
- # [19:17] <@askalski> tbsaunde, posted link, check it out
- # [19:19] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [19:20] <@firebot> bolterbugz@gmail.com requested review from marco.zehe@googlema il.com for attachment 592765 on bug 722248.
- # [19:20] <@firebot> bolterbugz@gmail.com requested feedback from surkov.alexander@ gmail.com for attachment 592765 on bug 722248.
- # [19:20] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=722248 nor, --, ---, bolterbugz, NEW, Some chrome accessibles report incorrect visibility states
- # [19:20] <@tbsaunde> davidb: MarcoZ if I want a js function to take an optional arg that can be true or false how do you do this?
- # [19:20] <@firebot> bolterbugz@gmail.com requested review from trev.saunders@gmail .com for attachment 592765 on bug 722248.
- # [19:20] * Quits: @hub (hub@moz-E2FCA694.figuiere.net) (Input/output error)
- # [19:21] <@davidb> tbsaunde: it is the wild west
- # [19:22] <@tbsaunde> davidb: ?
- # [19:22] <@davidb> suppose you have function foo(a)
- # [19:23] <@davidb> you can call foo("dog","cat");
- # [19:23] <@davidb> var optionalArg = arguments[1];
- # [19:23] <@davidb> inside foo def
- # [19:23] <@MarcoZ> tbsaunde: Or the other way around, define function foo(a, b, c), but only call foo(x, y);.
- # [19:23] <@davidb> yeah
- # [19:24] <@davidb> wild west
- # [19:24] <@firebot> markcapella@twcny.rr.com requested review from bolterbugz@gmail.co m for attachment 592768 on bug 589784.
- # [19:24] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=589784 nor, --, ---, markcapella, ASSI, Remove role="label" aria processing.
- # [19:25] * Parts: cbrodigan (christineb@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [19:27] <@MarcoZ> davidb: Build is kicked off, but may take some time due to the merge/backout mess I had to deal with earlier.
- # [19:27] <@davidb> np
- # [19:27] <@tbsaunde> davidb: sure, but I do I know if that argument is undefined
- # [19:29] <@davidb> tbsaunde: can you just check it?
- # [19:29] <@tbsaunde> davidb: ?
- # [19:30] <@davidb> well, inside foo(a,b) what doe b equal if we call foo("hi");?
- # [19:32] <@tbsaunde> davidb: uh, I'm not sure isn't it undefined or something?
- # [19:32] <@davidb> tbsaunde, MarcoZ I should probably tweak my patch to get and check the shell just before used.
- # [19:32] <@davidb> tbsaunde: yeah i think so
- # [19:33] <@davidb> undefined is a keyword in js IIRC
- # [19:33] <@tbsaunde> davidb: what about adding tests?
- # [19:34] <@davidb> tbsaunde: i can't do that before uplift i don't think
- # [19:35] <@MarcoZ> tbsaunde: Besides, the tests introduced in bug 591363 still pass.
- # [19:35] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=591363 nor, --, mozilla12, bolterbugz, RESO FIXED, (in)visible state is not always correct?
- # [19:35] <@davidb> i can try but i don't want this partial revert to hold on more tests
- # [19:35] <@askalski> ok, now I go for real. bye
- # [19:35] <@MarcoZ> davidb: You do check the shell and return if it is not defined.
- # [19:35] * Quits: @askalski (akuda@moz-1276DE7A.neoplus.adsl.tpnet.pl) (Quit: Wychodzi)
- # [19:35] <@davidb> MarcoZ: yeah, but I've moved it locally to be after the frame check
- # [19:36] <@MarcoZ> davidb: Oh!
- # [19:36] <@davidb> i don't use the shell until after the frame ancestor call
- # [19:36] <@MarcoZ> I see!
- # [19:36] <@davidb> build and running tests now
- # [19:36] <@davidb> *ing
- # [19:37] <@tbsaunde> davidb: ugh :(
- # [19:37] <@davidb> tbsaunde: ugh to what specificall?
- # [19:37] <@davidb> y
- # [19:39] * Joins: hub (hub@21B7B9F2.B87E9213.6E712CE2.IP)
- # [19:39] * ChanServ sets mode: +o hub
- # [19:40] <@tbsaunde> davidb: not adding tests
- # [19:40] <@davidb> fair
- # [19:41] <@davidb> tbsaunde: the tests will come.
- # [19:42] <@davidb> ok confirmed tests still pass with my local change.
- # [19:42] <@davidb> i'll upload the new patch
- # [19:43] <@firebot> bolterbugz@gmail.com cancelled review?(trev.saunders@gmail .com) for attachment 592765 on bug 722248.
- # [19:43] <@firebot> bolterbugz@gmail.com cancelled feedback?(surkov.alexander@ gmail.com) for attachment 592765 on bug 722248.
- # [19:43] <@firebot> bolterbugz@gmail.com requested review from trev.saunders@gmail .com for attachment 592774 on bug 722248.
- # [19:43] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=722248 nor, --, ---, bolterbugz, NEW, Some chrome accessibles report incorrect visibility states
- # [19:44] <@davidb> sigh
- # [19:44] <@davidb> forgot to qrefresh
- # [19:45] <@firebot> bolterbugz@gmail.com cancelled review?(trev.saunders@gmail .com) for attachment 592774 on bug 722248.
- # [19:45] <@firebot> bolterbugz@gmail.com requested review from marco.zehe@googlema il.com for attachment 592776 on bug 722248.
- # [19:45] <@davidb> tbsaunde: do you still want to review this one without patches?
- # [19:45] <@davidb> errr tests
- # [19:45] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [19:46] <@davidb> MarcoZ: the one thing I didn't test manually was context menus. I'm curious to know how that goes.
- # [19:49] <@tbsaunde> davidb: I'm not really sure
- # [19:50] <@MarcoZ> davidb: You can test that now if you want. Start NVDA, and on the nightly start page, tab to any of the links. Press the APPLICATIONS key. If NVDA says something about "Menu collapsed", the bug is still there. if it doesn't, you're good.
- # [19:50] <@MarcoZ> davidb: My build will still take a bit.
- # [19:50] <@davidb> MarcoZ: just tested. looks good.
- # [19:52] <@davidb> oh
- # [19:52] <@davidb> didn't see your comment about nvda… i'll try that as well (was using accprobe event watcher)
- # [19:54] <@davidb> MarcoZ: it just says "menu"
- # [19:55] <@MarcoZ> davidb: Nice!
- # [19:55] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [19:57] <@firebot> bolterbugz@gmail.com requested in-testsuite from the wind on bug 722248.
- # [19:57] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=722248 nor, --, ---, bolterbugz, NEW, Some chrome accessibles report incorrect visibility states
- # [19:57] <@tbsaunde> davidb: the only thing in that patch that I think concerns me is it appears you return offscreen if IsVisibleConsideringAncestors() returns false
- # [19:59] <@davidb> true
- # [19:59] * Quits: ehsan (ehsan@moz-A6FE435.build.sjc1.mozilla.com) (Ping timeout)
- # [20:00] <@tbsaunde> davidb: is that what you want to do? it seems that would make backround tabs invisible
- # [20:00] <@tbsaunde> which I thought nvda didn't like
- # [20:00] <@davidb> it makes the panels invisible but not the tabs
- # [20:01] <@tbsaunde> I'm not sure I understand that
- # [20:01] <@davidb> the tabs list is a sibling to the tab panels group
- # [20:01] <@davidb> tbsaunde: it would be safer to move the ancestor call to the end like we used to do.
- # [20:02] * Quits: mike5w3c (MikeS@moz-DAFE1A45.tokyo.ocn.ne.jp) (Quit: mike5w3c)
- # [20:03] <@tbsaunde> davidb: is there a reason not to do that?
- # [20:03] <@davidb> tbsaunde: I've now done it locally.
- # [20:03] <@davidb> tbsaunde: now the question is… do we want to check style visibility before all the more expensive rect stuff
- # [20:04] <@davidb> style visibility also happens to be checked in the frame ancestor call
- # [20:05] <@tbsaunde> I don't think I with it enough yet to be able to say
- # [20:05] * Quits: @hub (hub@21B7B9F2.B87E9213.6E712CE2.IP) (Ping timeout)
- # [20:06] <@tbsaunde> style not completely btter after cold :/
- # [20:06] <@davidb> ok
- # [20:07] <@tbsaunde> and I can't type either bleh
- # [20:07] * Joins: hub (hub@21B7B9F2.B87E9213.6E712CE2.IP)
- # [20:07] * ChanServ sets mode: +o hub
- # [20:09] * Quits: clown (clown@6EDD3027.A6295926.9D42CF23.IP) (Quit: Leaving.)
- # [20:09] <@tbsaunde> is there any reasonable way to debug tests?
- # [20:10] <@davidb> there isn't a great story for that
- # [20:10] <@davidb> trace is probably your best bet
- # [20:10] * Joins: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
- # [20:11] <@firebot> eitan@monotonous.org requested review from surkov.alexander@gm ail.com for attachment 592783 on bug 698823.
- # [20:11] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=698823 nor, --, mozilla10, eitan, NEW, Introduce virtual cursor/soft focus functionality to a11y API
- # [20:11] <eeejay> bbiab
- # [20:12] <@davidb> tbsaunde: don't worry too much about NVDA for this particular bug, the main thing is to not regress them.
- # [20:13] <@davidb> with respect to the NVDA concern with background tabs, that's complicated.
- # [20:16] <@tbsaunde> davidb: ok
- # [20:16] <@davidb> I'm not sure how to honour this: "Change Gecko to stop setting the invisible state on nodes in background tabs"
- # [20:18] <@MarcoZ> tbsaunde: I agree with davidb, the most important thing is that the new functionality remains intact and doesn't regress the newly introduced tests. And that is the case with this patch, so from the tests performed manually I'd say we're good.
- # [20:20] * Quits: victorporof (victorporo@C092FEB2.1C233438.79933D60.IP) (Connection reset by peer)
- # [20:20] * Joins: victorporo (victorporo@C092FEB2.1C233438.79933D60.IP)
- # [20:20] <@tbsaunde> davidb: so I guess I have no concrete objections, but I'm not sure I want to r+ something that complicated right now and I'm not really thrilled about taking patches without tests
- # [20:20] <@davidb> tbsaunde: that's fair.
- # [20:20] <@tbsaunde> sorry I'm not really useful right now
- # [20:20] <@davidb> np
- # [20:22] <@davidb> ok everything checks out okay with the ancestor call after the offscreen stuff
- # [20:23] <@tbsaunde> davidb: I wont object if you land it with Marco's r+ or ask surkov
- # [20:23] <@davidb> ok
- # [20:24] <@davidb> this feels rushed, but it is a lot like a partial revert
- # [20:24] <@davidb> visibility state is a can of worms though
- # [20:25] <@davidb> tbsaunde: want me to cancel your review?
- # [20:26] <@MarcoZ> davidb: Yes that's really tricky. And if we find something, we can still ask approval up the chain if needed.
- # [20:26] <@davidb> yes we have time
- # [20:26] <@davidb> i want to see if talos picks up any regression too, so this should be interesting
- # [20:27] <@davidb> MarcoZ: I'm going to attach the final? patch
- # [20:29] <@MarcoZ> davidb: Very good! And please, throw it against try-server and see what Talos results you get!
- # [20:29] <@tbsaunde> davidb: yea
- # [20:30] <@davidb> ok
- # [20:32] <@MarcoZ> davidb: Hub: Quite some retweets of my new Mac progress report, and a good number of reads on the blog itself, too!
- # [20:32] <@hub> cool
- # [20:33] <@davidb> woot
- # [20:33] <@firebot> bolterbugz@gmail.com cancelled review?(marco.zehe@googlema il.com) for attachment 592776 on bug 722248.
- # [20:33] <@firebot> bolterbugz@gmail.com requested review from marco.zehe@googlema il.com for attachment 592790 on bug 722248.
- # [20:33] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=722248 nor, --, ---, bolterbugz, NEW, Some chrome accessibles report incorrect visibility states
- # [20:34] <@tbsaunde> so, one more test question do we have tests where we change the text in a focused node?
- # [20:34] <@davidb> I don't know off hand.
- # [20:35] <@firebot> acelists@atlas.sk changed the Status on bug 537569 from NEW to RESOLVED.
- # [20:35] <@firebot> acelists@atlas.sk set the Resolution field on bug 537569 to FIXED.
- # [20:35] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=537569 min, --, ---, nobody, RESO FIXED, File menu and Filter field in about:config have same accesskey
- # [20:36] <@MarcoZ> tbsaunde: I don't know off-hand, either.
- # [20:36] <@MarcoZ> Is there a particular reason for your question?
- # [20:38] <@tbsaunde> MarcoZ: so, I'm trying to add a workaround for bug 686909 in which we claim the event is from user input if the target is focused and editable
- # [20:38] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=686909 nor, --, ---, nobody, NEW, The system suffix is for system generated events only
- # [20:38] <@tbsaunde> surkov wants tests
- # [20:39] <@MarcoZ> tbsaunde: So what's the test case? A focused textfield whose text is being changed?
- # [20:39] <@tbsaunde> MarcoZ: yes
- # [20:42] <@MarcoZ> tbsaunde: I would not think there's a difference to a non-focused textbox. Or is there?
- # [20:43] <@tbsaunde> MarcoZ: if and only if its focused event.isFromUserInput should be true
- # [20:43] <@tbsaunde> MarcoZ: so there will be a difference that should be tested
- # [20:44] <@MarcoZ> tbsaunde: OK!
- # [20:44] <@tbsaunde> or I gues I could just check non focused / editable things and make sure its not true
- # [20:44] <@tbsaunde> the coverage isn't as good but easy to do and might make Alex happy
- # [20:44] <@MarcoZ> tbsaunde: And add a test where you sppecifically focus the item first before making text changes.
- # [20:45] <@tbsaunde> MarcoZ: yeah, I guess I could do that but ugh work
- # [20:49] <@firebot> bolterbugz@gmail.com granted review for attachment 592768 on bug 589784.
- # [20:49] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=589784 nor, --, ---, markcapella, ASSI, Remove role="label" aria processing.
- # [20:53] <@MarcoZ> OK, build is done, rebuilding with latest pach to see if there is a difference. Confirming the fix so far.
- # [20:53] <@davidb> man
- # [20:54] <@davidb> the more I dive back into the visibility stuff, the sadder i get
- # [20:54] <@davidb> I think this should have been a lot less clever from the beginning.
- # [20:54] <@tbsaunde> davidb: yeah
- # [20:54] <@MarcoZ> davidb: Heritage from the very old days?
- # [20:55] <@davidb> Yeah. I think what really matters to screen readers is programmatic intended visibility. Not rendered visibility.
- # [20:56] <@davidb> So anyways, in the meantime, we need to take this revert because we exposed the wrong state in some cases. My bad.
- # [20:59] * Quits: @jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP) (Ping timeout)
- # [21:05] * Joins: ehsan (ehsan@F2D29657.F60B0462.67AC9B1.IP)
- # [21:05] <@firebot> New Core - Disability Access APIs bug 722417 filed by bolterbugz@gmail.com.
- # [21:05] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=722417 nor, --, ---, nobody, NEW, Expose intended visibility.
- # [21:06] <@firebot> markcapella@twcny.rr.com requested review from surkov.alexander@gm ail.com for attachment 592768 on bug 589784.
- # [21:07] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=589784 nor, --, ---, markcapella, ASSI, Remove role="label" aria processing.
- # [21:12] <@firebot> marco.zehe@googlemail.com granted review for attachment 592790 on bug 722248.
- # [21:12] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=722248 nor, --, ---, bolterbugz, NEW, Some chrome accessibles report incorrect visibility states
- # [21:13] <@davidb> MarcoZ: thanks
- # [21:14] <@davidb> MarcoZ: I made one small change locally
- # [21:14] <@davidb> hmmm
- # [21:14] <@davidb> naw
- # [21:14] <@MarcoZ> davidb: What change?
- # [21:15] <@davidb> MarcoZ: I added the style visibility check back near the top, because the early bail can save all that rect stuff, but i should really wait to see if talos tells me anything
- # [21:15] <@davidb> i haven't had an AT vendor tell me they use offscreen state yet
- # [21:16] <@MarcoZ> davidb: OK!
- # [21:16] <@MarcoZ> davidb: Anyway I'm off now. I'll check back in the morning, and if the patch has all it needs from surkov etc., I'll volunteer to land it and see that it makes its way through to Aurora.
- # [21:17] <@davidb> MarcoZ: thanks.
- # [21:17] <@MarcoZ> See you!
- # [21:17] <@davidb> ciao
- # [21:18] * Quits: @MarcoZ (marco.zehe@moz-1B071F06.dip.t-dialin.net) (Quit: g'night!)
- # [21:26] <@firebot> bolterbugz@gmail.com requested review from surkov.alexander@gm ail.com for attachment 592790 on bug 722248.
- # [21:26] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=722248 nor, --, ---, bolterbugz, NEW, Some chrome accessibles report incorrect visibility states
- # [21:28] * Joins: silvia (Adium@23AA46F7.E00E2F02.9C21467C.IP)
- # [21:29] * Joins: silvia1 (Adium@DCC5610.FC94C935.6E496AA7.IP)
- # [21:30] * Quits: silvia (Adium@23AA46F7.E00E2F02.9C21467C.IP) (Ping timeout)
- # [21:30] * Quits: silvia1 (Adium@DCC5610.FC94C935.6E496AA7.IP) (Quit: Leaving.)
- # [21:37] <@tbsaunde> davidb: you added some unneeded blank lines in that last patch
- # [21:38] <@davidb> tbsaunde: yeah i removed them locally
- # [21:38] <@tbsaunde> davidb: sounds good
- # [21:46] <@firebot> bolterbugz@gmail.com cancelled review?(bolterbugz@gmail.co m) for attachment 592711 on bug 589784.
- # [21:46] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=589784 nor, --, ---, markcapella, ASSI, Remove role="label" aria processing.
- # [21:46] <@firebot> bolterbugz@gmail.com cancelled review?(bolterbugz@gmail.co m) for attachment 592751 on bug 589784.
- # [21:47] <@firebot> bolterbugz@gmail.com granted review for attachment 592768 on bug 589784.
- # [21:49] * Joins: jprmc (jprmc@CB3EC984.A5F785DE.396E4C6D.IP)
- # [21:49] * ChanServ sets mode: +o jprmc
- # [22:08] * Quits: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [22:14] * @davidb considers adding a visibility state test to test_contextmenu.html
- # [22:16] <@davidb> hmm
- # [22:18] <@tbsaunde> hmm, I'm seeing a lot of doubled event failures in test_focus_autocomplete
- # [22:19] <@davidb> tbsaunde: with what patch?
- # [22:19] <@davidb> or in general?
- # [22:19] <@tbsaunde> davidb: with my state one, but I'm pretty sure its unrelated
- # [22:19] <@tbsaunde> anyway I'll try and investigate later
- # [22:23] <@tbsaunde> GOD DAM IT JS
- # [22:25] <@davidb> tbsaunde: maybe we should keep you in C++
- # [22:25] <@davidb> which bug is this for?
- # [22:25] <@tbsaunde> 686909
- # [22:26] <@tbsaunde> I figured it out, I was doing aEvent.isFromUserInput() instead of aEvent.isFromUserInput
- # [22:26] <@tbsaunde> I just made because a reasonable language would have this be a type erorr or something
- # [22:27] <@tbsaunde> instead of raising an exception that it is undefined
- # [22:27] * @davidb nods sagely
- # [22:29] <@tbsaunde> davidb: in the owe the poor people who don't understand dynamically typed langauges way? ;)
- # [22:29] <@davidb> hahah
- # [22:29] * Quits: @jprmc (jprmc@CB3EC984.A5F785DE.396E4C6D.IP) (Ping timeout)
- # [22:29] <@davidb> not this time
- # [22:29] <@tbsaunde> to be fair I try to avoid working on tests when I can
- # [22:29] <@davidb> however, dynamically typed languages really need TDD
- # [22:30] <@davidb> and we don't really do TDD
- # [22:30] <@davidb> speaking of js
- # [22:30] <@davidb> where the heck is synthesizeMouse defined?
- # [22:30] <@davidb> is that a language feature - i forget
- # [22:31] <@davidb> naw
- # [22:31] <@tbsaunde> grep is your friend ;)
- # [22:31] <@davidb> indeed
- # [22:31] <@hub> why debugging at compile time when you can have the users to it at runtime
- # [22:31] <@tbsaunde> I'd bet somewhere in our tests or simpletext
- # [22:31] <@davidb> yeah
- # [22:32] <@davidb> EventUtils.js
- # [22:32] <@tbsaunde> davidb: the trick is the main thing we have that's dynamically typed is the tests
- # [22:32] <@tbsaunde> and having tests for your tests is yoo dog
- # [22:32] <@davidb> true dat
- # [22:32] <@tbsaunde> I gues there's the build system, but most of that is string typed :p
- # [22:33] * Joins: jprmc (jprmc@CB3EC984.A5F785DE.396E4C6D.IP)
- # [22:33] * ChanServ sets mode: +o jprmc
- # [22:35] <@davidb> ok I gotta do the thing. back much later.
- # [22:35] * Quits: @davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: davidb)
- # [22:53] * Parts: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
- # [22:56] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
- # [23:08] * victorporo is now known as victorporof
- # [23:21] * Quits: @jprmc (jprmc@CB3EC984.A5F785DE.396E4C6D.IP) (Quit: Leaving)
- # [23:31] * Quits: khuey (khuey@moz-B77DEAEB.mozilla.org) (Ping timeout)
- # [23:34] * Joins: khuey (khuey@moz-B77DEAEB.mozilla.org)
- # [23:54] * Joins: surkov (surkov@DD563F5.F78D7EEB.34044A7F.IP)
- # [23:54] * ChanServ sets mode: +o surkov
- # Session Close: Tue Jan 31 00:00:01 2012
The end :)