/irc-logs / mozilla / #accessibility / 2015-10-09 / end
Options:
Previous day, Next day
- # Session Start: Fri Oct 09 00:00:00 2015
- # Session Ident: #accessibility
- # [00:15] <joanie> tbsaunde: weird hang thing? Are you using Orca master?
- # [00:17] <@tbsaunde> joanie: yeah, I wondered if that was the issue and pulled
- # [00:17] <@tbsaunde> still happened
- # [00:17] <joanie> can you reproduce the weird hang thing?
- # [00:18] <@tbsaunde> I think so, but I don't have str other than arrow around on your test page some
- # [00:19] <joanie> I wonder if when it's hanging, you're in the 'this tab has crashed' situation
- # [00:19] <joanie> you have a debug.out captured when the hang occurs?
- # [00:20] * joanie tries to repro it
- # [00:20] <@tbsaunde> joanie: no, I guess I can get one
- # [00:20] <joanie> it might help
- # [00:20] <@tbsaunde> I don't think the tab crashed, I attached with gdb, and it seemed like both processes where in ok states
- # [00:20] <joanie> ok
- # [00:22] <joanie> yeah, orca is not hanging when the tab crash happens
- # [00:22] <joanie> just arrow up and down??
- # [00:24] * Quits: @surkov (surkov@moz-an5vm5.cable.teksavvy.com) (Client exited)
- # [00:26] <joanie> tbsaunde: from just arrowing around on that page, I can crash the tab a bunch; I cannot hang orca at all. So I'm wondering if your build of Firefox has the tab crash fixed and is doing something new/different which Orca is not handling
- # [00:26] <joanie> I need to run some more errands, but if you manage to capture a debug.out from the hang, I'd love to see it. Emailing would work.
- # [00:29] <@tbsaunde> joanie: ah, seems I fine a different way to hang firefox
- # [00:31] <joanie> tbsaunde: hang firefox or orca?
- # [00:32] <joanie> for me, orca doesn't hang and the firefox web process crashes
- # [00:36] <@tbsaunde> joanie: firefox aparently
- # [00:38] <joanie> tbsaunde: phew
- # [00:38] <joanie> ok, running errands now. if it turns out it's orca hanging, send me the debug.out. if you cannot repro the web process crash, i'll try to do that tonight or tomorrow
- # [00:38] <joanie> bbiab
- # [00:43] <@tbsaunde> joanie: ok, now I can't make orca crash, but I can't make the tab crash either
- # [00:43] <@tbsaunde> maybe its worth you testing with the patch to not hang firefox that I just wrote?
- # [01:09] * Joins: davidb (davidb@moz-lsu4ch.dsl.bell.ca)
- # [01:09] * ChanServ sets mode: +qo davidb davidb
- # [01:09] * Quits: ~davidb (davidb@moz-lsu4ch.dsl.bell.ca) (Quit: peace out)
- # [01:52] * Quits: Justin_o (uid14648@moz-c6vipl.ealing.irccloud.com) (Quit: Connection closed for inactivity)
- # [01:59] <joanie> tbsaunde: is this patch for orca or gecko?
- # [01:59] <@tbsaunde> joanie: for gecko
- # [02:00] <joanie> i'd have to set up a build environment first
- # [02:00] <joanie> which I totally need to do
- # [02:00] <joanie> but not tonight since I'm feeling rather dead
- # [02:00] <@tbsaunde> joanie: yeah, or I can commit it since I think we should fix this anyway and wait for next week
- # [02:00] * joanie nods
- # [02:00] <@tbsaunde> joanie: heh, I always thought that was a good time for beating things into working
- # [02:00] <joanie> tbsaunde: are you thinking the patch you have for the hang also fixes my crash?
- # [02:01] <@tbsaunde> joanie: I suppose its possible, but probably not
- # [02:01] * Quits: smaug (chatzilla@moz-6kbmhh.rev.dnainternet.fi) (Ping timeout: 121 seconds)
- # [02:02] <joanie> ok, if you put the patch somewhere, I'll see about building firefox tomorrow
- # [02:02] <joanie> if you are going to commit it anyway, then.... i'll decide tomorrow
- # [02:02] <joanie> as for the web process crash, i'll try to make a simple, orca free, very specific test case tomorrow
- # [02:03] <joanie> my guess is that, aside from the race condition, it's a call Orca is making to the component iface's grab_focus()
- # [02:04] <joanie> part of me wonders if it's some combination of grab_focus() plus an immediate caret move
- # [02:04] <@tbsaunde> joanie: given its a race I don't know how much luck you'll have, but it'll be nice if I or smaug can reproduce
- # [02:04] <joanie> tbsaunde: I'm extremely lucky
- # [02:04] <joanie> ;)
- # [02:05] <joanie> besides, if I arrow nice and slow it doesn't happen
- # [02:05] <@tbsaunde> that's one possibility, another is that this is basically not actually related to accessibility, a11y just makes it much more easy to trip
- # [02:05] <joanie> which is why I've convinced myself that after some sleep I'll be able to make you a test case that you can reproduce
- # [02:05] * joanie nods at the a11y more easy to trip
- # [02:05] <@tbsaunde> heh, ok good luck :)
- # [02:05] <joanie> thanks
- # [02:06] <joanie> tbsaunde: so you working tomorrow?
- # [02:07] <@tbsaunde> joanie: that's the plan
- # [02:08] <joanie> ok, so let's touch base tomorrow about test cases, patches, etc.
- # [02:08] <@tbsaunde> yeah, I'm kind of dead too :)
- # [02:09] <joanie> ok then. g'nite!
- # [02:09] <joanie> and thanks
- # [02:09] <@tbsaunde> you too :)
- # [02:36] * Joins: surkov (surkov@moz-an5vm5.cable.teksavvy.com)
- # [02:36] * ChanServ sets mode: +o surkov
- # [02:37] * Quits: Jayflux (Jayflux@moz-9bf122.dsl.as9105.com) (Quit: Leaving)
- # [04:57] * Joins: yzen (yzen@moz-ql1r6k.cpe.pppoe.ca)
- # [04:57] * ChanServ sets mode: +o yzen
- # [05:10] * yzen is now known as yzen_
- # [05:12] * Quits: kkus (kkus@moz-7u2hvh.fios.verizon.net) (Ping timeout: 121 seconds)
- # [05:14] * Joins: kkus (kkus@moz-7u2hvh.fios.verizon.net)
- # [05:14] * Quits: @yzen_ (yzen@moz-ql1r6k.cpe.pppoe.ca) (Ping timeout: 121 seconds)
- # [05:35] * Joins: yzen (yzen@moz-ql1r6k.cpe.pppoe.ca)
- # [05:35] * ChanServ sets mode: +o yzen
- # [05:45] * yzen is now known as yzen_
- # [05:49] * Quits: @yzen_ (yzen@moz-ql1r6k.cpe.pppoe.ca) (Ping timeout: 121 seconds)
- # [06:28] * Quits: @surkov (surkov@moz-an5vm5.cable.teksavvy.com) (Client exited)
- # [08:04] * Tomcat is now known as Tomcat|sheriffduty
- # [09:16] * Quits: ioanachiorean (ioanachiore@moz-q23.52n.2.5.IP) (Quit: Leaving)
- # [09:16] * Joins: ioanachiorean (ioanachiore@moz-q23.52n.2.5.IP)
- # [10:39] * Joins: smaug (chatzilla@moz-6kbmhh.rev.dnainternet.fi)
- # [11:01] * Quits: smaug (chatzilla@moz-6kbmhh.rev.dnainternet.fi) (Ping timeout: 121 seconds)
- # [11:35] * Quits: mattur (sid16049@moz-4ms7mo.tooting.irccloud.com) (Ping timeout: 121 seconds)
- # [11:53] <@firebot> cbook@mozilla.com changed the Resolution on bug 1212880 from --- to FIXED.
- # [11:53] <@firebot> cbook@mozilla.com changed the Status on bug 1212880 from ASSIGNED to RESOLVED.
- # [11:53] <@firebot> cbook@mozilla.com set status-firefox44 to fixed on bug 1212880.
- # [11:53] <@firebot> cbook@mozilla.com changed the Target Milestone on bug 1212880 from --- to mozilla44.
- # [11:53] <@firebot> https://bugzil.la/1212880 — FIXED, ryanvm@gmail.com — TabParent.cpp build bustage with --disable-accessibility specified
- # [11:59] * Joins: mattur (sid16049@moz-4ms7mo.tooting.irccloud.com)
- # [12:07] * Quits: ioanachiorean (ioanachiore@moz-q23.52n.2.5.IP) (Ping timeout: 121 seconds)
- # [12:24] * Tomcat|sheriffduty is now known as Tomcat|sheriffduty|lunch
- # [13:47] * Joins: yzen (yzen@moz-ql1r6k.cpe.pppoe.ca)
- # [13:47] * ChanServ sets mode: +o yzen
- # [13:47] * Tomcat|sheriffduty|lunch is now known as Tomcat|sheriffduty
- # [13:53] * Joins: surkov (surkov@moz-an5vm5.cable.teksavvy.com)
- # [13:53] * ChanServ sets mode: +o surkov
- # [13:57] * yzen is now known as yzen_
- # [14:01] * Quits: @yzen_ (yzen@moz-ql1r6k.cpe.pppoe.ca) (Ping timeout: 121 seconds)
- # [14:22] * Quits: @surkov (surkov@moz-an5vm5.cable.teksavvy.com) (Client exited)
- # [14:35] * Joins: yzen (yzen@moz-ql1r6k.cpe.pppoe.ca)
- # [14:35] * ChanServ sets mode: +o yzen
- # [14:47] * Joins: surkov (surkov@moz-an5vm5.cable.teksavvy.com)
- # [14:47] * ChanServ sets mode: +o surkov
- # [14:56] * yzen is now known as yzen_
- # [14:56] * yzen_ is now known as yzen
- # [15:03] * Joins: sixtyten (quassel@moz-ulpjmu.sentoris.de)
- # [15:35] * Joins: davidb (davidb@moz-i5m.05u.207.66.IP)
- # [15:35] * ChanServ sets mode: +qo davidb davidb
- # [15:41] <@firebot> New Core - Disability Access APIs bug 1213281 filed by dbolter@mozilla.com.
- # [15:41] <@firebot> https://bugzil.la/1213281 — NEW, nobody@mozilla.org — crash in mozilla::a11y::DocAccessible::UpdateTreeOnInsertion(mozilla::a11y::Accessible*)
- # [15:44] * Joins: anvk (anovak@moz-040.mvk.11.204.IP)
- # [15:55] * Joins: clown (clown@moz-v6m.c3o.211.205.IP)
- # [16:11] * yzen is now known as yzen_
- # [16:12] * yzen_ is now known as yzen
- # [16:12] * yzen is now known as yzen_
- # [16:16] * Quits: @yzen_ (yzen@moz-ql1r6k.cpe.pppoe.ca) (Ping timeout: 121 seconds)
- # [16:27] <~davidb> oh err
- # [16:27] <~davidb> heyo
- # [16:36] * Joins: smaug (chatzilla@moz-m534ng.elisa-laajakaista.fi)
- # [16:40] <joanie> tbsaunde: Using the latest nightly, I'm not (so far) able to reproduce the web process crash we talked about yesterday. Last night, I could reproduce it pretty reliably. So I guess someone fixed something somewhere. :)
- # [16:40] <joanie> I'm also cloning Gecko and will build it so that in the future I can try patches, etc.
- # [16:42] <~davidb> :)
- # [16:44] <@tbsaunde> joanie: ok cool on both counts :)
- # [16:45] <joanie> btw, the bootstrap file y'all provide gives both the mercurial and git urls :)
- # [17:06] * Tomcat|sheriffduty is now known as Tomcat
- # [17:07] * Tomcat is now known as Tomcat|afk
- # [17:13] * Quits: sixtyten (quassel@moz-ulpjmu.sentoris.de) (A TLS packet with unexpected length was received.)
- # [17:16] * davidb is now known as davidb|afk
- # [17:30] * Joins: yzen (yzen@moz-i5m.05u.207.66.IP)
- # [17:30] * ChanServ sets mode: +o yzen
- # [17:36] * davidb|afk is now known as davidb
- # [17:45] * davidb is now known as davidb|afk
- # [18:00] <@firebot> New Core - Disability Access APIs bug 1213327 filed by tbsaunde+mozbugs@tbsaunde.org.
- # [18:01] <@firebot> https://bugzil.la/1213327 — NEW, nobody@mozilla.org — expose IAccessibleHyperlink on proxied accessibles
- # [18:24] * davidb|afk is now known as davidb
- # [18:45] * Quits: @surkov (surkov@moz-an5vm5.cable.teksavvy.com) (Client exited)
- # [18:46] <@firebot> chris@chrishardy.net changed the Resolution on bug 1163706 from INVALID to ---.
- # [18:46] <@firebot> chris@chrishardy.net changed the Status on bug 1163706 from RESOLVED to REOPENED.
- # [18:46] <@firebot> https://bugzil.la/1163706 — REOPENED, nobody@mozilla.org — Not possible to select text from different objects into a single selection using IAccessible2
- # [18:49] * Joins: surkov (surkov@moz-an5vm5.cable.teksavvy.com)
- # [18:49] * ChanServ sets mode: +o surkov
- # [18:49] * Quits: ~davidb (davidb@moz-i5m.05u.207.66.IP) (Ping timeout: 121 seconds)
- # [18:54] * Joins: davidb (davidb@moz-i5m.05u.207.66.IP)
- # [18:54] * ChanServ sets mode: +qo davidb davidb
- # [19:03] * Joins: milkywayprotein (milkywaypro@moz-8k5vuc.res.rr.com)
- # [19:42] * Quits: ~davidb (davidb@moz-i5m.05u.207.66.IP) (Quit: Blah blah blah)
- # [19:54] * Quits: milkywayprotein (milkywaypro@moz-8k5vuc.res.rr.com) (Client exited)
- # [19:55] * shu is now known as shu`pto
- # [21:11] <@firebot> New Core - Disability Access APIs bug 1213382 filed by tbsaunde+mozbugs@tbsaunde.org.
- # [21:11] <@firebot> https://bugzil.la/1213382 — NEW, nobody@mozilla.org — proxy more accessible selection methods
- # [21:11] <@tbsaunde> smaug: can you maybe take ^?
- # [21:18] * Joins: milkywayprotein (milkywaypro@moz-8k5vuc.res.rr.com)
- # [21:28] * joanie crosses her fingers and begins running orca's tests with e10s :)
- # [21:32] * wlach is now known as wlach|biab
- # [21:35] <@tbsaunde> joanie: thanks!
- # [21:43] * Joins: davidb (davidb@moz-lsu4ch.dsl.bell.ca)
- # [21:43] * ChanServ sets mode: +qo davidb davidb
- # [21:48] <@firebot> bugs@pettay.fi changed the Assignee on bug 1213382 from nobody@mozilla.org to bugs@pettay.fi.
- # [21:48] <@firebot> https://bugzil.la/1213382 — NEW, bugs@pettay.fi — proxy more accessible selection methods
- # [21:48] <smaug> tbsaunde: will do
- # [21:53] <joanie> tbsaunde: any chance that the value interface needs to be done like action and table?
- # [21:59] * Quits: ~davidb (davidb@moz-lsu4ch.dsl.bell.ca) (Quit: Blah blah blah)
- # [22:00] * Joins: yzen_ (yzen@moz-i5m.05u.207.66.IP)
- # [22:00] * ChanServ sets mode: +o yzen_
- # [22:03] * Quits: @yzen (yzen@moz-i5m.05u.207.66.IP) (Ping timeout: 121 seconds)
- # [22:05] <joanie> ok looks like we have the iface, but no signals
- # [22:06] * joanie makes a listener
- # [22:08] <@tbsaunde> joanie: let me look
- # [22:08] <@tbsaunde> smaug: thanks!
- # [22:08] * Joins: davidb (davidb@moz-lsu4ch.dsl.bell.ca)
- # [22:08] * ChanServ sets mode: +qo davidb davidb
- # [22:17] <joanie> tbsaunde: are you filing a bug? I was going to
- # [22:17] <@tbsaunde> joanie: I think atm you have more details
- # [22:17] * joanie nods
- # [22:17] <@tbsaunde> so you please do it or tell me what you know
- # [22:18] <joanie> i will do both. :)
- # [22:19] <@tbsaunde> hah
- # [22:25] <@firebot> New Core - Disability Access APIs bug 1213402 filed by jdiggs@igalia.com.
- # [22:25] <@firebot> https://bugzil.la/1213402 — NEW, nobody@mozilla.org — No ATK/AT-SPI2 object:property-change:accessible-value events for web widgets in e10s
- # [22:25] <joanie> tbsaunde: ^^ should explain it
- # [22:25] <joanie> executive summary: iface works; events missing
- # [22:26] <@tbsaunde> joanie: ah hrm lets see
- # [22:36] * Quits: anvk (anovak@moz-040.mvk.11.204.IP) (Quit: Leaving.)
- # [22:39] <@tbsaunde> joanie: this would be simpler if you didn't object to value change events when the value is text ;p ;)
- # [22:39] <joanie> heh
- # [22:40] <joanie> I think what objects is atk or at-spi2
- # [22:40] <joanie> Orca doesn't get those
- # [22:40] <@tbsaunde> joanie: yeah, they probably do warn
- # [22:40] <joanie> but something spews out errors
- # [22:40] <joanie> they warn *and* fail to pass along the event
- # [22:41] <@tbsaunde> joanie: oh maybe, I thought they warned and passed the message on
- # [22:42] <@tbsaunde> any way surkov any reason you'd mind breaking EVENT_VALUE_CHANGE up into EVENT_TEXT_VALUE_CHANGE and EVENT_NUMERIC_VALUE_CHANGED ?
- # [22:42] <joanie> if you suggest that I use the spewage as an event, I'll have to kill you. :P :P
- # [22:42] <@tbsaunde> joanie: lol no ;p that would be hard ;)
- # [22:44] * Quits: ~davidb (davidb@moz-lsu4ch.dsl.bell.ca) (Quit: peace out)
- # [22:45] <@surkov> I’m not sure I see a reason why we would need that, but I dont’ see anything particular bad in that
- # [22:46] <@tbsaunde> surkov: well, linux only wants the numeric ones, and that seems simpler and nicer? than filtering
- # [22:47] <@surkov> sounds good
- # [23:04] <joanie> tbsaunde: at least one of the failures is not a11y (yay), but....
- # [23:04] <joanie> see bug 1213439
- # [23:04] <@firebot> https://bugzil.la/1213439 — NEW, nobody@mozilla.org — Some web app key bindings fail to work if e10s is enabled
- # [23:04] <joanie> I filed it under Firefox:General because I didn't know where it really belonged
- # [23:05] <joanie> I also assume this is a bug and not a feature/correction
- # [23:06] * Quits: clown (clown@moz-v6m.c3o.211.205.IP) (Quit: Leaving.)
- # [23:07] <@tbsaunde> joanie: I'm not really sure, but probably :) hopefully somebody looks at firefox:general and sorts it, if not poke me tuesday and I'll figure out where to put it and what e10s flags need set
- # [23:07] <joanie> thanks tbsaunde!
- # [23:07] <joanie> i'm glad that failure isn't a11y
- # [23:07] <@tbsaunde> hah, me too ;)
- # [23:07] <joanie> the test results made me think it was lack of a11y events, but nope. Orca just wasn't where the test assumed
- # [23:08] * joanie triages the next one
- # [23:08] <@tbsaunde> heh
- # [23:10] <joanie> hmmm. another big failure might be due to that keyboard thang. it feels wrong to feel happy about that. ;)
- # [23:14] <@tbsaunde> yeah, one generally shouldn't feel good about bugs ;p
- # [23:15] <joanie> this one seems to be something else. running with and without e10s to compare, but I saw the selection-related issue you asked smaug to work on.
- # [23:15] <joanie> is that the selection interface?
- # [23:15] <@tbsaunde> joanie: no, that's totally windows only
- # [23:16] <joanie> ok, then I may have another one for y'all. /me dives in
- # [23:16] * Quits: @yzen_ (yzen@moz-i5m.05u.207.66.IP) (Ping timeout: 121 seconds)
- # [23:45] <@firebot> New Core - Disability Access APIs bug 1213452 filed by zebdeos@bayprogrammer.com.
- # [23:45] <@firebot> https://bugzil.la/1213452 — UNCONFIRMED, nobody@mozilla.org — [Mac] If option element has title attribute, VoiceOver only reads the title text, ignoring option te
- # [23:51] * wlach|biab is now known as wlach
- # Session Close: Sat Oct 10 00:00:00 2015
Previous day, Next day
Think these logs are useful? Then please donate to show your gratitude (and keep them up, of course). Thanks! — Krijn