/irc-logs / mozilla / #accessibility / 2014-12-18 / end
Options:
Previous day, Next day
- # Session Start: Thu Dec 18 00:00:00 2014
- # Session Ident: #accessibility
- # [00:00] * Quits: scott_gonzalez (scott_gonza@moz-7av2eq.hfc.comcastbusiness.net) (Client exited)
- # [00:00] * Quits: @yzen_ (yzen@moz-ql1r6k.cpe.pppoe.ca) (Ping timeout: 121 seconds)
- # [00:47] * Quits: Justin_o (uid14648@moz-c6vipl.ealing.irccloud.com) (Quit: Connection closed for inactivity)
- # [00:55] * Joins: surkov (surkov@moz-v8g78c.cpe.teksavvy.com)
- # [00:55] * ChanServ sets mode: +o surkov
- # [00:57] * Joins: yzen (yzen@moz-ql1r6k.cpe.pppoe.ca)
- # [00:57] * ChanServ sets mode: +o yzen
- # [01:00] * Joins: davidb (davidb@moz-nd2ii9.dsl.bell.ca)
- # [01:00] * ChanServ sets mode: +qo davidb davidb
- # [01:01] * Quits: @yzen (yzen@moz-ql1r6k.cpe.pppoe.ca) (Ping timeout: 121 seconds)
- # [01:04] * Quits: ~davidb (davidb@moz-nd2ii9.dsl.bell.ca) (Quit: Blah blah blah)
- # [01:06] * Joins: scott_gonzalez (scott_gonza@moz-5t7jkp.fios.verizon.net)
- # [01:30] * Joins: davidb (davidb@moz-4l2.hhb.15.74.IP)
- # [01:30] * ChanServ sets mode: +qo davidb davidb
- # [01:31] * Quits: ~davidb (davidb@moz-4l2.hhb.15.74.IP) (Quit: Blah blah blah)
- # [02:36] * Quits: @surkov (surkov@moz-v8g78c.cpe.teksavvy.com) (Client exited)
- # [02:45] * Joins: davidb (davidb@moz-nd2ii9.dsl.bell.ca)
- # [02:45] * ChanServ sets mode: +qo davidb davidb
- # [02:46] * Joins: yzen (yzen@moz-ql1r6k.cpe.pppoe.ca)
- # [02:46] * ChanServ sets mode: +o yzen
- # [02:50] * Quits: @yzen (yzen@moz-ql1r6k.cpe.pppoe.ca) (Ping timeout: 121 seconds)
- # [02:55] * Quits: ~davidb (davidb@moz-nd2ii9.dsl.bell.ca) (Quit: Blah blah blah)
- # [03:00] * Joins: isaacd (quassel@moz-btmifd.east.verizon.net)
- # [03:06] * Joins: yliao (yliao@moz-2hf62a.HINET-IP.hinet.net)
- # [03:36] * Joins: surkov (surkov@moz-v8g78c.cpe.teksavvy.com)
- # [03:36] * ChanServ sets mode: +o surkov
- # [03:43] * Quits: MrMazda (fmcz@moz-6psm3k.cable.mindspring.com) (Quit: ChatZilla 0.9.86.1 [SeaMonkey 1.1.18/2009082712])
- # [03:50] * Joins: MrMazda (fmcz@moz-6psm3k.cable.mindspring.com)
- # [05:03] * Joins: yzen (yzen@moz-ql1r6k.cpe.pppoe.ca)
- # [05:03] * ChanServ sets mode: +o yzen
- # [05:43] * Joins: icaaq (icaaq@moz-n1f1o3.cust.bredbandsbolaget.se)
- # [05:43] * icaaq is now known as icaaq|afk
- # [05:45] * icaaq|afk is now known as icaaq
- # [06:27] * Joins: maxli (maxli@moz-h8kluf.cs.uwaterloo.ca)
- # [06:31] * Quits: @surkov (surkov@moz-v8g78c.cpe.teksavvy.com) (Client exited)
- # [06:46] * Quits: a-865 (fmcz@moz-6psm3k.cable.mindspring.com) (Quit: ChatZilla 0.9.91 [SeaMonkey 2.31/20141202221108])
- # [07:09] * Joins: a-865 (fmcz@moz-6psm3k.cable.mindspring.com)
- # [07:31] * Quits: maxli (maxli@moz-h8kluf.cs.uwaterloo.ca) (Ping timeout: 121 seconds)
- # [08:11] * Tomcat|afk is now known as Tomcat|sheriffduty
- # [08:25] * yzen is now known as yzen_
- # [08:29] * Quits: @yzen_ (yzen@moz-ql1r6k.cpe.pppoe.ca) (Ping timeout: 121 seconds)
- # [09:18] * Joins: ioanachiorean (ioanachiore@moz-ohb.65e.79.82.IP)
- # [09:26] * Joins: yzen (yzen@moz-ql1r6k.cpe.pppoe.ca)
- # [09:26] * ChanServ sets mode: +o yzen
- # [09:30] * Quits: @yzen (yzen@moz-ql1r6k.cpe.pppoe.ca) (Ping timeout: 121 seconds)
- # [10:28] * Joins: agibson (agibson@moz-j04gi9.cable.virginm.net)
- # [10:51] * Joins: slee (chatzilla@moz-107ht8.range109-145.btcentralplus.com)
- # [11:05] * Joins: Gijs_away (chatzilla@moz-pnd.q6m.80.5.IP)
- # [11:06] * Gijs_away is now known as Gijs
- # [11:38] * Quits: yliao (yliao@moz-2hf62a.HINET-IP.hinet.net) (Connection closed)
- # [11:50] * icaaq is now known as icaaq|afk
- # [12:08] * Joins: yliao (yliao@moz-aj9ejj.HINET-IP.hinet.net)
- # [12:11] * Joins: davidb (davidb@moz-nd2ii9.dsl.bell.ca)
- # [12:11] * ChanServ sets mode: +qo davidb davidb
- # [12:12] * Quits: ~davidb (davidb@moz-nd2ii9.dsl.bell.ca) (Quit: peace out)
- # [12:13] * Quits: yliao (yliao@moz-aj9ejj.HINET-IP.hinet.net) (Ping timeout: 121 seconds)
- # [12:24] * icaaq|afk is now known as icaaq
- # [12:25] * icaaq is now known as icaaq|afk
- # [12:25] * icaaq|afk is now known as icaaq
- # [13:07] * agibson is now known as agibson|afk
- # [13:36] * Joins: surkov (surkov@moz-v8g78c.cpe.teksavvy.com)
- # [13:36] * ChanServ sets mode: +o surkov
- # [14:07] * Joins: Justin_o (uid14648@moz-c6vipl.ealing.irccloud.com)
- # [14:22] * Joins: MarcoZ (marco.zehe@moz-tb35md.dip0.t-ipconnect.de)
- # [14:22] * ChanServ sets mode: +ao MarcoZ MarcoZ
- # [14:29] * agibson|afk is now known as agibson
- # [14:33] * Quits: slee (chatzilla@moz-107ht8.range109-145.btcentralplus.com) (Ping timeout: 121 seconds)
- # [14:40] * Joins: yzen (yzen@moz-ql1r6k.cpe.pppoe.ca)
- # [14:40] * ChanServ sets mode: +o yzen
- # [14:40] * Quits: @yzen (yzen@moz-ql1r6k.cpe.pppoe.ca) (Connection closed)
- # [14:41] * Joins: yzen (yzen@moz-ql1r6k.cpe.pppoe.ca)
- # [14:41] * ChanServ sets mode: +o yzen
- # [14:56] * Joins: anvk (anovak@moz-871.mvk.11.204.IP)
- # [14:56] * yzen is now known as yzen_
- # [15:00] * Quits: @yzen_ (yzen@moz-ql1r6k.cpe.pppoe.ca) (Ping timeout: 121 seconds)
- # [15:05] * Joins: slee (chatzilla@moz-107ht8.range109-145.btcentralplus.com)
- # [15:06] * Joins: yliao (yliao@moz-2hf62a.HINET-IP.hinet.net)
- # [15:06] <Gijs> MarcoZ: ping?
- # [15:07] <Gijs> MarcoZ: I have a question about the search UI a11y story
- # [15:11] * Quits: yliao (yliao@moz-2hf62a.HINET-IP.hinet.net) (Ping timeout: 121 seconds)
- # [15:13] * Quits: SteveF (chatzilla@moz-80j6qs.cable.virginm.net) (Quit: ChatZilla 0.9.91 [Firefox 34.0/20141125180439])
- # [15:18] <&MarcoZ> Gijs: Pong! What's up with it?
- # [15:23] <Gijs> MarcoZ: so right now, the buttons for the one-off searches get some kind of synthetic focus, ish
- # [15:23] <Gijs> MarcoZ: because we want all keypresses to still go to the inputfield, that's where the 'real' focus stays
- # [15:24] <Gijs> MarcoZ: what's the appropriate aria-markup to indicate stuff is changing when you switch between the different one-off search buttons?
- # [15:24] <Gijs> MarcoZ: is that even possible?
- # [15:24] <Gijs> (while keeping the 'real' focus on the inputbox)
- # [15:31] * Joins: yzen (yzen@moz-i5m.05u.207.66.IP)
- # [15:31] * ChanServ sets mode: +o yzen
- # [15:35] <&MarcoZ> Gijs: That's a case for aria-activedescendant. Or at least we can experiment with that.
- # [15:35] * Joins: davidb (davidb@moz-i5m.05u.207.66.IP)
- # [15:35] * ChanServ sets mode: +qo davidb davidb
- # [15:36] <&MarcoZ> Gijs: e. g. if keyboard focus is on the actual input, aria-activedescendant is not present. As soon as the user starts tabbing, aria-activedescendant is set on the input field with the value being the ID of the button that is the "active" button.
- # [15:36] <~davidb> heyo
- # [15:37] <&MarcoZ> Gijs: When tabbing cycles back to the input itself, e. g. none of the buttons has pseudo-focus, aria-activedescendant gets removed again.
- # [15:43] <Gijs> MarcoZ: makes perfect sense, thanks!
- # [15:43] <&MarcoZ> Gijs: Feel free to ask me to run with a certain patch if you have it. Or ask me to build with it myself, I can do that, too.
- # [15:43] <Gijs> MarcoZ: and sorry for the delay - listening to your podcast! It's really nice, thanks for doing one :)
- # [15:44] <Gijs> (I mean, not the first one, of course - but either way)
- # [15:44] <Gijs> *it's not the first one
- # [15:53] <&MarcoZ> Gijs: Glad you're enjoying it! :)
- # [15:54] * Quits: slee (chatzilla@moz-107ht8.range109-145.btcentralplus.com) (Quit: ChatZilla 0.9.91 [Firefox 34.0/20141125180439])
- # [15:56] * Joins: clown (clown@moz-v6m.c3o.211.205.IP)
- # [15:57] * clown is now known as clown_mtg
- # [16:06] * davidb is now known as davidb|afk
- # [16:08] * Joins: slee (chatzilla@moz-107ht8.range109-145.btcentralplus.com)
- # [16:17] * Quits: slee (chatzilla@moz-107ht8.range109-145.btcentralplus.com) (Ping timeout: 121 seconds)
- # [16:28] * davidb|afk is now known as davidb
- # [16:40] <joanie> surkov: davidb: I just learned something interesting from clown_mtg
- # [16:40] <~davidb> amazing
- # [16:40] <joanie> In particular, he says that the fix for https://bugzilla.mozilla.org/show_bug.cgi?id=1013584 doesn't actually expose those elements as if they were divs
- # [16:40] <@firebot> Bug 1013584 — FIXED, surkov.alexander@gmail.com — Children of tables and lists with role='presentation' should each be exposed as if they were div ele
- # [16:40] <joanie> at least not for ATK
- # [16:41] <joanie> in which case the bug I filed wasn't actually fixed
- # [16:41] <joanie> and the implementation is not correct
- # [16:41] <joanie> :(
- # [16:41] * ~davidb steps out of the way, rolls out the carpet for surkov
- # [16:41] <joanie> was that by design?
- # [16:41] <joanie> fortunately, wrt the AAMs, WebKitGtk's implementation is correct
- # [16:44] <@surkov> joanie: they just get an accessible and should have similar semantics with div, but they aren’t actual div. but what problem do we have?
- # [16:45] <joanie> if you expose a non-paragraph as a paragraph, Orca's navigation by paragraph will move to non-paragraphs
- # [16:45] <joanie> if you expose stuff as ROLE_TEXT, you're just plain wrong
- # [16:45] <joanie> ;)
- # [16:46] <joanie> I didn't file a bug asking for presentational children to be exposes as if they were divs (aka ROLE_SECTION) because I was bored
- # [16:46] <joanie> :)
- # [16:46] <joanie> so I'm curious as to why, given the stated expectations for ATK
- # [16:46] <joanie> the decision was made to do something different
- # [16:47] <joanie> in general -- let alone using paragraph (wrong) and text (really, really wrong)
- # [16:47] <@surkov> joanie: what is expected role?
- # [16:47] <joanie> SECTION
- # [16:47] <joanie> as stated in the bug
- # [16:47] <joanie> several times I think
- # [16:47] * joanie looks
- # [16:48] <joanie> btw, several times in that bug, I explicitly state why ROLE_TEXT is completely wrong
- # [16:48] <@surkov> mm, ok, joanie can you file a bug pls?
- # [16:48] <joanie> so if that really is the mapping, then it's just.... mind boggling to me
- # [16:48] <joanie> a new bug stating what?
- # [16:48] <@surkov> that we expose accessible iwth wrong role
- # [16:48] <joanie> that presentational children of lists and tables should be exposed as if they were divs
- # [16:49] <@surkov> is that the issue, right?
- # [16:49] <joanie> because the bug that I filed stated that
- # [16:49] <joanie> davidb (I think) repurposed it to not be specific to ATK
- # [16:49] <joanie> then everything I stated should not be done was done anyway
- # [16:49] <joanie> so I don't think we need a new bug filed
- # [16:50] <joanie> I think we need to reopen the old bug
- # [16:50] <joanie> and fix it correctly
- # [16:50] <joanie> but I'm not sure if you fixed it the way you did deliberately or not
- # [16:50] <@surkov> opening old bugs is not good for tracking
- # [16:50] <joanie> no?
- # [16:50] <joanie> ok, I'll clone the old bug
- # [16:51] <@surkov> I much prefer to have dependants and blcoking
- # [16:51] <@surkov> to keep things smaller
- # [16:51] <joanie> surkov: the old bug is the bug you are asking me to file :)
- # [16:51] <@surkov> joanie: from impl point of view it will never be a div
- # [16:51] <@surkov> it’s just terms
- # [16:51] <@surkov> the issue you want same role as div has
- # [16:51] <joanie> bug it's terms (or roles) that impact Orca
- # [16:51] <joanie> div is the generic block text element
- # [16:51] <@surkov> orca doesn’t deal with divs
- # [16:52] <@surkov> it deals with roles
- # [16:52] * Joins: slee (chatzilla@moz-107ht8.range109-145.btcentralplus.com)
- # [16:52] <@surkov> so say that
- # [16:52] <joanie> well, ROLE_SECTION
- # [16:52] <@surkov> but not div
- # [16:52] * joanie sighs
- # [16:52] * joanie opens a new bug
- # [16:52] <@surkov> thanks :)
- # [16:54] <~davidb> i'll take blame here - I should have looked for a test with section children.
- # [16:55] * Joins: SteveF (chatzilla@moz-80j6qs.cable.virginm.net)
- # [16:55] <SteveF> davidb: hi
- # [16:56] * clown_mtg is now known as clown
- # [16:56] <~davidb> hi SteveF
- # [16:56] <SteveF> davidb: what happened with moz discussion about getcomputedRole
- # [16:57] <SteveF> the pf action item on you seaid it was closed
- # [16:57] <~davidb> SteveF, i like it except just have a property not a method so element.computedRole, surkov still likes to explore a more complicated approach :)
- # [16:57] <SteveF> see blinks intent to implement https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/WToTEn4lveY
- # [16:58] <~davidb> (complicated but powerful yet i can't see it happening)
- # [16:58] <~davidb> SteveF, yeah so they are going for properties - which i like. i haven't thought about label
- # [16:58] <~davidb> but seems reasonable
- # [16:59] <~davidb> I can't help but think I need surkov on board before I can truly 100% endorse it.
- # [16:59] <SteveF> ok what we don't have is any spec text for the feature, anybody here up for sketching it out?
- # [17:00] <SteveF> would like some point of reference for discussion
- # [17:00] <~davidb> i can review
- # [17:00] <~davidb> is DomenicM interested?
- # [17:00] <~davidb> or James
- # [17:00] <SteveF> if someone can write up soemthing simple i can add to spec somewhere
- # [17:01] <~davidb> can we wait until new year?
- # [17:01] <SteveF> deberybody appears interested but nobody has written it down :-)
- # [17:01] <SteveF> everybody
- # [17:01] <~davidb> is there anyone pushing?
- # [17:01] <SteveF> oh sure re new year
- # [17:01] <~davidb> ok cool
- # [17:01] <SteveF> I am pushing :-)
- # [17:02] <~davidb> no more importantly,... SteveF how old was that facebook pic with you chris mills etc?
- # [17:02] <@firebot> New Core - Disability Access APIs bug 1113153 filed by jdiggs@igalia.com.
- # [17:02] <~davidb> no/now
- # [17:02] <@firebot> https://bugzil.la/1113153 — NEW, nobody@mozilla.org — [ATK] Children of tables and lists with role='presentation' should each be exposed with ATK_ROLE_SEC
- # [17:02] <SteveF> last night
- # [17:02] <~davidb> jealous
- # [17:03] <SteveF> btw davidb: i just donated to mozilla so now I won your ass ;-)
- # [17:03] <SteveF> won/own
- # [17:03] <SteveF> yeah it was a nice night
- # [17:04] <~davidb> lol
- # [17:04] <~davidb> thank you!
- # [17:04] <~davidb> I'm sorry there is no tax receipt for anywhere except US right now
- # [17:05] <~davidb> MarcoZ, NDA... sigh (re email)
- # [17:05] * Tomcat|sheriffduty is now known as Tomcat|afk
- # [17:09] * Joins: yliao (yliao@moz-aj9ejj.HINET-IP.hinet.net)
- # [17:11] <SteveF> surkov: can you have a think and feedback on the intent to implement https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/WToTEn4lveY ?
- # [17:12] <@surkov> SteveF: in general I don’t like this approach at all
- # [17:12] <@surkov> and I’m not sure I’m aware of use case
- # [17:13] <SteveF> surkov: ok so can you propose an alternative?
- # [17:13] <@surkov> let’s start from the use case
- # [17:13] <@surkov> I’m not sure what problem we want to solve
- # [17:13] * Quits: yliao (yliao@moz-aj9ejj.HINET-IP.hinet.net) (Ping timeout: 121 seconds)
- # [17:17] <@surkov> SteveF: ^
- # [17:17] <~davidb> SteveF, the question is about why? not how.
- # [17:18] <~davidb> (e.g. easier testing etc)
- # [17:18] <~davidb> (webby AT)
- # [17:18] <SteveF> surkov: yeah need james etal to supply use cases
- # [17:18] <~davidb> I'm also curious about "who" is driving this
- # [17:19] <SteveF> james craig, alice/dominic and domenic denicola its seems
- # [17:19] <@surkov> SteveF: ok, maybe it makese sene to run thread to figure out use cases
- # [17:20] <Domenic> Testing is the biggest UC I can think of off the top of my head. Especially for custom elements.
- # [17:22] <~davidb> gecko currently doesn't have the ability to shutdown our engine, once started... so there is a lot of potential work for us.
- # [17:22] <~davidb> (a11y engine)
- # [17:23] <@surkov> if testing is the use case then we need to end up with full API, these two methods aren’t helpful
- # [17:25] <Domenic> What is "full API" and why aren't these methods subsets of it
- # [17:26] <SteveF> surkov: have started thread asking for use cases :-) http://lists.w3.org/Archives/Public/public-html/2014Dec/0022.html
- # [17:27] <~davidb> excellent
- # [17:28] <@surkov> SteveF: thanks!
- # [17:29] <@surkov> Domenic: if those methods are subset then basically Element should be extended by all IAccessible2 method and friends, that’s sort of weird
- # [17:29] <&MarcoZ> davidb: I saw that thingie re NDA. I opt for you to reply that Mozilla is all about open-source and we don't generally have NDA's, and we only sign them if absolutely necessary, and project-based, not per se up front.
- # [17:30] <@surkov> and then you cannot even think of non DOM a11y
- # [17:30] <&MarcoZ> davidb: And you because you're the official team lead. ;)
- # [17:30] <@surkov> like if you want to test list item then you cannot test its bullet because it may not have DOM
- # [17:30] <@surkov> thus I think we should have something separate from Element
- # [17:30] <Domenic> Hmm
- # [17:33] <Domenic> Well my interest is in being able to construct and test custom elements that appear to a11y tech the same as built-in elements. So e.g. for list item I'd envision a shadow DOM with the bullet, with the bullet having its own stuff.
- # [17:33] <~davidb> MarcoZ, agreed. I replied with something similar.
- # [17:35] <&MarcoZ> davidb: Just saw it.
- # [17:36] * davidb is now known as davidb|afk
- # [17:39] <tbsaunde> Domenic: why is it important to be able to test your element works using js?
- # [17:40] * Joins: maxli (maxli@moz-favcmk.cs.uwaterloo.ca)
- # [17:43] * Quits: slee (chatzilla@moz-107ht8.range109-145.btcentralplus.com) (Quit: ChatZilla 0.9.91 [Firefox 34.0/20141125180439])
- # [17:50] * SteveF really wish PF would get with the program so its mailing lists/tracker could be fully public and used for discussion :-(
- # [17:57] * Quits: &MarcoZ (marco.zehe@moz-tb35md.dip0.t-ipconnect.de) (Quit: n8)
- # [18:02] * Quits: ioanachiorean (ioanachiore@moz-ohb.65e.79.82.IP) (Ping timeout: 121 seconds)
- # [18:21] * Quits: maxli (maxli@moz-favcmk.cs.uwaterloo.ca) (Quit: Leaving.)
- # [18:21] * Joins: maxli (maxli@moz-favcmk.cs.uwaterloo.ca)
- # [18:34] * Quits: maxli (maxli@moz-favcmk.cs.uwaterloo.ca) (Quit: Leaving.)
- # [18:38] * icaaq is now known as icaaq|afk
- # [18:40] * davidb|afk is now known as davidb
- # [18:41] * Quits: clown (clown@moz-v6m.c3o.211.205.IP) (Quit: Leaving.)
- # [18:44] * Joins: clown (clown@moz-v6m.c3o.211.205.IP)
- # [18:46] <~davidb> SteveF++
- # [18:49] * davidb is now known as davidb|afk
- # [18:53] <@firebot> New Core - Disability Access APIs bug 1113210 filed by dkeeler@mozilla.com.
- # [18:53] <@firebot> https://bugzil.la/1113210 — NEW, nobody@mozilla.org — accessible/atk/AtkSocketAccessible.cpp: unused function 'mai_atk_socket_get_instance_private'
- # [18:54] * davidb|afk is now known as davidb
- # [18:59] * Quits: clown (clown@moz-v6m.c3o.211.205.IP) (Quit: Leaving.)
- # [19:00] * Joins: jongund (chatzilla@moz-emft4q.dres.uiuc.edu)
- # [19:00] * Joins: clown_mtg (clown@moz-v6m.c3o.211.205.IP)
- # [19:10] * Gijs is now known as Gijs_away
- # [19:10] * Quits: SteveF (chatzilla@moz-80j6qs.cable.virginm.net) (Ping timeout: 121 seconds)
- # [19:10] * Joins: jamesn (jnurthen@moz-4cflac.oracle.com)
- # [19:22] <@yzen> eeejay hi
- # [19:23] <eeejay> hey
- # [19:23] <@yzen> eeejay remember we used to have Gip failures for gaia because of accessibility being always on ?
- # [19:27] * Gijs_away is now known as Gijs
- # [19:28] * Joins: Ryuno-Ki (Ryuno-Ki@moz-06f9am.dip0.t-ipconnect.de)
- # [19:29] <eeejay> brb
- # [19:31] <tbsaunde> surkov: so, sometimes CacheChilrenInSubtree called from ProcessContentInsertions() can cause accessibles to be added to the tree without show events. any how good ideas on how to fix that?
- # [19:32] <@surkov> tbsaunde: not sure, file a bug describing the problem?
- # [19:35] <tbsaunde> surkov: ok, let me understand why CacheChildrenInSubtree does this
- # [19:35] <tbsaunde> (yet again) I'm so sick of dealing with this
- # [19:36] <@surkov> right, usually these are tricky problems
- # [19:49] * Joins: SteveF (chatzilla@moz-80j6qs.cable.virginm.net)
- # [19:58] * agibson is now known as agibson|afk
- # [20:07] * agibson|afk is now known as agibson
- # [20:07] * Joins: SteveF_ (chatzilla@moz-80j6qs.cable.virginm.net)
- # [20:08] * Quits: SteveF (chatzilla@moz-80j6qs.cable.virginm.net) (Ping timeout: 121 seconds)
- # [20:08] * SteveF_ is now known as SteveF
- # [20:24] * Quits: clown_mtg (clown@moz-v6m.c3o.211.205.IP) (Quit: Leaving.)
- # [20:24] * Quits: agibson (agibson@moz-j04gi9.cable.virginm.net) (Quit: )
- # [20:33] <eeejay> yzen: sotty. yes
- # [20:33] <eeejay> sorry
- # [20:34] <@yzen> eeejay so somehow im getting some very crazy failures (not im mine, nor in marionette unit test) because i m creating an accessible retreival
- # [20:34] <eeejay> yzen: hm.
- # [20:36] <@yzen> https://treeherder.allizom.org/ui/#/jobs?repo=mozilla-inbound&revision=eb6428c5e590
- # [20:43] * davidb is now known as davidb|afk
- # [20:51] * davidb|afk is now known as davidb
- # [20:57] <eeejay> yzen: which failures, specifically?
- # [20:57] * davidb is now known as davidb|afk
- # [20:57] * Quits: jamesn (jnurthen@moz-4cflac.oracle.com) (Quit: Leaving)
- # [20:57] * davidb|afk is now known as davidb
- # [20:58] <@yzen> the web-platform ones for example
- # [20:58] <@yzen> so if get the accessible retrieval lazily they will pass
- # [20:58] <@yzen> W(...)
- # [21:02] * davidb is now known as davidb|afk
- # [21:03] * Quits: jongund (chatzilla@moz-emft4q.dres.uiuc.edu) (Quit: ChatZilla 0.9.91 [Firefox 34.0.5/20141126041045])
- # [21:08] * Quits: Ryuno-Ki (Ryuno-Ki@moz-06f9am.dip0.t-ipconnect.de) (Quit: WeeChat 1.0.1)
- # [21:09] * Joins: yliao (yliao@moz-2hf62a.HINET-IP.hinet.net)
- # [21:12] * Joins: clown (clown@moz-v6m.c3o.211.205.IP)
- # [21:13] * Quits: yliao (yliao@moz-2hf62a.HINET-IP.hinet.net) (Ping timeout: 121 seconds)
- # [21:20] * Quits: SteveF (chatzilla@moz-80j6qs.cable.virginm.net) (Ping timeout: 121 seconds)
- # [21:27] * davidb|afk is now known as davidb
- # [21:33] <~davidb> tbsaunde, hate to distract ya but do you have any ideas about bug 1113210?
- # [21:33] <@firebot> https://bugzil.la/1113210 — NEW, nobody@mozilla.org — accessible/atk/AtkSocketAccessible.cpp: unused function 'mai_atk_socket_get_instance_private'
- # [21:38] <tbsaunde> davidb: I thought I commented
- # [21:39] <~davidb> tbsaunde, ah then i guess we take the patch?
- # [21:40] <tbsaunde> davidb: didn't see the patch, I can review in a bit
- # [21:40] <~davidb> ok i'll bump it to you
- # [21:45] * Joins: SteveF (chatzilla@moz-80j6qs.cable.virginm.net)
- # [21:50] * Joins: agibson (agibson@moz-j04gi9.cable.virginm.net)
- # [22:01] <tbsaunde> moveable rr recordings would be sweet right about now
- # [22:04] * Quits: agibson (agibson@moz-j04gi9.cable.virginm.net) (Quit: )
- # [22:08] <tbsaunde> surkov: have a few minutes to see if you can reproduce something?
- # [22:09] <@surkov> sure, what do I need for that?
- # [22:10] <tbsaunde> surkov: I think just opening google with a debugger open
- # [22:11] <tbsaunde> I'm seeing ProcessContentInserted() called with an array of one element which is a br in a div in a html input and when its called for that it creates an accessible from CacheChildrenInsubtree
- # [22:13] <@surkov> ok, next?
- # [22:13] * icaaq|afk is now known as icaaq
- # [22:14] <tbsaunde> surkov: you see that too?
- # [22:15] <@surkov> tbsaunde: I didn’t try but that sounds reasoable, what’s wrong do you see in that?
- # [22:16] <@surkov> ProcessContentInserted() is supposed to call CacheChildrenInsubtree to create a subtree
- # [22:16] <tbsaunde> surkov: well, it means accessibles are created and added to subtree, but there's no show event
- # [22:16] <@surkov> UpdateTree should take care of it
- # [22:16] <@surkov> is that CacheChildrenInsubtree called from UpdateTree()?
- # [22:16] <tbsaunde> surkov: should all accessible creation happen through UpdateTree() ?
- # [22:17] <tbsaunde> surkov: no, directly
- # [22:17] <@surkov> oh, that one
- # [22:18] <@surkov> and that UpdateTree() fails to handle that?
- # [22:18] <tbsaunde> surkov: yeah, seems so
- # [22:18] <@surkov> quite strange
- # [22:18] <tbsaunde> the inserted content UpdateTree is called for is just the br, but some random other div gets a new accessible
- # [22:20] <tbsaunde> surkov: can you into it please?
- # [22:20] <~davidb> is there an editable involved?
- # [22:20] <@surkov> I guess that div is a parent
- # [22:21] <@surkov> I guess I can take a look, can you file a bug?
- # [22:21] <tbsaunde> davidb: yeah, there's some inputs around
- # [22:21] * Quits: Gijs (chatzilla@moz-pnd.q6m.80.5.IP) (Quit: sleep)
- # [22:21] <tbsaunde> surkov: saying what exactly? just a c&p of this log?
- # [22:22] * Quits: SteveF (chatzilla@moz-80j6qs.cable.virginm.net) (Ping timeout: 121 seconds)
- # [22:22] <@surkov> describing a problem
- # [22:23] <~davidb> "accessibles are created and added to subtree, but there's no show event" + log
- # [22:23] <~davidb> plus details
- # [22:24] <tbsaunde> davidb: other than a non portable rr recording there isn't really more details
- # [22:25] <~davidb> ack
- # [22:40] * Quits: ~davidb (davidb@moz-i5m.05u.207.66.IP) (Quit: Blah blah blah)
- # [22:43] * Quits: anvk (anovak@moz-871.mvk.11.204.IP) (Quit: Leaving.)
- # [22:53] * Parts: clown (clown@moz-v6m.c3o.211.205.IP)
- # [22:59] * Joins: yzen_ (yzen@moz-i5m.05u.207.66.IP)
- # [22:59] * ChanServ sets mode: +o yzen_
- # [22:59] * Quits: icaaq (icaaq@moz-n1f1o3.cust.bredbandsbolaget.se) (Quit: 10-79 Notify coroner, over and out)
- # [23:00] * Quits: @yzen (yzen@moz-i5m.05u.207.66.IP) (Ping timeout: 121 seconds)
- # [23:12] * Quits: @yzen_ (yzen@moz-i5m.05u.207.66.IP) (Ping timeout: 121 seconds)
- # [23:15] * Quits: @surkov (surkov@moz-v8g78c.cpe.teksavvy.com) (Client exited)
- # [23:37] * Joins: SteveF (chatzilla@moz-80j6qs.cable.virginm.net)
- # [23:46] * Quits: SteveF (chatzilla@moz-80j6qs.cable.virginm.net) (Ping timeout: 121 seconds)
- # Session Close: Fri Dec 19 00:00:00 2014
Previous day, Next day
Think these logs are useful? Then please donate to show your gratitude (and keep them up, of course). Thanks! — Krijn