/irc-logs / mozilla / #accessibility / 2012-03-26 / end
Options:
- # Session Start: Mon Mar 26 00:00:00 2012
- # Session Ident: #accessibility
- # [00:02] * Quits: sam (sam@moz-7813285F.aztw.cable.virginmedia.com) (Quit: Leaving)
- # [00:54] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [02:22] * Quits: fxa90id (fxa90id@moz-77DA422D.neoplus.adsl.tpnet.pl) (Quit: Leaving)
- # [03:04] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Quit: ChatZilla 0.9.88.1 [Firefox 14.0a1/20120325082828])
- # [03:07] <@firebot> New Firefox - Keyboard Navigation bug 739114 filed by sharat48909@gmail.com.
- # [03:07] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=739114 maj, P2, Firefox 5, nobody, UNCO, mozilla webbrowser 5.0 is crashing in windows 8 version
- # [03:11] <@firebot> tanner.sumo.bugs@gmail.com changed the Component on bug 739114 from Keyboard Navigation to Untriaged.
- # [03:24] * Joins: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com)
- # [03:49] * Joins: brambles (brambles@4CBAB088.F3076E90.1822ACA6.IP)
- # [04:20] * khuey is now known as khuey|away
- # [05:03] * Joins: surkov (surkov@38A7FA2F.65585E1B.33A1AC3C.IP)
- # [05:03] * ChanServ sets mode: +o surkov
- # [05:56] * Joins: icaaq (Adium@moz-7D09279.cust.bredbandsbolaget.se)
- # [06:27] * khuey|away is now known as khuey
- # [06:28] <@tbsaunde> surkov: how do you feel about making GetDocumentNode() inline on nsDocAccessible?
- # [06:29] <@surkov> mm, tbsaunde, you could keep it inline on nsAccessNode like return mContent->OwnerDoc()
- # [06:29] <@surkov> in long term it doesn't work
- # [06:29] <@surkov> anyway if we need it on nsDocAccessibe mostly then I'm fine
- # [06:32] <@tbsaunde> surkov: why doesn't it work long term?
- # [06:33] <@tbsaunde> surkov: I think the only thing we actually call it on is doc accessibles, and they already know there doc is why I suggested that
- # [06:37] <@surkov> tbsaunde: I think we should allow document accessible without mContent
- # [06:37] <@surkov> tbsaunde: then it makes sense
- # [06:39] <@tbsaunde> surkov: ok, but would document accessibles still have a pointer to their content doc?
- # [06:40] <@surkov> nsIDocument? yes
- # [06:41] <@tbsaunde> surkov: yeah
- # [06:41] <@tbsaunde> so then afaik it'll be fine since currently GetDocumentNode() on doc accessibles just returns mDocument
- # [06:42] <@surkov> yes
- # [06:59] <@firebot> New Core - Disability Access APIs bug 739130 filed by trev.saunders@gmail.com.
- # [06:59] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=739130 nor, --, ---, nobody, NEW, make GetDocumentNode() inline on nsDocAccessible
- # [07:12] <Mark_Capella> tbsaunde: do you mind if I copy some of the above log to the bug comments? If might like to work this one later and the extra info would help
- # [07:15] * Quits: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com) (Client exited)
- # [07:26] <@tbsaunde> Mark_Capella: sure
- # [07:30] <@firebot> surkov.alexander@gmail.com granted in-testsuite on bug 726071.
- # [07:30] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=726071 nor, --, mozilla14, markcapella, RESO FIXED, get rid nsAccUtils::GetPositionAndSizeForXULSelectControlItem
- # [07:50] <@tbsaunde> Mark_Capella: what was the number of the bug for enums in nsAccessible?
- # [07:50] <Mark_Capella> bug738886
- # [07:50] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=738886 enh, --, ---, nobody, NEW, Address / Improve enums in nsaccessbile.h
- # [08:15] * khuey is now known as khuey|away
- # [08:30] * Quits: @surkov (surkov@38A7FA2F.65585E1B.33A1AC3C.IP) (Quit: surkov)
- # [08:54] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [08:57] * Joins: aaronlev (aaronlev@BCFC8969.9EBA4B5B.A641CD9D.IP)
- # [08:58] * Joins: NimeshNeema (u2689@moz-160C58C6.com)
- # [09:03] * Joins: victorporof (victorporo@C5030E4F.2484615.4A6B528C.IP)
- # [09:05] * Quits: victorporof (victorporo@C5030E4F.2484615.4A6B528C.IP) (Quit: Linkinus - http://linkinus.com)
- # [09:06] * Quits: @hub (hub@moz-E2FCA694.figuiere.net) (Ping timeout)
- # [09:19] * Joins: victorporof (victorporo@C5030E4F.2484615.4A6B528C.IP)
- # [09:37] * Joins: surkov (surkov@38A7FA2F.65585E1B.33A1AC3C.IP)
- # [09:37] * ChanServ sets mode: +o surkov
- # [09:45] * Joins: MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net)
- # [09:45] * ChanServ sets mode: +o MarcoZ
- # [09:57] * Quits: icaaq (Adium@moz-7D09279.cust.bredbandsbolaget.se) (Quit: Leaving.)
- # [09:57] * Joins: icaaq (Adium@714E29CB.13DB46CE.3B93FF6D.IP)
- # [09:57] <@firebot> marco.zehe@googlemail.com granted review for attachment 608785 on bug 727942.
- # [09:57] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=727942 nor, --, ---, surkov.alexander, ASSI, Windows app can't read Firefox setting for "zoom level"
- # [10:03] * Joins: margle (margle@moz-FD55C096.dsl.mweb.co.za)
- # [10:12] <Mark_Capella> surkov: can i nag @ u for reviews on bug734982 and bug737724 ? (maybe u already saw)
- # [10:12] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=734982 nor, --, ---, markcapella, ASSI, Map ARIA role "form".
- # [10:12] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=737724 nor, --, ---, markcapella, ASSI, make IsDefunct() inline
- # [10:13] <@surkov> Mark_Capella: I'm on it
- # [10:13] <Mark_Capella> :) very nice
- # [10:16] <Mark_Capella> tbsaunde: could i also get a more lengthy explanation of what you're looking for on bug 735645 ?
- # [10:16] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=735645 nor, --, ---, nobody, NEW, expose sub and sup elements in text attributes
- # [10:16] * Quits: @MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net) (Ping timeout)
- # [10:17] * Joins: MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net)
- # [10:17] * ChanServ sets mode: +o MarcoZ
- # [10:19] <@tbsaunde> Mark_Capella: sure
- # [10:20] <Mark_Capella> :) more nice
- # [10:20] <@tbsaunde> Mark_Capella: anything in particular you want to know?
- # [10:22] <Mark_Capella> just something in english ... im reading the bug, and the one it refers back to ... whats missing? sub/sup seems to work ...
- # [10:22] <@tbsaunde> Mark_Capella: what do you mean "seems to work"?
- # [10:27] <@tbsaunde> Mark_Capella: do you mean you can add a mochitest for the sup and sub html elements having the correct text attribute and it passes with no changes to c++?
- # [10:29] <Mark_Capella> dang ... I'm probably coming @ this back-asswards ... I can't see the test that fails ...
- # [10:30] <@tbsaunde> Mark_Capella: well, faik there isn't a test yet
- # [10:30] <@tbsaunde> *afaik
- # [10:33] <@tbsaunde> Mark_Capella: you should add tests that the text in <sub>some subscript text</sub> has the textPosition "sub"
- # [10:33] <@tbsaunde> text attribute
- # [10:33] <@tbsaunde> and the reverse for <sup> and the sup text position
- # [10:37] <Mark_Capella> ok ... I'll wait to "take" the bug until I understand it better I'm still a [good first bug] kinda guy :)
- # [10:37] <@tbsaunde> Mark_Capella: what's hard to understand?
- # [10:46] <@tbsaunde> Mark_Capella: you see what GetTextPosValue() does right?
- # [10:47] <Mark_Capella> yah - i believe i follow that routine
- # [10:47] <@firebot> surkov.alexander@gmail.com denied review for attachment 608961 on bug 737724.
- # [10:47] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=737724 nor, --, ---, markcapella, ASSI, make IsDefunct() inline
- # [10:47] <Mark_Capella> :( dang
- # [10:48] <@surkov> Mark_Capella: you might need a help from tbsaunde to fix concerns
- # [10:48] <Mark_Capella> surkov: ok ... i saw the deny ... brb
- # [10:49] <@surkov> Mark_Capella: because we should regress in crashes I think
- # [10:49] <@tbsaunde> Mark_Capella: so, that function handles the case that text has been made a superscript or a subscript using css, but we should also handle the case it is super script or subscript because of native markup
- # [10:50] <Mark_Capella> tbsaunde: ok
- # [10:50] <@tbsaunde> so, in that method you should get the content for the frame that was passed in and get the tag to see if it is a html sub or sup element
- # [10:51] <@tbsaunde> and then return an element of the enum acordingly
- # [10:52] <Mark_Capella> ok - i had that germ of an idea somewhere along the line ... "frame" is new/foreign concept ...
- # [10:53] <Mark_Capella> internal structures still get me
- # [10:54] * Quits: victorporof (victorporo@C5030E4F.2484615.4A6B528C.IP) (Quit: Leaving...)
- # [10:56] <Mark_Capella> i had a bug with alex and I couldnt translate between nsIContent* and a nsAccessible* for example
- # [10:56] <Mark_Capella> too me forever to make the leap
- # [10:59] <@firebot> surkov.alexander@gmail.com granted review for attachment 608962 on bug 734982.
- # [10:59] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=734982 nor, --, ---, markcapella, ASSI, Map ARIA role "form".
- # [11:04] <@tbsaunde> Mark_Capella: make sense now?
- # [11:05] <Mark_Capella> tbsaunde: a heck of a lot moreso ! :) I've got more to go on now ... thats all I need for the moment
- # [11:06] <Mark_Capella> and I just had a patch ripped up :) so I have to look back to that first ...
- # [11:09] <@tbsaunde> sure
- # [11:19] <Mark_Capella> oh i was wrong it passed review with nits :)
- # [11:24] <@MarcoZ> Hi gentlemen!
- # [11:24] <@tbsaunde> hey MarcoZ
- # [11:25] <@tbsaunde> surkov: you still wnat to densify class names right?
- # [11:26] <@tbsaunde> actually, I don't want to worry about that until we actually have a generic folder in the tree :)
- # [11:46] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [12:09] * Quits: margle (margle@moz-FD55C096.dsl.mweb.co.za) (Quit: Computer has gone to sleep.)
- # [12:13] <icaaq> Hi, I'm looking for a open source screen magnification app that works for osx, any suggestion?
- # [12:18] <@firebot> trev.saunders@gmail.com cancelled superreview?(neil@httl.net) for attachment 608942 on bug 648265.
- # [12:18] <@firebot> trev.saunders@gmail.com requested superreview from neil@httl.net for attachment 609270 on bug 648265.
- # [12:18] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=648265 nor, --, ---, trev.saunders, NEW, provide dexcomed table interface version
- # [12:22] * Joins: victorporof (victorporo@C5030E4F.2484615.4A6B528C.IP)
- # [12:31] <@surkov> tbsaunde: what does densify class names mean?
- # [12:33] <@firebot> surkov.alexander@gmail.com requested review from trev.saunders@gmail .com for attachment 608887 on bug 718627.
- # [12:33] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=718627 nor, P1, ---, hub, ASSI, [Mac] Navigating by character, or interacting with, the text in the awesome bar does not speak the c
- # [12:35] <@surkov> MarcoZ, did you hear about open sourse osx screen magnifiers (cc icaaq)?
- # [12:36] * Joins: askalski (akuda@moz-4C8A107E.pool85-48-91.dynamic.orange.es)
- # [12:36] * ChanServ sets mode: +o askalski
- # [12:36] <@MarcoZ> surkov: Nope. Apple provide their own Zoom functionality in OS X, and as far as I know, like for screen readers, nobody developed a screen magnifier.
- # [12:37] <icaaq> MarcoZ: ok, thanks. was looking for alternatives :) cc surkov
- # [12:42] * Joins: margle (margle@moz-FD55C096.dsl.mweb.co.za)
- # [12:42] <@MarcoZ> icaaq: I guess since Apple includes its integrated solution, nobody bothered.
- # [12:42] <icaaq> MarcoZ: yes
- # [12:44] <@tbsaunde> surkov: I meant do things like nsRootAccessible -> RootAccessible, but realized we should wait for the table patch to land before worrying about it
- # [12:44] <@surkov> tbsaunde: that's right
- # [12:48] <@tbsaunde> surkov: so, I'm still not clear on how a mTreeView goes away before we shut the accessible down
- # [12:49] <@surkov> tbsaunde: TreeViewChanged I think
- # [12:49] <@surkov> do something like treeBoxObj.view = null in java script and you do that
- # [12:53] <@askalski> hi everyone
- # [12:53] <@tbsaunde> hi askalski you were looking for me yesterday?
- # [12:53] <@askalski> tbsaunde, yes. I have a question
- # [12:54] <@askalski> tbsaunde, is there a way to do check what binary created a particular accessible node?
- # [12:55] <@askalski> tbsaunde, eeejay in his project look for firefox window by looking for "document:loaded" event, which is a possible source of instability in general case
- # [12:55] <Mark_Capella> surov: you did review+ on bug734982 but also asked for changes ... do I post another attachment and ask for review again?
- # [12:55] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=734982 nor, --, ---, markcapella, ASSI, Map ARIA role "form".
- # [12:55] <@askalski> tbsaunde, I would like to look for a simple "give me /home/andrzej/bin/firefox-bin window"
- # [12:55] <Mark_Capella> surkov: ---^
- # [12:59] <@askalski> tbsaunde, any ideas?
- # [13:00] <@tbsaunde> askalski: well, you can get the name of the application by crawling up the tree if you have an accessible
- # [13:01] <@tbsaunde> but I don't believe you can get the binary that that application is for
- # [13:01] <@askalski> tbsaunde, ok. thanks.
- # [13:02] <@tbsaunde> askalski: or are you trying to get the application accessible for an app given its binary?
- # [13:04] <@askalski> tbsaunde, that would work even better
- # [13:05] <@tbsaunde> askalski: well, then the simple solution is to arrange for there to be no other accessible pplications other than the one you want to test
- # [13:05] <@askalski> tbsaunde, now the only way I know I can grab an accessible right now is to get desktop and dive in the tree :)
- # [13:05] <@askalski> askalski, arrange? You mean - run in virtualized environment, right?
- # [13:05] <@askalski> tbsaunde, ^
- # [13:05] <@tbsaunde> askalski: that and grabing accessibles as targets of events is the way ats usually do it...
- # [13:05] <@askalski> tbsaunde, and what about virtual desktops?
- # [13:06] <@askalski> tbsaunde, I could create a virtual desktop, call it some way etc
- # [13:06] <@tbsaunde> askalski: well, by arrange I mean set up the world so its true, virtualizing the enviroment would be one way to do that
- # [13:07] <@tbsaunde> askalski: if you want to test the firefox tree what's wrong with just walking through the kids of the deskot until you find one with name "firefox" and then you have the rapplication accessible for us
- # [13:07] * Quits: @MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net) (Ping timeout)
- # [13:08] <@askalski> tbsaunde, yes, I imagined this option. I just thought there might be more friendly ways to do it
- # [13:08] * Joins: MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net)
- # [13:08] * ChanServ sets mode: +o MarcoZ
- # [13:08] <@tbsaunde> askalski: na, its perfectly fine
- # [13:09] <@askalski> tbsaunde, in the beginning of my employment, someone told me to use some x??? screen virtualizer for dekstop. do you remember it's name?
- # [13:09] <@tbsaunde> askalski: to do what?
- # [13:09] <@askalski> tbsaunde, there was some 4-letter command containing x that created a virtual x-display
- # [13:10] <@tbsaunde> askalski: well, there's xvfb and xnest, but I'm really confused what your trying to do here...
- # [13:10] <@firebot> markcapella@twcny.rr.com requested review from surkov.alexander@gm ail.com for attachment 609279 on bug 734982.
- # [13:10] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=734982 nor, --, ---, markcapella, ASSI, Map ARIA role "form".
- # [13:11] <@askalski> tbsaunde, in pyatspi I have a way to list desktops. If I created one artificial one for firefox (and gave it a very strange name) that would be a good approach to ensure I will not grab coincidentally some other "firefox" window
- # [13:13] * Quits: victorporof (victorporo@C5030E4F.2484615.4A6B528C.IP) (Ping timeout)
- # [13:13] <@tbsaunde> askalski: well, pyatspi only has one desktop with multiple applications under it
- # [13:14] <@tbsaunde> you might be able t get two completely seperate setups of pyatspi and the at-spi registryd, but that would be tricky and take a lot of work
- # [13:14] <@surkov> Mark_Capella: you don't need to ask for review again I think, those changes are nits
- # [13:14] <@askalski> tbsaunde, then why does it have a "getDesktopCount" method at all?
- # [13:15] <Mark_Capella> wasnt sure of the rules, i changed it, so thought new patch needed new review
- # [13:15] <@tbsaunde> askalski: becuase some sun engineers decided to over engineer things 10 years ago :/
- # [13:15] <@tbsaunde> askalski: there's a bug to clean that up somehow iirc
- # [13:16] <@tbsaunde> askalski: its kind of a dumb simple minded approach, but for now I think you should just grab the first firefox application you find
- # [13:16] <Mark_Capella> surkov: id still have to re-attach the new patch right ...? I haven't found a way to edit the patch directly in the bugzilla
- # [13:16] <@surkov> Mark_Capella: sure
- # [13:17] <@tbsaunde> tht'll always be good enough for build machines, and if someone running linux cares enough they can figure out how to make it deal with other firefoxen running at the same time
- # [13:17] <Mark_Capella> ok ... then add the new attachment ... dont bother to flag with review? request
- # [13:17] <@askalski> tbsaunde, you're probably right, I am overthinking it. I will run binary with a argument to a prepared page and grab the firefox window that opens that page
- # [13:18] <@firebot> surkov.alexander@gmail.com granted review for attachment 609279 on bug 734982.
- # [13:18] <Mark_Capella> whoever checks it in figures out which attachment to use
- # [13:18] <@tbsaunde> askalski: sounds good
- # [13:18] <Mark_Capella> (in this case you, so you already know)
- # [13:19] <@askalski> tbsaunde, another question
- # [13:19] <@askalski> tbsaunde, it's dumb, but I couldn't find any way to interact with firefox window via mozbase
- # [13:19] <@tbsaunde> askalski: in fact you could make the prepared page be something like /tmp/aasdasfewfwsdas.html and then grab the first firefox that has that as the name of one of its windows
- # [13:20] <@tbsaunde> askalski: ok, well, I don't really know anything about mozbase, but go on
- # [13:20] <@askalski> tbsaunde, I can grab stuff via pyatspi (like buttons) and just "click" them via pyatspi, right?
- # [13:21] <@tbsaunde> askalski: yes, hrough actions on the accessible
- # [13:22] <@askalski> tbsaunde, great, thanks
- # [13:22] <@tbsaunde> np
- # [13:47] <@firebot> New Core - Disability Access APIs bug 739179 filed by markcapella@twcny.rr.com.
- # [13:47] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=739179 min, --, ---, nobody, NEW, Provide test for strong table landmark "application" in test_aria_roles.html
- # [14:09] * Quits: margle (margle@moz-FD55C096.dsl.mweb.co.za) (Quit: Computer has gone to sleep.)
- # [14:13] * Quits: @MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net) (Ping timeout)
- # [14:14] * Joins: margle (margle@moz-FD55C096.dsl.mweb.co.za)
- # [14:15] * Joins: MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net)
- # [14:15] * ChanServ sets mode: +o MarcoZ
- # [14:16] <@MarcoZ> Geez, crappy connection today.
- # [14:18] <@askalski> tbsaunde, is it normal, that I found a lot of unrecent a11y nodes in my desktop tree?
- # [14:18] <@askalski> tbsaunde, like nodes from a website I haven't open for a long time now?
- # [14:21] * Quits: margle (margle@moz-FD55C096.dsl.mweb.co.za) (Quit: Computer has gone to sleep.)
- # [14:23] * Quits: @MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net) (Connection reset by peer)
- # [14:23] * Joins: MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net)
- # [14:23] * ChanServ sets mode: +o MarcoZ
- # [14:32] * Quits: @MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net) (Quit: Need to take Hexe to the vet. bbl!)
- # [14:34] * Joins: fxa90id (fxa90id@moz-77DA422D.neoplus.adsl.tpnet.pl)
- # [14:35] <@tbsaunde> askalski: off hnd that seems bad
- # [14:36] <@askalski> tbsaunde, what do you mean?
- # [14:37] <@firebot> New Core - Disability Access APIs bug 739190 filed by surkov.alexander@gmail.com.
- # [14:37] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=739190 nor, --, ---, nobody, NEW, change AppendTarget(nsIContent* aContent) to take nsDocAccessible* argument
- # [14:38] <@tbsaunde> askalski: I mean I'd think that should normally happen
- # [14:38] <@tbsaunde> err, ;shouldn't normally happen
- # [14:45] <@firebot> New Core - Disability Access APIs bug 739191 filed by surkov.alexander@gmail.com.
- # [14:45] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=739191 nor, --, ---, nobody, NEW, get rid nsAccUtils::GetDocAccessibleFor(const nsIPresShell* aPresShell)
- # [14:49] <@askalski> tbsaunde, yeah, I thought so too
- # [14:51] <@firebot> New Core - Disability Access APIs bug 739193 filed by surkov.alexander@gmail.com.
- # [14:51] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=739193 nor, --, ---, nobody, NEW, nsTextEquivUtils::AppendTextEquivFromContent shouldn't use GetAccService()->GetAccessible
- # [14:51] <@tbsaunde> askalski: I have no idea why though, you'd have to debug and find out who keeps holding refs somewho
- # [14:52] <@tbsaunde> *somehow
- # [14:52] <@askalski> tbsaunde, yeah. the funny thing is that these pages were not loaded for couple restarts now
- # [14:52] <@askalski> tbsaunde, that's super weird
- # [14:54] <@tbsaunde> askalski: oh, so you can see them in pyatspi tree but not in domi one?
- # [14:56] <@askalski> tbsaunde, I have seen them in Accerciser, which uses pyatspi, and nowhere on the desktop!
- # [14:57] <@tbsaunde> askalski: well, if you can't see them with domi I'd guess its somesort of pyatspi issue
- # [14:57] <@tbsaunde> so you might want to poke mgorse about it in #a11y on irc.gimp.org
- # [14:58] <@askalski> tbsaunde, cannot reproduce :(
- # [14:58] <@askalski> tbsaunde, and 2) I need to work on something else today
- # [14:58] <@tbsaunde> askalski: sure
- # [14:59] * tbsaunde is now known as tbsaunde|afk
- # [15:06] * Joins: margle (margle@moz-FD55C096.dsl.mweb.co.za)
- # [15:12] <@firebot> New Core - Disability Access APIs bug 739198 filed by surkov.alexander@gmail.com.
- # [15:12] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=739198 nor, --, ---, nobody, NEW, stop GetAccService()->GetAccessible usage in AccEvent::GetAccessibleForNode
- # [15:15] * Quits: margle (margle@moz-FD55C096.dsl.mweb.co.za) (Quit: Textual IRC Client: http://www.textualapp.com/)
- # [15:20] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
- # [15:22] * Joins: davidb (davidb@moz-5E2C9899.eng.wind.ca)
- # [15:32] * Joins: margle (margle@moz-FD55C096.dsl.mweb.co.za)
- # [15:33] * Joins: hub (hub@moz-E2FCA694.figuiere.net)
- # [15:33] * ChanServ sets mode: +o hub
- # [15:35] * Quits: davidb (davidb@moz-5E2C9899.eng.wind.ca) (Quit: Yaaic - Yet another Android IRC client - http://www.yaaic.org)
- # [15:35] * Joins: davidb_ (davidb@F2D29657.F60B0462.67AC9B1.IP)
- # [15:35] <davidb_> hi all!
- # [15:35] * davidb_ is now known as davidb
- # [15:36] * Quits: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: blast off!)
- # [15:36] * Joins: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP)
- # [15:36] * ChanServ sets mode: +qo davidb davidb
- # [15:37] <@davidb> let's try this again
- # [15:41] <@davidb> askalski: did you get the info you needed?
- # [15:41] <@davidb> re platform testing
- # [15:45] * Joins: shorlander (shorlander@moz-853043D6.dhcp.insightbb.com)
- # [15:50] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [15:52] * Quits: NimeshNeema (u2689@moz-160C58C6.com) (Ping timeout)
- # [15:53] * Joins: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
- # [16:02] <@askalski> davidb, yes I did
- # [16:02] <@davidb> good to hear
- # [16:02] <@askalski> davidb, generally it was a very fruitful weekend
- # [16:02] <@davidb> great to hear :)
- # [16:02] <@askalski> davidb, I used Accerciser, and finally got the thing visualised
- # [16:02] <@davidb> in your head you mean?
- # [16:03] <@askalski> davidb, entire accessibility tree. I got into some documentation of Accessible (linux) interface
- # [16:03] <@davidb> ok
- # [16:03] <@askalski> davidb, learned a lot of stuff, understood huge parts of specleniu
- # [16:03] <@davidb> groovy
- # [16:04] <@askalski> davidb, ok. the meeting is in 5 hours, right?
- # [16:04] * @davidb checks
- # [16:04] <@askalski> davidb, then I need to get my workout done soon, so I will go now
- # [16:04] <@davidb> yes in 5 hours
- # [16:04] <@askalski> and be back in 2-3h
- # [16:04] <@davidb> see ya later
- # [16:06] * Quits: @askalski (akuda@moz-4C8A107E.pool85-48-91.dynamic.orange.es) (Quit: Wychodzi)
- # [16:08] * clown is now known as clown_mtg
- # [16:17] * Quits: aaronlev (aaronlev@BCFC8969.9EBA4B5B.A641CD9D.IP) (Quit: aaronlev)
- # [16:18] * Joins: jprmc (jprmc@A0505678.6E78B65D.CDBA6951.IP)
- # [16:18] * ChanServ sets mode: +o jprmc
- # [16:18] * Joins: victorporof (victorporo@37FE3F80.AB163232.79933D60.IP)
- # [16:19] * Joins: NimeshNeema (u2689@moz-160C58C6.com)
- # [16:20] * Joins: arky (arky@DA3C0EDC.8ED53E9C.FA662B63.IP)
- # [16:21] * Quits: logbot (logbot@moz-622AFC27.glob.com.au) (Ping timeout)
- # [16:25] * Joins: logbot (logbot@moz-622AFC27.glob.com.au)
- # [16:28] * Joins: MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net)
- # [16:28] * ChanServ sets mode: +o MarcoZ
- # [16:31] <@MarcoZ> Hi all!
- # [16:31] <@davidb> hi!
- # [16:32] <@davidb> MarcoZ: how is Hexe?
- # [16:34] * Quits: victorporof (victorporo@37FE3F80.AB163232.79933D60.IP) (Connection reset by peer)
- # [16:41] * tbsaunde|afk is now known as tbsaunde
- # [16:42] <@MarcoZ> davidb: Getting better. The wound is healing well.
- # [16:43] <@hub> what happened?
- # [16:46] * Quits: fxa90id (fxa90id@moz-77DA422D.neoplus.adsl.tpnet.pl) (Ping timeout)
- # [16:50] <@MarcoZ> hub: She tore off part of her left-mot nail from her left paw.
- # [16:51] <@MarcoZ> s/mot/most/
- # [16:51] * Joins: aaronlev (aaronlev@BCFC8969.9EBA4B5B.A641CD9D.IP)
- # [16:51] <@MarcoZ> Like tearing a fingernail.
- # [16:51] <@MarcoZ> And the skin below that is very soft and vulnerable, so she needs to wear protective gear around it so the paw doesn't get infected by dirt.
- # [16:52] * Quits: margle (margle@moz-FD55C096.dsl.mweb.co.za) (Quit: Computer has gone to sleep.)
- # [16:53] <@tbsaunde> surkov: is there anything left to be reviewed on bug 718627? it seems like you got to all of it
- # [16:53] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=718627 nor, P1, ---, hub, ASSI, [Mac] Navigating by character, or interacting with, the text in the awesome bar does not speak the c
- # [16:53] <@surkov> tbsaunde: I think I'd like you to take a look, probably at next iteration
- # [16:57] <@hub> MarcoZ: ouch
- # [16:58] <@hub> surkov, tbsaunde: I won't be able to touch it before Wednesday anyway
- # [16:58] <@tbsaunde> surkov: ok, sounds good
- # [16:58] <@tbsaunde> hub: ok
- # [16:59] <@hub> I'm in PTO until Wednesday
- # [16:59] <@tbsaunde> hub: ok, enjoy it :)
- # [16:59] <@surkov> ok
- # [17:00] <@tbsaunde> surkov: it looks like we could optimize AccEvent::GetDocAccessible() in the case of there being an mAccessible
- # [17:00] <@tbsaunde> should be a good first bug and probably worth to do
- # [17:01] <@surkov> that's right
- # [17:03] <@MarcoZ> Hub: Enjoy your PTO! :)
- # [17:04] * @davidb waves
- # [17:04] <@firebot> trev.saunders@gmail.com changed the Target Milestone on bug 734982 from --- to mozilla14.
- # [17:04] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=734982 nor, --, mozilla14, markcapella, ASSI, Map ARIA role "form".
- # [17:11] * Quits: JulienP (julien.pic@moz-DFEC0675.opera.com) (Quit: )
- # [17:14] * Joins: JulienP (julien.pic@moz-DFEC0675.opera.com)
- # [17:19] * Quits: aaronlev (aaronlev@BCFC8969.9EBA4B5B.A641CD9D.IP) (Quit: aaronlev)
- # [17:21] * Joins: victorporof (victorporo@37FE3F80.AB163232.79933D60.IP)
- # [17:22] * Quits: arky (arky@DA3C0EDC.8ED53E9C.FA662B63.IP) (Quit: Leaving)
- # [17:25] * @MarcoZ waves back.
- # [17:29] * Quits: @surkov (surkov@38A7FA2F.65585E1B.33A1AC3C.IP) (Quit: surkov)
- # [17:34] <@davidb> MarcoZ: I wonder why Dennis is barfing on the text inputs
- # [17:51] <@firebot> New Core - Disability Access APIs bug 739253 filed by trev.saunders@gmail.com.
- # [17:51] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=739253 nor, --, ---, nobody, NEW, optimize AccEvent::GetDocAccessible
- # [18:00] * Quits: @jprmc (jprmc@A0505678.6E78B65D.CDBA6951.IP) (Ping timeout)
- # [18:05] * Joins: nhirata (nhirata.bu@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [18:07] * Quits: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com) (Ping timeout)
- # [18:09] * Joins: peteb-away (ptbrunet@moz-E9B02845.austin.res.rr.com)
- # [18:14] * clown_mtg is now known as clown
- # [18:14] * Joins: shorlande (shorlander@moz-853043D6.dhcp.insightbb.com)
- # [18:14] * Quits: shorlande (shorlander@moz-853043D6.dhcp.insightbb.com) (Input/output error)
- # [18:17] * khuey|away is now known as khuey
- # [18:32] * Joins: jongunderson (chatzilla@moz-528042D5.near.uiuc.edu)
- # [18:39] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [18:43] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [18:57] <@davidb> eeejay: hi
- # [18:57] <eeejay> davidb, hey
- # [18:58] <@davidb> eeejay: you'll be on the call with clint right?
- # [18:58] <eeejay> davidb, it is at 12 PDT?
- # [18:58] <@davidb> yeah
- # [18:59] <@davidb> in 2 hours 10 minutes
- # [18:59] <eeejay> yeap. i'll be there
- # [18:59] <@davidb> groovy
- # [19:00] * Joins: ehsan (ehsan@93B4174C.6E78B65D.CDBA6951.IP)
- # [19:23] <@MarcoZ> davidb: I have no clue. Perhaps he's running a checker over the site and notices that thre are hardly any labels.
- # [19:23] <@davidb> although he did report NVDA and FF versions
- # [19:33] <@MarcoZ> Yeah. It *is* true that the stuff that's in the value attribute is not spoken as the label, but as the contents of the field.
- # [19:33] * khuey is now known as khuey|away
- # [19:33] <@MarcoZ> So one has to wait for "edit" to pass before hearing "first name" in NVDA.
- # [19:33] <@MarcoZ> And I agree, labels *are* the most robust way to label fields. :)
- # [19:34] <@MarcoZ> Or they should use the placeholder attribute.
- # [19:39] * Quits: ehsan (ehsan@93B4174C.6E78B65D.CDBA6951.IP) (Input/output error)
- # [19:45] * Joins: margle (margle@moz-9542A19B.dsl.mweb.co.za)
- # [19:58] <@MarcoZ> Good night!
- # [19:58] * Quits: @MarcoZ (marco.zehe@moz-5E065B2.dip.t-dialin.net) (Quit: l8er)
- # [19:59] * Joins: ehsan (ehsan@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [20:03] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [20:04] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [20:07] * khuey|away is now known as khuey
- # [20:09] * Joins: jprmc (jprmc@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [20:09] * ChanServ sets mode: +o jprmc
- # [20:18] * Quits: jongunderson (chatzilla@moz-528042D5.near.uiuc.edu) (Ping timeout)
- # [20:28] * Joins: askalski (akuda@moz-4C8A107E.pool85-48-91.dynamic.orange.es)
- # [20:28] * ChanServ sets mode: +o askalski
- # [20:30] <@askalski> hi I am back :)
- # [20:31] * Quits: victorporof (victorporo@37FE3F80.AB163232.79933D60.IP) (Connection reset by peer)
- # [20:35] * Joins: victorporof (victorporo@37FE3F80.AB163232.79933D60.IP)
- # [20:35] <@firebot> mbrubeck@mozilla.com changed the Status on bug 736059 from NEW to RESOLVED.
- # [20:35] <@firebot> mbrubeck@mozilla.com set the Resolution field on bug 736059 to FIXED.
- # [20:35] <@firebot> mbrubeck@mozilla.com changed the Target Milestone on bug 736059 from --- to mozilla14.
- # [20:35] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=736059 nor, --, ---, nobody, NEW, do some cleanups of nsAccessible::GetAttributesInternal
- # [20:35] <@davidb> askalski: good. Note I'm not going to drive this meeting but am hopeful we can derive next steps.
- # [20:36] <@askalski> davidb, ok. I got 5 questions listed
- # [20:36] <@davidb> heh good
- # [20:36] <@askalski> davidb, technical mostly
- # [20:36] <@davidb> ok
- # [20:36] <@askalski> davidb, other than "what should I test" which I forward to everyone within the team :)
- # [20:36] <@davidb> i'll be sitting with Clint and others here in Toronto
- # [20:37] <@askalski> ok, so I am the only remote?
- # [20:37] <@davidb> eitan
- # [20:37] <@davidb> etc
- # [20:37] <@askalski> davidb, ok
- # [20:37] <@davidb> askalski: vidyo is working for you right?
- # [20:37] <@askalski> davidb, btw, Eitans code is strongly hacked, but after hanging on various IRC it turned out a lot of that is necessary :(
- # [20:37] <@askalski> davidb, good you asked. I will charge tablet now
- # [20:38] <@davidb> please do, and have headset ready
- # [20:38] <@davidb> if you want to do a trial i can
- # [20:39] <@askalski> davidb, we can do now, good idea
- # [20:39] <@davidb> ok
- # [20:40] <@davidb> askalski: i'm in my room
- # [20:40] <@askalski> joining soon
- # [20:42] * Quits: margle (margle@moz-9542A19B.dsl.mweb.co.za) (Quit: Computer has gone to sleep.)
- # [20:43] <@askalski> davidb, yep. the battery was dead
- # [20:43] <@davidb> yikes
- # [20:43] <@davidb> ok
- # [20:43] <@askalski> I will use the cable for the conference
- # [20:44] <@askalski> but it will take me a moment to join
- # [20:44] <@davidb> cable?
- # [20:44] <@askalski> is our headset (the USB one) compatible with android?
- # [20:44] <@askalski> cable = power supply
- # [20:44] <@davidb> ok
- # [20:45] <@davidb> askalski: dunno about headset
- # [20:45] <@davidb> you might need to go with just earphones
- # [20:45] <@askalski> that's ok
- # [20:46] <@askalski> davidb, btw, does firefox support UIA?
- # [20:46] <@davidb> askalski: not directly no. not yet.
- # [20:46] <@davidb> there is a proxy on systems that support com
- # [20:46] <@askalski> davidb, but we plan it eventually, right?
- # [20:47] <@davidb> askalski: we are in discussions right now - i am talking to a lead at MS tomorrow
- # [20:47] <@askalski> davidb, I was looking for backends for a11y testing and I stumbled on a subject again
- # [20:47] <@davidb> my opinion is yes.
- # [20:47] <@davidb> yep
- # [20:47] <@askalski> davidb, right. you want to make sure that the standard will remain open?
- # [20:47] <@askalski> (become?)
- # [20:48] * Quits: clown (clown@67828CC7.C1A51174.9D42CF23.IP) (Connection reset by peer)
- # [20:48] <@davidb> i have concerns around that yes
- # [20:48] <@davidb> but really, it is windows, i can't expect too much
- # [20:49] <@tbsaunde> davidb: why? how is it different from msaa?
- # [20:49] <@askalski> davidb, but users will eventually switch to it sooner or later. that's the power of Microsoft :)
- # [20:49] <@askalski> standard enforcement
- # [20:50] <@davidb> tbsaunde: the design is different, and also it allows a non-COM infrastructure
- # [20:50] <@davidb> tbsaunde: we probably won't see COM on Windows 8 mobile arm.
- # [20:50] <@tbsaunde> davidb: oh, hmm, interesting
- # [20:50] <@tbsaunde> ok
- # [20:50] <@davidb> Also Narrator support for the MSAA proxy is tanking
- # [20:50] * Joins: clown (clown@67828CC7.C1A51174.9D42CF23.IP)
- # [20:51] <@tbsaunde> davidb: hm?
- # [20:51] <@davidb> Narrator on Windows 8 is UIA only and only tested with UIA.
- # [20:52] * Joins: margle (margle@moz-9542A19B.dsl.mweb.co.za)
- # [20:52] <@davidb> and if Windows is jealous of OSX … Narrator will continue to see improvements
- # [20:52] <@davidb> I need to talk to Jamie and Mick
- # [20:52] <@davidb> sent them email
- # [20:53] <@tbsaunde> davidb: ok, but narator only using uia is fine if the proxy between msaa and uia works well enough or we could write a ia2 uia proxy
- # [20:54] <@tbsaunde> although at one point surkov thought we might want to do uia for perf reasons
- # [20:54] <@davidb> yep but we still have to deal with non-COM systems
- # [20:54] <@davidb> there are some aspects of UIA that help perf
- # [20:55] <@tbsaunde> sure, but its some api
- # [20:55] <@tbsaunde> where the api presumably has a pretty liberal license so people can use it from their proprietary apps
- # [21:03] * @davidb adds some agenda items https://wiki.mozilla.org/Accessibility/Meetings/2012-03-28
- # [21:03] * @davidb hunts clint and mdas
- # [21:03] <@tbsaunde> davidb: want me to call in?
- # [21:05] <@davidb> tbsaunde: sure
- # [21:05] <@davidb> tbsaunde: are you talking about the mozbase call?
- # [21:05] * Quits: Mark_Capella (chatzilla@moz-DD0C7E4F.twcny.res.rr.com) (Ping timeout)
- # [21:06] <eeejay> anyone in the mozbase call yet?
- # [21:06] <@davidb> eeejay: we're here
- # [21:06] <@davidb> c'mon in
- # [21:07] * Joins: ctalbert (ctalbert@moz-B77DEAEB.mozilla.org)
- # [21:07] <@tbsaunde> davidb: which video room? and yes
- # [21:07] <eeejay> clint's room?
- # [21:07] <ctalbert> eeejay: you can join my vidyo room now
- # [21:07] <ctalbert> we have it working
- # [21:07] <@davidb> 9304
- # [21:09] * Joins: mcote (mcote@moz-FD8EB826.mc.videotron.ca)
- # [21:09] * Joins: mdas (mdas@F2D29657.F60B0462.67AC9B1.IP)
- # [21:09] * Joins: wlach (wlach@F2D29657.F60B0462.67AC9B1.IP)
- # [21:09] <ctalbert> Hello, we can take notes up here: https://etherpad.mozilla.org/a11y-mozbase
- # [21:10] <@davidb> askalski, eeejay ^
- # [21:16] * Quits: @jprmc (jprmc@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
- # [21:19] <@davidb> askalski: https://tbpl.mozilla.org/
- # [21:21] <@davidb> eeejay: can you mute
- # [21:21] <eeejay> done
- # [21:22] <@davidb> thanks - just in case that's the echo
- # [21:25] * Joins: icaaq1 (Adium@moz-7D09279.cust.bredbandsbolaget.se)
- # [21:25] * Quits: icaaq (Adium@714E29CB.13DB46CE.3B93FF6D.IP) (Ping timeout)
- # [21:28] * Quits: ehsan (ehsan@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Input/output error)
- # [21:30] <eeejay> marrionete
- # [21:30] <mdas> https://developer.mozilla.org/en/Marionette
- # [21:31] * Joins: ehsan (ehsan@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [21:31] <@davidb> eeejay: you look like you want to speak?
- # [21:32] <eeejay> a bit
- # [21:32] <@davidb> heh
- # [21:34] <@davidb> resolution: all automation in mozbase
- # [21:35] <@davidb> resolution: for driving UI use marionette
- # [21:35] <eeejay> ok, i didn't understand the scope of mozbase..
- # [21:36] <eeejay> i thought it did automation
- # [21:36] <@davidb> me too
- # [21:36] <eeejay> so we need to use marionette...
- # [21:36] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [21:37] <wlach> eeejay: you can think of mozbase as a set of tools for driving automation that's outside the browser
- # [21:37] <wlach> not sure if that makes any sense
- # [21:37] * Joins: jprmc (jprmc@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [21:37] * ChanServ sets mode: +o jprmc
- # [21:37] <eeejay> and.. it sounds like marionette will support selenium driver API, so migrating speclenium might not be that hard
- # [21:37] <eeejay> wlach, sure
- # [21:38] <mdas> eeejay: yeah we're implementing the webdriver API and it was written so we can run selenium commands without changes required
- # [21:39] <eeejay> yeah... just grab them and make them local
- # [21:39] <eeejay> re: remote tests ^
- # [21:42] <@davidb> ok
- # [21:44] <@davidb> tbsaunde: are you on the call?
- # [21:50] <@tbsaunde> davidb: yeah, but my volume was low and andriod aparently doesn't let you change that while on a call... so I only heard half what you said
- # [21:51] <@davidb> ok
- # [21:52] <@davidb> askalski: clear as mud?
- # [21:52] <@davidb> kidding
- # [21:52] <@davidb> i think we have a plan
- # [21:53] <@tbsaunde> davidb: ok, I hope its small and reasonable :)
- # [21:53] <@davidb> i think so
- # [21:54] <@tbsaunde> so, I actually ended up icking around an idea I've kicked around before of jsut hacking the tests we want into libxul
- # [21:55] <@davidb> tbsaunde: how would that work?
- # [21:55] <@tbsaunde> (during the call for some reason)
- # [21:55] <@tbsaunde> davidb: when we start a11y we install event handlers for platform a11y events
- # [21:56] <@tbsaunde> then, js does some stuff and our event handlers track they've been called and crawl around the tree when invoked
- # [21:56] <@davidb> hum
- # [21:56] <@tbsaunde> done
- # [21:57] <@tbsaunde> knid of hacks, but I'd think would work with fiarly little effort
- # [21:57] <@davidb> i'm unsure how much extra code this would add
- # [21:57] <@tbsaunde> probably not much other than the tests
- # [21:58] <@tbsaunde> but you say eeejay askalski nd the ateam people have a plan that they're probably better at implementing than they would be at those hacks
- # [21:58] <@davidb> yep
- # [21:59] * @tbsaunde shrugs
- # [21:59] * @tbsaunde is really afraid of over engineering this though
- # [21:59] * Quits: ehsan (ehsan@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Input/output error)
- # [22:02] * Joins: aaronlev (chatzilla@BCFC8969.9EBA4B5B.A641CD9D.IP)
- # [22:04] * Quits: nhirata (nhirata.bu@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: nhirata)
- # [22:06] * Joins: nhirata (nhirata.bu@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [22:06] <@davidb> yeah
- # [22:07] * Quits: nhirata (nhirata.bu@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: nhirata)
- # [22:07] <@davidb> eeejay: what do you think about keeping speclenium the way it is, just improving it here and there. and for mozbase ultimately just taking what we need and building something smaller for testing platform regressions?
- # [22:07] <@davidb> tbsaunde, askalski ^
- # [22:08] * Joins: nhirata (nhirata.bu@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [22:09] <eeejay> davidb, askalski's call. but i think it would be nice to keep it all under a common codebase. and not abandon the multi-browser case
- # [22:09] <@davidb> eeejay: would we abandon it?
- # [22:09] <@tbsaunde> davidb: I'm not sure I follow
- # [22:09] * @davidb wants to understand clearly
- # [22:10] <eeejay> davidb, i mean, support mozbase/marionette, but also keep support for selenium for other browsers
- # [22:10] <@tbsaunde> eeejay: by multi browser case you mean multiple browsers on the same machine at the same time? I'd call that nice to have but I don't see why we can't add it later
- # [22:10] <eeejay> ie. extend
- # [22:10] <eeejay> tbsaunde, i mean the ability to run chrome
- # [22:10] <@davidb> or chromium
- # [22:11] <@tbsaunde> eeejay: oic
- # [22:11] <eeejay> tbsaunde, when i wrote speclenium the vision was an a11y acid test
- # [22:11] <@tbsaunde> eeejay: sure
- # [22:11] <@davidb> eeejay: that's what i think speclenium should be
- # [22:11] <@tbsaunde> eeejay: but I don't want to solve that problem here, if it helps solve our problem great, if not and its easier to to just write our wn thing for firefox thats absolutely fine too
- # [22:11] <@davidb> i'm just want to be sure we don't try to fit a square peg in a round hole
- # [22:12] <@tbsaunde> yeah
- # [22:12] <@tbsaunde> the a11y browser acid test is a good idea and I think it would be great if someone does it
- # [22:13] <@davidb> actually i think it is more of a compat test than pure acid test
- # [22:13] <@tbsaunde> but if a gecko specific thing solves our problem with less effort then I think we should do that instead of making the a11y acid test solve our problem
- # [22:13] <@davidb> but i dunno the full power of speclenium
- # [22:13] <eeejay> tbsaunde, yeah. askalski's call. but i think 80% of speclenium is reusable in this project. so instead of just reusing bits, it would be good to extend it
- # [22:13] <@tbsaunde> eeejay: sure
- # [22:14] <@davidb> eeejay: 80% sounds enticing, what is that 80%?
- # [22:14] <@davidb> i mean if you take away the selenium part
- # [22:15] <@tbsaunde> I don't know much of anything about speclenium, so up to askalski
- # [22:16] <@davidb> eeejay: also, you should eat :)
- # [22:16] <@askalski> eeejay, honestly I would prefer to do two things at the same time: get speclenium working again and create a testing app (that might use some portions of speclenium code)
- # [22:17] <eeejay> davidb, testing for accessible objects and events across machine boundaries with python
- # [22:17] <@askalski> eeejay, davidb, do we need "across machine"?
- # [22:18] <eeejay> askalski, it sounds like mozbase has "slaves", etc. so i think so
- # [22:18] <@askalski> eeejay, davidb, I believed that test slaves are single machine env anyway
- # [22:18] <@askalski> cut off from network
- # [22:18] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [22:18] <@tbsaunde> askalski: if its anything like our other test setups no
- # [22:19] <@askalski> eeejay, I agree that tree format and diffing and stuff are very attractive
- # [22:19] <@tbsaunde> eeejay: the usually way we use slaves is less fien grined than that, slaves run a whole test suite and report a log and failed or passed
- # [22:19] <@askalski> eeejay, though - to my reckon - test slaves (target machines) are 1) limited in setup 2) testing only FF in cut-off environment
- # [22:20] <eeejay> tbsaunde, gotcha
- # [22:20] <@askalski> eeejay, deploying entire speclenium there is 1) complex 2) would not give us much advantage
- # [22:20] <@tbsaunde> I can see how diffing trees is useful for the compat testing, but it doesn't seem that useful for our purpose
- # [22:20] <@askalski> tbsaunde, agree
- # [22:21] <@askalski> my suggestion would be to 1) maintain speclenium 2) build something else for slaves
- # [22:21] <@askalski> reusing code wherever possible
- # [22:21] <@askalski> and sensible
- # [22:21] <@davidb> I will say that the simpler the better for stuff running in our test automation. Especially if we want to enlist any non-a11y team support down the road.
- # [22:21] <eeejay> tbsaunde, it is useful in regression testing too. that is how orca does most of its tests
- # [22:22] <@davidb> (I mean simpler as in not too large codebase)
- # [22:22] <@askalski> eeejay, what you think about my idea?
- # [22:22] <@tbsaunde> eeejay: sure, being able to diff to some degree is useful, but one side can be hard coded
- # [22:22] <eeejay> askalski, go with what seems right, and correct the course as you proceed :)
- # [22:22] <@askalski> eeejay, ok
- # [22:23] <@askalski> I believe we have a consensus now :)
- # [22:23] <@davidb> Now that we've bullied poor eeejay.
- # [22:23] * @davidb hands eeejay some ointment
- # [22:23] <@tbsaunde> eeejay: I mean something similar to our mochitests for trees we have a jsobject that is the tree we want to find and we compare that against what we find
- # [22:23] <eeejay> oh, and about pyia. it is just small layer that makes msaa work like at-spi via pyatspi
- # [22:24] <@davidb> clever
- # [22:24] <eeejay> davidb, am i supposed to eat it?
- # [22:24] * @tbsaunde remains terrified by the idea of touching COM from python
- # [22:24] <@askalski> eeejay, yes, I read it, it's nice. I need to learn some stuff about COM
- # [22:24] <eeejay> askalski, so pyatspi is not sacred either, just if you want to extend the tests to at-spi one day, it will make it easier
- # [22:24] * eeejay knows nothing about COM
- # [22:24] <@askalski> eeejay, also, I there is a file "constants.py" - I could try to generate it from .idl
- # [22:25] <eeejay> tbsaunde, it is called NVDA!
- # [22:25] <@davidb> eeejay: i rub it on my wounds and bruises.
- # [22:25] <eeejay> askalski, they haven't changed in 15 years.
- # [22:26] <@tbsaunde> eeejay: I'm not saying it doesn't work
- # [22:26] <@davidb> lol
- # [22:26] * khuey is now known as khuey|away
- # [22:27] <@tbsaunde> eeejay: we have xpcom/ds/nsIByteBuffer.h =P
- # [22:27] <@askalski> eeejay, ok, I just address the comment I found in the beginning of the file
- # [22:27] <eeejay> oh :)
- # [22:27] * @tbsaunde is saddened by the thought of that file
- # [22:27] <@askalski> anyway, the way you like it :) I can leave it as it is
- # [22:27] <eeejay> it is really weird looking at stuff i wrote 4 years ago
- # [22:28] <eeejay> to TODOs suddenly look so vain
- # [22:28] <@tbsaunde> eeejay: yeah, tht can be weird
- # [22:28] <@tbsaunde> eeejay: atleast you don't have 10 year old todos ;)
- # [22:29] * Quits: margle (margle@moz-9542A19B.dsl.mweb.co.za) (Quit: Textual IRC Client: http://www.textualapp.com/)
- # [22:29] <eeejay> tbsaunde, i do. but it is abandonware. i thought speclenium was too
- # [22:29] * Joins: Mark_Capella (chatzilla@moz-EF9E3E68.twcny.res.rr.com)
- # [22:29] <@tbsaunde> eeejay: oh, ok
- # [22:29] * Joins: ehsan (ehsan@93B4174C.6E78B65D.CDBA6951.IP)
- # [22:36] <@askalski> there is a plague of demotivotional posters in Poland about Firefox. all positive about the browser
- # [22:37] <@askalski> want some
- # [22:37] <@askalski> ?
- # [22:37] <@davidb> i'm not sure haha
- # [22:37] <@davidb> fine
- # [22:37] * khuey|away is now known as khuey
- # [22:38] <@askalski> http://demotywatory.pl/3740761/Firefox this is "Firefox, and it's problems with windows"
- # [22:38] <@davidb> heheh
- # [22:39] * Quits: Mark_Capella (chatzilla@moz-EF9E3E68.twcny.res.rr.com) (Ping timeout)
- # [22:39] <@davidb> i'm not sure what to think of a fox licking a window
- # [22:39] <@askalski> davidb, it tries to get in
- # [22:39] <@askalski> or something
- # [22:39] <@davidb> yeah
- # [22:39] <@askalski> http://demotywatory.pl/3743727/Slyszelismy-ze-zainstalowales-najnowszego-IE and this is "we heard you turned down Firefox for new IE"
- # [22:40] <@davidb> heheh
- # [22:40] <@davidb> ok i gotta step offline
- # [22:40] <@davidb> will be back in 4 hours
- # [22:40] <@askalski> me too
- # [22:40] <@davidb> ciao
- # [22:40] <@askalski> me in 10 :D
- # [22:40] <@askalski> bye
- # [22:40] * Quits: @davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: davidb)
- # [22:40] * Quits: @askalski (akuda@moz-4C8A107E.pool85-48-91.dynamic.orange.es) (Quit: Wychodzi)
- # [22:42] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [22:45] * Quits: ehsan (ehsan@93B4174C.6E78B65D.CDBA6951.IP) (Ping timeout)
- # [22:47] * Joins: ehsan (ehsan@93B4174C.6E78B65D.CDBA6951.IP)
- # [22:48] <@tbsaunde> ok, I'm off to eat too
- # [22:48] * tbsaunde is now known as tbsaunde|afk
- # [22:52] * Quits: clown (clown@67828CC7.C1A51174.9D42CF23.IP) (Quit: Leaving.)
- # [22:55] * Quits: aaronlev (chatzilla@BCFC8969.9EBA4B5B.A641CD9D.IP) (Ping timeout)
- # [23:00] * Quits: @jprmc (jprmc@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: Leaving)
- # [23:01] * Joins: jprmc (jprmc@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [23:01] * ChanServ sets mode: +o jprmc
- # [23:02] <@firebot> akeybl@mozilla.com granted approval-mozilla-beta for attachment 605820 on bug 735666.
- # [23:02] <@firebot> akeybl@mozilla.com granted approval-mozilla-aurora for attachment 605820 on bug 735666.
- # [23:02] <@firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=735666 cri, --, mozilla14, surkov.alexander, RESO FIXED, crash [@ nsIDocument::GetContainer() ]
- # [23:21] * Joins: aaronlev (chatzilla@BCFC8969.9EBA4B5B.A641CD9D.IP)
- # [23:25] * tbsaunde|afk is now known as tbsaunde
- # [23:42] * Quits: icaaq1 (Adium@moz-7D09279.cust.bredbandsbolaget.se) (Quit: Leaving.)
- # [23:58] * Quits: aaronlev (chatzilla@BCFC8969.9EBA4B5B.A641CD9D.IP) (Ping timeout)
- # [23:59] * Joins: aaronlev (chatzilla@BCFC8969.9EBA4B5B.A641CD9D.IP)
- # Session Close: Tue Mar 27 00:00:00 2012
The end :)