UbuntuIRC / 2008 /08 /04 /#ubuntu-mozillateam.txt
niansa
Initial commit
4aa5fce
[07:30] <BigUrsis> Do you guys also track thunderbird?
[09:10] <XioNoX> hi !
[09:10] <gnomefreak> hi
[09:24] <gnomefreak> asac: The following packages will be REMOVED: enigmail thunderbird thunderbird-gnome-support
[09:24] <gnomefreak> The following NEW packages will be installed: hunspell-fr
[09:24] <gnomefreak> that is new
[09:27] <XioNoX> firefox is using hunspell since firefox 3, but the ubuntu version keep myspell because it was used by many other program
[09:27] <asac> gnomefreak: yes. we are on it
[09:27] <gnomefreak> i know
[09:28] <gnomefreak> i thought Ubuntu was supposed to migrate to hunspell and drop myspell if we could
[09:28] <gnomefreak> not sure what requires myspell anymore
[09:47] <gnomefreak> as is there a reason there is no hunspell-en?
[10:17] <XioNoX> asac ?
[10:22] <gnomefreak> seems scrollkeeper is keeping back ubuntu-desktop
[10:27] <gnomefreak> where is trash now located?
[10:28] * gnomefreak cant remember
[10:33] <zbraniecki> hi all
[10:33] <XioNoX> zbraniecki, hi
[10:41] <[reed]> zbraniecki: apparently, I can't send mail to Florian
[10:42] <[reed]> his mail provider is blocking me for some reason
=== zbraniecki is now known as gandi
[10:47] <gnomefreak> bug 201655
[10:47] <ubottu> Launchpad bug 201655 in mozilla-thunderbird "send email with attachement from openoffice through thunderbird fails: "unable to open the temporary file"" [Low,Confirmed] https://launchpad.net/bugs/201655
[11:16] <gnomefreak1> where the hell is english
[11:16] <gnomefreak1> and size sucks
[11:16] <gnomefreak> shows english here maybe its just the font is so small
[11:38] <asac> [reed]: florian?
[11:38] <asac> hi gandi
[11:39] <asac> XioNoX: whats up ;)
[11:39] <[reed]> asac: another mozilla europe intern
[11:40] <gandi> asac: hi man!
[11:40] <asac> [reed]: ah ok.
[11:40] <gandi> asac: how's your jetlag?
[11:40] <asac> gandi: thats well ;) ... but i had issues getting to airport as my float flight was cancelled and when i arrived i found water hazard in my flat :(
[11:41] <asac> now living at my mothers place till thats sorted
[11:41] <gandi> ouch
[11:41] <asac> but well ... at least i am at home ;)
[11:41] <gandi> I'm trying to survive my jetlag
[11:41] <asac> yeah ;) ... maybe the jetlag will come through when adrinaline gets settled
[11:49] <Jazzva> hey asac, welcome back. Too bad for your apartment... I hope it will soon be fine :)
[11:49] <Jazzva> anyway, I'm off to a friend to work on a documentation for some school project... need to write a processor simulator.
[11:49] <Jazzva> So, see you tonight.
[11:51] <XioNoX> asac, is is working, without the ainstallButtonPressed, with the help of gandalf
[11:53] <XioNoX> I've to get something to eat
[11:56] <gandi> asac: yea, I just switched XioNoX to go with DOM travelling instead of ID querying
[11:59] <armin76> asac: bumb!
[12:08] <asac> gandi: dom travelling? so convert RDF to structured XML first?
[12:09] <gandi> no
[12:09] <gandi> he's just placing a variable from RDF into XUL tree
[12:09] <gandi> http://pastebin.mozilla.org/506114
[12:09] <gandi> bottom
[12:09] <gandi> label with ID valueURI
[12:09] <gandi> (I asked him to remove ID since it's not unique)
[12:09] <gandi> and then buttInstall
[12:10] <gandi> buttInstall.setAttribute("oncommand", "ainstallButtonPressed(this.previousSibling.getAttribute('value'));");
[12:10] <gandi> erm
[12:10] <gandi> buttInstall.setAttribute("oncommand", "installPlugin(this.previousSibling.getAttribute('value'));");
[12:10] <gandi> anyway, it's all about DOM travel
[12:11] <gandi> nextSibling, previousSibling, firstChild, lastChild, items...
[12:11] <gandi> you know
[12:18] <asac> gandi: ah ok. why not set the value as a "new" attribute to the button?
[12:18] <asac> so we dont need to rely on the "previous" sibling being the label
[12:19] <asac> <button valueURI="?..." oncommand="this.getAttribute..."/> ?
[12:20] <asac> wont wont that work?
[12:22] <asac> gandi: ?
[12:23] <gandi> asac: breaking the schema?
[12:23] <gandi> despite, the label was there before, so I just suggested how to use it
[12:24] <asac> gandi: well. i dont think that the URI should be displayed in general
[12:24] <gandi> oh, maybe
[12:24] <asac> if it is now, it probably just for debuggin purpose
[12:24] <gandi> I didn't parse the code in my head
[12:24] <gandi> :(
[12:25] <asac> gandi: actually i think that it should work to generate javascript code with variables evaluated _before_ ... most likely thats merely and escaping thing. do you know anyone that knows for sure?
[12:26] <gandi> how can it evaluate before?
[12:26] <gandi> we would have to parse RDF template in memory, am I right?
[12:26] <asac> well ... the template is evaluated to create dom, right? I dont think that the variable is not evaluated during that step
[12:27] <asac> so you should be able to use the variable in any text ... which should include javascript
[12:27] <gandi> it may be
[12:27] <asac> that was always my understanding
[12:27] <gandi> from what I see in arzhel's code
[12:27] <asac> if it really gets evaluated not during template evaluation, but during display time
[12:27] <asac> then well. thats borked
[12:27] <gandi> he's just creating a XUL template that feeds data from RDF
[12:28] <asac> right
[12:28] <gandi> which means that the code is evaluated exactly after plugging the table into DOM
[12:28] <gandi> line 93/94
[12:29] <asac> yes. but why wont ?XPIInstall be expanded to its value in javascript if its replaced in arbitrary texts
[12:29] <gandi> well, it's replaced after line 94
[12:29] <gandi> in XUL template
[12:29] <gandi> then JS can feed from it
[12:29] <gandi> that's how I see it
[12:30] <gandi> if you would like to have it purely in JS then the code doesn't have to build XUL template
[12:30] <gandi> but pure XUL
[12:30] <asac> gandi ... what i want is that the xul template generates javascript _with_ the variable in place
[12:30] <gandi> and all variables should be queried directly from RDF JS object
[12:30] <gandi> ah
[12:30] <gandi> ]ok
[12:30] <asac> like oncommand="javascript: alert(?XPIInstall)"
[12:31] <asac> that was the idea
[12:31] <gandi> of course
[12:31] <asac> but apparently it dident work
[12:31] <gandi> ok
[12:31] <asac> which mus be some escaping
[12:31] <gandi> sorry
[12:31] <gandi> I'm tired
[12:31] <asac> unless the variable is never evaluated until getting displayed
[12:31] <gandi> let me rethink what you said
[12:31] <gandi> no, it's evaluated in line 94
[12:31] <gandi> I'm just not sure if you can get it that easily in template
[12:31] <asac> right. but for XioNoX it wasnt when he tried. he inserted the alert above and saw "?XPIInstall" in the alert dialog
[12:31] <gandi> so, you definitely can do this in the way you described
[12:31] <asac> and not the value
[12:32] <gandi> with new variable in <button>
[12:32] <gandi> but that will break XUL sanity
[12:32] <asac> right. that was the workaround i gave him friday to unblock
[12:32] <asac> him
[12:32] <asac> gandi: true.
[12:32] <asac> gandi: the other way would be to use <input type="button" value="...")?
[12:32] <asac> so it could look for value attribute
[12:33] <gandi> yea, that seems to be clean to me
[12:33] <asac> not real xul though i guess
[12:33] <asac> but html
[12:33] <gandi> aagh
[12:33] <gandi> right
[12:33] <gandi> then we'd have to use another namespace
[12:34] <gandi> so for me, label with display:none sounds easiest
[12:34] <gandi> and I can look for a better solution now
[12:34] <asac> gandi: problem is that it will introduce fragility in code if we look for previousSibling
[12:34] <asac> so moving the label somewhere else will break
[12:34] <gandi> well, this can be easily solved by
[12:34] <asac> gandi: do you know someone who can tell us why the generate-into-javascript code option wont work?
[12:35] <gandi> class="valueURI"
[12:35] <gandi> this.parentNode.getElementsByClassName('valueURI')[0]
[12:35] <gandi> yea, looking for him now
[12:35] <asac> ok. still we rely on the element being in the parentNode. all not nice ;)
[12:36] <gandi> are you sure that
[12:36] <asac> lets first check the real solution (javascript code)
[12:36] <asac> if that doesnt work its hacky, but well.
[12:36] <gandi> oncommand=installPlugin(?XPIocation) won't work?
[12:36] <asac> gandi: that was XioNoX told me ... i thought it did
[12:36] <gandi> ok
[12:37] <gandi> let me try to reproduce that
[12:37] <asac> he said it didnt for him. but we shoulod be sure that its not an escaping issue or something
[12:37] <gandi> and I'm looking for a template guy
[12:37] <asac> cool
[12:37] <gandi> I'd say that it probably replaces the variables only if the whole value is the variable name
[12:37] <gandi> so oncommand "?XPILocation" would work
[12:38] <asac> gandi: there are examples where there is:
[12:38] <asac> "my name is ?lastname and please dont bother me"
[12:38] <gandi> hmm
[12:38] <asac> at least i am pretty sure i saw things like that
[12:39] <gandi> yea
[12:39] <gandi> let's see
[12:39] <gandi> maybe there have to be spaces around variable?
[12:39] <asac> maybe
[12:39] <asac> should be easy to try
[12:43] <XioNoX> back
[12:46] <asac> gandi: the other way would be to define a installpluginbutton element in xbl which would basically be a button, but allows the installpluginuri attribute dom whise
[12:46] <asac> wise
[12:46] <gandi> yea
[12:46] <asac> at least in that way we could just say event.target.getAttribute("installpluginuri")
[12:47] <gandi> right
[12:47] <asac> i wouldnt rely on how the dom is layouted
[12:47] <gandi> and that would be sane from the code standpoint
[12:47] <gandi> but it would be harder to maintain
[12:47] <gandi> and read
[12:48] <asac> gandi: well. in xul it would be easy to read. and xbl-templates are used all over the place anyway
[12:49] <gandi> I disagree. Creating xbl for one button would raise the complexity of code comparing to just adding one attribute to <button> manually
[12:49] <asac> <installpluginbutton uri="?XPIInstallLocation" text="&install;" oncommand="installPlugin(event.target.getAttribute("uri"));"/>
[12:49] <asac> owell ...escaped of course
[12:49] <asac> ;)
[12:49] <gandi> but it would be interesting challenge for Arzhel to learn XBL ;)
[12:49] <asac> hehe
[12:49] <asac> yeah.
[12:49] <gandi> especially it's a really easy XBL case
[12:50] <asac> gandi: but please try to get a confirm from a template guru that it really doesnt work to generate it into javascript code
[12:50] <gandi> will do
[12:50] <gandi> first, trying to confirm from XioNoX that it doesn't
[12:50] <asac> good
[12:50] <asac> thanks
[12:50] <asac> have to do something else now (and lunch)
[12:50] <asac> be back in 1-2 hours
[12:53] <XioNoX> ok, i've read everything :D
[12:53] <XioNoX> buttInstall.setAttribute("oncommand", "installPlugin(' ?XPILocation ');");
[12:53] <XioNoX> seem to work
[12:53] <asac> XioNoX: cool ... is ' needed?
[12:54] <asac> maybe ' ?XPILocation '.trim() ;)
[12:54] <asac> hehe
[12:54] <XioNoX> " give a total bug of the window
[12:54] <asac> XioNoX: sure you would have to escape the inner " like \" i guess
[12:54] <XioNoX> and without ' i have an error on the console and it don't work
[12:54] <asac> ok
[12:54] <gandi> yea
[12:54] <asac> XioNoX: try the .trim()
[12:54] <gandi> \" should work
[12:55] <gandi> and trim will help avoid whitespaces
[12:55] <asac> in this case whitespaces might be no problem, but in gneral we dont want whitepsaces
[12:55] * gandi would be a good parser... I understand how they think
[12:55] <asac> haha
[12:55] <XioNoX> what does trim() exactly ? just remove whitespace ?
[12:55] <gandi> well, it's logical if you think about
[12:55] <gandi> XioNoX: yea
[12:55] <gandi> from beginning and end
[12:56] <gandi> asac: imagine <button label="something?valuesomething"/>
[12:56] <gandi> how it can guess which one is variable?
[12:56] <gandi> without spaces?
[12:57] <XioNoX> chrome://mozapps/content/extensions/extensions.xul, line 1: " apt:flashplugin-nonfree?section=multiverse ".trim is not a function
[12:58] <XioNoX> ?XPILocation isn't a variable when trim(); work
[12:58] <XioNoX> buttInstall.setAttribute("oncommand", "installPlugin(' ?XPILocation '.trim());");
[13:00] <gandi> yea
[13:00] <gandi> it seems JS has no native trim() function
[13:00] <gandi> you can use this:
[13:00] <gandi> ''.replace(/^\s+|\s+$/g, '')
[13:00] <gandi> instead
[13:00] <gandi> ' ?XPILocation '.replace(/^\s+|\s+$/g, '')
[13:00] <gandi> should work
[13:01] <XioNoX> k
[13:02] <XioNoX> no error, but still spaces :D
[13:11] <gandi> .replace(/ /g,'')
[13:12] <XioNoX> good!
[13:16] <XioNoX> asac, next steep ? :D
[18:50] <Jazzva> asac, ping
[18:51] <asac> Jazzva: hey
[18:51] <asac> ;)
[18:51] <Jazzva> hey hey :)
[18:51] <asac> i think i am starting to feel better now ;)
[18:51] <Jazzva> Really?
[18:51] <Jazzva> And your home?
[18:51] <asac> not sure :-D
[18:51] <Jazzva> ouch... I hope you'll manage to fix it :)
[18:51] <asac> tomorrow i meet with a guy that wants to repair things
[18:52] <Jazzva> report a bug on LP ;)
[18:52] <asac> lets see how much it costs
[18:52] <asac> yeah me too ... otoh, my mother has 2MB/s synchronous line
[18:52] <Jazzva> hope it won't be too much...
[18:52] <asac> she lives in one of the two small cities where you have kind of fiber to home
[18:52] <asac> in germany
[18:52] <Jazzva> MB or Mb?
[18:53] <asac> not sure ... i did an ssh copy and it gave me 1 Megabyte per-second
[18:53] <asac> but thats the server that cant encrypt more at the same time
[18:53] <asac> get the same in internal net
[18:53] <Jazzva> and I hoped it was just a typo...
[18:53] <asac> hehe
[18:53] * Jazzva is frustrated about his net connection
[18:53] <asac> finally i can upload as hell ;)
[18:53] <Jazzva> yeah... :)
[18:53] <asac> now i just need to convince her that my equipment can stay here ;)
[18:54] <asac> then i work from home with my laptop and build remotely ;)
[18:54] <Jazzva> haha :)
[18:54] <asac> not sure if she likes the idea to get another heating ;)
[18:54] <Jazzva> get her an AC ;)
[18:55] <asac> maybe not a bad deal :)
[18:55] <asac> at least she never complained about broken connectivity :(
[18:55] <asac> in the last 4 years or so
[18:55] <Jazzva> btw, did fta told you the results of the yesterday's meeting?
[18:55] <asac> i read the backlog in channel
[18:55] <asac> thanks for running it
[18:55] <asac> i completely forgot about it while being over-exhausted
[18:56] <asac> from travel-hell into real-life hell :(
[18:57] <asac> Jazzva: i found that we failed to do the XPI.TEMPLATE for thunderbird
[18:57] <asac> (which should be easy)
[18:58] <Jazzva> yep... it should be easy :)
[18:58] <Jazzva> will you do it? I can add them, if you have something else to do...
[18:59] <asac> Jazzva: want to write the team report ;) instead?
[18:59] <asac> hehe
[19:00] <asac> for jul
[19:00] <Jazzva> team report?
[19:00] <Jazzva> m-e-d?
[19:00] <asac> that would be a good start :)
[19:00] <Jazzva> ouch... :)
[19:00] <asac> i think m-e-d is just a statistic table
[19:00] <asac> e.g. how many extensions, how many updated? and how many need work
[19:00] <asac> and how many need more info
[19:00] <Jazzva> yep... still would need to get extensions by date :)
[19:00] <asac> or something like that
[19:01] <asac> Jazzva: what exactly do you mean by that?
[19:01] <Jazzva> hmm... i guess i could just go over pages for extensions and look at the date of the last changelog entry
[19:01] <asac> Jazzva: you could look at ~ubuntu-dev extensions
[19:01] <asac> and see which was touched there
[19:01] <Jazzva> asac, just thinking how to find only extensions that were changed in july :)
[19:01] <asac> -> uploaded
[19:01] <Jazzva> right
[19:01] <asac> then we need the number of extensions in the "need info" table
[19:01] <asac> and those that "are not in the archive"
[19:02] <asac> and those that have an upstream update pending (xpi
[19:02] <asac> )
[19:03] <Jazzva> asac, btw... are you willing to upload firefox-sage and ...something else that failed to build :)?
[19:03] <asac> sure
[19:03] <Jazzva> ok, just let me see what packages :). I know there are two of them
[19:03] <asac> what is with lp:~bdrung/firefox-extensions/stanford-pwdhash.ubuntu ?
[19:03] <asac> that appears to be new contribution
[19:03] <asac> https://code.edge.launchpad.net/firefox-extensions
[19:03] <Jazzva> yeah, and I think I looked at it
[19:04] <Jazzva> I think there was a bug report for it's addition to the repo
[19:04] <Jazzva> s/repo/archive/
[19:05] <Jazzva> nope... you asked him to package it with xpi.template
[19:05] <Jazzva> bug 216892
[19:05] <ubottu> Launchpad bug 216892 in ubuntu "[needs-packaging] pwdhash" [Wishlist,Confirmed] https://launchpad.net/bugs/216892
[19:05] <asac> lets see if the bug is linked to the branch
[19:05] <Jazzva> it isn't
[19:06] <Jazzva> linked... not sure if it's done (though, the bug is set to "Confirmed"), so I put the status Fix in progress
[19:07] <Jazzva> it's foxyproxy
[19:07] <Jazzva> https://code.edge.launchpad.net/~jazzva/firefox-extensions/foxyproxy.ubuntu
[19:07] <Jazzva> and firefox-sage https://code.edge.launchpad.net/~jazzva/firefox-extensions/firefox-sage.ubuntu
[19:08] <asac> ok ... so both?
[19:08] <Jazzva> mhm
[19:09] <asac> foxyproxy is latest in ~ubuntu-dev
[19:09] <asac> e.g. nothing to do
[19:09] <Jazzva> oh, right
[19:10] <Jazzva> i thought you didn't upload the fix for ftbfs :)
[19:12] <asac> firefox-sage isnt merged
[19:12] <asac> doing that now
[19:12] <Jazzva> cool :). thanks
[19:13] <asac> Jazzva: so does it built now?
[19:14] <asac> ;)
[19:14] <Jazzva> hmm... it should :)
[19:14] <asac> pushed
[19:14] <Jazzva> yay :)
[19:14] <Jazzva> let's see the results :)
[19:14] <Jazzva> asac, it seems this is merged https://code.edge.launchpad.net/~jazzva/firefox-extensions/mozilla-livehttpheaders.ubuntu
[19:14] <asac> merged or mergable?
[19:14] <Jazzva> but it reports "On hold" for the merge proposal
[19:14] <Jazzva> merged
[19:15] <Jazzva> the report is close
[19:15] <Jazzva> d
[19:15] <asac> why on hold?
[19:15] <Jazzva> dunno...
[19:15] <asac> marked as merged
[19:15] <asac> better now?
[19:15] <Jazzva> yep, the last version in the changelog is -0ubuntu1, so it was merged :)
[19:16] <asac> ok ... at least i know again where i can mark it as merged
[19:16] <Jazzva> my room has a bug... I have to debug it... :)
[19:16] <asac> didnt find it in the new interface
[19:16] <asac> probabyl thats why it was "on hold"
[19:16] <asac> uh ... a real "bug" ?
[19:19] <Jazzva> yep... the one that goes "bzzz *thomp* bzzz *thomp*"
[19:20] <asac> ha ... just used win95 ;)
[19:20] <Jazzva> you?
[19:23] <Jazzva> 5 updated extensions... is that good?
[19:41] <Jazzva> asac, https://wiki.ubuntu.com/MozillaTeam/Extensions/ReportJuly08
[19:41] <Jazzva> check if it's ok
[19:41] <Jazzva> gnomefreak mentioned that he did some changes to the template and that he will post it to the wiki, but nothing so far... will he appear soon?
[19:42] <Jazzva> (if there is no report, then just wait until the wiki saves it
[19:42] <Jazzva> )
=== Moot2 is now known as MootBot
[20:10] <asac> Jazzva: nice
[20:11] <Jazzva> yay :)
[20:12] <Jazzva> ok, now just to wait for gnomefreak to post it on the blog :)
[20:12] <asac> hmm ... was he here today?
[20:12] * asac scrolls back
[20:12] <Jazzva> dunno... I think I saw him in the morning
[20:12] <Jazzva> but then I was out for the most of the day
[20:13] <asac> yes 12 am
[20:13] <Jazzva> dinner time. I'll see to add examples for TB to XPI.TEMPLATE, if you don't want to do it :).
[20:13] <Jazzva> brb
[20:30] <Jazzva> back
[20:31] <Jazzva> asac, will you work on xpi.template?
[20:49] * asac dinner
[20:49] <Jazzva> have fun
[22:01] <fta> asac, why did you disable jemalloc in xul intrepid ?
[22:05] <fta> hmm, Launchpad web service, sounds nice
[22:06] <Jazzva> lp web service? new name for lp?
[22:06] <fta> no, an API
[22:06] <Jazzva> ah
[22:06] <fta> https://help.launchpad.net/API/
[22:08] <fta> well, i've just got an email announcing that
[22:08] <Jazzva> i see the mail now...
[22:13] <asac> fta: i didnt
[22:14] <asac> did i?
[22:14] <fta> -jemalloc_in_xul.patch
[22:14] <fta> +#jemalloc_in_xul.patch
[22:14] <fta> +#jemalloc_static.patch
[22:15] <fta> - * fix "jemalloc not enabled in --with-xul-sdk= builds": we fix this by
[22:15] <fta> - linking xulrunner-bin and xulrunner-stub against libjemalloc.so and
[22:15] <fta> - adding the DEBIAN_XUL_DIR into /etc/ld.so.conf.d/xulrunner-1.9; run
[22:15] <fta> - ldconfig in postinst accordingly.
[22:15] <fta> + * (disabled in intrepid) fix "jemalloc not enabled in --with-xul-sdk= builds": we
[22:15] <fta> + fix this by building libjemalloc as a static lib and linking xulrunner-bin and
[22:15] <fta> + xulrunner-stub against it.
[22:15] <fta> - add debian/patches/jemalloc_in_xul.patch
[22:15] <fta> + - add debian/patches/jemalloc_static.patch
[22:15] <fta> - update debian/patches/series
[22:18] <asac> fta: yes, read rev 309
[22:18] <asac> changelog
[22:18] <asac> err, bzr log ;)
[22:18] <asac> fta: works fine in hardy, but in intrepid remote client crashed
[22:19] <asac> maybe is now fixed automagically
[22:38] <fta> checking for sqlite3 >= 3.6.0... Requested 'sqlite3 >= 3.6.0' but version of SQLite is 3.5.9
[22:41] <fta> checking for snd_pcm_open in -lasound... yes
[22:42] <asac> fta: latest 3.1?
[22:42] <fta> yep, 1.9.1
[22:42] <asac> good
[22:42] <asac> maybe pull in the jemalloc patches there too
[22:43] <fta> even debian doesn't have sqlite3 >= 3.6.0
[22:43] <asac> yes. they need the latest
[22:43] <asac> you can try to dump that, but afaict 3.5.9 has bad perf
[22:43] <asac> if its only that it should at least work
[22:45] <fta> not sure what to do, either package 3.6.0+, or patch configure, or wait
[22:45] <Jazzva> asac, how's this comment for MOZ_XPI_MOZILLA_DIRS in xpi.template's rules file?
[22:45] <Jazzva> http://paste.ubuntu.com/34195/
[22:47] <Jazzva> I don't know if I should put there the part "take care of Depends in control file"...
[22:47] <fta> http://www.sqlite.org/35to36.html
[23:38] <fta> Jazzva, http://paste.ubuntu.com/34207/
[23:41] <Jazzva> nice :)