=== rickspencer3 is now known as rickspencer3-afk [00:20] bye all ... off for a quick mini holiday - back on Sunday [01:58] robert_ancell - regarding bug 420234 - usually if the symbols differ from your symbols file, you will see that in the build log (with a diff as well) [01:58] Launchpad bug 420234 in pango1.0 "Update to 1.25.5" [Wishlist,In progress] https://launchpad.net/bugs/420234 [02:00] chrisccoulson, which log? The output of bzr-builddeb? [02:00] bzr-buildpackage [02:01] yeah, you should see it somewhere when you build it [02:02] robert_ancell - there is an example of what you should see in here: [02:03] (search for dpkg-gensymbols) [02:03] http://launchpadlibrarian.net/25389911/buildlog_ubuntu-jaunty-amd64.tracker_0.6.93-0ubuntu1_FULLYBUILT.txt.gz [02:03] (URL would help) [02:03] ;) [02:04] chrisccoulson, ah, thanks [02:04] you're welcome === Amaranth_ is now known as Amaranth [06:53] mac_v: you there? [06:57] mac_v: have a look at this - http://launchpadlibrarian.net/30923475/out-2.ogv [06:59] rugby471: hi... just a sec [06:59] kl [07:00] rugby471: i dont understand , does that mean you have found a way to use multiple png/svg? in the GTKtree? [07:00] oh yeah! [07:00] that animated icon is made up of 9 seperate pngs [07:01] we aren't constrained to gifs now :-) [07:01] rugby471: hmm... great , ust one think the speed is too fast, can the animation be slower? [07:01] just* [07:01] yeah ofcourse [07:01] basically I set a timeout [07:01] source_id = gobject.timeout_add(50, self.timeout) [07:01] rugby471: thats great , can i have a look at the code? [07:01] where 50 is the number of milliseconds [07:01] sure [07:01] though some of it could be improved :-) [07:02] mav_v: http://pastebin.com/d15c17617 [07:02] mac_v: http://pastebin.com/d15c17617 [07:03] a lot fo the code could be improved, however at the most basic level, it just keeps updating a gtk.Image every n milliseconds with a new pixbuf [07:04] rugby471: i think it would be better to use 18 images , for 10 degrees each , but first lets try with 20 degrees and with 9 images and timer 1000 millisecs and see how it works [07:04] sure, all the variables can be changed [07:04] this was just a 5-minute quick demo that I slapped up [07:05] just send me the icons when you are done and I can do it :-) [07:06] rugby471: just give me a couple of hours , i'm in the middle of something else ... will get back to you when i have the images :) [07:06] hehe sure [07:49] mvo: yo [07:55] hey rugby471 [07:56] I have nearly cracked the icon thingy [07:56] have a look here: http://launchpadlibrarian.net/30923475/out-2.ogv [07:57] * mvo looks [07:58] rugby471: sweet! [07:58] hehe [07:58] I am just integrating it into the software-store code [07:58] rugby471: is the code in a repo yet? [07:58] * mvo jumps in delight [07:58] thanks [07:58] :) [07:58] :-) [07:58] I will upload it as soon as it is done [07:59] I already have a repo here - https://code.launchpad.net/~rugby471/software-store/software-store-andrew [07:59] it has some small changes === tkamppeter_ is now known as tkamppeter [08:00] mvo : I might need you to test something in a few minutes [08:01] I am testing software-store through virtualbox [08:01] and the throbber looks kinda jumpy [08:01] but it might just be virtualbox [08:02] rugby471: cool, I'm merging now and have a look [08:02] ok [08:02] but just wait, I ned to push the icon changes [08:03] ok, there is also currently a hardcoded path i(/mnt/software-store :) [08:03] yeah :-) [08:03] where is it again, I need to change that [08:05] mvo : where is that hardcoded path again? [08:05] rugby471: softwarestore/view/viewswitcher.py [08:05] line 134 (in my merge) [08:06] ah yes [08:06] look at my branch now https://code.launchpad.net/~rugby471/software-store/software-store-andrew [08:07] mvo : that has been corrected and the proper icon code is in there [08:07] sweet! thanks [08:07] * mvo merges [08:07] I need you to test that revision [08:07] see if the icon is jumpy :-) [08:07] btw I have put in a placeholder progress icon [08:08] mac_v said : just give me a couple of hours , i'm in the middle of something else ... will get back to you when i have the images :) [08:10] mvo : btw does the new policykit support that checkbox where it said 'remeber authorisation for session' ? [08:10] otherwise that dialog is going to get pretty annoying [08:10] (unless we support a queue of app installs/removes and then only apply change when the user says 'go') [08:15] rugby471: it should only ask once per action (i.e. once for install, once for removal) [08:16] ah, my mistake :-) [08:16] :) [08:17] I guess its arguable if it should only ask once and not once per install/remove [08:17] yeah [08:17] anyway [08:17] mvo : have you had a chance to look at that progress icon stuttery thing [08:17] (sorry if I am being impatient) [08:18] rugby471: yes, its doing it for me too, I'm just checking if I can make sense of it [08:18] oh.. damn it [08:19] are you running on karmic (native) [08:19] rugby471: http://filebin.ca/jyxqcw/Progress.tar , i'v just labeled them 1~9 see how this works with 1000 millisecs , if need we can make more [08:19] ah goody [08:20] i'll put them in my branch then [08:20] mac_v: would you mind uploading this tar to the wiki page about icons, as I am only going to use them as png's [08:21] rugby471: i thought mpt wanted them as svg ? does svg not work? [08:21] converting to pmg is not prob [08:21] png* [08:22] well it's going to take up memeroy & cpu converting them on the fly [08:22] mvo : what do we want to do, leave them as SVG and convert on the fly, or make them pngs (we are only using them at one res so there isn't much need for svg) [08:23] mvo: rugby471: hmmm... ok.. i'll upload both versions , which ever works better we can use , i'v no probs with png , its just mpt wanted them as vector images [08:23] oh okay [08:25] I think it makes sense to have the svg in the source, but for the display, we can just use the png [08:25] rugby471: just to make sure how does it work now? do we need more icons? or is 9 enough? [08:26] I shall just check [08:26] rugby471: ok... pls upload a quick video of it :) [08:27] to be honest, the more there are, the smoother it will be [08:27] so maybe 9 is not enough [08:27] the firefox throbber is 31 images :-) [08:27] I shall just test it out [08:29] my thoughts too [08:34] actually it looks quite good [08:35] however it we want it at slow speeds then we need more [08:35] mac_v : could you try 20? [08:35] I shall upload a video [08:37] rugby471: I moved your code into its own widget now, could you have a look at http://paste.ubuntu.com/260759/ please? [08:37] mac_v : here is a video http://videobin.org/+cu/f2.html [08:37] sure [08:37] hehe that is a lot better than my code :-) [08:38] rugby471: all build on top of your work! [08:39] now when I run it standalone its smooth, I try it in software-store now. its quite possible that the main loop there is already pretty busy (with doing dbus stuff etc) [08:39] yeah it is smooth for me [08:40] mvo : how slow do we want to progress icon, or is that mpt's job ? [08:41] hehe software-store is going to be awesome :-) [08:41] rugby471: uh, good question, it should not be too fast otherwise people get nervous :) [08:41] yeah that's what I was thinking, spinning icons are being to annoy me already this morning :-) [08:42] mac_v: we need about 20 as the icon needs to be quite slow [08:44] looking at it now, my icon code was pretty dirty :-) [08:45] mac_v: don't worry actually, if you are busy I'll create the 20 form your SVG [08:45] rugby471: hmm... just a sec. how many do we need exactly? 18 or 20? [08:46] 20 - a rough estimate [08:46] 18 is fine also [08:48] rugby471: if you notice the video the 100 millisec is quite correct speed , just addin more icons will fix it... i'll get you the icons in half hr , or if you are in a hurry ;) ... [08:49] don't worry I shall do them :-) [08:49] rugby471: no probs then , do you know how to convert them to png? [08:49] yup [08:50] ok.. great [08:57] mvo : is this the speed we want (FPS of 5) ? - http://videobin.org/+cv/f3.html [08:57] (ignore gitteryness) [08:58] yeah, that looks good [08:58] I like that [08:58] ok [08:58] well that is about 40 icons :-) [08:59] should I create them? [08:59] wait amin [08:59] damn it :-) [08:59] we can rotate them in gtk [08:59] http://www.pygtk.org/docs/pygtk/class-gdkpixbuf.html#method-gdkpixbuf--rotate-simple [09:01] thats makes it a lot easier [09:01] we only have to ship one image [09:01] mvo : I'll try and do that [09:10] mvo : oops it seems we can't we can only do 90 degrees, oh well :-) [09:11] rugby471: heh :) that is a real drawback for this function :P [09:11] oh well :) [09:11] * mvo wonders if a thread would help with the stuttering [09:13] * rugby471 wonders if he will every get his head around what threading is [09:14] yeah, good point - I usually try to avoid it too [09:15] hm, maybe its something else, it seems like the timout is called often enough (I just checked with time.time()) [09:27] mvo: though its nice to see rugby471 is enthusiastic to fix this quickly , it is better to get a review of the icon from either kwwii or mat_t , before we push this into main... what do you think? [09:28] btw kwwii will be coming back to work today [09:29] I just tried 45 icons, it was good but I think too much :-) [09:29] I#m going to try with 30 icons [09:30] make that 36 (nice round number :-) [09:32] rugby471: about the number of icons , i think its a bit waste of space to have so many , is it possible to compress all of them and extract the images to /tmp on the fly and use them? [09:33] dunno [09:33] that is something to look into [09:33] I recon we could do it with PIL [09:33] but for the moment we are just trying to get it working :-) [09:33] but I will look into other methods don't worry [09:39] mvo : is this okay? (with 36 images, 12 FPS) http://videobin.org/+cv/f4.html [09:45] rugby471: looks great :) [09:46] cool [09:47] rugby471: I checked the stuttering a bit more and it seems like it has something to do with the treeview and/or the image renderer. when I add the normal gtk.Image based widget to the main app, there is no stuttering during install or removal [09:47] (well, no is not quite right, I saw it once very briefly) [09:47] right [09:50] well that gtkImageCellRenderer.py is not mine [09:50] I found it in the pygtk FAQ [09:50] there is a bit about animaiton in there [09:50] line 29 [09:50] mvo : maybe it has something to do with it? [09:52] mvo: is software store not supposed to have "report a problem" [apport support] ? or ... [09:53] mac_v: it should have it, I need to look how I can add it [09:53] rugby471: yeah, I was suspecting that too [09:53] hehe [09:54] meh, I really like the animatedimage class now .) I hope we can keep it [09:54] mvo: ok ,i'll just report a bug as reminder? [09:54] I was wondering if we should maybe have a AnimatedCellRenderer or something [09:54] mac_v: there is one open already IIRC [09:54] oh... ok [09:55] bug #420127 [09:55] Launchpad bug 420127 in software-store "software-store lacks a "Report a Problem" function" [Undecided,New] https://launchpad.net/bugs/420127 [09:55] :) [09:57] great ;) [09:57] mvo : I recon it has to be that animation_timeout function, but the pygtk in there is too advanced for me :-) [10:01] oops I'm back [10:02] maybe I shouldn't have run a karmic virtualmachine and 10 other windows since 8:00AM :-) [10:02] mvo : gotta go for a bit now, don't worry you will crack it :-) [10:03] :) [10:03] * mvo waves [10:03] mac_v: launchpad integration is now added [10:05] mvo : apport? [10:06] yes [10:06] kl [10:06] and get help online etc [10:07] * mac_v wonders why he upgrade to kernel -8 is being blocked ! [10:07] the* [10:10] mvo : just before I go, could a possible solution be to set the cellrenderer back to a pixbuf [10:10] and simply update the pixbuf rather than updating an image? [10:11] image > gtk.Image [10:12] rugby471: yes, definitely worth a try [10:13] or simply everytime we update the image, tell the cellrender to redraw [10:13] anyway I shall leave these things to you :-) [10:14] I will work on this in the afternoon again :) [10:14] cool [10:14] mvo: hey! I tried to open the .ui file of software-store yesterday in glade and I got in multiple warnings (additionnally to the traditional "no icon in.."): http://paste.ubuntu.com/260791/ Is it normal? [10:14] so feel free to look into it again some more [10:16] didrocks: no - I think there is a problem in the ui file somewhere [10:16] mvo: ok, but you didn't add anything outside glade, right (just to check)? [10:16] didrocks: you mean I did not modify the file by hand :) ? no [10:16] I don't think so [10:17] mvo: ok, thanks. For launchpad integration, I can do some work there. I've already done that in both C and python. [10:20] didrocks: meh, sorry. I just added it :( [10:20] mvo: oh ok, no pb. I'll work more on the interface itself :) [10:20] didrocks: the bug why the search looses focus is also anoying [10:20] didrocks: great, thanks :) [10:22] mpt: well, I was thinking about your animation (for "where is it") and maybe we'll have some solution using GNOME accessibility layer. (but this animation can't be in clutter as we want to control something outside the current application) [10:23] mpt: also, I looked at the ui yesterday. Ok, we can do some stuff in clutter, but you have to be aware of two things: [10:24] 1/ clutter needs a 3D driver. It will not be an issue with GNOME 3 as it uses clutter (with mutter, which is metacity + clutter) by default, so, they enforce a 3d driver support already [10:25] didrocks, yes, Ted Gould also thought we'd need to use the accessibility layer for that. So we have discussed turning it on by default early in the Karmic+1 cycle, to shake out the bugs so that it can be always-on. [10:25] But we can't add clutter in the store before GNOME3 (we can, but everyone who wants to use it have to have a 3d handling graphic driver) [10:25] mpt: ok, perfect. I'll give a try, so :) [10:26] 2/ clutter-gtk is still in early development phase (so binding I rebuilt for adding advanced effect is even not in trunk) [10:31] thanks for investigating this, didrocks [10:38] mvo : did you see the other changes in my branch? [10:40] rugby471: no, sorry. I will look at them now and merge them individually [10:40] cheers [10:40] there is one change that mpt probably needs to look at [10:41] it is about the behaviour of the Get free software and Installed Software buttons [10:41] mpt: someojne filed a bug that clicking the buttons should take you to the home screen [10:41] I agreed with this [10:41] Also I put in place that when you clicked these the navigation bar is set back to home [10:42] mpt & mvo : is this inline with the behaviour that you wanted when you designed it? [10:42] rugby471, no, they're like tabs in a tabbed window, they should take you to exactly the same part of the section you were in before [10:43] If you want to go up to the top of the section, you can use the path button [10:44] mvo, has rugby471 signed the Contributor Agreement? [10:44] oh okay, don't merge that change then :-) [10:45] mpt : no not yet, but I can [10:45] cool, thanks [10:45] I shall do it in half an hour, I have to go do something now :-) see ya [10:48] mpt: I'm wondering what's the difference of context between animation 1 et 2 in "Animation of the main pane" section. For me, navigating from the lobby screen to a department screen or changing from one item to another in the navigation pane is the same thing. I should miss something :) [10:49] didrocks, I need to go through the spec and make my terminology consistent, I've been all over the place [10:49] didrocks, but by "department" I mean Accessories, Education, Games, Graphics, etc [10:50] Those departments aren't shown in the navigation pane, whereas the "Get Free Software" and "Installed Software" sections are in the navigation pane. [10:51] mpt: ok, oh, I guessed first that the navigation pane was the pane at the top, not the one on the left. Ok, understood now, thanks :) [10:51] The subtle metaphor is that this is a department store, and you're going in an elevator to (for example) the floor where all the Games are, and (ding!) the doors open [10:51] sweet :) [10:52] I'll try to get something, but again, it will not be mergeable (just proof of concept) until clutter isn't mandatory for GNOME. In the meanwhile I can work also on the pygtk side. === al-maisan_ is now known as al-maisan [11:45] hrmm .. I cannot print at present and see the following line in /var/log/messages upon plugging in the printer cable: [11:45] Aug 28 12:38:57 Px4 kernel: [ 168.321039] type=1503 audit(1251455937.238:16): operation="open" pid=4408 parent=4403 profile="/usr/sbin/cupsd" requested_mask="::r" denied_mask="::r" fsuid=7 ouid=0 name="/dev/bus/usb/" [11:46] Is this a know issue? [11:46] What should I add to /etc/apparmor.d/usr.sbin.cupsd to ix it? [11:47] *fix [12:06] Hmm .. I added a "/dev/bus/usb/** rw," line to /etc/apparmor.d/usr.sbin.cupsd; now the printer state is: "Waiting for printer to become available" .. for the last 10 minutes. [12:07] BTW, /var/log/messages now says: [12:07] Aug 28 12:58:57 Px4 kernel: [ 1368.053230] usblp0: USB Bidirectional printer dev 7 if 0 alt 0 proto 2 vid 0x0482 pid 0x0015 === MacSlow is now known as MacSlow|lunch === Tm_T_ is now known as Tm_T [12:57] mvo: are you busy, or could I ask you something about software-store's code [13:02] rugby471: just ask [13:02] cool [13:02] mvo: I am trying to pass an optionla package argument form the commandline [13:03] I have set it up so that the arguement gets to app.py fine [13:03] however when I try to run self.app_details_view.show_app(package) [13:03] I get the error File "/mnt/softwarestore/apt/aptcache.py", line 30, in has_key [13:03] return self._cache.has_key(key) [13:03] AttributeError: 'NoneType' object has no attribute 'has_key' [13:04] rugby471: its probably still opening the cache in the background in this case, see the refresh() code [13:04] I think it is becuase I am trying to view this page before a cache has been built, where should I put the code so that it behaves corrcetly? [13:04] hehe [13:04] rugby471: in this case (when using apt:) we could just not do the cache opening in the background [13:05] ok [13:05] what function is opening the cache? [13:07] mvo : ah I think I have got past that blocker :-) [13:07] thanks [13:07] rugby471: apt/aptcache.py adds itself via a timeout that opens the cache, I add a "cache-opened" signal now, then you can just subscribe to that - what do you think? [13:07] oh, you solved it already :) ? [13:07] a bit :-) [13:07] one last question [13:08] mvo : on the wiki page, it says that if software-store PACKAGE is run, then we should open that package [13:08] however while looking on the OptionParser page (the module that I am using for parsing our arguements) [13:09] it says that optional arguements are typically given a switch (ie. -p PACKAGE) [13:09] only compulsory arguements are given without a switch [13:09] do we still want to go with the functionality specified on the wiki? [13:10] from a technical point of view, there is no problem [13:10] .. either way [13:10] just a convention (thats the word I was looking for) [13:11] I think in this case we should follow our spec [13:11] ok [13:11] t would be nice if you could add a --debug switch as well [13:11] sure [13:11] that just sets logger to debug :) [13:11] great! [13:11] would that correspond to logging.basicConfig(level=logging.DEBUG) [13:11] ? [13:12] mvo : ? [13:12] hello [13:12] hi [13:13] is Ted Gould here ? [13:14] rugby471: yes [13:14] kl === MacSlow|lunch is now known as MacSlow [13:22] c_korn, not yet, but ted should be around within the next hour [13:22] kenvandine: k. thanks [13:28] mvo : for handling of errors (for example the package supplied on the commandline doesn't exist) do you want me to put in a error dialog function that can be reused? [13:28] in app.py [13:30] rugby471: please :) [13:30] rugby471: simple wrappers around a gtkmessagedialog should be fine [13:30] I like something like "error(summary, text)" [13:30] error(primary, secondary) - I'm not fuzzed aobut the exact variable names [13:32] kenvandine, hi, i saw you've committed some stuff to ~gwibber-team/gwibber/packaging, could you please update ~gwibber-team/gwibber/packaging.trunk? [13:32] yeah that was what I was going for :-) [13:32] fta, humm... what is that branch for? [13:33] kenvandine, well, it's the branch that is supposed to track trunk, it's base for the 2 PPAs [13:33] oh [13:33] i see [13:33] i'm not sure about the one you touched [13:33] i touched the one referenced in Vcs-Bzr [13:37] fta, ~gwibber-team/gwibber/packaging.trunk hasn't been touched since may [13:38] kenvandine, indeed, yet that the branch used for the dailies, so it's stable ;) [13:38] +'s [13:38] ok :) [13:48] fta, merged and pushed [13:49] thanks [13:53] any one knows when messaging indicator will get empathy support? [13:53] mpt: i'v shown the icon to kwwii too , he said its fine , now where and how many icons are needed is confusing! could you clarify? [13:55] mvo : I have half-completed the commandline package handling, it probably needs someone who knows more about the code to complete it (wink, wink) :-) However the debug, help message and error dialog thing all work and are implemented [13:56] rugby471: nice, is it in the same branch that you posted before? [13:57] not yet, just pushing [13:57] mac_v, sory, what specifically is confusing? [14:00] mpt: rugby471 and mvo have been doing the animation using 36 or 42... i dont know , i wanted to know exactly how many icons you guys wanted. once you are done with the code stuff [14:00] mvo : nearly there, by the way, only merge the contributions of the latest rev as the rest is probably out of sync with your branch [14:00] mac_v, 36 or 42 what? Pixels, or frames, or something else? [14:01] mac_v: 36 is the number we are going with and the 36 icons are available in my branch [14:01] 36 - number of rotations [14:01] mpt: 32px but 36 number of icons [14:01] mac_v, mvo is really the best person to answer that question. My only requirement is that it doesn't look jumpy. :-) [14:01] hehe ;) [14:02] mpt : I did some testing and 36 is the best number [14:02] it doesn't look jumpy at the speed we want to go with but it isn't too many [14:02] rugby471: so i dont need to send any icons? all is good then? [14:02] rugby471, a period of about two seconds? [14:03] mac_v : yes all is good [14:03] mpt: (with 36 images, 12 FPS) http://videobin.org/+cv/f4.html [14:03] mpt: now that rugby471 has completed the 36 icons from the 9 i'v sent do i need to sign the contributions [14:04] yup [14:04] http://www.canonical.com/contributors [14:04] * mpt is surprised Chromium apparently "does not support HTML5 video" [14:04] rugby471, that looks good [14:05] mpt : thx, HTML5 video is coming though [14:05] The icon will fairly often be spinning around for hundreds of seconds while you do other stuff, so it should be fairly sedate :-) [14:05] mpt: rugby471: hehe.. i was confused , would it be my contribution or rugby471's , since he completed the set ;) [14:05] yeah thats what me and mvo were talking about [14:06] if it was whizzing around we'd have loads of emails from annoyed users rather than happy ones :-) [14:06] mpt: i send the mail to mvo right? [14:06] mac_v : definitely yours, all I did was rotate the image in inkscape and export it [14:06] mac_v, mvo and to the address on the Web page [14:06] mpt: I added a text for packages that have other packages depending on them (so that the user knows about that). does "%s is installed on this computer. %s installed pieces of software use it." is a good sentence? [14:06] queuing installs is a long awaited features, people would like it with software-store :) [14:07] chrisccoulson: have you done much polkit1 debugging yet? [14:07] rugby471, we could be sneaky and do what Microsoft did with Internet Explorer's throbber, and Apple did with the bouncing Dock in Mac OS X: make it fast in early versions, but slow it down in later versions, so that people who count the speed of loading by the number of spins/bounces are impressed at how much faster it is in the new version [14:08] james_w - not yet. how come? [14:08] mpt : haha that is great :-) [14:08] users-admin is broken for me [14:08] (users-admin:1717): polkitgtk-WARNING **: Error sync-checking authorization for action id `org.freedesktop.systemtoolsbackends.set': Remote Exception invoking org.freedesktop.PolicyKit1.Authority.CheckAuthorization() on /org/freedesktop/PolicyKit1/Authority at name org.freedesktop.PolicyKit1: org.freedesktop.PolicyKit1.Error.NotAuthorized: Only trusted callers can use CheckAuthorization(), unix-user:jw2328 is not trusted [14:09] i haven't tried users-admin yet actually [14:09] one second, i will try it too [14:09] no worries though, just wondered if you would have some pointers :-) [14:09] mvo, so that's a case where we really do need to send people to Synaptic? [14:09] mvo, I see an example of this is Sylpheed mail client [14:10] mpt: depends if we want to do that or not, we can just let people remove it anyway [14:10] mpt: or show them a dialog or something with details. I don't mind either way [14:10] mvo, well, what are we trying to protect people from here? [14:11] mpt: removing stuff by accident. [14:11] mvo : you have a dialog function :-) [14:11] mpt: I see two cases here: "I removed bla and that also removed blub" and "I removed hearts and now all my gnome-games are gone" [14:11] * mpt looks at Synaptic's equivalent warning: "The chosen action also affects other packages" [14:12] mvo, how are those two cases different? [14:12] mpt: the former is for packages that are a dependency of something already installed [14:12] mpt: the later is for packages that have multiple apps in the same package (gnome-games) [14:12] hmmmm [14:12] mpt: currently the user may think that that he can remove "hears" only [14:12] mvo, is it easy to get from a package name to the application name+icon? [14:12] when he really can only remove all of them nor none [14:13] mvo, e.g. if someone tries to remove Hearts, to give them a list of the other *applications* that will also be removed [14:13] mpt: that is doable (but not implemented right now) [14:13] james_w - i don't see the error here [14:14] oh [14:14] thanks for testing [14:14] mvo, are you comfortable doing it in time? Or would you rather just show a list of packages? [14:14] mvo, or just failing and sending people to Synaptic? [14:15] mpt: not all pkgs will have desktopfiles/icons so we will need both [14:15] well, sure [14:15] mpt: as a fallback (if it can not be done) we can sent people to synaptic [14:17] mvo, I'm happy designing any of those three (confirmation with list of applications/packages > confirmation with list of packages > sending people to Synaptic), it's just dependent on how much time you want to spend on it for 1.0. [14:20] mpt: lets aim for "confirmation with list of applications/packages" [14:20] We choose to go to the moon! [14:20] ok [14:21] going to space is enough :P [14:21] mvo, I'll start drawing that up in about 1 hour 40 min, I have a meeting with the other spaceman first :-) [14:21] james_w - the error message suggests that the caller is not UID 0 [14:22] that shouldn't be the case though should it? (system-tools-backends runs as root)? [14:22] but it is g-s-t that queries isn't it? [14:22]