[00:20] mkarnicki: good news, I can reproduce the error using beuno's token, bad news that I have now idea what's going on...yet [00:22] verterok: oops. how did you reproduce it? [00:22] mkarnicki: using beuno's oauth token [00:22] verterok: when did you pull the branch? quite much has changed, unless you're talking about protocol side [00:22] mkarnicki: protocol side [00:23] mkarnicki: the bug that prevents aquarius, beuno and statik from using the app [00:23] verterok: oh.. I saw that the problem was even before log-in screen popped up [00:23] verterok: crap.. those are one of these ppl I'd love to have check out the app ;( [00:23] mkarnicki: FWIW, I'm using the latest code from 0.x branch [00:24] verterok: no idea what's wrong.. I have Android 2.1 -update1 running, too. [00:24] mkarnicki: looks like a server/data issue [00:25] I have a guy that might have this problem in exactly one folder (strange!), is it possible? [00:25] when he browsed one particular foldeer, he didn't get the content from the server [00:28] verterok: that's bad :( [00:28] verterok: I don't like feeling that I can do nothing to help [00:29] mkarnicki: no, it looks like the server is sending something the client doesn't expect [00:29] mkarnicki: I think it might be a bug in the client code that build the message [00:30] * mkarnicki acknowledges [00:32] verterok: maybe you should get some rest, you probably worked hard enough today. you seem to be quite a busy person :) [00:32] mkarnicki: yes, but now that I know where the bug might be...how stop? :) [00:33] verterok: ^ ^ [01:02] mkarnicki: yay! found it! [01:03] * verterok dances [01:04] verterok: you're amazing!?!?! what was it?! [01:04] mkarnicki: bug in the OIO code [01:04] verterok: oh! [01:04] verterok: I'm very glad you found it! :) [01:05] verterok: no wonder! I was pushing you so hard you worked under stress =D [01:06] * mkarnicki is happyy [01:06] ? [01:06] mkarnicki: bug in the OIO codetually sit down and debug it [01:06] verterok: you know, I asked for OIO client and really looked forward to it, etc etc, I'm kidding that you were in rush ;) [01:06] oops [01:06] :) [01:06] mkarnicki: I wasn't been able to actually sit down and debug it [01:06] (just kdding :) ) [01:11] mkarnicki: I'll push the fix in a while, I [01:12] I'll let you know when it's fixed, so can re-build the client once it's fixed in order to allow the other guys to play with it? [01:13] verterok: not sure why '?' - yes, please. take your time, no worries :) [01:13] :) [01:13] ok, thanks [01:14] verterok: no, thank you again :) [01:16] mkarnicki: pushed, revno 24 [01:17] verterok: fast! xD [01:17] verterok: I'll pull that tomorrow, I don't want to break anything (did you apply my one patch? [merge, whatever it's called]) [01:18] mkarnicki: not yet, I'll merge it later tonight [01:19] mkarnicki: I was trying to avoid using the android/InflaterOutputStream.java when isn't needed [01:19] sure! np [01:19] no, not that one [01:19] but I couldn't find a way to do it [01:19] there was one rev up an NPE fix I think [01:19] mkarnicki: which one? [01:19] I only have two revisions in my u1-sp branch [01:19] u1-java-sp [01:19] in my +junk [01:19] hmm, I missed that [01:20] I'll look it up [01:20] mkarnicki: I'll merge that revno now [01:21] https://code.launchpad.net/~mkarnicki/+junk/ubuntuone-java-storageprotocol-fix rev 22 [01:21] sorry for the link [01:21] * mkarnicki blushes [01:21] mkarnicki: pushed revno 25 :) [01:21] verterok: great job :) you're fast [01:22] ok, now I'm off to get dinner [01:22] later! [01:23] verterok: goodnight! [01:23] verterok: I'll be leaving soon. take care, thanks for all your work [01:23] and suport [01:24] *support === cpg is now known as cpg|away === yofel_ is now known as yofel [02:16] * mkarnicki went offline === cpg|away is now known as cpg === dendro-afk is now known as dendrobates === cpg is now known as cpg|away [05:11] anyone around? === cpg|away is now known as cpg === cpg is now known as cpg|away === cpg|away is now known as cpg [06:03] test [07:55] morning! [08:39] morning [08:41] morning all === JUMPA is now known as jumpa [08:44] hi duanedesign [09:03] * duanedesign is thinkning of updating the stickies in the Ubuntu One section on the forum. [09:04] if anyone has any input on what might be good current info to get to users [09:11] morning rye [09:11] duanedesign, morning! [09:13] i was just mentioning that I was going to update the stickies in the Ubunt One section of the forums. If you have ny ideas for information that would be good for this let me know. [09:54] is there a way to upload to u1 vie ftp? [09:55] is there a way to upload to u1 via ftp? [10:04] eckibox: no there is not [11:08] is this the right channel to ask about the ubuntu single sign in service? [11:14] and471: yes, though I may not help you [11:15] and471: what's up? [11:15] mkarnicki, I work on the ubuntu software-center and we were wondering whether with the Ubuntu SSO login, (using lazr.restfulclient) whether it was possible to distunguish on an authentication error, whether it was the email that was wrong, or the email and the password [11:16] oh, maybe duanedesign or rye will be able to help/redirect that question ( and471 ^ ) [11:16] and471: but it is the place to ask that one, so hang around for an answer :) [11:16] mkarnicki, thanks [11:16] and471: you're welcome. [11:30] rye: who would be best to answer and471s SSO question? [11:32] duanedesign: hi :) do you ever sleep? [11:32] :) [11:34] mkarnicki: my hours vary so it just gives the illusion that i am always up :P === cpg is now known as cpg|away [11:35] duanedesign: aha ;) I'm rarely availible 5:00-10:00UTC which here is 7-12, I code well during night :) [11:55] duanedesign, rye I have to go now, so if you can answer it, could you redirect the answer to mpt (on #ubuntu-desktop) === jumpa is now known as JUMPA [12:44] i have got a privacy question: why is google analytics active when i am signed in on the webinterface? I do not what google to know, what files i have, what contacts etc. are saved in my account. Is the service necessary? [12:49] interesting question (though I doubt they have access to any data on the page just like that..) (sorry, I have no answer for you) [12:52] CardinalFang: may I stalk you with another question :> ? if I want my files to be accessible via my content provider, what method should I use to open that file for writing? [12:55] soney, if you are voried about google i strongly sugest that you block there urls, either on a dns, or browser level. [12:56] soney, firefox has lots of plugins for doing just that [12:56] AJenbo: nevertheless, this is indeed an interesting question. I don't think analytics has direct access to page content, does it? [12:57] it has access to all contents of the page, which are displayed [12:57] mkarnicki, no all they get is the URL [12:57] exactly, so soney shouldn't be worried :) [12:58] if you embed js, it has access to the full document [12:58] Well there script dosn't read out the content, but in theory if they turned evil they could rewrite there script to scrape the content. [12:58] If u1 used a local copy of the GA script they could avoide google going evil [12:59] google does support this, but it's not recomended as it will not automatically be updated [13:00] who needs the statistics of GA? [13:00] mkarnicki, I don't have much insight into that. Never used local files before. [13:00] CardinalFang: np! I'm looking at the docs :) I have seen it somewhere in the dev guide [13:02] The statistic page woun't show individual files, but it could be used to see folderes. [13:03] (folder names) [13:03] and contactnames [13:03] yes [13:03] why? [13:03] soney, I know Canonical uses those statistics to design web pages. "Do we care about IE? What's the color depth of visitors? How many users are mobile phones? et c." Analytics helps answer those questions. [13:03] CardinalFang++ exactly, that's what analytics is fore [13:03] *for [13:04] soney, there is no reason why, it is just a consiquence of how browsing your files work, and how GA works [13:04] all this information are possible to get without google [13:04] soney yes but it is harder [13:04] i know GA and how it works [13:05] soney, all this information is easy with Google, though. Would you prefer to wait a few months for Feature X while Canonical writes their own analytics engine? Ooo, how about paying another $10 per month for a new salary? [13:05] but it is more secure for the users, because GA is not a free service! The Users of u1 are paying with there privacy for this service! [13:06] soney in therory any one with access to the server could get to your files, isn't that a bigger issue for you? [13:07] GA is free as in gratis but not as in freedom, the same goes for U1 storagte [13:07] if i pay for a service, i don't want my information to got to another service like GA [13:08] AJenbo, I think those arguments are red herrings. [13:08] I think the Ubuntu One privacy policy and its therefore dependent GOOG Analytics privacy policy applies here. Please understand them. [13:09] I'll ask Canonical lawyers to make sure Ubuntu One's privacy policy is compatible with the Google service U 1 uses. [13:09] That is, if U 1 makes a promise, all its tools should follow it also. [13:10] Will that satisfy you, soney? [13:10] yes! That would be very nice! [13:10] sound good [13:11] :) [13:17] soney, do you mind if someone wants to follow up with you in email? [13:18] mkarnicki: hi, g'morning! fyi, I just pushed new revision, with a fix to the node attribute handling in GetContent [13:18] verterok: thank you, neat! I was wondering in the morning if you have applied that too, but didn't have time to check it yet :) great! [13:19] verterok: sorry, good morning! =D [13:19] is there a support mail address? Then i'll write a short abstract to it! [13:19] mkarnicki: :) revno 26 [13:19] verterok: I'll pull and apply the android FileInflater from my branch, right? [13:19] soney, there's a bug-tracking system. [13:19] mkarnicki: yes, that's the only missing bit in my branch [13:19] rodrigo_: land https://code.edge.launchpad.net/~rodrigo-moya/ubuntuone-client/scroll-when-searching-contacts/+merge/29971 :) [13:20] verterok: niceeee! :) [13:20] ok [13:21] soney, I'm emailing someone about it. Give me the bug number or a email address to reply to, if you care. [13:21] mkarnicki: I have been thinking in you problem with the async client code, and maybe I have a solution without the need to refactor...need to think about it a bit more and will reply the mail :) [13:22] nessita, Chipaca, buenos dias. (Is there an es greeting for mornings, specifically?) [13:22] verterok: you're great verterok, really :) you remember about my every question and e-mail :) [13:22] CardinalFang: buenos días is literally trasnlated to Good morning [13:23] nessita: not *literally*, no. But, yes, it translates like that. [13:23] verterok: wait, you mean the node.getParentNode() functionality or the concurrent refresh of directories? [13:23] Oh, I thought it was a general "daytime" greeting. It doesn't sound funny at 1600, though, right? [13:23] nessita: literally it translates as "good days" [13:23] Chipaca: true, I'm sleepy [13:23] :-) [13:23] mkarnicki: the deferreds one [13:23] CardinalFang: only if you have a very sleepy pillow-lined face [13:24] mkarnicki: I wasn't able to start with the parent node mail yet [13:24] verterok: the latter is not very important, plus - I think creating more then one deferred would solve the thing, would it not? ok! [13:24] verterok: no problem!! take your time, it's not critical [13:24] ok [13:24] CardinalFang: "buenas tardes" is used after lunch [13:25] CardinalFang: if you say "buenos dias" and they're grumpy/badgery, they'll come back with something along the lines of "buenas tardes, or didn't you have lunch already?" [13:28] Ah, thanks. === teknico is now known as teknico_away === JUMPA is now known as jumpa [13:38] Hello. Maybe anyone can remember me and my problem from yesterday. The problem was, that ubuntuone seems to not be uploading a folder, which i have added to synchronization and which contains over 33k files together having 700mb. here is the syncdaemon.log: http://paste.ubuntu.com/464516/ [13:39] I have also made a log on my own, which shows the current time followed by the output of `u1sdtool --waiting-content | wc -l` followed by the output of `u1sdtool --waiting-metadata | wc -l`. Here is the file: http://paste.ubuntu.com/464517/ === teknico_away is now known as teknico [13:41] I have noted, that since ubuntuone has stopped looking for new files (line 2529 first file) the number of waiting metadata does not change. (line 64 second file) [13:49] facundobatista or verterok, would you have some time to see the log from jumpa? seems like SD doesn't move on with the metadata queue [13:50] nessita, ok [13:51] Hola jumpa [13:51] hi [13:52] jumpa, it seems that the client got stuck because of the server didn't answer to last petition [13:52] jumpa, one way to unstuck it is to disconnect it and reconnect it again [13:52] jumpa, we can not investigate further unless the log is in DEBUG mode [13:52] but that would force the client to review every of the 30k files, doesnt it? [13:53] jumpa, so, if you want, you can put the log in DEBUG and if it fails again, we can look at it better [13:53] jumpa, it will go through local and server rescan, yes [13:53] jumpa, which version of ubuntuone-client do you have? [13:54] that took around 3 hours the last time [13:54] jumpa, it may be that you're out of space (and in some versions we didn't log that in INFO) [13:54] 0 bytes Used (0.0%) says webinterface [13:54] jumpa, if you go to System -> Preferences -> Ubuntu One, how is your account? [13:55] 0 KB used [13:55] ¿? [13:55] tell me how to switch to debug mode for logging and i will reconnect [13:55] http://imagebin.ca/view/bJeLGB.html <--- does anyone know why the syncdaemon actually causes peaks when throttling is enabled? [13:55] jumpa, do you see files in the web ui? [13:56] just empty folders [13:57] jumpa, mmm... this is very strange... I see files created in the log [13:57] jumpa, for example: dev/tests/ogre/test/media/packs/skybox.zip [13:58] jumpa, if you walk that path in the web ui, do you see the .zip? [13:58] apachelogger: peaks in the network? [13:58] * verterok looks before keep talking [13:58] facundobatista: no, its just an empty directory [13:58] apachelogger: yes [13:58] :) [13:59] verterok: tcp rate shaping? [13:59] verterok, see jumpa's log... he has files in the log as "created", but he does not see them in the we ui, and free space is zero (?) [14:00] used space is zero, free space is 2 GB [14:00] apachelogger, could you please do: cat ~/.config/ubuntuone/syncdaemon.conf [14:00] verterok: anyhow, what surprised me was that after some time the daemon died with 2010-07-16 14:56:04,798 - ubuntuone.SyncDaemon.StorageClient - INFO - Connection lost, reason: [Failure instance: Traceback (failure with no frames): : Connection was closed cleanly. [14:00] verterok, jumpa, sorry, yes, used space is zero, like no files created [14:01] apachelogger: the current implementaion of the throttling causes those peaks, we are not using tcp rate shaping [14:01] facundobatista: no files uploaded ;) [14:01] * verterok looks at the logs [14:01] verterok, but in the logs: ''dev/tests/ogre/test/media/packs/skybox.zip'' | Called new_local_file_created (In: T:LOCAL:F) [14:02] (for example) [14:02] ah! stupid of me [14:02] the file *is* created [14:02] facundobatista: if the file was created but not uploaded the size is still 0 [14:02] we didn't put content into it yet [14:02] yes, that :) [14:02] verterok, oh, the freezing cold outside got into my brain [14:03] jumpa, so, no strange thing at all... just put the logs into debug, and try again [14:03] facundobatista: it's too cold to think, we should move to a place with constant 25C :) [14:03] facundobatista: http://paste.ubuntu.com/464524/ [14:03] facundobatista: and how do i put them into debug? [14:04] jumpa, furthermore... we fixed some very nasty details that are important when you are working with tens of thousands of files [14:04] verterok: hm, ok :) [14:04] jumpa, so you really should use our PPA, not Lucid's version [14:04] jumpa, (if you put the PPA, it has debug logs by default) [14:04] ok, what is the ppa? [14:05] apachelogger: yes, twisted throttling is completely broken :( [14:05] apachelogger, could you please do: cat ~/.config/ubuntuone/syncdaemon.conf [14:05] facundobatista: that is the content of syncdaemon.conf :P [14:05] verterok: so that crash might be related to that? [14:06] jumpa, one minute please [14:06] a bit unfortuante, but at least one can now configure trottling in the KDE ui :D [14:06] apachelogger: were you uploading/downloading stuff? [14:06] verterok: downloading [14:06] apachelogger, are you really telling me that http://paste.ubuntu.com/464524/ is the content of syncdaemon.conf ??? [14:06] intial to be precise [14:06] apachelogger, didn't you confuse of pastebin? [14:07] apachelogger: yes, it might be causing starvation and the server drops the connection [14:07] apachelogger: this is all handwaving and guessing :) [14:07] ^^ [14:07] I guess I am happy as long as it is not my fault :D [14:07] verterok, facundobatista: thanks [14:10] jumpa, this is it: https://edge.launchpad.net/~ubuntuone/+archive/stable [14:11] jumpa, you can do: sudo add-apt-repository ppa:ubuntuone/stable [14:11] yes [14:11] will sudo apt-get upgrade install the new version? [14:11] jumpa, after an apt-get update, yes [14:12] W: Failed to fetch http://ppa.launchpad.net/ubuntuone/stable/ubuntu/dists/lucid/main/binary-amd64/Packages.gz 404 Not Found [14:12] jumpa, you'll need to quit and restart the client, though [14:13] jumpa, mmm.. wait a second, maybe I was wrong [14:13] Ubuntu One Desktop+ kanban walkthrough starting on Mumble right now [14:15] rodrigo, dobey, which is the PPA for stable builds for Lucid? jumpa need it, and I found something that it seems is for Jaunty and Karmic :| [14:16] jumpa, I'm not finding it... maybe rodrigo_ or dobey, which are the specialists for this, could help you [14:16] jumpa, maybe rye or joshuahoover have proper instructions for this, though [14:16] i could start to be a betatester and take ubuntuone/beta :P [14:18] facundobatista: if the fixes are in the stable-1-2 branch, the changes are in lucid-updates, not in the ppa [14:18] jumpa, verterok is right... you should work with beta version [14:19] jumpa, the fixes I mean should be there [14:19] ppa:ubuntuone/stable is the ppa [14:20] but i need to get the latest stable stuff built in there still [14:20] but yeah, there is a package in lucid-proposed with a lot of fixes [14:20] ah, -proposed is the name [14:20] thanks dobey! [14:21] verterok: yeah, i don't think it has made it to -updates yet (or it likely would have shown up as an update already) :) [14:21] dobey: is lucid-updates repo enabled by default? [14:23] updates is [14:23] proposed isn't [14:23] ah, ok [14:39] I have added the ppa now, updated and upgraded, but im not sure if ubuntuone-client has been updated. How can i verify that it has been updated? [14:40] jumpa: apt-cache policy ubuntuone-client [14:40] -proposed isn't a ppa btw [14:40] Installed: 1.2.1-0ubuntu3 [14:40] dobey: I think jumpa is using the beta ppa [14:41] CardinalFang: any ideas why my program get's stuck at final FileOutputStream os = ctx.openFileOutput("/sdcard/u1test", Context.MODE_PRIVATE); ? [14:41] the newest beta version is 1.1.91+r487-0ubuntu1~ppa2~lucid [14:41] don't use the beta ppa [14:41] just remove it [14:41] confusing :D [14:42] and then? [14:42] in the software sources preferences, enable the proposed repo [14:42] it's on the Updates tab [14:43] okay [14:44] so im using the proposed packages of every application now? [14:44] thisfred, lp:~cmiller/desktopcouch/replication-more-polite [14:44] CardinalFang, so, your basic http auth branch is done? [14:44] CardinalFang: thx, will look at it in a few [14:44] mkarnicki, "stuck", like it hangs in that instruction? [14:45] CardinalFang: yes, line below Log.d does not print anything. I've just asked over #android-dev [14:45] CardinalFang: will let you know [14:45] afk, downloading upgrades :D [14:45] mkarnicki, nothing unusual in "adb logcat" around then? [14:46] CardinalFang: no, just silence :) [14:46] Wow. Does that file exist on the sdcard? Suppose you remove it or create an empty file? [14:46] mkarnicki, ^ [14:47] CardinalFang: Open a private file associated with this Context's application package for writing. Creates the file if it doesn't already exist. [14:47] There could be two steps that fail, creating the metadata and its perms, and returning an open file descriptor. [14:47] CardinalFang: it should create, if non-existent [14:47] nessita, alecu: another branch for your reviewing pleasure -> https://code.edge.launchpad.net/~rodrigo-moya/ubuntuone-client/no-enabling-on-special-udfs [14:47] ok [14:47] mkarnicki, right. I'm just trying to poke at it for debugging. [14:48] mkarnicki, it *should* not hang, though. ;) [14:48] CardinalFang: =) [14:49] nessita, so, any task on the sso thing you want us to work on? [14:49] CardinalFang: it may be the case that this is used for private Context associated files, thus should not land on /sdcard/here directly [14:49] CardinalFang: I'll check this out after lunch :) thanks! [14:49] mkarnicki, I'm getting breakfast too. Back in a bit. [14:49] :) [14:51] rodrigo_: well, is hard to say because we don't have a clear design to use to split task as per their dependencies. Maybe completing the design is the task I need more help on? [14:52] nessita, the design from the design team you mean? or the architecture? [14:52] rodrigo_: the architecture :-) [14:53] nessita, ok, shall we have a call today, or next week in Prague? [14:53] rodrigo_: if we have people avialable *now* we should have that call now [14:53] nessita, who do we need? [14:53] rodrigo_: if people will be eventually available, we can postpone [14:54] hmm [14:54] rodrigo_: for the call? [14:55] nessita, for the discussion, yeah [14:55] rodrigo_: well, you, Chipaca (maybe), verterok if possible, and me, at least [14:56] * verterok looks around [14:56] ok, let's have it today, if Chipaca and verterok can, and next week we discuss you and me the details, if any, ok? [14:56] verterok: if you have some time we may ask for your input on a dbus-related call (same topic as we talked yesterday) [14:56] sure [14:57] nessita: let me ask my manager [14:57] hehehe [14:57] nessita: sure! [14:57] :D [14:57] jajaja [14:57] Chipaca: are you available? [14:57] haha [14:58] i wonder if it's a good thing, or a bad thing, if i am not in that meeting, as it were :) [14:58] nessita: I've got to talk about cards with mandel a little, but then I'm ok [14:59] nessita, I wont keep him for too long [14:59] Chipaca: no problem, would you ping us when ready? [14:59] dobey: I'd say bad, unless you're feeling particularly contrary today :) [15:00] well, contrariness depends on the design :) [15:03] CardinalFang: how best to test this branch? setup.py build and then run the desktopcouch-service from the bin directory in the build? [15:14] verterok: you look busy with guys here today, but if you have a minute sooner or later, please have a look at http://paste.ubuntu.com/464554/ - probably won't take you long to spot what's wrong/missing [15:14] unless it again turns out to be platform issue (which I don't believe. getContent for directories works) [15:15] note to y'all desktop+ folks: the daily standups as of next monday or at 13z every day mon-thu, and the kanban walkthrouh is at 13z on fri [15:16] s/or/are/ [15:16] verterok: I'll be back later :) lunch time [15:16] mkarnicki: ok [15:20] verterok: nessita: dobey: rodrigo_: ready when you arr [15:21] mumble? [15:22] I am ready [15:24] dobey: are we waiting for you? [15:25] i don't think so? [15:25] dobey: oh, ok. Then we won't :) [15:26] i'll just yell at you all later if you mess it up ;) [15:26] dobey: sounds like a sany, forward-looking, sustainable, green, proactive plan! [15:27] s/sany/sane/, but zany works too :) [15:28] sany is a mix of sane and zany, i suspect [15:29] or a hygiene product [15:29] nah, that would be a zany-sani [15:46] alecu, when you finish the contacts picker bug, talk with nessita about sso, ok? [15:47] rodrigo_, ok [15:49] Chipaca, did you ping pfibiger about QA for the finished tasks? [15:50] rodrigo_: on it [15:50] rodrigo_: thanks for the reminder [15:50] ok === jumpa is now known as JUMPA [16:13] rodrigo_, ping [16:13] alecu, pong [16:18] rodrigo_, one question... I'm finding I need to do complex code for the case when the user is searching for something with spaces inside [16:19] alecu, hmm, I don't think the searching code in e-d-s does that, it just searches for the exact string, although I'm not sure [16:19] alecu: did johnlea reply to the bug to resolve the issues i mentioned? [16:20] dobey, yes, he did [16:21] dobey, although only on the "1st and 2nd" letters issues [16:21] dobey, not on the "other fields" stuff [16:22] rodrigo_, it seems that e-s-d splits on spaces and tries to search each part in one of the fields [16:22] ah [16:22] I mean, in all fields [16:23] then split the search string and try to highlight that, I guess [16:23] rodrigo_, so, if you type "ro mo", it will find your name [16:23] ah, /me tests [16:23] alecu: right [16:23] yeah, right [16:24] alecu: and wasting time writing the complex code to deal with 'hilighting' results is best avoidable if the final result should be 'don't bother hilighting characters in the name' [16:24] which, i think, is the better route to take anyway [16:24] yeah, I don't think highlighting is a great idea, given we just show the results that match [16:25] alecu, so, try to do it as easy as possible, so that we can remove it easily if we decide to [16:25] also, it would show names not highlighted, given we search all fields, not just names [16:25] rodrigo_: the easiest possible path for doing the hilighting, is quite complex :) [16:26] it's not that complex [16:26] dobey, rodrigo: but highlighting seems the current trend... both chrome and firefox highlight on the url bar [16:26] rodrigo_: constantly inserting and removing markup in a string is pretty complex [16:27] alecu: chrome and firefox don't show an icon view, they pop up a drop-down list, and hilighting isn't exactly what they do [16:27] dobey, the way I'm thinking of doing it is by only redoing the string, never removing stuff. [16:27] also, they only search in visible strings [16:27] dobey, well, right: we are calling "highlighting" to what we should be calling "bolding" [16:28] dobey, and also rodrigo has enabled a list view instead of the icon view [16:28] i still maintain that it doesn't really make sense here [16:29] alecu: but searching in our contacts picker is always a narrowing of the view, that isn't the case in other places [16:29] dobey, yes: on both those places it is a narrowing of your search history [16:29] firefox for example always shows N results at most, and those results can change drastically [16:30] alecu: no, if you type "h" the drop-down never shows your entire history :) [16:30] alecu, you don't need to redo the string, just make a new one and set the markup column on the model [16:31] rodrigo_, yes, I'm doing exactly that [16:31] alecu, oh, although the list view doesn't have the markup column though, you'll have to add it and hide/show the normal one or the markup one [16:31] rodrigo_, but the complex part comes when typing two or more words separated by spaces [16:32] alecu, I'd say just split the search string by spaces, and try to hightlight one or both [16:32] rodrigo_, because the parts could overlap... [16:32] oh, right [16:32] so, I'm thinking of doing a search for each part [16:32] i think "de la P" should match "de la Pena" rather than being separate searches [16:33] dobey, but I think "ro mo" should match rodrigo [16:33]