IRC Logs for #crux-devel Tuesday, 2008-08-05

*** sepen has joined #crux-devel01:06
*** Rotwang has joined #crux-devel03:10
*** fetid has quit IRC04:56
sepenwho will send the meeting remainder mail?05:37
Rotwangyou ;]05:37
sepencptn is on holidays05:37
sepenRotwang, I'm not in the devteam05:37
sepenbut maybe its better than don't have it05:38
sepentilman, ping05:38
aonwhat do you mean by "devteam" :)05:39
sepenwell the coreteam05:39
aonah, right05:39
sepentilman, jue, sip and cptn05:40
aoni never got any memo about the notify sender having to be a core member :)05:40
sepenah ok05:40
aoni don't think anyone will be disrupted if you send it05:40
sepenjust I suppose after cptn assumes this role05:40
sepenok I'll send05:41
teKmind putting me on cc:?05:42
sepenI've a reason for don't doing it05:42
aonteK: you're not on the list?05:42
sepenthe next week Ill take vacations too05:42
teKaon: crux-dev?05:42
aonsepen: well whatever, i'll send it :)05:42
sepenaon, ok better05:43
teKI doubt.. another point on my todo. Is it public?05:43
aoni think05:43
teKI will try to subscribe05:43
aonwell people are able to send maintainer applications to it so it must be public :)05:43
teKwhich explains why I did not attend any meeting up to now05:43
sepenaon, there aren't any items on the Wiki Agenda page ATM05:44
teKI knew there was a conspiracy going on behind my back :>05:44
aonsepen: i know05:44
sepenaon, but I think portdb-cli should be a nice item to add05:44
sepenand also the last tickets appeared05:44
sepennice! I've some questions for the meeting but I think isn't necessary to add them in the Wiki05:55
sepenlike number of days in the Timeline page05:55
sepenATM 30 seems too big05:56
aon-a :)05:56
sepenalso I noted that they're still a ton of ports using x11 as a dependency05:56
sepenI'm preparing my notes :)05:57
aongood :)05:57
sepenaon, you known my repoverify page?
sepenI changed the skin to feel like crux05:58
Rotwangsomeone should adopt hplip05:59
Rotwangor drop it05:59
sepenhmm still its maintained by predatorfreak06:02
Rotwangnot any more ;]06:03
sepen/usr/ports/opt/hpijs seems similar, right?06:03
sepenwell if you're interested feel free to adopt it06:04
Rotwangsepen: im not06:04
Rotwangbut there shouldnt be any orphaned ports in contrib06:04
sepenI don't have HP printers06:04
sepenwell so we will change the maintainer line to unmaintained06:05
sepenand open a ticket06:05
sepenimho one or two week should be enough06:06
sepenif not it should be purged from contrib, no?06:06
sepenwhat do you think?06:06
Rotwanggood for me06:07
Rotwang8 ports still orphaned by predatorfreak06:08
Rotwangsorry less06:09
sepenok Ill modify them06:09
Rotwangmd5deep and hplip06:09
nipuLare we meeting tomorrow?07:28
nipuLi said something funny?07:28
nipuLah, atti just posted on the ML07:29
Rotwangyou said tomorrow, its not that funny tho07:30
nipuLwell it is tomorrow for me07:41
*** Rotwang has quit IRC07:42
nipuLok, this one has me confused07:44
nipuL!FS#340 url07:44
spraybotFS#340 -
*** sepen has quit IRC08:07
*** fetid has joined #crux-devel08:49
*** j^2 has quit IRC09:17
*** fetid has quit IRC09:28
*** fetid has joined #crux-devel09:29
*** treach has joined #crux-devel10:06
*** Rotwang has joined #crux-devel10:20
*** fetid has quit IRC10:25
tilmanaon: good job :D10:40
*** treach has left #crux-devel10:45
*** tilman has quit IRC10:54
*** nipuL has quit IRC10:54
*** teK has quit IRC10:54
*** Rotwang has quit IRC10:54
*** aon has quit IRC10:54
*** rehabdoll has quit IRC10:54
*** aon has joined #crux-devel11:02
*** Rotwang has joined #crux-devel11:02
*** rehabdoll has joined #crux-devel11:02
*** teK has joined #crux-devel11:02
*** tilman has joined #crux-devel11:03
*** nipuL has joined #crux-devel11:04
*** treach has joined #crux-devel11:08
*** Rotwang has quit IRC11:10
*** sip has joined #crux-devel12:53
*** Rotwang has joined #crux-devel12:59
tilmanhi sip! :)13:00
*** jue has joined #crux-devel13:00
tilmanhi jue13:00
sipfinally on vacation \o/13:00
Rotwanghi all \o/13:01
juenice to see you, sip13:01
sipthanks. really sorry abut my lacking of contribution during the last moths, but things got messy at wwork13:02
sipI'll lurk a bit and try to get in touch again with the latest news ;)13:02
teKhello people13:02
tilmansip: that's okay, though it was a bit annoying at times, too ;D13:03
sipI know, I know  *throws a stone to himself*13:03
tilmansip: can you (eventually :D) write a short howto on updating
tilmanwe really need to do that :]13:04
sipall the machine?13:04
sipthou shall use the wonderful pkg-get ;-)13:05
tilmanwe were mostly wondering about stuff like: is there  a rescue system we can boot into?13:05
tilmanor do we have to call the kalmar guys to weave some magic?13:05
sipno, there's a daily backup of etc, home and few other stuff afair13:06
tilmanso you're saying we can just pkgadd new stuff while the system is running? :D13:06
sipwell that worked last time(TM)13:06
sipbut I was pretty much sure cause I updated a few customers machine that way13:07
sipobviously you have to fix some stuff by hand (ie the syslog var in /etc/rc.conf)13:07
siphehe, somebody messed up with the toolbar while updating flyspray13:09
tilmanthat was me13:13
tilmanthe FS update didn't go that smooth13:13
tilmansomehow we managed to break mysql and had to update that as well etc ;)13:14
sipyeah there was some custom code inside13:14
sipoh :-)13:14
aonsip: btw, there's an xterm 23613:14
sipthanks, lokking at my ports atm13:15
sipand looking for a spell checker too13:15
aonoh, and sudo 1.6.9p17 :)13:16
siphold on, I'm still at crdao ;-)13:17
sipback in a few mins13:18
tilmanjue: btw, i'm holding off on the udev update until viper updates opt/cryptsetup. cryptsetup calls 'udevsettle' atm, which isn't available in the new udev anymore (replaced by 'udevadm settle' or somesuch)13:19
*** thrice` has joined #crux-devel13:20
jueok, please have a look at FS#270 as well13:21
tilmangngngng yes13:21
jueso if we are there, we have two new tickets13:23
spraybotFS#336 - pkgmk and source files downloaded by hand13:24
tilmani don't know about that one13:26
tilmannot sure it's an issue that you have to put file:// in front of the filename :)13:26
jueI've added my comment to the ticket, looks like sepen mostly agree on that13:26
jueyou've read my comment?13:26
tilmanthe way i understood sepen's comment he still doesn't like that solution much13:27
juehmm, I'd say he complains that pkgmk dosn't error out anything else than http:,ftp:, file: etc13:28
tilmantoo bad sepen isn't around right now13:29
tilmanre. #34013:32
tilmanif we apply that patch, we'd have to do the same for every other pkgmk switch that also has a variable in /etc/pkgmk13:32
tilmanactually, just -no-strip i guess13:32
jueyep, removing the PKGMK_IGNORE_FOOTPRINT would be the best solution IMHO ;-)13:33
tilmani've never heard of anyone setting that variable to yes ;p13:33
jueme too, i don't get why we need it13:33
Rotwangmismatches occurs quite often i noticed13:33
jueno for me13:34
tilmanwe could ignore the bug for now and see whether anyone else votes for it13:36
sipthat's the spirit :P13:36
aon+1 :)13:36
juetilman: agreed13:36
sipbtw, any comment on #277 and #278?13:36
tilmanother than 'java sucks anyway'?13:37
sipI mean, do you prefer a stripped jdk/jre version too?13:37
sipnah, I'm learning to love it13:37
jueit's your port, and you have good reasons, so it's ok for me13:37
aoni think we should keep them the way they are13:37
aoni don't want to use a broken java for my "Programming 101" class in a month :)13:38
sipgood. also, i'm not sure about the legal implications of removing stuff13:38
tilmanjue: i cannot promise that i'll have time for eg #335 in the near future13:50
juedosn't matter13:53
teKI do have some questions/proposals, anyone ready for listening? :)13:53
teKwho of you ever did the pkginfo -i | grep foo?13:54
juetilman: seems that no one has noticed that bug13:54
tilmanjue: yup13:54
RotwangteK: prt-get listins is also very nice13:54
tilmanteK: the bug? or what?13:54
tilmani do pkginfo -i13:54
tilmani do pkginfo -i |grep foo all the time13:54
teKyes, listinst does work, too. Why don't we add a feature that pkginfo -i invokes grep automatically if called with more params13:55
tilmanhack ^ 123 ;p13:56
teKjust a suggestion13:56
tilmanteK: but yes, in principle :)13:56
tilmanlet me look at the bug again13:56
teKI figure, you guys do like patches more than suggestions :>13:56
tilmanhow about this;13:57
spraybotFS#220 - pkginfo -i foo13:57
tilman"pkginfo -i" will list everything, as it does now13:57
tilman"pkginfo -i some_regexp" will do the expected thing ;)13:57
teKyes, the [foo] should be merely optional13:57
tilmanthat's what sepen hinted at in the original summary13:57
teKcopycat, got me *rolleyes*13:57
tilmanteK: but really, let's not touch pkgutils5 with that :]13:58
tilmanit would be very easy to add it in pkgutils613:58
teKdue for 2.5?13:58
sipnice idea tho13:58
tilmanteK: dunno, depends on how much we will delay 2.5 :D13:58
teKokay. Who will take over the patching13:59
juepatching of what?13:59
tilmando you know c? :p13:59
tilmanjue: adding support for regular expressions/grepping in pkginfo-i14:00
juein pkgutils5?14:00
teKtilman: a little bit14:00
teKjue: no14:00
tilmanteK: basically you'd have to check for an additional argument when -i was detected14:01
tilmanthen, run regcomp() on that argument to compile the regular expression14:02
tilmanif it fails, tell the user that he's an idiot ;)14:02
tilmanteK: basically read src/pkginfo.c ... then you'll realize how it could be done :D14:02
teKthe basic picture is clear to me14:03
teKmy next question regards lvm2 and cryptoloop support during installation-time14:05
tilmancryptoloop? you mean cryptsetup, right?14:06
*** RyoS has joined #crux-devel14:06
RyoSo hai14:06
tilmanhi RyoS14:07
RyoShow are you guys?14:07
teKI have a somewhat working base regarding the encrtyption during setup; the last thing to complete the puzzle will be the initrd for the first (and next) boot14:07
teKtilman: yes, cryptsetup14:07
tilmaninitrd :x14:07
tilmanoh, you're talking about an encrypted /14:07
teKcrux-2.4.iso does use one, too.14:07
tilmanand /usr and whatnot14:07
teKyou may encrypt /home only, too. But then again cryptsetup won't be required during installation14:08
teKopions, anyone?14:08
tilmangood morning nipuL :D14:08
nipuLit couldn't hurt to add them to the install cd14:10
RyoSteK: crux for paranoids? i am in14:10
tilmanports-wise it would mean we had cryptsetup in core, right?14:10
tilmanand the iso would have to have extended initrd support?14:10
RyoSid make it opt14:10
teKthere are dependencies14:10
teKlet me have a quick look14:11
tilmanRyoS: btw, re. your application. i wanted to ask you a bunch of questions, but a) i forgot what i wanted to know and b) i'm busy now ;D14:11
tilmanteK: libdevmapper14:11
tilmanech, libgcrypt, libgpg-error, popt14:12
RyoStilman: no problem. only thing is my ipv6 broker is unreliable and i am too arrogant to connect via v4 lol14:12
teKtilman: right14:12
RyoSi am just here for the meeting with xchat right now :)14:12
RyoSbut you can mail me or jabber maybe anytime14:12
tilmanRyoS: ah, i wondered why you weren't logged in14:12
tilmanteK: well, i wonder whether that stuff really has to be in core if we do this :]14:13
tilmanit would be nice to keep core minimal of course14:13
RyoScount that not everybody is paranoid like hell14:13
teKtilman: opt should be fine?14:13
RyoSso making it opt would fit perfectly fine i guess14:13
tilmanyeah, i guess14:13
teKthere will be a separate, optional section in the CRUX handbook, too14:14
juetilman: don't think so14:14
teKRyoS: +114:14
jueteK: how about adding a ticket?14:15
tilmanwell, i know nothing about initrds, so i cannot really comment there14:15
tilmanin general: the smaller the diff for iso.git is, the more likely it is to get merged :D14:15
teKjue: sure14:16
tilmanteK: having support for LVM would be a good thing, i remember that a buddy of mine chose arch over crux because of this :D14:16
nipuLtoo lazy to make a custom boot disk14:16
teKI'd like to take over those two parts14:16
teKso I will open a ticket (and post to crux-dev?)14:17
tilmango ahead14:17
teKfirst I will phinish this phonecall :-)14:17
aontilman: that's not necessarily a bad thing, at least you don't have to hear him complaing "z0mg it lacks BASICS"14:21
treachlvm isn't really "teh basics", but it would be nice if the process was supported during setup14:22
aoni was talking about the basics that would've been missing if installing it with lvm would've been possible :)14:23
teKthe cool thing is: this stuff is merely optional and won't change anything if you don't want it to14:23
tilmani like those best14:23
aoni wonder if an army buddy of mine is still running that installation of Argh he installed after he had tried crux :)14:23
nipuLi suppose now is a good time to mention
nipuLi put the ports for it here
sip0.0a? :-)14:26
tilmansounds scary14:26
nipuLsource is 3k14:26
tilmanoh, it's actually there14:26
nipuLmake that 1.8k14:27
tilmancould you run httpup-repgen there? :p14:28
tilman"Of course there are bugs, but I prefer to call them features." haha14:31
sipis that a full-featured project? ^^14:32
nipuLthere's no error handling and the command line argument parsing is a bit cludgy14:32
tilmancould we swap the current portdb with your rewrite now, btw?14:34
tilmanor is there still stuff missing/broken/..?14:34
nipuLi've implemented everything the original did14:34
nipuLmy only concern is if there's any "features" i haven't noticed14:35
nipuLthe db is read only to the script so it should be safe14:36
tilmanwe could ask for testers on the crux mailing list and see whether anyone actually cares =)14:37
*** Rotwang has quit IRC14:37
nipuLi announced on crux@14:37
juetilman: you mean care about pdb?14:38
tilmanno, the rewritten ports database14:38
tilmannipuL: sorry, i forgot about that14:39
nipuLwell the main reason for the portdb rewrite was to make pdb easier to write14:40
juemaybe we should ask sip as well, IIRC he wrote the old db14:40
sipare you talking about ?14:41
tilmani'm confused about that all the time. i thought tillb wrote the db14:41
tilmansip: +
sipI re-implemented it to make it small and cache data in the bg14:41
sippmwiki wants do display it14:42
sipgot it14:42
tilmanyeah, stupid me copied from the wrong mail14:42
sipyeah got it from the ML update14:43
nipuLstupid me posted the wrong url14:43
tilmansip: but the git link that nipuL included isn't tillb's version that added (halfway completed) xml support, is it?14:43
sipno I think it's the one in git14:45
sipI mean14:45
sipmaybe till did not push changes into the tree14:45
tilmansince tillb didn't work on the portdb/the xml thing in ages it would be nice to switch to nipul's version14:46
tilmansip: unless there's stuff in it that you object to :D14:46
nipuLthe original on it's own is ok, but when you see what was required to just add xml the need for a rewrite is apparent14:46
tilmani think "maintainable" wins over "clever code" here in case php-idiots like me have to fix something :)14:47
sipI think it's a good idea, as long as somebody wants to maintain it ;-)14:47
tilmannipuL: did you contact tillb?14:47
nipuLnot as such14:47
sipheh you kept the sanitize() function14:48
nipuLi noticed to didn't do much14:48
nipuLbut i wasn't sure, it might have been some gross php hack14:48
sipyeah it was there to validate input abck at the time14:49
tilmanlike, prevent sql injection?14:49
tilmanthe db is opened read-only, right? :]14:49
tilman  function doQuery() {14:50
tilman    $this->lazy_programmer();14:50
tilman  }14:50
sipthat pictures what happened :-)14:50
tilmanmmh, there's zero comments in that code14:51
teKit was hard to write it, hell it should be hard to read, too14:51
nipuLit needs documenation?14:52
sipno, it was 200 lines and wrote it on the train14:52
sip<insert other lame excuses>14:52
nipuLmine bloated out to 455 lines14:53
nipuLand i wrote it will i should have been working14:53
nipuLso essentially i got paid to write it14:53
* tilman runs off15:02
nipuLkick that pillows arse15:02
spraybotFS#341 - Add support for dmcrypt to the installation ISO15:05
spraybotFS#342 - Add support for lvm2 to the installation ISO15:05
sip /o/15:07
teKreminds me of:
sipodd I can't connect to bash.org15:09
nipuL\o| _o_ \o/ o<15:09
teKsip: Italy picks up censorship, too? *duck*15:10
treachdoesn't work here either.15:10
* treach blames the FRA15:10
*** j^2 has joined #crux-devel15:17
sipgood night15:28
*** sip has quit IRC15:28
*** thrice` has left #crux-devel15:29
teKmeeting seems over15:33
teKgn8, ppl15:33
RyoSnn teK15:48
*** treach has left #crux-devel16:20
*** jue has quit IRC16:33
*** viper_ has joined #crux-devel16:58
*** treach has joined #crux-devel17:11
*** RyoS has quit IRC18:05
*** Rotwang has joined #crux-devel18:05
*** viper_ has quit IRC18:34
*** treach has quit IRC18:55
*** Rotwang has quit IRC20:00
*** fetid has joined #crux-devel21:14

Generated by 2.11.0 by Marius Gedminas - find it at!