IRC Logs for #crux-devel Thursday, 2006-10-19

*** pitillo has joined #crux-devel00:07
*** Romster has quit IRC00:07
*** cptn has joined #crux-devel01:16
*** ChanServ sets mode: +o cptn01:16
*** sip has joined #crux-devel02:02
sipmorning. before we go officially live with git, do you think we should organize the repos in subdirs? (ie: tools/, ports/)02:11
cptnhi sip02:11
cptnsounds sensible02:11
cptnmmmh, is Falk going to be in charge of public BTW?02:12
sipI think so02:12
cptnI've just seen in #crux that it's in the navigation already02:12
cptnand publically editable02:12
sipthers's also a news item on crux.nu02:13
cptnah yeah02:13
sipgah, now we have to change all the headers for web apps :)02:13
cptnyes :-)02:14
cptndo I have to be in a special group for contrib?02:19
cptnsince I already have an account anyway...02:20
sipjust put yourself in the contrib group02:20
cptnheh, okay02:20
sipcptn, did you have any other solution in mind?02:24
cptnno, that's perfectly fine02:25
cptnI was just wondering whether there are groups at all02:25
cptnbut it's definitely a good idea02:25
sepensip, 'news' item could show latest 10 added ports02:27
sipnot a bad idea. There was some proposal to add a news box with latest ports/commits/etc02:28
sipa sort of compact timeline
cptnhowever, what we really need are patches :-)02:29
cptnthe ideas are already there02:29
sepenand ... are there any rule for style naming ports?02:29
sepeni.e: foo-pluggin-bar  or  foo-bar-pluggin ?02:30
sipplus we already have rss:
sipno formal rule AFAIK02:30
sepenI think would be a good job to write a document rules and also a modification of 'prtverify' to check port names02:32
cptnwe tend to trust the maintainers02:32
cptnthe more rules you have, the more people you need to enforce them02:32
sipsepen, I think the best option is to contact the maintainers of ports with inconsistent names and ask for a change02:34
sepenyes of course, but a minimal guideline also02:35
sepenavoiding: foo_bar , Foo-Bar, ... ->> foo-bar02:36
sepenprtverify could perform this feature02:37
sipthe only half-inconsistent ports that I can spot on core/opt are sdl_*02:39
sipbut the port name reflects the upstream name in tis case02:39
sipdo you have other examples in core/opt? (or maybe in contrib?)02:40
cptnthere's also p5-* vs perl-*, and perl vs ruby vs python02:42
sipah, right, such as in ruby-cairo Vs. mysql-ruby02:45
cptnOTOH, many of the inconsistencies come from upstream02:45
cptni.e. python-pygtk sounds silly :-)02:46
sipnot sure about forcing changes in a case like that, since people woul likelly search for pygtk02:47
cptnyou'd still find it :-)02:47
sipwas thiking about python-gtk02:47
sipgah, I can't spell today02:47
cptnis it just me, or did the forum attract a wave of newbies?02:48
sepensip, tcp_wrappers (core)02:49
cptnhow is that inconsistent?02:49
sipwell, that's the upstream name02:49
sepencddb_get, libart_lgpl, libtxc_dxtn,  mod_*, ... (opt)02:50
sepen$ ls -1 /usr/ports/opt | grep '_'02:51
cptnsepen: please02:51
cptnthe ports are called like they're called upstream02:51
cptnwe don't want '-' everywhere02:52
cptnwe don't rename the packages, if the author calls it mod_*, we call it mod_* too02:52
sepenwell, I also have '_' ports in my repo, it's only a opinion02:56
cptnwell, what are you trying to say?02:57
sepenI think that we need naming rules02:58
cptnbut you've shown only examples which follow a rule already02:58
cptn"if the author calls it X, call it X too"02:59
cptnfor example, I don't understand why you call fglrx 'ati_driver'03:02
sipinstead of ati-driver? :-P03:04
sepencptn, yes I have _ ports but my repo ins't contrib-core-opt03:07
cptnbut using '_' is perfectly legal03:08
cptneverywhere, in core, opt, contrib too03:08
cptnlet me repeat what I said before:03:08
cptn"if the author calls it X, call it X too"03:08
cptnif it's called mod_php, we call the port 'mod_php'03:09
cptnif the software is called libart_lgpl, we call the port 'libart_lgpl'03:09
cptnand so on03:09
cptnthat's what most people understand when we talk about the 'name' of a package, as denoted in the Pkgfile03:10
cptnit's not a description, or a fantasy name03:10
cptnit's the name of the sortware which this port builds03:10
cptntherefore, it's strange that you make a port for fglrx_6_8_0-$version-$release.i386.rpm, and call it 'ati_driver'03:10
cptnand not 'fglrx'03:10
sepenthat's correct03:11
cptnit's not about underscores, using those is perfectly fine03:11
sepenbut BitchX ->> bitchx03:11
sepennon capital letters03:11
cptnas stated in the rules03:11
sepenwell, I'm satisfied now xD03:12
cptn says:03:12
cptn"The name of a package should always be lowercase (e.g. name=eterm and not name=Eterm)"03:13
sepencptn, and what about to install in /var ??03:14
cptnwhat about it?03:14
sepencan I install on /var ?03:15
cptn"In general packages should install files in these directories. Exceptions are of course allowed if there is a good reason."03:15
sepen/var/webfs ??03:15
cptnall web servers install to /var IIRC03:15
sepenI ported webfs such as daemon03:16
cptnyes, I remember that03:16
sepenbut now a new release03:16
*** sip has quit IRC06:03
*** sip has joined #crux-devel06:10
*** sip has quit IRC06:18
*** cptn has quit IRC06:32
*** cptn has joined #crux-devel06:33
*** ChanServ sets mode: +o cptn06:33
*** jaeger has joined #crux-devel08:11
*** j^2 has joined #crux-devel09:02
j^2morning guys09:05
cptnhey j^209:06
cptnjaeger: I disabled immediate notifications for crux-devel09:08
cptnI hope that's okay with you09:08
aonhi j^209:10
aon& others :)09:10
jaegercptn: shouldn't be a problem09:15
jaegerand hello, all09:15
cptnhi :-)09:15
aoncups is slow to print big images :)09:15
cptnthat's just cups way to help safe the forest09:16
aonprinting just a line of text is more of a waste than printing a whole-page image imo :)09:18
cptnbut nothing prevents you better from printing at all than making it a PITA09:20
cptnare you printing pictures of your car again? ;-)09:20
*** sip has joined #crux-devel09:20
aoni've never printed any of those09:24
jaegeranyone tried flash9 yet?09:27
cptnI haven't, but apparently Han made a port09:28
jaegerbah. sound sync is still awful09:29
jaegerweird... it gets better as the flash plays09:30
cptnI think they promised this to get better in v9, right?09:30
sipit's a diesel09:30
j^2ahh yes the last day of the conferance09:35
j^2then back to simple admining :P09:35
j^2yep, how is everyone today?09:48
sipj^2, I'm fine thanks. What about you? Did you remeber do bring a pillow for the conference?09:51
j^2sip, hehe, great hopped up on caffiene myself, making everyone's day like always, yah i've actaully just been wowing instead of paying any attention :P09:52
cptnmmmh, I think it's time to write down a description what crux-devel@ is supposed to be10:14
*** treach has joined #crux-devel10:16
cptnhow does that sound?10:27
jaegerlooks ok to me10:30
pitilloseems short and clear for me. (May be good to add one for this channel too?... only a opinion)10:30
cptnpitillo: yeah, good idea10:31
j^2i like it....though i know my pull isnt much :P10:34
cptnI think you're underestimating your pull :-)10:34
j^2aww thanks cptn10:36
cptnokay, I've uploaded it10:41
cptnalso added to
cptngotta love the git hashes in the timeline :-)10:50
cptnhow geeky is that10:50
siptoo much ;)10:50
sipbtw the timeline could need some refactoring if we want to make it flexible enough for other uses (ie a news column)10:51
cptnjaeger: flash9's sounds sync looks pretty good here10:53
*** sip has left #crux-devel10:53
tilmancptn: are you series about this 'app spotlight' thing?10:54
tilmanserious, wtf10:54
cptnif there's anyone to do it, why not10:54
cptndo you think it's a bad idea?10:54
tilmanseems a bit silly, tbh10:55
tilmanbut sure, if people like it, go ahead10:55
pitillocptn, ummm what about put that url at channel`s topic? It`s interesting keep that descriptions in mind for a good use.10:56
*** cptn changes topic to "CRUX developer channel ยท"10:59
cptntilman: I've discussed with Simone earlier today whether it's cruxy or not11:00
cptnbut personally I've found so many interesting ports not via the ports tree but via the web that I believe it could be nice11:00
cptnwell, I found the software in the net, not the ports of course :-)11:01
cptnbut the ports were there11:01
cptnjaeger: it's still a bit of, but much less11:02
*** prologic has quit IRC11:03
*** prologic has joined #crux-devel11:10
sepenI was build new release for bluez-* (2.25 to 3.7), where I can make changes for this port, Im newbie with flyspray, I must use flyspray for that?11:33
*** sepen has quit IRC12:40
*** sepen has joined #crux-devel12:45
*** j^2 has quit IRC14:30
*** j^2 has joined #crux-devel14:32
*** sepen has quit IRC14:36
*** j^2 has quit IRC14:53
*** cptn has quit IRC14:53
*** jaeger- has joined #crux-devel15:34
*** jaeger- has quit IRC15:37
*** jaeger has quit IRC15:45
*** jaeger has joined #crux-devel15:45
*** sip has joined #crux-devel15:55
*** sip has quit IRC16:58
*** jaeger has quit IRC17:38
*** jaeger has joined #crux-devel18:26
*** treach has quit IRC18:33
*** pitillo has quit IRC19:14
*** jaeger- has joined #crux-devel20:13
*** jaeger has quit IRC20:30

Generated by 2.11.0 by Marius Gedminas - find it at!