--- Log opened Tue Jan 01 00:00:28 2008
00:48 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit [Remote closed the connection]
00:54 -!- dionak [n=dkidd@cpe-024-074-132-143.carolina.res.rr.com] has joined #webgui
00:54 -!- dionak [n=dkidd@cpe-024-074-132-143.carolina.res.rr.com] has left #webgui []
01:28 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit [No route to host]
02:22 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
02:22 -!- mode/#webgui [+o preaction] by ChanServ
03:40 -!- dionak [n=dkidd@cpe-024-074-132-143.carolina.res.rr.com] has joined #webgui
03:40 -!- dionak [n=dkidd@cpe-024-074-132-143.carolina.res.rr.com] has left #webgui []
05:18 < CIA-19> WebGUI: colin * r5217 /WebGUI/t/Asset/Asset.t: add tests for getTitle and getMenuTitle
06:18 < CIA-19> WebGUI: colin * r5218 /WebGUI/t/Asset/Asset.t: tests for getToolbarState, toggleToolbar and getUiLevel
06:21 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit ["Leaving"]
08:21 -!- SDuensin [n=Scott@12-217-75-184.client.mchsi.com] has quit [Read error: 110 (Connection timed out)]
10:17 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
10:17 -!- mode/#webgui [+o preaction] by ChanServ
14:03 < CIA-19> WebGUI: doug * r5219 /branch/doug-experimental/lib/WebGUI/ (Asset/Report.pm i18n/English/Asset_Report.pm): Nightly checkin because you forgot
14:50 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit [Connection timed out]
15:10 -!- SDuensin [n=Scott@12-217-75-184.client.mchsi.com] has joined #WebGUI
18:11 -!- Radix__ [n=Radix@203.161.71.161.static.amnet.net.au] has joined #webgui
18:30 -!- Radix_ [n=Radix@203.161.71.161.static.amnet.net.au] has quit [Read error: 110 (Connection timed out)]
19:38 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
19:55 -!- Haarg [n=haarg@66-188-98-191.dhcp.mdsn.wi.charter.com] has quit ["This computer has gone to sleep"]
20:23 < perlmonkey2> Hello?
20:23 < perlmonkey2> Everyone still hung over?
20:40 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit [Remote closed the connection]
21:38 -!- BartJol [n=Administ@54-167.surfsnel.dsl.internl.net] has joined #webgui
21:39 < BartJol> well happy new year everybody
21:45 -!- BartJol [n=Administ@54-167.surfsnel.dsl.internl.net] has left #webgui []
21:48 < CIA-19> WebGUI: colin * r5220 /WebGUI/t/Asset/Asset.t: 
21:48 < CIA-19> WebGUI: Add the failing test for assetAddPrivileges back in, but don't run it.
21:48 < CIA-19> WebGUI: Add tests for assetUiLevel and getUiLevel.
23:48 < CIA-19> WebGUI: colin * r5221 /WebGUI/t/Asset/Asset.t: Coverage tests for assetExists.
--- Day changed Wed Jan 02 2008
00:10 -!- dannymk [n=danny@68.15.55.161] has joined #webgui
00:12 < dannymk> cap10morgan, please contact me when you get a chance.  I would like to add your patch to my module, however I have a few questions.
00:33 -!- perlmonkey2 [n=perlmonk@ip68-97-51-70.ok.ok.cox.net] has quit ["Leaving."]
00:43 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
00:43 -!- mode/#webgui [+o preaction] by ChanServ
01:18 < dannymk> Hi preaction
01:18 < dannymk> happy new year.
01:18 <@preaction> indeed
01:19 < dannymk> I am in the process of adding Authentication 2.0 to the OpenId module
01:20 <@preaction> nice
01:21 < dannymk> as you requested.   However, the rest will have to wait.  Just started a new Perl training job that is taking up a lot of my time.
01:22 < dannymk> Have you seen cap10morgan on this channel in the last month?
01:22 <@preaction> no problem, i've finally got time and my own branch to start experimenting with, but i've got some productivity things to do first (since everything I do would be so much easier if I didn't copy/paste it every single time, and instead subclassed it)
01:22 <@preaction> yes, once in a while
01:25 < dannymk> I need to talk to him about his patch.  Where can I find his email?
01:26 <@preaction> is his profile private?
01:26 < dannymk> yes
01:29 <@preaction> you might want to just reply to the thread, he'll get an e-mail if he's subscribed to it
01:29 < dannymk> Done!
01:30 < dannymk> I also would like to work on the WChat object.  Since I have been away a while I expect someone else took on the module.
01:30 < dannymk> has it been worked on?
01:30 <@preaction> nope, we've been busy on other things
01:31 < dannymk> Darn, I need to work on that too.
01:31 <@preaction> I've got a little ajax-enabled Shoutbox asset though, if you want the code for that
01:31 < dannymk> Really would love to have that Chat WObject
01:31 <@preaction> i made it as part of the WUC last october
01:31 <@preaction> you still haven't gotten it? we tried about 10 different ways
01:32 < dannymk> yes, I have it now.  Just have not worked on the code
01:32 <@preaction> ah
01:32 < dannymk> would like it to be done!  :-P
01:32 < dannymk> Guess, I have to work on it.
01:32 <@preaction> it'd be nice, to be sure
01:33 <@preaction> but imho there are more important things. the chat wobject is a flashy eye-candy thing, i've got internal mechanical things that i need to work on to make webgui better
01:33 <@preaction> that isn't to say that the chat wobject won't bring more people into webgui than my work
01:35 < dannymk> you guys have your hands full and that is why I help when I can.
01:35 < dannymk> I appreciate the software.  It is clean, simple and somewhat well documented.
01:35 < dannymk> :-)
01:35 <@preaction> have you seen the new URL / Content handlers, they're insanely awesome
01:36 <@preaction> the new developments in 7.5 (HEAD) are going to kick even more ass than 7.4 did
01:38 <@preaction> perlbot crud
01:38 <@preaction> perlbot learn crud as Create, Retrieve, Update, Delete -- the four corners of a proper database management tool
01:38 < perlbot> added crud to the database
01:40 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
01:47 < dannymk> Not yet, very immersed in this new project I have in my new job
01:47 <@preaction> understandable, new jobs are always fun
01:47 < dannymk> I hate sysadmin/developer work but I hae to make a living :-)
01:47 < dannymk> hae=have
01:47 <@preaction> why do you hate sysadmin/developer work? what would you rather be doing?
01:47 < dannymk> that is why I don't login during the day any longer
01:47 < dannymk> I just rather be just a developer
01:48 <@preaction> ah
01:48 < dannymk> in these jobs I have to be a jack of all trades.  Sysadmin, DBA, Developer, Designer, etc...
01:48 <@preaction> eh, i'm a generalist. i learned how to be a developer, and now that i've got a very decent job doing that, i'm teaching myself how to sysadmin / netadmin / network, etc..
01:48 <@preaction> i like knowing as much as possible about everything
01:49 < dannymk> I have been doing it almost 20 years and my mind can't take all the junk :-(
01:49 < dannymk> just too much to learn
01:50 < dannymk> don't mind working on stuff a bit but when it becomes part of my job then it becomes a problem
01:50 < dannymk> that is why I left my last job after 18 years
01:50 < dannymk> was tired of carrying the DBA
01:51 <@preaction> another middle manager trying to cut corners by making you work three jobs?
01:52 < dannymk> NO, we had a sysadmin which I had to train then we had an older guy that really did not understand Oracle working as an Oracle DBA
01:55 <@preaction> much fun
01:55 <@preaction> as they say, everyone gets promoted just to their level of incompetence
01:55 <@preaction> i feel that some people are better at faking competence than others
01:56 < dannymk> Peter principle all the way, but that is the motto of the federal govt.
01:56 < dannymk> oops, did I say that :-)
01:57 <@preaction> surprisingly we've got a few gov't clients, and they're very helpful when it comes to testing / exercising the code we write for them
01:58 < dannymk> Ah, not every office is like that
01:58 <@preaction> i mean, we've probably doubled the size of WebGUI with one of our projects, and there remain little-to-no bugs in their code since they're so thorough in tracking them down
01:59 < dannymk> Good to know.
01:59 <@preaction> but, it just means they've got competent people. that and their entire department lives and breathes their website
02:00 <@preaction> which is another testament to the stability and usefulness of WebGUI ;)
02:00 < dannymk> well, that may be a well designed office.  Can't say the same for the place I worked.
02:00 < dannymk> yup.
02:02 <@preaction> it's almost as if the bigger an office gets, the more people rubbing shoulders and bumping elbows, the more BS gets floated around, the more market-speak team-building d'ya-have-a-second-to-help-this-guy-every-day-for-his-entire-career, and in a creative business like software development, the best people can't thrive
02:10 < dannymk> ah, you have experience in this area :-)
02:11 < dannymk> LOL.  Well, I hope the new year brings us better circumstances.  Take care and again,  Happy New Year....
02:12 < dannymk> later...
02:12 -!- dannymk [n=danny@68.15.55.161] has quit ["Leaving"]
03:45 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit ["Exit- stage left"]
03:59 < ascii> how ya like dstat
04:00 < ascii> erp
04:48 < CIA-19> WebGUI: colin * r5222 /WebGUI/t/Asset/Asset.t: 
04:48 < CIA-19> WebGUI: Cover several other small utility routines. More getNotFound
04:48 < CIA-19> WebGUI: tests need to be written.
06:17 -!- Haarg [n=haarg@66-188-98-191.dhcp.mdsn.wi.charter.com] has joined #webgui
06:18 -!- mode/#webgui [+o Haarg] by ChanServ
07:38 -!- perlmonkey2 [n=perlmonk@ip68-97-51-70.ok.ok.cox.net] has joined #webgui
08:13 -!- perlmonkey2 [n=perlmonk@ip68-97-51-70.ok.ok.cox.net] has quit ["Leaving."]
09:36 -!- AMH_bob [n=bob@alphamega-dmp.xs4all.nl] has joined #webgui
10:11 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit [Success]
10:43 -!- AMH_henry [n=henry@alphamega-dmp.xs4all.nl] has joined #webgui
12:19 -!- Haarg [n=haarg@66-188-98-191.dhcp.mdsn.wi.charter.com] has quit ["This computer has gone to sleep"]
12:19 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
12:19 -!- mode/#webgui [+o preaction] by ChanServ
12:48 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit [Read error: 110 (Connection timed out)]
12:50 -!- BartJol [n=Administ@host1.procolix.nl] has joined #webgui
14:08 -!- SDuensin [n=Scott@12-217-75-184.client.mchsi.com] has quit ["Leaving"]
15:45 -!- SDuensin [n=Scott@49.sub-75-207-93.myvzw.com] has joined #WebGUI
15:47 < SDuensin> Greetings.
15:50 < BartJol> hi
15:52 < AMH_bob> HAPPY NEWYEAR!!!!
15:52 < BartJol> dito
15:53 < Radix__> Yeah, a belated Happy New Year from me too guys :)
16:12 < SDuensin> Blah.  New Year means I had to go back to work.  :-P
16:26 < Radix__> I'm off until monday
16:27 < Radix__> but I wish I was back at work
16:27 < Radix__> 40'C tomorrow, no aircon at home, and power shortages to boot
16:29 < Radix__> tempted to go in tomorrow and take monday off in lieu ;)
16:30 < Radix__> Oh no.. maybe not.. monday is forecast for 40'C too
16:30 < BartJol> Australian whether isn't much fun, I understand
16:30 < BartJol> wheather
16:31 < Radix__> Actually for most of the year it's great
16:31 < Radix__> just summer sucks
16:31 < Radix__> you get days where it's 40-45'C
16:31 < Radix__> last week we had a temp of 45.5'c
16:31 -!- Haarg [n=haarg@66-188-98-191.dhcp.mdsn.wi.charter.com] has joined #webgui
16:31 -!- mode/#webgui [+o Haarg] by ChanServ
16:31 < Radix__> highest on record for december since 1963 
16:32 < Radix__> jan/feb always sucks for temps
16:32 < Radix__> http://www.bom.gov.au/cgi-bin/wrap_fwo.pl?IDW12300.txt
16:32 < BartJol> here it's about 1 C
16:32 < Radix__> yah, well our temp doesn't really go that low generally.. or very rarely ;)
16:37 < nuba> summer here in Rio sucks a bit too
16:37 < nuba> too hot sometimes
16:38 < nuba> thanks god for A/C
16:50 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has joined #webgui
16:50 -!- mode/#webgui [+o rizen] by ChanServ
17:55 -!- khenn [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has joined #WebGUI
17:55 -!- mode/#webgui [+o khenn] by ChanServ
17:55 <@khenn> happy new year =)
18:03 -!- AMH_henry [n=henry@alphamega-dmp.xs4all.nl] has left #webgui []
18:07 -!- AMH_bob [n=bob@alphamega-dmp.xs4all.nl] has left #webgui []
18:08 -!- BartJol [n=Administ@host1.procolix.nl] has left #webgui []
18:21 -!- wgGuest26 [n=wgGuest2@168.8.72.205] has joined #webgui
18:22 -!- wgGuest26 is now known as teflond0n
18:23 < teflond0n> anyone here have any luck importing users into a database for webgui?
18:23 < teflond0n> still havent had any luck with the userImport.pl
18:24 < Radix__> I had no problems at all with userImport.pl
18:25 < Radix__> What's happening exactly? are you getting errors?
18:25 < teflond0n> its saying DateTime isnt defined....   for some reason the perl mods arent wanting to cooperate
18:30 < teflond0n> is there another way to just have it import a csv file or something
18:30 <@rizen> you need to set a TZ environment variable
18:31 <@rizen> like TZ=America/Chicago
18:36 < Radix__> tzselect should help with that
18:36 < Radix__> or tzconfig
18:37 < Radix__> you're going to have other issues later on if you don't sort out the DateTime issue now anyway
18:38 < nuba> spectre would fail for me with 'Cannot determine local time zone' 
18:38 < nuba> until I did a "cp /usr/share/zoneinfo/America/Sao_Paulo /etc/localtime"
18:39 < nuba> do your /etc/localtime file exists?
18:50 < Radix__> actually /etc/localtime should be a symlink - and the correct way to configure it is to use tzselect or tzconfig
18:52 -!- perlmonkey2 [n=perlmonk@casr-dev.fwsm2ppuser.ou.edu] has joined #webgui
18:53 < nuba> makes sense
18:54 < nuba> cp worked fine for me tho ;)
18:56 < nuba> hey, freebsd's tzsetup prefers "cp" too, just tested and it makes a copy, not a symlink
18:57 < Radix__> it'll work.. just not as nice as a symlink ;)
18:57 < Radix__> ubuntu/debian/centos all create symlinks
18:58 < nuba> maybe it's a "Junior" sindrome, these linux kids, always choosing to do different than their daddy (freebsd) ;D
18:59 < Radix__> pfft
19:04 -!- perlDreamer [n=ckuskie@nat039.mxim.com] has joined #webgui
19:17 -!- perlmonkey2 [n=perlmonk@casr-dev.fwsm2ppuser.ou.edu] has quit ["Leaving."]
19:21 -!- perlmonkey2 [n=perlmonk@casr-dev.fwsm2ppuser.ou.edu] has joined #webgui
19:24 < teflond0n> i do see the etc/localtime file
19:25 < teflond0n> but I am running it on Mac osx
19:25 < teflond0n> not sure if that will cause any issues with it
19:38 < teflond0n> hey nuba since I live near the Atlanta, GA area would should it be /usr/share/zoneinfo/America/New_York ?
19:44 < Radix__> if it's in the same timezone as you, then yeah, that'd work
19:44 < perlDreamer> ... and WebGUI slowly wakes up from its week-long slumber
19:46 < nuba> 'morning, webgui!
19:46 -!- Radix__ [n=Radix@203.161.71.161.static.amnet.net.au] has quit ["Windows doesn't have users, it has hostages."]
19:56 < teflond0n> anyone know what the equivalent would be in osx for the tzconfig or tzsetup
19:56 < perlDreamer> sorry, I don't use osx
19:59 <@rizen> you shouldn't need it on osx
20:00 < nuba> i think you can find that under "system preferences"
20:00 <@rizen> osx sets it automatically when you edit your clock
20:00 <@rizen> perhaps you're not using the right perl
20:00 <@rizen> did you do
20:00 <@rizen> . /data/wre/sbin/setenvironment.sh
20:00 <@rizen> note the dot and the space that come before the command
20:02 < teflond0n> i did the command from that directory so it was ./data/wre/sbin/setenvironment.sh
20:02 < teflond0n> no spaces
20:15 < teflond0n> would it be bad to comment that out in the userImport where it asks for the DateTime
20:17 < teflond0n> er possible
20:18 < teflond0n> rizen any more suggestions
20:18 <@rizen> that's bad
20:18 <@rizen> and you need the space
20:19 <@rizen> it's absolutely important
20:19 <@rizen> that's almost certainly where your problem is at this point
20:19 <@rizen> and if it's not, i'd recommend you get plain black support
20:25 -!- torben78 [n=torben@pD95EFDFD.dip.t-dialin.net] has joined #webgui
20:30 < perlmonkey2> Making the YUI widgets look good is killing me.
20:30 <@khenn> which ones?
20:30 < perlmonkey2> rizen: I'm not that great at layout/design.  Is there a way to make regular html inputs look sexy, because the YUI widgets are too hard to manipulate given they are buried in like 3 objects when created with JS.
20:30 < perlmonkey2> khenn: Menu and buttons.
20:31 <@khenn> the newest yui menu is all CSS based
20:31 <@khenn> never played with the buttons
20:31 <@khenn> but we've made some pretty cool looking drop down menus
20:32 < perlmonkey2> khenn: When selecting a question type in the menu, I dynamically generate options for that question type.  The divs grows and shrinks and I have to do a bunch of crazy stuff to make them look like I want.
20:33 <@khenn> hmm, I guess I'm not following.  You lost me with "question type"
20:34 <@rizen> he's working on the survey
20:34 <@rizen> it's a new ajax style interface for the survey
20:34 <@khenn> ah
20:34 < perlmonkey2> If they had a menu option that was multiple select, I'd just use that.  
20:34 <@rizen> pm2, i'm not a good one to ask. both khenn and preaction have played with yui a lot more than me
20:35 <@rizen> i just use it for adding functionality to form controls
20:35 < perlmonkey2> There button objects, while prettier than radio buttons, are still large and clunky.
20:35 < perlmonkey2> s/there/their/
20:35 < perlmonkey2> I could use css to make them pretty, but that isn't what I do :)
20:35 < perlDreamer> rizen: I think maxscience/iPhoneGuy had a 3rd alias that was also mac related
20:36 <@khenn> css is how to make stuff in yui pretty though
20:36 <@khenn> so you will definitely find it difficult to do w/o it
20:37 <@rizen> perlmonkey2 if you make it work and then give me some sort of documentation on one css classes and ids you've exposed
20:37 <@rizen> i can have our designer come up with the pretty css for you
20:37 <@rizen> or rather, i'll just introduce you to our designer when he comes back from vacation
20:37 < perlmonkey2> rizen: that would be great.  
20:38 < perlmonkey2> rizen: I guess that means I can stop futzing with the layout and just throw it on the page and worry about functionality.
20:38 <@rizen> yup
20:39 <@rizen> what's your email address, i'll send an introductory email to steve 
20:49 < teflond0n> can anyone let me know where the sql database is that contains the list of users ?
20:49 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
20:49 -!- mode/#webgui [+o preaction] by ChanServ
20:49 < perlDreamer> it called "users"
20:50 < perlDreamer> and the groups one is called "groups"
20:50 < ckotil> what about settings?
20:50 < perlDreamer> hmmmm
20:50 < perlDreamer> I think it's called settings
20:50 < ckotil> i dont get it.
20:50 < perlDreamer> and the one for assets is called asset
20:50 < ckotil> who came up with that scheme?
20:50 < perlDreamer> some guy named JT
20:51 < ckotil> he's too smart.
20:51 <@rizen> he's an idiot
20:51 <@rizen> you give him too much credit
20:51 < perlDreamer> what?  you mean he didn't come up with that schema?
20:51 < ckotil> coworker of mine installed the new wre on RH5.
20:52 < ckotil> he said it the installation was a breeze.
20:52 < ckotil> and it was really easy to setup 3 sites.
20:52 < ckotil> he rarely touches webgui, but since i talk about it all the time he decided to try it out.
20:52 <@rizen> nice
20:53 -!- torben78 [n=torben@pD95EFDFD.dip.t-dialin.net] has left #webgui []
20:54 < perlmonkey2> I set of the WRE on cent5.1 a few weeks ago and it was extremely nice.
20:56 <@rizen> you guys have no idea how glad it makes me to hear that
20:56 < perlmonkey2> I wanted to throw out an idea and see what people thought.  Since disk is becoming really cheap, but CPU/Ram is still expensive as far as hosting is concerned.  What about a Wobject that would allow files to be remotely hosted, but WebGUI would treat them like local files?  Maybe build in DAV support, or something like it?
20:56 <@rizen> it means that the 400+ hours that I put into WRE 0.8 was worth it
20:56 < perlmonkey2> rizen: Yeah, I would never use anything but the WRE for production now.
20:57 <@rizen> pm2: re your idea, i'd need further description to understand what you'd want and why before i could make a judgement
20:57 < perlmonkey2> rizen: Like the file pile but would store the files on a remote file server.
20:59 <@rizen> why would that be good?
20:59 <@rizen> that would actually increase the use of the CPU and memory because then all those files would have to be served through mod_perl so it could request them from the dav server
21:00 <@rizen> that is unless the dav server also had a straight http url for the file and you were willing to expose that url directly to your users
21:00 < perlmonkey2> rizen: I was actually thinking that WebGUI would do a one time authen for the users ip, and the remote file server would directly serve to them that once.
21:01 < perlmonkey2> This may be a solution that doesn't really fit any given problem though.
21:01 <@rizen> hrmm.. i guess off the top of my head i can't see anyone actually using that so it's not something i'd allow in the core
21:01 <@rizen> i do want to go the other way around however
21:02 <@rizen> i want to create a dav interface into webgui
21:02 < nuba> i've been toying about the idea of sending static content somewhere else, too
21:02 <@rizen> so you can use webgui as a file store
21:02 < nuba> keeping only the webgui brains running in my dedicated box, 
21:02 < nuba> and push uploads etc to a cheap hosting elsewhere
21:02 <@rizen> nuba, you can already do that
21:02 < nuba> you can do that with modproxy and rewrites, yeah
21:02 <@rizen> you can just push all the static uploads to a file server using a remote mount
21:02 < nuba> but how about sync etc?
21:03 < nuba> suppose there's no remote mounting
21:03 <@rizen> and then change your uploads folder to point to http://files.example.com/uploads/
21:03 < nuba> you won't see that in the feature list of these mega-cheap hosting plans
21:03 < perlDreamer> you also won't see WebGUI there, nuba
21:03 < perlDreamer> unfortunately
21:03 <@rizen> you can use ftp as a remote mount
21:03 < nuba> yeah, thats why i dont even think about pushing webgui there
21:03 < nuba> just the uploads etc
21:03 <@rizen> you can also use FUSE scpfs
21:03 < perlmonkey2> those mega cheap hosters allow ftp
21:04 < perlmonkey2> I've found that those mega cheap hosters are also extremely slow.  They brag about 1 million TB connections to the net, but the reality is like 128kb
21:04 < nuba> yeah, then theres the latency between uploading the file to the webgui host and it being pushed to the cheapass server
21:06 < perlDreamer> rizen: tonight could you please turn on the -v switch to testCodebase so I can diagnose the failing DeleteExpiringWorkflows test?
21:06 < perlDreamer> actually, doesn't have to be tonight.  Any night would be good.
21:10 < nuba> to be honest, i didnt stop to think thoroughly about this yet, of coupling webgui dox + cheapass server
21:11 < nuba> for now a single dedicated box is good enough for me
21:12 < nuba> a seducing "next step" lately has been checking the amazon's EC2 thing
21:13 < perlDreamer> I looked at that as a smoke testing platform
21:13 < perlDreamer> Seemed expensive for a volunteer budget
21:13 < nuba> instead of ugly hacking and glueing of scattered hosting services here and there
21:16 -!- Radix_ [n=Radix@203.161.71.161.static.amnet.net.au] has joined #webgui
21:16 -!- mode/#webgui [+v Radix_] by ChanServ
21:16 < nuba> rizen: wont PB.com be interested in funding the smoke test box?
21:18 < perlDreamer> they run a smoke test now
21:18 < perlDreamer> but I was thinking bigger.  Multiple OS'es, multiple WRE's
21:18 < nuba> i see. like a testing farm.
21:18 < perlDreamer> yeah
21:18 < perlDreamer> it's been really hard to get other people into testing wG
21:21 < nuba> what are the requirements if one wants to help?
21:22 < perlDreamer> there are two main testing needs
21:22 < perlDreamer> 1) We need lots more tests to be written
21:23 < perlDreamer> and that can be done piecemeal
21:23 < perlDreamer> 2) I think we need a testing aggregator to take smoke reports from lots of users and coallate them
21:23 < perlDreamer> but we don't have the lots of users part yet so it's a lower priority
21:24 <@rizen> pd: just turned on the -v flag, please let me know when we can turn it off again
21:24 < perlDreamer> okay.  Tomorrow should be good to turn off.  It will just make a BIG smoke report tonight
21:24 < perlDreamer> hm.  Maybe I should add in diagnostics so we can truly do it in one shot...
21:25 < perlDreamer> I think a test may be leaking Session information, which is why the test DeleteExpiredSession test is failing
21:25 <@rizen> pd, i have a spare box running vmware that i use for compiling and testing the wre on multiple platforms (everything except mac and windows)
21:26 <@rizen> however, the problem with using that is when it comes time to build a new WRE, i'll have to resetup all the tests again
21:26 <@rizen> that's more work for me
21:26 < perlDreamer> right
21:26 <@rizen> and i'm already a big bottlekneck
21:26 < perlDreamer> it would be good if it was scriptable
21:27 < perlDreamer> the host would instanciate the virtual sessions, run a test on them, submit the smoke report for each
21:27 < perlDreamer> but I don't know if that is possible on VMware or not
21:27 < perlDreamer> besides, it would be good to get more people in the community involved if possible
21:27 < nuba> vmware has a perl api
21:28 < nuba> so you could switch machines on, off, etc with a script if you want
21:28 <@rizen> all the vmware hosts are up all the time
21:28 <@rizen> so we don't need to turn them on or off
21:28 <@rizen> it's a beefy box
21:28 <@rizen> dual proc xeon with 4gigs of ram
21:28 <@rizen> so it can actually run about 6 os'es at the same time reliably
21:29 < nuba> nice
21:29 <@rizen> pd, i could give you access to the box if you would like to set it up as a test system for multiple platforms
21:29 < nuba> sounds like you're proud of you muscle car^H^H^Hserver ;)
21:30 < perlDreamer> sounds like I need start researching aggregators then :)
21:30 < nuba> webgui@home
21:30 < perlDreamer> smolder
21:30 <@rizen> we could have them all post to the smoke tests forum
21:31 < perlDreamer> yes, but aggregators can also generate reports with the data, like "This test started failing on this platform at that SVN rev"
21:31 < perlDreamer> but while the research gets done, it wouldn't hurt to post them
21:32 <@rizen> ah
21:32 <@rizen> that's very cool
21:32 <@rizen> i didn't know such a thing was even possible
21:32 <@rizen> you rock pd
21:33 <@rizen> btw, hop on IM and i'll give you the login info for that box
21:38 < nuba> rizem, why is it a good idea for spectre to handle tasks to a mod_perl enabled httpd instead of launching a WebGUI::Session-enabled, lean, script?
21:38 < nuba> rizen*
21:39 <@rizen> for lots of reasons
21:39 <@rizen> 1) spectre is single threaded
21:39 <@rizen> so if something crashes
21:39 <@rizen> it would crash spectre
21:39 <@rizen> by keeping spectre small and clean, less chance of crash
21:40 <@rizen> 2) the mod_perl processes are already running so why not use them instead of loading all that code into ram yet again
21:40 <@rizen> 3) you can load balance workflows by adding more webgui nodes
21:40 <@rizen> and thusly the workflow engine can grow with your web site
21:40 <@rizen> is that good enough?
21:40 < nuba> you could work around 1 with 'perl spectre_worker_instance.pl "do stuff"' with return code
21:41 <@rizen> i could, and i did in the early versions of it
21:41 <@rizen> but 2 and 3 made me realize that was dumb
21:42 < nuba> ok, thanks for the reply
21:48 < nuba> another thing: i was editing some branches having many posts (about 3k total), and when clicking on 'save' the mod_proxy httpd would timeout while waiting for the mod_perl httpd's reply to the post. I didnt check the code but I risk guessing a branch edit means finding every descendant asset and creating a revision, creating the workflow instances, etc, all in a single request, is that right? 
21:50 <@preaction> yeah
21:53 < nuba> and then, these timeouts became such a pita, and the toones of workflows became such a pita, that i deleted the entire site, and migrated again from scratch, this time adjusting all I wanted while creating the asset, to avoid "branch edits"
21:57 < nuba> but basically, the idea is that "branch edit" dont seem to be scaling the way it works now
21:58 <@rizen> they don't scale well right now
21:58 <@rizen> it's on my todo list to fix eventualy
21:58 <@rizen> they work fine if you have less than 100 children
21:58 <@rizen> but if you have more than 1000 children they are a concern
22:00 < nuba> k
22:02 < nuba> i wonder if slicing the work with a recursive approach would be better, doing stuff like 1) edit me, 2) queue the task of editing for my children.
22:03 <@rizen> it needs to be fixed in one of two ways
22:03 < nuba> just my $0.02
22:03 <@rizen> either generate a workflow that will make all the edits to children
22:03 <@rizen> or change the screen to stream informtation back to the browser
22:03 <@rizen> as it makes changes
22:03 <@rizen> so that it doesn't time out
22:04 <@rizen> and the user gets immediate feedback
22:04 < nuba> yup
22:04 -!- rizen is now known as rizenisafk
22:22 -!- teflond0n [n=wgGuest2@168.8.72.205] has quit []
22:22 -!- wgGuest53 [n=wgGuest5@168.8.72.205] has joined #webgui
22:23 < nuba> i had more questions but can wait 'till hes not afkb anymore ;)
22:23 -!- wgGuest53 is now known as spanishinquisiti
22:24 < nuba> spanishinquisiti: nobody expected you here
22:25 < nuba> NOBODY expects the Spanish Inquisition!
22:25 < spanishinquisiti> si`
22:26 < nuba> well i'll paste the questions here, so rizen can asnwer later if he wants, or not at all..
22:29 < nuba> 1) is WebGU::Session memory-hungry or its just me? While importing the 3k posts (photo gallery, each post w/ at least one pic), i had to do it in chunks (of about a hundred posts each), otherwise the import script would grow to eat all available memory and then die.
22:34 < nuba> thats all for now :)
22:41 <@preaction> WebGUI::Session is huge, yes. but it has everything inside of it
22:41 <@preaction> there are things though, that might be able to be taken out when developing/running utility scripts, such as the query log, if it's possible
22:56 < perlDreamer> preaction: have you ever tinkered with Test::Builder?
22:56 <@preaction> not as such, no
22:56 < perlDreamer> I've been trying to fix the problem with TODO tests with wg::Test::Maker::* and haven't had any luck
23:02 <@preaction> do we assume that anyone running SVN HEAD does a resetdev after every svn update? (or at least, if things break spectacularly)?
23:02 < perlDreamer> if things break spectacularly, resetdev is the first thing I do
23:02 <@preaction> or should we start instituting a plan to make running HEAD more managable, if/when the new upgrade system is implemented
23:03 <@preaction> oh, dur, new upgrade system negates the need for it
23:03 < perlDreamer> what's the new upgrade system?
23:04 <@preaction> when it gets implemented: it's basically a way for each commit to SVN to be a "release"
23:04 <@preaction> releases are based on patches, which may or may not have associated collateral / executables
23:05 <@preaction> of course, there's a -stable and -unstable branch, etc... full-featured, etc...
23:11 < perlDreamer> rizenisafk: amazon is getting into the self-publishing biz.  I don't know how their rates compare to Lulu
23:17 < perlDreamer> That should help locate the extra session leaker.
23:18 < CIA-19> WebGUI: colin * r5223 /WebGUI/t/lib/WebGUI/Test.pm: add more diagnostics to help find the leaking session. These will be removed or hidden behind conditionals later
23:34 <@preaction> it's possible to test the operation of a utility script, correct?
23:34 <@preaction> i suppose the better question is should I? if it's being added to core
23:38 < perlDreamer> I would, but I'm prejudiced that way
23:38 < perlDreamer> I think new "Perl" way to do script testing is to wrap the guts in a module
23:38 < perlDreamer> and both the script and test exercise the guts
23:38 < perlDreamer> kind of MVC for scripts
23:44 <@preaction> it's for migrating different types of content into a Gallery, i don't like the idea of putting "addAlbumFromCollaborationThread" in the Gallery, though the idea does have merit
23:44 <@preaction> it's just the modular way seems messy and the monolithic way seems messy
23:44 < perlDreamer> or make a Gallery::Utility module?
23:44 <@preaction> ah, true
23:45 <@preaction> that sounds good, i'll do that
23:45 <@preaction> then i can also add the addAlbumFromFilesystem and i think there was another script that they wanted
23:47 < perlDreamer> cool!
23:49 < perlDreamer> you're right about the messiness.
23:50 < perlDreamer> Either it's all monolithic and you have to drive it with system calls
23:50 < perlDreamer> or you pull it into two pieces
23:50 < perlDreamer> which both need documentation, etc
23:52 <@preaction> it'll be good writing the Gallery::Utility framework though, and provide a way for people to make their own migration things
--- Day changed Thu Jan 03 2008
00:07 < perlDreamer> So hopefully by tomorrow I'll have the failing DeleteExpiredSessions test diagnosed
00:07 < perlDreamer> and that will just leave us with the TODO'ed and SKIP'ed Gallery* tests
00:07 < perlDreamer> the eternally failing i18n label tests
00:07 < perlDreamer> and the help.t test which is failing on empty Gallery* keys
00:07 <@preaction> the failing i18n/help tests are false keys
00:07 <@preaction> yeah
00:08 <@preaction> i'll be working on all that for the next week or two, so no need to worry about those
00:08 < perlDreamer> some of the i18n are false keys, the rest are from WebGUI::Setup and nested scope with i18n calls
00:08 <@preaction> (admittedly, in addition to new dev)
00:08 <@preaction> but there is one i'm still wondering about: one test in the gallery still fails
00:08 <@preaction> Test::Deep says "expected 6, recieved 6"
00:08 <@preaction> so afaik it should pass
00:09 < perlDreamer> Well, yes and no.
00:09 < perlDreamer> That's actually a false warning
00:09 < perlDreamer> I worked on that a little yesterday
00:09 < perlDreamer> Check the commit log and that test for more info
00:09 <@preaction> k
00:09 < perlDreamer> The "real" problem is that asset sizes are wrong between the two methods of fetching the file_loop
00:10 <@preaction> oh, ew
00:10 < perlDreamer> yeah, one says 311 bytes, the other says 313
00:10 < perlDreamer> only two bytes different
00:10 < perlDreamer> very strange
00:10 <@preaction> so we leave it for now?
00:10 <@preaction> i mean, hiding it behind TODO or SKIP seems wron
00:10 <@preaction> g
00:11 < perlDreamer> I TODO'ed it for now, so that when it gets fixed it will "Unexpectedly pass"
00:11 <@preaction> sounds good
00:14 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
00:25 < perlDreamer> vayde: can you take this guy? http://www.zedshaw.com/rants/rails_is_a_ghetto.html
00:26 < vayde> sure.  want me to rip his lungs out?
00:26 < perlDreamer> Nah.  He's only ranting at ruby people.
00:26 < perlDreamer> besides, if you break his toes one by one it's better
00:27 < vayde> dunno if I have the patience for that.  I tend to prefer one good hard cross to the throat
00:27 <@preaction> this post, from the first few paragraphs, sounds like pure win
00:28 < vayde> though as I glance down this page, I feel the urge to do a similar page for the morons who originally wrote the code I'm hacking these days
00:28 < vayde> doStuff(\%$ref);
00:29 < vayde> long time no chat preaction, how's things?
00:29 < perlDreamer> That might actually be useful, as I think it creates a "safe" ref back to $ref that doesn't allow stomping on it.
00:29 < perlDreamer> but it could be written more clearly
00:29 <@preaction> doStuff sounds like fun
00:29 <@preaction> things going about as normal, much new ideas but no time to implement them
00:29 < vayde> yeah, it *could* be useful like that perlDreamer, but they thought they had the original $ref to wrik with
00:30 < vayde> sweet.  Just before christmas I finished rewriting this one site's code a 3rd time cos the morons supporting it don't think that changing a site I'm migrating will affect the schedule
00:31 < vayde> even the project managers are starting to talk about buying me a ticket to go 'talk' to the bozos
00:32 < nuba> heh
00:32 <@preaction> lol
00:33 < nuba> >Their ideas were horrendously lame. I swear if someone says they.re starting a social network I.m gonna beat them with the heel of my shoe.
00:33 < nuba> poor guy
00:33 < CIA-19> WebGUI: graham * r5224 /branch/WebGUI_7.4/ (docs/changelog/7.x.x.txt lib/WebGUI/AssetTrash.pm): fix: Assets with no committed versions may be left as orphans when parent is purged
00:33 < perlDreamer> Haarg: You beastie.  Hack on!
00:35 -!- SDuensin [n=Scott@49.sub-75-207-93.myvzw.com] has quit ["Leaving"]
00:36 -!- khenn_ [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has joined #WebGUI
00:39 < vayde> jeez this guy's got a head of steam going.  where did you find this?
00:39 -!- khenn_ [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has quit [Read error: 104 (Connection reset by peer)]
00:39 < perlDreamer> slashdot
00:39 -!- khenn_ [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has joined #WebGUI
00:40 < vayde> hmm, musta missed it this morning.  Bank woulda filtered it out anyway
00:40 < perlDreamer> it just popped up within the last hour or so
00:40 < perlDreamer> you woulda seen it tomorrow
00:42 <@preaction> the argument on ThoughtWorks and the Consulting thing is spot-on
00:48 < CIA-19> WebGUI: graham * r5225 / (3 files in 3 dirs): 
00:48 < CIA-19> WebGUI: Merged revisions 5224 via svnmerge from
00:48 < CIA-19> WebGUI: https://svn.webgui.org/plainblack/branch/WebGUI_7.4
00:48 < CIA-19> WebGUI: ........
00:48 < CIA-19> WebGUI:  r5224 | graham | 2008-01-02 16:18:53 -0600 (Wed, 02 Jan 2008) | 1 line
00:48 < CIA-19> WebGUI:  fix: Assets with no committed versions may be left as orphans when parent is purged
00:48 < CIA-19> WebGUI: ........
00:48 < CIA-19> WebGUI: chrisn * r5226 /WebGUI/ (6 files in 5 dirs): 
00:48 < CIA-19> WebGUI: Add fine-grained export controls. You can now specify whether you want specific
00:48 < CIA-19> WebGUI: assets to be exportable. If any of an asset's parents aren't exportable, that
00:48 < CIA-19> WebGUI: asset also won't be exportable.
00:49 -!- perlmonkey2 [n=perlmonk@casr-dev.fwsm2ppuser.ou.edu] has quit ["Leaving."]
00:51 < perlDreamer> apeiron: with a default of 0, won't that disable exporting for the whole site when the user upgrades?
00:51 <@apeiron> perlDreamer, hm.
00:52 < perlDreamer> or am I reading the sense of the bit backwards?
00:52 <@apeiron> No, you've got it right.
00:53 -!- khenn [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has quit [Read error: 110 (Connection timed out)]
00:55 <@apeiron> perlDreamer, But on the other hand, if it defaults to 1, those who *don't* want to export the entire site when they upgrade have to change a lot of assets.
00:56 < perlDreamer> Exactly, but currently wG allows exporting the entire site.  At the very least I think it would need something in the gotchas.txt file for the release.
00:56 < perlDreamer> and, they'd only need to change the website root (defaultAsset), since it checks all parents
00:57 < perlDreamer> so, if the default is 1, then you set any "node" to 0 to disable all of its children from exporting
00:57 <@apeiron> Good point, sir.
01:01 <@preaction> yeah, i'm with colin: every upgrade must maintain the status quo. if they want change, they can do the work
01:01 <@preaction> is there an edit branch entry for the exportable property?
01:02 <@apeiron> Not yet anyway.
01:04 < perlDreamer> preaction: I'm in violent agreement with you preaction, with the caveat "maintain the status quo whenever possible".
01:04 < perlDreamer> In any case, I think this is a cool feature to add to wG
01:04 <@apeiron> Violent agreement? 
01:04 <@apeiron> You're taking up arms? :)
01:05 < perlDreamer> You ever watch an arguement where two people totally agree with one another but disagree on semantics?
01:05 < perlDreamer> "violent agreement"
01:18 < CIA-19> WebGUI: chrisn * r5227 /WebGUI/ (lib/WebGUI/Asset.pm docs/upgrades/upgrade_7.4.18-7.5.0.pl): Maintain status quo and keep the default of exporting all assets.
02:13 -!- SDuensin [n=Scott@12-217-75-184.client.mchsi.com] has joined #WebGUI
02:22 -!- rizenisafk is now known as rizen
02:23 <@rizen> nuba: WebGUI::Session is not memory hungry that I know of. Are you sure it wasn't just your script? Could you provide me with your script so I could use it as a test case?
02:24 <@rizen> preaction: yes we assume that svn update means reset dev in a lot of cases
02:25 <@rizen> perldreamer, do you have a link to the amazon self publish?
02:28 < perlDreamer> re amazon: AP article: http://news.yahoo.com/s/ap/20080102/ap_on_hi_te/business_of_life;_ylt=AhD5AwPjqeEzMAkaVBdrWqMDW7oF
02:28 < perlDreamer> amazon in particular: createspace.com
02:33 < CIA-19> WebGUI: colin * r5228 /WebGUI/docs/upgrades/upgrade_7.4.18-7.5.0.pl: reorder upgrade steps so that addIsExportable happens before we try to update the templates
02:33 < CIA-19> WebGUI: colin * r5229 /WebGUI/t/Asset/Asset.t: update plan and invert test for new default for isExportable
02:41 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit [Connection timed out]
02:46 -!- rizen is now known as rizenisafk
02:53 -!- perlDreamer [n=ckuskie@nat039.mxim.com] has quit ["Download Gaim: http://gaim.sourceforge.net/"]
04:33 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit [Remote closed the connection]
04:38 -!- khenn [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has joined #WebGUI
04:38 -!- mode/#webgui [+o khenn] by ChanServ
04:38 -!- khenn_ [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has quit [Connection reset by peer]
05:17 -!- rizenisafk [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has quit []
06:01 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
06:01 -!- mode/#webgui [+o preaction] by ChanServ
06:28 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit [Success]
06:49 -!- perlDreamer [n=colink@pool-71-117-209-28.ptldor.fios.verizon.net] has joined #webgui
06:49 -!- mode/#webgui [+v perlDreamer] by ChanServ
07:58 -!- perlDreamer [n=colink@pool-71-117-209-28.ptldor.fios.verizon.net] has quit ["Leaving."]
08:28 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has joined #webgui
08:28 -!- mode/#webgui [+o rizen] by ChanServ
08:28 -!- rizen is now known as rizenisafk
08:54 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
08:54 -!- mode/#webgui [+o preaction] by ChanServ
09:02 -!- CIA-48 [n=CIA@208.69.182.149] has joined #WebGUI
09:03 -!- CIA-19 [n=CIA@208.69.182.149] has quit [Client Quit]
09:14 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit [Connection timed out]
09:31 -!- AMH_bob [n=bob@alphamega-dmp.xs4all.nl] has joined #webgui
09:39 -!- Radix__ [n=Radix@203.161.71.161.static.amnet.net.au] has joined #webgui
09:46 -!- Radix|- [n=Radix@203.161.71.161.static.amnet.net.au] has joined #webgui
09:57 -!- Radix_ [n=Radix@203.161.71.161.static.amnet.net.au] has quit [Read error: 110 (Connection timed out)]
10:01 -!- Radix__ [n=Radix@203.161.71.161.static.amnet.net.au] has quit [Read error: 110 (Connection timed out)]
10:27 -!- Radix_ [n=Radix@203.161.71.161.static.amnet.net.au] has joined #webgui
10:27 -!- mode/#webgui [+v Radix_] by ChanServ
10:45 -!- Radix|- [n=Radix@203.161.71.161.static.amnet.net.au] has quit [Read error: 110 (Connection timed out)]
11:08 -!- Radix__ [n=Radix@203.161.71.161.static.amnet.net.au] has joined #webgui
11:11 -!- BartJol [n=Administ@host1.procolix.nl] has joined #webgui
11:26 -!- Radix_ [n=Radix@203.161.71.161.static.amnet.net.au] has quit [Read error: 110 (Connection timed out)]
11:33 -!- diakopter [n=diakopte@pdpc/supporter/active/diakopter] has joined #webgui
13:39 -!- BartJo1 [n=Administ@host1.procolix.nl] has joined #webgui
13:45 -!- BartJol [n=Administ@host1.procolix.nl] has quit [Read error: 110 (Connection timed out)]
14:18 -!- SDuensin [n=Scott@12-217-75-184.client.mchsi.com] has quit ["Leaving"]
14:27 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has joined #webgui
14:27 -!- mode/#webgui [+o rizen] by ChanServ
14:27 -!- rizenisafk [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has quit [Read error: 104 (Connection reset by peer)]
14:27 -!- crythias [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has quit [Read error: 113 (No route to host)]
15:48 -!- crythias [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has joined #webgui
15:48 -!- mode/#webgui [+v crythias] by ChanServ
15:55 -!- SDuensin [n=Scott@74.sub-75-207-208.myvzw.com] has joined #WebGUI
15:56 < SDuensin> Greetings.
17:18 -!- spanishinquisiti [n=wgGuest5@168.8.72.205] has quit []
17:22 -!- nuba [n=nuba@hq.imaterial.org] has quit ["Reconnecting"]
17:22 -!- nuba [n=nuba@hq.imaterial.org] has joined #webgui
17:23 -!- nuba [n=nuba@hq.imaterial.org] has quit [Client Quit]
17:23 -!- nuba [n=nuba@hq.imaterial.org] has joined #webgui
17:34 -!- perlmonkey2 [n=perlmonk@casr-dev.fwsm2ppuser.ou.edu] has joined #webgui
17:57 < BartJo1> yeah, just made my first sql report with template and it works!
18:15 -!- AMH_bob [n=bob@alphamega-dmp.xs4all.nl] has quit ["Leaving."]
18:19 -!- khenn [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has quit [Remote closed the connection]
18:20 -!- khenn [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has joined #WebGUI
18:20 -!- mode/#webgui [+o khenn] by ChanServ
18:34 -!- BartJo1 [n=Administ@host1.procolix.nl] has left #webgui []
19:11 -!- perlDreamer [n=ckuskie@nat039.mxim.com] has joined #webgui
19:12 < perlDreamer> Now, that's what I call a test log!
19:27 < perlmonkey2> Setting visibility to hidden in a div should make all the children of that dom be inivisible.  That is unless you are using YUI which will laugh at your css and make you cry.
19:28 < nuba> heh
19:29 < perlmonkey2> Even doing this after all the YUI objects have been created does nothing: var temp = document.getElementById("editanswer");
19:29 < perlmonkey2> 146             temp.setattribute("visibility","hidden");
19:29 < perlmonkey2> I need a stick to beat YUI with.
19:30  * nuba handles perlmonkey2 a large trout
19:31  * perlmonkey2 wraps it in tinfoil and butter and places it over a camp fire.
19:32 < nuba> very good, indeed its better to tell horror stories at night about YUI than resorting to violence :)
19:33 < nuba> YUI the CSS eater
19:34 < perlmonkey2> heh
19:35 < perlmonkey2> I can't be too mad at it, I have some groovy drag and drop boxes with some totally cool menus that change dynammically depending on what you click.  Looks slick.
19:44 < perlmonkey2> My JS-foo was rusty.  setattribute is not synonymous with .style.
19:50 < perlDreamer> rizen: ping
19:50 <@rizen> pong
19:51 < perlDreamer> does each child apache process end up with a session?
19:51 <@rizen> no
19:51 <@rizen> each page request ends up with a session
19:51 <@rizen> and the session is destroyed at the end of the request
19:51 < perlDreamer> how 'bout each spectre runner?
19:51 <@rizen> when i say destroyed i mean closed by the way, not deleted
19:52 < perlDreamer> right
19:52 <@rizen> spectre does the same thing because as far as webgui is concerned it's just another web browser
19:52 <@rizen> and spectre keeps track of the session id cookie just like a browser, so it reuses the cookie for subsequent requests
19:53 <@rizen> must get lunch now, be back in 10 min
19:53 < perlDreamer> it's okay, you can have 30
20:00 <@rizen> back
20:00 <@rizen> i can eat lunch whilst answering your questions
20:01 <@rizen> mmmmmm. leftover meatloaf sandwiches
20:11 < perlDreamer> Well, instead of digging, let me tell you what I see from the big test log of 2008
20:11 < perlDreamer> The tests start with 3 pre-existing sessions.
20:12 < perlDreamer> Sessions also seem to be added and deleted, regardless of whether or not the test creates and cleans them up
20:12 < perlDreamer> so I was trying to figure out where they might come from
20:13 < perlDreamer> here's the summary of the session info:
20:13 < perlDreamer> http://webgui.pastebin.com/m61f41342
20:13 < perlDreamer> so that's the mystery
20:13 < perlDreamer> the good news is that it has little, if anything to do with the DeleteExpiredSessions test failing
20:14 <@rizen> is that session objects or session ids?
20:14 < perlDreamer> session ids
20:14 < perlDreamer> select sessionId from userSession
20:14 <@rizen> all the tests get the session from WebGUI::Test->getSession though don't they?
20:14 < perlDreamer> yes and no
20:15 <@rizen> so if there's a problem wouldn't it be there?
20:15 < perlDreamer> they all get their main session from getSession
20:15 < perlDreamer> but some of them create auxiliary sessions for testing using Session->opn
20:15 < perlDreamer> open
20:16  * perlDreamer digs a little
20:16 <@rizen> there shouldn't be many of those though, so it should be easy to locate and clean up problematic ones, right?
20:16 < perlDreamer> yes, but the first test run does not create 3 additional sessions
20:17 < perlDreamer> they were there before
20:17 < perlDreamer> the test started
20:18 < CIA-48> WebGUI: jt * r5230 /tools/makerelease.pl: added verbose option and turned on <pre> blocks in message board formatting
20:18 < perlDreamer> to me it looks like something else aside from the test is using the database
20:20 <@rizen> well the build script uses the database
20:20 <@rizen> to construct the create.sql
20:21 <@rizen> the tests actually run as part of the nightly build process
20:21 <@rizen> which has to generate a create.sql script
20:22 <@rizen> that means it also runs the upgrade script
20:22 <@rizen> it does that before the test process starts running
20:32 < perlDreamer> that would explain it
20:32 < perlDreamer> I did find a test that is leaking sessions
20:32 < perlDreamer> Session/Scratch.t
20:44 < nuba> rizen: do we stick with the Locator name for the map asset?
20:44 < nuba> maybe Map instead?
20:45 <@rizen> don't care
20:45 < nuba> k
20:45 < nuba> just so you know, im back working on it since yesterday
20:46 <@rizen> nice
20:47 < nuba> ive cleaned up my notes, and now I happen to know webgui's api better, so im rebooting the project and reusing some of the stuff i wrote hastly in aug/07
20:47 < nuba> what do you recommend, check out the head or work on 7.4.19-stable?
20:48 < perlDreamer> usually, all new dev is supposed to happen on HEAD
20:48 < CIA-48> WebGUI: colin * r5231 /WebGUI/t/ (Session/Scratch.t Workflow/Activity/DeleteExpiredSessions.t): 
20:48 < CIA-48> WebGUI: Clean up all temporary session info in Session/Scratch.t
20:48 < CIA-48> WebGUI: Add a preliminary run of DeleteExpiredSessions to DeleteExpiredSessions.t
20:48 < CIA-48> WebGUI: to clean up any old, funky sessions which might just be lying around
20:48 < CIA-48> WebGUI: and interfering with the test.
20:48 <@rizen> check out head
20:49 <@rizen> feel free to branch head for your own work if need be like preaction did for the calendar
20:49 <@rizen> that way you can continuously commit
20:49 <@rizen> and not lose any work if your computer dies
20:49 <@rizen> then you can merge back to head when you're done
21:12 -!- wgGuest40 [n=wgGuest4@c-24-5-38-30.hsd1.ca.comcast.net] has joined #webgui
21:12 -!- wgGuest40 [n=wgGuest4@c-24-5-38-30.hsd1.ca.comcast.net] has quit [Client Quit]
21:12 -!- wgGuest99 [n=wgGuest9@c-24-5-38-30.hsd1.ca.comcast.net] has joined #webgui
21:12 -!- wgGuest99 [n=wgGuest9@c-24-5-38-30.hsd1.ca.comcast.net] has quit [Client Quit]
21:13 < perlDreamer> rizen: please pull the -v switch from the nightly smoke test run.  That test should pass now.
21:31 <@rizen> ok
21:33 < CIA-48> WebGUI: jt * r5232 /tools/makerelease.pl: don't need the -v anymore
22:12 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has quit []
22:17 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has joined #webgui
22:18 -!- mode/#webgui [+o rizen] by ChanServ
22:30 -!- crythias [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has quit ["Leaving."]
22:37 -!- crythias [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has joined #webgui
22:37 -!- mode/#webgui [+v crythias] by ChanServ
22:54 < nuba> checking out HEAD to my SVK mirror..
22:54 < nuba> will branch from there
22:55 -!- crythias [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has left #webgui []
23:13 -!- crythia3 [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has joined #webgui
23:13 -!- crythia3 [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has left #webgui []
23:16 < nuba> actually, head, body, limbs, etc.
23:18 < perlDreamer> nuba: was it hard to plug WebGUI's SVN into SVK?
23:18 < perlDreamer> I've wanted to try for years, but haven't spent the time to actually do it
23:18 < nuba> just followed http://www.webgui.org/community-wiki/subversion
23:19 < perlDreamer> and then import that into your SVK repo?
23:19 < perlDreamer> nm, I see there's a new section in there
23:20 < nuba> yeah, the only thing missing there is "svk depotmap --init"
23:23 < nuba> it me took about 30 mins to go from rev 1 to rev 5232
23:25 < nuba> it was like history flashing before my eyes! :D
23:30 < perlDreamer> it imported the whole thing?  That would be huge!
23:31 < nuba> 157M    .svk/
23:32 < nuba> not as mammothy as i expected, too
23:33 < perlDreamer> I kinda wish we also had imported the CVS history into there, too.
23:46 < nuba> is that still available somewhere?
23:46 < perlDreamer> maybe.  if it's anywhere, it would be on sourceforge
23:47 < perlDreamer> project name pbwebgui
23:48 < CIA-48> WebGUI: colin * r5233 /WebGUI/t/Asset/Post.t: 
23:48 < CIA-48> WebGUI: AssetTrash bugfix exposes bad test code. There's no need to
23:48 < CIA-48> WebGUI: purge when you can roll back a commit container asset like the CS.
23:52 < nuba> its still there http://pbwebgui.cvs.sourceforge.net/pbwebgui/
23:53 < perlDreamer> cool
23:53 < nuba> goes as far as beta 3.7.0 
23:53 < perlDreamer> researching the history in SVN is a little tricky because there was a repo reorg last year
23:54 < nuba> yep
--- Day changed Fri Jan 04 2008
00:15 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has quit []
00:24 < nuba> hey, how do you guys go about running HEAD? manually check upgrade_7.4.18-7.5.0.pl for changes every update?
00:25 < nuba> i just got a fatal cause column isExportable didnt exist yet..
00:26 < perlDreamer> what rev did you pull?
00:26 < nuba> is this a mistake (whoever added it to upgrade_7.4.18-7.5.0.pl forgot to add it to create.sql) ?
00:26 -!- SDuensin [n=Scott@74.sub-75-207-208.myvzw.com] has quit ["Leaving"]
00:26 < perlDreamer> It's not a mistake, since create.sql is for VERSION-1
00:27 < perlDreamer> you install create.sql and then run the updater
00:27 < nuba> 5234 
00:27 < perlDreamer> that's also what I'm running, and it installs fine
00:28 < perlDreamer> can you paste your resetdev script output please?
00:28 < nuba> ok. now, what about keeping it in sync?
00:28 < nuba> its manual tracking of upgrade_7.4.18-7.5.0.pl, isnt it?
00:29 < nuba> i mean, the way to go
00:29 < perlDreamer> you mean between your branch and HEAD?
00:29 < perlDreamer> yes, manually
00:29 < nuba> ok, thx
00:31 < nuba> i've no resetdev below /data/
00:32 < nuba> is that edible?
00:32 < nuba> :)
00:32 < perlDreamer> it's a script that everyone has made.
00:32 < perlDreamer> there should be one in the core
00:33 < nuba> this is on wre 0.7.
00:33 < nuba> is it a wre script?
00:33 < perlDreamer> I don't use the wre, so I'm not sure
00:33 < perlDreamer> see this: http://webgui.pastebin.com/m681280b
00:38 < nuba> thanks
00:39 < nuba> i'll modify this to fit my setup
00:39 < nuba> so this is THE resetdev?
00:40 < perlDreamer> no, it's just mine
00:41 < perlDreamer> for THE resetdev, I'd ask a Plain Black staffer
00:41 < nuba> looks like something useful to have at  https://svn.webgui.org/svnweb/plainblack/browse/tools/
00:51 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
00:54 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
00:54 -!- mode/#webgui [+o preaction] by ChanServ
00:55 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit [Remote closed the connection]
01:14 < nuba> perlDreamer: you may want to look into upgrade.pl's --skipBackup --skipDelete --skipMaintenance for line 18 of http://webgui.pastebin.com/m681280b
01:15 < nuba> to make it faster
01:16 < perlDreamer> thanks, nuba!
01:17 < perlDreamer> preaction: does the WRE have a builtin resetdev script?
01:17 <@preaction> not that i'm aware of
01:17 <@preaction> there's a better solution that graham was working on that didn't require so many passwords, but dunno how far he gots
01:17 <@preaction> i made some of my own improvements to resetdev, but it's still a hack
01:18 < perlDreamer> care to share?
01:18 <@preaction> one sec
01:18 < perlDreamer> just don't paste it directly channel.  It makes Doug mad.
01:21 <@preaction> http://webgui.pastebin.com/m96d02d1
01:24 < nuba> good, incorporating lines 16-20 into my hacked version of pD's resetdev lookalike
01:24 < perlDreamer> nuba: you only need to do that if you want to go through the initial site setup
01:25 < perlDreamer> if I understand what's going on correctly
01:25 < nuba> yeah, i got it
01:25 <@preaction> 16 is the site setup, 17 is show debug more, 18-20 is start with auto-commit and real-time workflows
01:26 < perlDreamer> I have an idea
01:26 < perlDreamer> Why don't we make one that is universal, in two flavors
01:26 < perlDreamer> WRE and non-WRE?
01:26 < perlDreamer> and then check them into the /tools area in SVN?
01:27 <@preaction> there's a tools area?
01:27 < perlDreamer> Yeah
01:27 < perlDreamer> It's at the top-level
01:27 < nuba> https://svn.webgui.org/svnweb/plainblack/browse/tools/
01:28 < nuba> or maybe, if you're going to try doing it WRE/non-WRE neutral, at WebGUI/sbin
01:30 <@preaction> it should be possible to do one resetdev for both WRE and non-WRE, with some decent sanity checks
01:32 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has joined #webgui
01:32 -!- mode/#webgui [+o rizen] by ChanServ
01:32 < perlDreamer> hail, rizen
01:33 <@rizen> only here for 32 seconds
01:33 <@rizen> don't hail me =(
01:33 < perlDreamer> ah
01:33 < perlDreamer> not hailed like a taxi-cab, hail like greeted with honor and respect
01:34 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has quit [Client Quit]
01:52 < perlmonkey2> If anyone cares the survey system is coming along nicely.  Tomorrow I'll put something public up for those who care to see how it is going.
01:52 < perlmonkey2> It isn't tied to WebGUI yet, just javascript scalfolding for creating surveys.
01:52 < perlmonkey2> g'night.
01:52 -!- perlmonkey2 [n=perlmonk@casr-dev.fwsm2ppuser.ou.edu] has quit ["Leaving."]
01:59 <@preaction> should've gotten him his own branch
02:00 < perlDreamer> yeah
02:12 -!- dionak [n=dkidd@cpe-024-074-132-143.carolina.res.rr.com] has joined #webgui
02:12 < dionak> perlDreamer, are you awake?
02:17 < perlDreamer> I'm here
02:22 < perlDreamer> well, kinda here anyway
02:22 < perlDreamer> dionak: I'll be bouncing back and forth between two cubes, but I'm around.
02:23 < dionak> perlDreamer: Hm. Sounds like you're busy. You had asked the other day if I have read Perl Best Practices and I just wanted to answer. It is on my list but has not been completely read yet. 
02:31 < perlDreamer> yes, I remember that. I had a question about stringy eval, but it's since been answered by JT checking in WebGUI::Pluggable
02:36 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has joined #webgui
02:36 -!- mode/#webgui [+o rizen] by ChanServ
02:41 < dionak> interesting...
02:41 <@rizen> what is?
02:41 < dionak> WebGUI::Pluggable
02:43 <@rizen> oh, i'm quite proud of how it turned out
02:44 < dionak> I can see why. :) I didn't realize you were planning a plugin system. How are you envisioning this to be used? 
02:45 < dionak> it looks like you've been doing quite a bit of work.
02:45 <@rizen> most of the work i'm doing these days hasn't shown up in svn yet
02:45 <@rizen> only little things
02:46 <@rizen> but if you look at the pluggable urls, content handlers, or auth system (WebGUI::Operation::Auth) you'll see how it's used
02:46 < dionak> looking..
02:46 <@rizen> tonight i'll convert macros and some other plugin points to use it as well
02:46 < dionak> thats really cool, JT
02:47 <@rizen> WebGUI::Pluggable is meant to be used at all plugin points to provide a speedy, easy, and secure way of loading plugins
02:47 < dionak> so we could create our own plugin for a new auth method, for instance...
02:47 <@rizen> you always could, it's just that now there's a standard way that all plugins are loaded, rather than having each plugin system write it's own mechanism for loading them
02:48 <@rizen> WebGUI::Pluggable is really meant to be used by core developers that create plugin entry points
02:48 <@rizen> not so much for people who write plugins
02:48 < dionak> ah, ok.
02:49 <@rizen> so if at some point you decided "JT, I think that session sub-objects should be pluggable?" 
02:49 <@rizen> i'd say, great, let's use WebGUI::Pluggable to make that happen
02:50 < dionak> gotcha
02:56 <@rizen> have you seen the pluggable URL and content handlers
02:56 <@rizen> i would think that those would probably be quite useful to Knowmad
02:57 < dionak> i'm trying to envision how to use that
02:58 < perlDreamer> you could use it to call PHP
02:58 < dionak> so we could setup urls that handle content in a certain way?
02:58 <@rizen> well for URL handlers
02:58 -!- perlDreamer [n=ckuskie@nat039.mxim.com] has quit ["Download Gaim: http://gaim.sourceforge.net/"]
02:59 <@rizen> you can write a custom application at a specific URL that is not an asset
02:59 <@rizen> therefore it doesn't load the asset system on those requests
02:59 <@rizen> and thusly is much faster
02:59 <@rizen> and less resource intensive
02:59 < dionak> nice...
02:59 < dionak> so we could use that for integrating software 
02:59 < dionak> which we do a lot
02:59 <@rizen> also, some people don't want to write assets because they are too complicated, or you never want more than one instance of it on a site
02:59 <@rizen> absolutely
02:59 <@rizen> that's a great example
03:00 <@rizen> you still have all of webgui's resources available, but you choose how to use them
03:00 <@rizen> rather than being confined to the asset architecture
03:00 <@rizen> for reusable content objects, asssets are great
03:00 < dionak> wow, that's really very useful. 
03:00 <@rizen> but for integration, assets are rarely great
03:00 <@rizen> and for content handlers...have you ever wished you could have  your own pluggable operations?
03:01 < dionak> yea, 
03:01 <@rizen> well with content handlers you could make your own operations system that would do whatever you wanted it to do
03:02 <@rizen> instead of op= you could have knowmad=
03:02 <@rizen> as in /page?knowmad=dothis
03:02 <@rizen> all the existing ops will still work
03:02 <@rizen> but now you can add your own things
03:02 <@rizen> and since it's not using the op= interface
03:02 <@rizen> you don't have to worry about conflicting with future changes in webgui
03:02 <@rizen> because knowmad= is your own namespace
03:03 < dionak> so if i wanted to use another piece of software for say, forums or blogging or whatnot, I could just hook it in? That will certainly open up the possibilities for selling the CMS. 
03:03 <@rizen> oh yeah
03:03 < dionak> Yea, that will work well. I'm looking forward to trying that
03:03 <@rizen> ever since i built the asset system into webgui it has become more of a cms and less of a framework
03:03 <@rizen> it's moving away from it's framework roots
03:03 <@rizen> so i decided it needs to get back tot he basics
03:04 <@rizen> which is why i introduced pluggable url and content handlers
03:04 <@rizen> if you're interested, there's an article in the black blog that explains it a little
03:05 < dionak> cool, i'll share it with the group and check out the blog. Btw, I just noticed how few articles are on Sitepoint.com for perl.  Looks like a good writing opportunity.
03:06 <@rizen> never even heard of that site
03:06 <@rizen> but if it's low on perl articles, you're right
03:06 <@rizen> a great opportunity
03:07 < dionak> it's a top site on the web
03:07 < dionak> check it out. i have to grab some dinner. thanks for the info!
03:07 -!- dionak [n=dkidd@cpe-024-074-132-143.carolina.res.rr.com] has left #webgui []
03:07 -!- dionak [n=dkidd@cpe-024-074-132-143.carolina.res.rr.com] has joined #webgui
03:08 -!- dionak [n=dkidd@cpe-024-074-132-143.carolina.res.rr.com] has quit []
03:12 -!- SDuensin [n=Scott@12-217-75-184.client.mchsi.com] has joined #WebGUI
03:18 < CIA-48> WebGUI: jt * r5234 /WebGUI/lib/WebGUI/Macro.pm: switched macro system to use WebGUI::Pluggable
03:33 < CIA-48> WebGUI: jt * r5235 /WebGUI/lib/WebGUI/Operation.pm: converted WebGUI::Operation to use WebGUI::Pluggable
03:50 -!- nuba [n=nuba@hq.imaterial.org] has quit [Remote closed the connection]
04:59 -!- nuba [n=nuba@hq.imaterial.org] has joined #webgui
05:15 < nuba> 'night folks
05:15 <@rizen> howdy
05:15 < nuba> off to bed!
05:16 <@rizen> bye
07:15 <@preaction> although i get annoyed by it, i am liking how the writing of tests are revealing bugs in almost completely unrelated areas
07:15 -!- crythias [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has joined #webgui
07:15 -!- mode/#webgui [+v crythias] by ChanServ
07:16 <@preaction> latest bug: I seem to get intermittent "Cannot call purgeRevision on an undefined value" when rolling back a version tag containing Threads, but rolling back by going into the site itself works just fine
08:09 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has quit []
08:33 < CIA-48> WebGUI: doug * r5236 /WebGUI/ (6 files in 6 dirs): Added migration utility for Gallery
09:35 -!- AMH_bob [n=bob@alphamega-dmp.xs4all.nl] has joined #webgui
10:54 -!- BartJol [n=Administ@host1.procolix.nl] has joined #webgui
12:20 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit ["Leaving"]
12:23 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
12:23 -!- mode/#webgui [+o preaction] by ChanServ
13:48 -!- crythias [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has quit [Read error: 104 (Connection reset by peer)]
14:03 < CIA-48> WebGUI: doug * r5237 /branch/doug-experimental/ (3 files in 3 dirs): Nightly checkin because you forgot
14:12 -!- SDuensin [n=Scott@12-217-75-184.client.mchsi.com] has quit ["Leaving"]
15:38 -!- SDuensin [n=Scott@120.sub-75-205-40.myvzw.com] has joined #WebGUI
15:38 < SDuensin> FrIdAy!
15:55 < BartJol> good morning to you
16:48 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has joined #webgui
16:48 -!- mode/#webgui [+o rizen] by ChanServ
16:54 -!- wgGuest01 [n=wgGuest0@user-3e888277.tcl115.dsl.pol.co.uk] has joined #webgui
16:57 -!- wgGuest01 [n=wgGuest0@user-3e888277.tcl115.dsl.pol.co.uk] has quit [Client Quit]
16:58 -!- wgGuest03 [n=wgGuest0@user-3e888277.tcl115.dsl.pol.co.uk] has joined #webgui
17:00 -!- wgGuest03 is now known as lrobinson
17:05 -!- wgGuest40 [n=wgGuest4@adsl190-28-33-172.epm.net.co] has joined #webgui
17:05 < wgGuest40> hello
17:05 < wgGuest40> i have a question
17:06 < wgGuest40> a macro can return a array of elements?
17:07 < BartJol> well I think so
17:09 < BartJol> it might complicate your return statement
17:13 -!- perlmonkey2 [n=perlmonk@casr-dev.fwsm2ppuser.ou.edu] has joined #webgui
17:13 < wgGuest40> why?
17:14 < wgGuest40> why the return statement will be complicated?
17:16 < BartJol> well, just a layout thingy, if all the output is placed on one line without delimiters
17:17 < BartJol> I'm not an expert in these matters, just trying to help...
17:18 <@rizen> macro can't return an array
17:18 <@rizen> it must return a string
17:18 <@rizen> that said, if you want it to return a comma delimited string
17:18 <@rizen> that string could then be passed in to to another macro
17:22 < wgGuest40> ok
17:23 < wgGuest40> thanks
17:28 -!- wgGuest40 [n=wgGuest4@adsl190-28-33-172.epm.net.co] has quit []
17:29 -!- crythias [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has joined #webgui
17:29 -!- mode/#webgui [+v crythias] by ChanServ
17:42 < perlmonkey2> It could return a Storable object which could be any complex data structure.
17:42 < perlmonkey2> oh, they left.
17:42 < perlmonkey2> freeze is so underutialized :P
17:42 <@rizen> freeze is bad
17:42 <@rizen> and not allowed in webgui
17:43 < perlmonkey2> serializing is bad?
17:43 <@rizen> actually storable isn't allowed in webgui
17:43 <@rizen> no serializing is good
17:43 <@rizen> storable is bad
17:43 <@rizen> because it's binary storage, which is architecture dependant
17:43 <@rizen> so it's not portable
17:43 <@rizen> we recommend serializing to json
17:44 < perlmonkey2> hmm......so freezing on my x64-86 won't thaw on my x86?
17:45 < perlmonkey2> json is good too.  json is also underutialized.  it is especially good for moving data to/from JS :)
17:52 < perlmonkey2> FYI: Storable has 'nstore' which is arch independant.
18:08 < BartJol> well, it's weekend for me, beers, here I come
18:09 -!- BartJol [n=Administ@host1.procolix.nl] has left #webgui []
18:09 <@rizen> it's supposed to be arch independent
18:09 <@rizen> but we had some problems with it a while back
18:10 <@rizen> people transitioning from PPC to intel on mac lost a bunch of poll data as a result
18:10 <@rizen> so we decided then that it was time for it to go away
18:14 -!- AMH_bob [n=bob@alphamega-dmp.xs4all.nl] has quit [Read error: 110 (Connection timed out)]
19:00 < perlmonkey2> Wow, well that sucks.  I won't be using Storable anymore then.
19:24 -!- perlDreamer [n=ckuskie@nat039.mxim.com] has joined #webgui
19:24 < perlDreamer> yeah!  "Clean" test results again!
19:26 -!- lrobinson [n=wgGuest0@user-3e888277.tcl115.dsl.pol.co.uk] has left #webgui []
19:54 -!- ascii [n=ascii@host-64-139-233-19.nctv.com] has quit ["leaving"]
20:00 <@rizen> wahoo pd
20:00 <@rizen> nice job
20:09 -!- perlmonkey2 [n=perlmonk@casr-dev.fwsm2ppuser.ou.edu] has quit ["Leaving."]
20:21 < perlDreamer> rizen: Do the modules in Operation/*.pm need to be Plug'ed, too?
20:21 <@rizen> are there any other than auth that load plugins?
20:21 <@rizen> i don't think so
20:22 <@rizen> i already made Operation.pm use pluggable
20:22 < perlDreamer> Several of the modules load things using eval.
20:22 <@rizen> oh
20:22 < perlDreamer> SpellCheck, Workflow
20:22 <@rizen> then yes we should do it
20:22 <@rizen> oh good point
20:22 <@rizen> i forgot about those
20:22 <@rizen> you want me to do it, or are you going to do it?
20:22 <@rizen> i suppose i should
20:22 < perlDreamer> I can do a few of them today
20:22 <@rizen> i shouldn't make you clean up my mess
20:22 <@rizen> that's bad
20:23 <@rizen> i'll do it
20:23 < perlDreamer> as you wish
20:23 <@rizen> i'll do it now
20:23 < perlDreamer> ack eval Operation/*
20:23 <@rizen> i'm going to leave commerce as is for now
20:23 <@rizen> since i'm working on replacing it
20:24 <@rizen> ooh
20:25 <@rizen> i should be using $class->can in WebGUI::Pluggable
20:25 <@rizen> that would be a good extra check
20:25 < perlDreamer> would that work on procedural code?
20:25 < perlDreamer> or just on objects?
20:25 <@rizen> no just on the objects
20:25 <@rizen> i don't think that would work on procedural
20:25 <@rizen> but if it would, that would be sweet
20:25 <@rizen> could you look that up for me?
20:26 < perlDreamer> yup
20:26 < perlDreamer> no procedural code
20:26 < perlDreamer> and doesn't work on AUTOLOADs
20:30 <@rizen> oh crap
20:30 <@rizen> then maybe i shouldn't do it
20:30 <@rizen> because then i can't use instanciate() on form controls
20:31 < perlDreamer> not unless you call the form directly, instead of going through WebGUI::Form
20:31 < perlDreamer> iirc, WGBP says that AUTOLOAD is not recommended
20:31 < perlDreamer> but I don't know if we've thought that far ahead yet
20:32 <@rizen> yeah, but autoload makes the form system usable
20:32 <@rizen> without it, the form system wouldn't be nearly as cool
20:42 <@rizen> wow there are a lot of plugin points in webgui that i forgot about
20:43 <@khenn> like what?
20:43 <@rizen> form controls, workflow activities, form helpers, workflow helpers
20:43 <@rizen> just to name 4
20:43 < perlDreamer> Asset constructors
20:43 < perlDreamer> i18n
20:43 < perlDreamer> Help
20:44 <@rizen> no, you did the last 2
20:44 <@rizen> and i know about asset, i just don't know how to do that one yet
20:44 < perlDreamer> I was just adding to the list of plugin points
20:44 <@rizen> it's a bit more complicated
20:44 <@rizen> oh, i thought this was a list of ones i forgot
20:44 <@rizen> not a complete list
20:44 <@khenn> yeah the ones you forgot
20:44 < perlDreamer> my bad
20:44 <@khenn> pd just started listing them all =)
20:44 <@rizen> for a complete list we also have auth modules, macros, url and content handlers
20:47 <@rizen> oh and any object for that matter
20:47 <@rizen> as a parameter to a workflow activity
20:53 < perlDreamer> whoa
20:53 < perlDreamer> I think I found the first WebGUI constructor that doesn't take a session object
20:54 <@rizen> which is?
20:54 < perlDreamer> WebGUI::Search::Index
20:54 <@rizen> that's a mistake if that's the case
20:55 <@rizen> it certainly should have a session
20:55 < perlDreamer> hm
20:56 < perlDreamer> I guess it does get a session, but it takes an asset as an argument
20:56 < perlDreamer> it must get the session from that
20:56 <@rizen> ah
20:56 <@rizen> yeah it doesn't make sense to pass it in twice
21:01 <@rizen> pd could you look at line 323 of WebGUI::Workflow::Instance for me
21:02 < perlDreamer> sure
21:02 -!- perlmonkey2 [n=perlmonk@ip68-97-51-70.ok.ok.cox.net] has joined #webgui
21:02 <@rizen> it seems like that shouldn't work
21:02 <@rizen> because $params is an array reference
21:03 <@rizen> and i think it should be @{$params} to make it work
21:03 <@rizen> but there haven't been any complaints, so i don't know what's up
21:03 < perlDreamer> maybe they don't take params yet?
21:03 < perlDreamer> or all expect array references?
21:07 < perlDreamer> my Workflow/Spectre-fu isn't that great
21:07 < perlDreamer> but it looks like it's creating itself over again
21:07 < perlDreamer> because $class and $method come from $self, rather than the activity
21:10 < perlDreamer> obviously, I'm not paying enough attention in the Workflow class :)
21:10 <@rizen> i think i'm going to stop second guessing myself
21:11 <@rizen> it's obviously working or workflows everywhere would be going to hell in a handbasket
21:12 < perlDreamer> in my Workflow Instance table, all classes and methods are NULL
21:13 <@rizen> you probably only have maintenance workflows up right now
21:13 <@rizen> which have no objects
21:13 <@rizen> most workflows are maintenance type
21:13 <@rizen> but approval workflows work on version tags
21:13 <@rizen> and lots of other workflows work on groups and users
21:14 < perlDreamer> you're right, I don't have any of those running right now
21:14 <@rizen> workflow is kind of an insane bitch
21:14 <@rizen> but i'm glad i wrote it
21:14 <@rizen> it makes so many other things in webgui so much nicer
21:14 < perlDreamer> I'll understand it better once I've tested it
21:14 <@rizen> like the new commerce system for example
21:17 < perlDreamer> I'll understand that once I've tested it, too
21:17 <@rizen> you won't have to test that
21:17  * perlDreamer briefly considers changing his nick to V'GER
21:18 <@rizen> i'm building it doing test driven dev
21:18 <@rizen> holy crap, i just cut about 40 lines out of WebGUI::Workflow::Instance due to WebGUI::Pluggable
21:19 < perlDreamer> refactoring rocks
21:19 <@rizen> overall I'd say WebGUI::Pluggable has eliminated about 250 lines of code
21:27 <@rizen> oh wait, ->can will work with the forms system
21:27 <@rizen> because it's in the AUTOLOAD method that we call can
21:28 <@rizen> we're not calling it on an autoload method
21:34 < CIA-48> WebGUI: jt * r5238 /WebGUI/lib/WebGUI/ (4 files in 3 dirs): converted workflow system to use WebGUI::Pluggable
21:41 -!- cap10morgan [n=cap10mor@vc1-868-3.adsl.indra.com] has joined #webgui
21:41 < cap10morgan> should ?op=spectreGetSiteData ever return an empty json object? (i.e. just "{}")
21:42 <@rizen> no, at the very least it should have 2 nodes, the workflow and schedule nodes
21:42 <@rizen> i forget the exact labeling
21:42 < cap10morgan> ok, that's what i thought
21:42 < cap10morgan> hmm, my sites are returning empty objects
21:42 < cap10morgan> on 7.4.14
21:42 < cap10morgan> i'll try restarting
21:42 <@rizen> those nodes may be empty if you have no schedules or workflows
21:42 <@rizen> but they should at least exist
21:42 < perlDreamer> is there anything in your webgui.log?
21:43 < cap10morgan> right
21:43 < cap10morgan> no, i need to turn up the logging level though
21:43 < perlDreamer> there is one way for it to return {}
21:43 < perlDreamer> and that's if you make an out of subnet request
21:44 < cap10morgan> perlDreamer: ah, that could be
21:44 < perlDreamer> see line 62 of Operation/Spectre.pm in HEAD for an example
21:44 < perlDreamer> if the request is out of subnet, it will issue a security warning and then fall through to the bottom
21:44 < cap10morgan> yep, that was it
21:44 < cap10morgan> thanks pD
21:44 < perlDreamer> no sweat
22:11 <@rizen> pd you around?
22:11 <@rizen> i don't understand what i'm looking at in WebGUI::Operation::Help _loadHelp
22:12 <@rizen> oh nevermind
22:12 <@rizen> now i do
22:14 <@rizen> crap
22:14 <@rizen> i don't know how to load it though
22:18 < CIA-48> WebGUI: jt * r5239 /WebGUI/lib/WebGUI/ (4 files in 2 dirs): migrated forms system to use WebGUI::Pluggable
22:18 < CIA-48> WebGUI: jt * r5240 /WebGUI/lib/WebGUI/Operation/ProfileSettings.pm: converted to use WebGUI::Pluggable
22:32 <@preaction> would non-508-compliant asset templates in the core be considered a minor/cosmetic bug?
22:33 <@rizen> depends upon the asset
22:33 <@preaction> or since WebGUI is international, is there an international standard we can achieve?
22:33 <@preaction> i seem to be getting bunches of questions about 508 and accessibility these days
22:33 <@rizen> the internatlonal standard is WAI compliance put forth by the W3C, however many countries (including the US) require their own compliance standards
22:34 <@rizen> for example, the project manager asset as far as i'm concerned, doesn't need to be WAI compliant because it is an intranet style app
22:34 <@preaction> my reply is invariably: depends on what the designer does, but a minor audit i conducted gave a few places where we could improve (if we're not, in fact, breaking 508)
22:34 <@rizen> however, any app used by the general public should be WAI or section 508 compliant
22:34 <@preaction> ah, of course
22:34 <@preaction> the Collab System being one of those that might need a little 508 help
22:35 <@rizen> probably
22:35 <@rizen> it's such a big ass app these days
22:35 <@preaction> inaccessible tables, there needs to be all manners of weird tags in tables
22:35 <@preaction> that too
23:01 -!- perlmonkey2 [n=perlmonk@ip68-97-51-70.ok.ok.cox.net] has quit ["Leaving."]
23:03 < CIA-48> WebGUI: doug * r5241 /WebGUI/ (10 files in 4 dirs): Removed old .tmpl files from upgrades. All upgrade collateral must now be in .wgpkg packages
23:04 <@rizen> soooooooooooo cooool
23:04 <@rizen> i'm going to have to bring that up in one of the black blog NEXT posts
23:32 < perlDreamer> rizen: I'd split it into two parts
23:32 < perlDreamer> 1) the load
23:32 < perlDreamer> 2) the data access
23:33 <@rizen> i'm thinking about leaving it as is
23:33 <@rizen> since it's a special case anyway
23:33 <@rizen> or does that seem bad?
23:36 < perlDreamer> "It's always easy to stay within strong boundaries."
23:36 < perlDreamer> I'd covert the load to use Pluggable
23:38 <@rizen> your mom is a ______
23:38 <@preaction> when a workflow activity has an error, it's supposed to increment the priority level, correct?
23:38 < perlDreamer> ex-nurse
23:39 <@rizen> in spectre yes
23:39 <@rizen> not in the workflow table
23:39 <@preaction> ok
23:40 <@preaction> seems to not be happening, which is causing the same 5 activities to keep running over and over. i'll try to find out why
23:41 <@rizen> blame apeiron
23:41 <@rizen> it's fun
23:42 <@apeiron> No, see, that's apeiron v1.0, the 2.0 model isn't fun any more.
23:42 <@apeiron> However, recent versions of khenn are still fun to blame.
23:42 < perlDreamer> does it throw an exception when falsely blamed?
23:43 <@apeiron> Of course.
23:44 < perlDreamer> must be good code then
23:45 <@rizen> let's try
23:46 <@rizen> my $apeiron = $pb->getEmployee("chris");
23:46 <@rizen> say $apeiron->blame;
23:46 < perlDreamer> you forgot to catch the exception
23:47 <@rizen> i'm waiting to see die output
23:47 <@rizen> so far nothing is printing
23:47 <@rizen> this is a slow program
23:47 <@rizen> hrmmm...maybe it's hung
23:47 <@preaction> it's backed up with errors from other processes
23:47 <@rizen> i guess i'll have to kill it
23:47 <@rizen> `killall apeiron`
23:48 -!- apeiron was kicked from #webgui by rizen [rizen]
23:48 < perlDreamer> it's a remote process, all you did was detach it
23:48 <@rizen> that seemed to work
23:49 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
23:49 <@rizen> oh no
23:49 <@rizen> kill one employee object and an old one pops up in it's place
23:49 -!- cap10morgan [n=cap10mor@vc1-868-3.adsl.indra.com] has left #webgui []
23:50 <@rizen> ping vayde
23:50 -!- apeiron [n=apeiron@c-71-230-67-187.hsd1.pa.comcast.net] has joined #webgui
23:50 -!- mode/#webgui [+o apeiron] by ChanServ
23:51 <@rizen> hrm
23:51 <@preaction> resurrected!
23:51 <@rizen> it's not responding either
23:51 <@rizen> but the apeiron process spawned again
23:51 <@rizen> it appears that killall needs a --stay-dead option
23:51 <@preaction> so the getSuspendedQueue in spectre is supposed to have the same priority as the original workflow. and the "workingPriority" is the new priority that spectre itself uses?
23:51 < perlDreamer> I told you that this was just a remote connection
23:52 <@apeiron> Error: caught fatal MessingWithApeiron exception! Aborting...
23:52 <@rizen> finally some output from that damn program
23:53 <@rizen> i guess he was right, you can't blame him without trapping the exception
23:53 < vayde> hey rizen, how you been?
23:53 <@rizen> preaction: that sounds right, but i don't know unless i look at the code
23:53 <@rizen> terrible
23:53 <@rizen> wow i must need a new computer
23:53 <@rizen> my ping command took forever to respond
23:53 <@rizen> and it didn't follow protocol either
23:54 <@rizen> the vayde program is broken
23:54 < vayde> lol.  yeah.  I get that alot
--- Day changed Sat Jan 05 2008
00:04 <@preaction> looks like it's just the spectre.pl --status message, it doesn't show the right working priority for suspended workflows, i expect because they're added to the suspended queue at their original priority
00:15 -!- khenn_ [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has joined #WebGUI
00:16 -!- rizen [n=rizen@71-86-227-90.static.mdsn.wi.charter.com] has quit []
00:27 -!- khenn_ [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has quit [Remote closed the connection]
00:28 -!- khenn [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has quit [Read error: 104 (Connection reset by peer)]
00:28 -!- khenn [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has joined #WebGUI
00:28 -!- mode/#webgui [+o khenn] by ChanServ
00:29 <@preaction> there's still legacy workflow activities that have a 60-second run-time, would it be cool if I made that configurable? i've got 26000 e-mails and a site that seems to add them faster than they can be removed
00:30 < perlDreamer> I think 60 seconds might be an "important" number
00:30 < perlDreamer> like, the spectre time ping interval
00:30 <@preaction> it was, but that was changed iirc
00:31 <@preaction> because now we have the workingPriority thing
00:31 < perlDreamer> well, if that's so (and admittedly, my spectre-fu is weak) then it should be okay to configure it.  TMRFE buy-in is always good.
00:32 <@preaction> i updated the time, the job runs every minute now, but it still only gets rid of about 20 each time, which takes (say) two minutes (due to spectre overhead and other factors)
00:33 <@preaction> perlbot math (26000 / 20 * 2) / 60 / 24
00:33 < perlbot> 1.80555555555556
00:33 < perlDreamer> 20 email/minute seems very, very slow
00:33 <@preaction> it'll take two DAYS to get through these...
00:33 < perlDreamer> which server is that?
00:33 <@preaction> the Alumni site
00:34 <@preaction> they continue to stress parts of WebGUI that I don't think anybody else does
00:34 < perlDreamer> I meant, which email server?  sendmail, qmail, exchange .. ?
00:34 <@preaction> we use sendmail iirc
00:35 < perlDreamer> and it will take two days as long as no new ones are added
00:35 <@preaction> they have applications that allow them to send large HTML-formatted mails
00:35 < perlDreamer> and they're using the collab as an email list server?
00:35 <@preaction> to every user in their database
00:35 <@preaction> 20,000+
00:35 <@preaction> so i think they used it once, maybe twice (since they can also configure who gets the messages, so not everyone has to)
00:35 -!- crythias [n=gyoung@fl-69-68-148-96.sta.embarqhsd.net] has left #webgui []
00:36 <@preaction> so i was thinking some sort of message priority / first-in-first-out system
00:36 <@preaction> since important messages like lost passwords and welcome messages with important instructions are not getting sent out because they're behind (ordered by what though) other, less important messages
00:37 < perlDreamer> A FIFO system will still block, but the messaging priority sounds good.
00:37 <@preaction> ordered by priority first, FIFO second (to make sure there's absolutely no chance that this kind of thing can happen again, unless the priorities are messed up
00:38 < perlDreamer> I see. ORDER BY priority, dateSubmitted
00:38 <@preaction> so i propose both really, because if you send a huge mass at the same priority, and then add another mass, but expect the first to be send before the second, you'll be disappointed
00:39 <@preaction> laugh, i'm up to 28000 now :p
00:39 < perlDreamer> see, that's 10% more in 5 minutes
00:40 < perlDreamer> we need to find a better way to do large messages/large recipient sets
00:41 <@preaction> since most of there are the same message? but it's possible to personalize, no?
00:41 <@preaction> isn't that what the "toGroup" thing in mailQueue is for?
00:42 < perlDreamer> sure, but would a forking dispatcher help?  Is sendmail the limiting factor, or is it Spectre?
00:42 <@preaction> the application generates a list, but it could generate a Group instead
00:42 <@preaction> it seems like some get sent faster than others
00:43 < perlDreamer> I'll be back in a while
00:43 -!- perlDreamer [n=ckuskie@nat039.mxim.com] has quit ["Download Gaim: http://gaim.sourceforge.net/"]
01:25 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit [Remote closed the connection]
01:29 <@preaction> perlbot math (28000 / 20 * 2) / 60 / 24
01:29 < perlbot> 1.94444444444444
02:48 < CIA-48> WebGUI: chrisn * r5242 /WebGUI/lib/WebGUI/Asset.pm: 
02:48 < CIA-48> WebGUI: Add a check in update() for whether isExportable exists, preventing problems
02:48 < CIA-48> WebGUI: with users upgrading from previous versions of WebGUI. Tested in an upgrade
02:48 < CIA-48> WebGUI: from 7.4.0 -> SVN head (i.e., 7.5.0).
02:52 -!- SDuensin [n=Scott@120.sub-75-205-40.myvzw.com] has quit ["Leaving"]
03:33 < CIA-48> WebGUI: chrisn * r5243 /WebGUI/docs/upgrades/upgrade_7.4.18-7.5.0.pl: Add a missing \t in the output of addIsExportable.
04:33 < CIA-48> WebGUI: colin * r5244 /WebGUI/t/lib/WebGUI/Test.pm: 
04:33 < CIA-48> WebGUI: Keep the session tracking, but hide it behind an environment
04:33 < CIA-48> WebGUI: variable to keep the line noise down. Also, add user and
04:33 < CIA-48> WebGUI: group tracking.
05:49 -!- SDuensin [n=Scott@12-217-75-184.client.mchsi.com] has joined #WebGUI
06:18 < CIA-48> WebGUI: colin * r5245 /WebGUI/t/Asset/Asset.t: 
06:18 < CIA-48> WebGUI: Add a stub test for newByDynamicUrl, that really should
06:18 < CIA-48> WebGUI: have been a stub test for update.
06:18 < CIA-48> WebGUI: Add tests for getNotFound, testing what is returned for both
06:18 < CIA-48> WebGUI: a page and a Snippet.
06:19 -!- perlDreamer [n=colink@pool-71-117-209-28.ptldor.fios.verizon.net] has joined #webgui
06:19 -!- mode/#webgui [+v perlDreamer] by ChanServ
06:19 <+perlDreamer> preaction: are you awake and willing to answer inane testing related questions?
06:19 <@preaction> that depends, are you inane enough to handle the answer?
06:19 <+perlDreamer> maybe
06:23 <@preaction> then fire away
06:23 <+perlDreamer> I'm having no luck trying to use Test::Builder to TODO-ify tests in Test::Maker
06:23  * nuba wonders if hes inane enough to watch the exchange..
06:24 <+perlDreamer> So I'm wondering why it isn't working
06:24 <+perlDreamer> the only thing I can think of is that it's OO instead of procedural
06:24  * diakopter lurks
06:25 < nuba> diakopter: want some popcorn?
06:25 <@preaction> so because Test::More doesn't see that it's currently inside of a named TODO: block at some nesting depth, it doesn't truly label them as TODO?
06:25 < diakopter> nuba: i have plenty, thank you
06:25 <+perlDreamer> Right
06:25 <+perlDreamer> Maybe I should commit the code
06:26 <+perlDreamer> It's harmless (useless)
06:26 <+perlDreamer> so you can look
06:27 <@preaction> k
06:27 <+perlDreamer> it's in
06:27 <+perlDreamer> the code is straight out of the Test::Builder::Module docs
06:29 <@preaction> perhaps todo_skip would be better? does that work?
06:29 <+perlDreamer> yes and no
06:29 <+perlDreamer> skip will make the tests not run
06:29 <+perlDreamer> I want these tests to run
06:32 <+perlDreamer> I wonder
06:32 <+perlDreamer> since nothing is exported, import is never called
06:32 <@preaction> look at Test::Builder and look for <level> and tell me if that's what we're looking for
06:33 <+perlDreamer> I did.  The Builder docs are bad
06:33 <+perlDreamer> That's why I chose Builder::Module
06:33 <+perlDreamer> but maybe the whole premise is wrong
06:33 < CIA-48> WebGUI: colin * r5246 /WebGUI/t/lib/WebGUI/Test/ (Maker.pm Maker/Permission.pm): add prototype Test code to allow Test::Maker modules to handle TODO
06:33 <+perlDreamer> and I should include that stuff in new
06:35 <+perlDreamer> I'll try retrofitting to tht
06:35 <+perlDreamer> that
06:35 <@preaction> uhm, is there two package statements in WebGUI::Test::Maker now?
06:36 <+perlDreamer> No.  Why do you ask?
06:36 <@preaction> it looks like the entire module got duplicated and pasted into itself
06:36 <@preaction> weird
06:37 <@preaction> i think it might've been there for a while
06:37 <@preaction> it should only be 192 lines, not 384
06:38 <@preaction> same thing with Test::Maker::Permission
06:38 <+perlDreamer> the svn repo looks okay
06:38 <+perlDreamer> my local copy looks okay
06:38 <+perlDreamer> for Test::Maker
06:38 <+perlDreamer> I'll check Permission next
06:38 <+perlDreamer> It looks okay, too.
06:39 <@preaction> ha, i think it's because i'm looking at a different codebase :p
06:39 <@preaction> yeah :p
06:39 <+perlDreamer> phew
06:40 <@preaction> is there a test i can use to play with? I'm going to start messing around with Test::Builder and/or Test::More to see why it's happening
06:40 <+perlDreamer> Asset/Asset.t
06:41 <+perlDreamer> see the commented out code for canAddMaker2
06:41 <@preaction> k
06:41 <+perlDreamer> that's also what I'm using, so it should be easy to duplicate results
06:41 <@preaction> so i uncomment the code and try that?
06:41 <+perlDreamer> yes
06:41 <+perlDreamer> and also the TODO code down below that calls the run method
06:42 <@preaction> k
06:44 <+perlDreamer> level looks right.  It should kick things up 1 level
06:44 <@preaction> so 92 and 93 are failing, when in fact they should be TODOing
06:44 <+perlDreamer> right, they should fail AND be TODO'ed
06:45 <@preaction> right, but it's not one level, they've decended a level because they're being called inside the Test::Maker
06:45 <@preaction> though i really don't quite understand it
06:46 <@preaction> ok() is called in Test::Builder, up one level would be the TODO: block, up one more would be main::, in a normal situation
06:47 <@preaction> so if ok in Test::Builder is instead called in Test::Maker, which is called in a TODO: block, which is called from main, it would need to go up one more, no?
06:47 <+perlDreamer> yes
06:47 <+perlDreamer> so adding 1 to Test::Builder::Level, localized?
06:47 <@preaction> lets try
06:48 <@preaction> omg it works
06:48 <+perlDreamer> using Builder::Module?
06:50 <@preaction> adding 'local $Test::Builder::Level = $Test::Builder::Level+1;' as line 189 (or somewhere at the top of the run() sub)
06:50 <@preaction> i think it's using Builder::Module, it's using runUsers()
06:51 <+perlDreamer> You're a genius
06:51 <+perlDreamer> You know what happens now right?
06:51 <+perlDreamer> I have to bow to you.
06:51 <@preaction> but then, wouldn't that runUsers introduce another level of depth?
06:51 <@preaction> so why does it work?
06:52 <+perlDreamer> maybe it's caller depth, not stack depth
06:52 <+perlDreamer> we're two packages deep, regardless of how many subroutines are called
06:53 <@preaction> that sounds weird. Test::Builder uses caller() to try to find it, isn't that the stack depth?
06:53 <+perlDreamer> I'm not sure.
06:53 <@preaction> meh, i expect it'll be an exercise for a later day, after we find another way to break things ;)
06:54  * perlDreamer bows
06:54 <+perlDreamer> Thanks for collaborating, preaction
06:55 <@preaction> np, interesting problems are always fun, i don't come across nearly enough in my daily work ;)
06:55 <+perlDreamer> any luck with the email bomb?
06:55 <@preaction> new ways on how to manage site-wide collateral data and reporting are so boring i'm making some superclasses that will cut my work by at least 70%
06:56 <@preaction> yes and no, the e-mails have stopped increasing, but there's nothing i can do to keep things going faster really
06:56 <@preaction> it only came to our attention because of some problems with version tags, perpetual.
06:56 <+perlDreamer> yeah, I found another good buglet today
06:56 <+perlDreamer> but it's not really fixable
06:56 <+perlDreamer> If you instanciate an object inside a version tag, then commit the object
06:56 <@preaction> i've put enough logging info it's pouring out of my eyeballs and i can't find the exact piece of code causing a workflow activity to be created and the version tag and all assets gets destroyed, but the activity remains
06:57 <+perlDreamer> weird
06:57 <@preaction> the bug graham fixed last week might cause it, so i'll be putting that patch in
06:57 <+perlDreamer> yeah, it mucks things up if you purge a parent before a child now
06:57 <@preaction> they're getting upgraded to the first 7.5.0 beta as soon as it gets released, so there's that blessing (right now they're running a patched 7.4.13)
06:58 <@preaction> i think there might also be a problem with trying to create a child while the parent is still in a pending state, but i could've swore i tried that one already
06:58 <+perlDreamer> is it a collab system?
06:59 <+perlDreamer> if so, there's definitely a problem
06:59 <@preaction> custom application allowing users to post photos, stores those photos in a folder
06:59 <@preaction> seems that most of the problems happen when the folder is created
06:59 <+perlDreamer> if a child ever has to get getParent, it will puke
06:59 <+perlDreamer> getParent can't find uncommitted parents
06:59 <@preaction> that... doesn't sound good
07:00 <+perlDreamer> it's exactly what we found in the Gallery/Album tests
07:00 <@preaction> getParent should, imho, always get the parent. if there's no committed parent it should get the uncommitted one
07:01 <@preaction> but then, how could there be a committed child ... because i don't test for that before allowing children to be added to Gallery
07:01 <+perlDreamer> so long as you don't access a method that calls getParent, you can do anything you like
07:02 <+perlDreamer> I think all the containers should override and extend addChild to make sure of it
07:02 <+perlDreamer> right now it's checked in the interface, but not the API
07:03 < nuba> hey guys, quick question: on the maps asset im generating some javascript which I'd think would be better if using a template instead of mixing perl code and many chunks of HEREDOCs or string concats..
07:04 <+perlDreamer> templating tends to be slow
07:04 <+perlDreamer> how about sprintf?
07:04 < nuba> the thing is, if I create a template, chances are the users could delete it and break the asset
07:05 <@preaction> that's the chance you take with all assets
07:06 <@preaction> Time Tracker, Project Manager, huge parts have javascript that unless copied perfectly will break
07:06 <@preaction> i would suggest a snippet though, or if it's absolutely necessary for all asset functionality, a script in the Extras folder
07:06 <+perlDreamer> but I think it's variable, that's why he wants to template it
07:07 <@preaction> you could either include it yourself using session->http->setScript or have the template do it
07:07 <@preaction> oh
07:07 <@preaction> no way to make it an Object of some kind?
07:07 <+perlDreamer> nuba: for stuff like that I use sprintf with a heredoc.  It's the lightest weight template available in pure, core perl.
07:08 < nuba> i wanted to throw it in the headBlock of the template, just to figure out that headBlock isnt processed, just included as is in the head.
07:08 <@preaction> i would offer up the Event www_edit method as an example of how that can go horribly horribly wrong
07:09 <@preaction> put it on top of the markup in the Template area, or make it an object of some kind that you instanciate in the template part
07:09 < nuba> personally, i'm inclined to embed a template in a $javascript_template scalar, and process it
07:10 < nuba> possibly with TT instead of HTML::Template if it starts to get more complicated. would that be ok?
07:10 < nuba> or thats not the webgui way?
07:10 <@preaction> no way to just write out some JSON? 
07:10 <+perlDreamer> preaction: I see what you mean.  But you're still using quoting and string concat vs a HEREDOC/sprintf
07:11 < nuba> well if I write JSON then i'll move the complexity to javascript. 
07:11 <+perlDreamer> nuba: there's no "good" way to do it.  It's either slow or ugly.
07:11 < nuba> i'd rather keep it in perl
07:12 < nuba> by complexity i mean the decisions of what javascript to run
07:12 < nuba> i'd rather decide at the perl code what javascript to create
07:13 <@preaction> not sure on the policy of TT v. HTML::Template. I hear rumors of moving to TT exclusively using an updated WebGUI::Asset::Template
07:13 <@preaction> but i don't know what'll happen to them
07:13 <@preaction> i expect HTML::Template will remain the defacto standard for now
07:13 < nuba> i think i'll stick with heredocs for now
07:13 < nuba> but thanks guys for the input
07:16 <+perlDreamer> I heard rizen talk about the templating, too, preaction.  There's a HTML::Template dialect in TT now.
07:16 <@preaction> perlDreamer, imma commit this, the test is commenting up
07:16 <@preaction> yeah, that's what he was talking about i think
07:16 <+perlDreamer> already committed
07:16 <@preaction> oh k
07:18 < CIA-48> WebGUI: colin * r5247 /WebGUI/t/lib/WebGUI/Test/ (Maker.pm Maker/Permission.pm): make Maker/Permission.pm handle TODO tests
07:18 < CIA-48> WebGUI: colin * r5248 /WebGUI/t/Asset/Asset.t: convert the commented out code into TODOs
07:20 < nuba> looks like $session->style->setRawHeadTags is my new friend..
07:23 <+perlDreamer> it's good stuff
07:23 <+perlDreamer> it even has a test suite
07:33 < CIA-48> WebGUI: doug * r5249 /WebGUI/t/lib/WebGUI/Test/ (Maker.pm Maker/HTML.pm Maker/Permission.pm): fixed Test::Maker::HTML Level. added some comments for future generations
07:48 < CIA-48> WebGUI: colin * r5250 /WebGUI/t/Asset/Asset.t: refine the one newByDynamicClass test
07:51 <+perlDreamer> now if only I could get the related bug fixed with addAssetPrivilege, we could resolve the test and unTODO it
07:52 -!- perlDreamer [n=colink@pool-71-117-209-28.ptldor.fios.verizon.net] has quit ["Leaving."]
09:27 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit ["Leaving"]
12:10 -!- CIA-48 [n=CIA@208.69.182.149] has quit [Remote closed the connection]
12:17 -!- CIA-20 [n=CIA@208.69.182.149] has joined #WebGUI
16:37 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
17:22 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit [Remote closed the connection]
17:48 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
18:29 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit ["Exit- stage left"]
19:55 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
19:55 -!- mode/#webgui [+o preaction] by ChanServ
19:56 < diakopter> JT: you around?
20:18 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit [Connection timed out]
20:59 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
21:37 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit [Remote closed the connection]
22:18 < CIA-20> WebGUI: colin * r5251 /WebGUI/lib/WebGUI/Asset/Wobject/Gallery/Utility.pm: remove whitespace interfering with POD formatting
22:18 -!- danny_mk [n=danny@68.15.55.161] has joined #webgui
--- Day changed Sun Jan 06 2008
00:42 -!- danny_mk [n=danny@68.15.55.161] has quit ["Leaving"]
01:27 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
01:27 -!- mode/#webgui [+o preaction] by ChanServ
02:45 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
02:54 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit [Remote closed the connection]
05:57 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
06:45 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit ["Exit- stage left"]
09:52 -!- khenn_ [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has joined #WebGUI
10:08 -!- khenn [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has quit [Read error: 113 (No route to host)]
12:25 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has quit [Read error: 110 (Connection timed out)]
13:04 -!- preaction [n=doug@static-72-1-4-143.ntd.net] has joined #webgui
13:04 -!- mode/#webgui [+o preaction] by ChanServ
15:59 -!- BartJol [n=Administ@54-167.surfsnel.dsl.internl.net] has joined #webgui
16:58 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
17:18 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit ["Exit- stage left"]
18:04 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has joined #webgui
18:08 -!- BartJol [n=Administ@54-167.surfsnel.dsl.internl.net] has left #webgui []
19:21 -!- khenn_ [n=khenn@24-159-239-186.dhcp.mdsn.wi.charter.com] has quit []
--- Day changed Mon Jan 07 2008
00:21 -!- danny_mk [n=danny@68.15.55.161] has joined #webgui
00:26 -!- danny_mk [n=danny@68.15.55.161] has quit ["Leaving"]
02:23 -!- SynQ [n=koen@host1.procolix.nl] has left #webgui []
02:27 -!- Radix-wrk [n=Jesse@203.161.68.67] has joined #webgui
02:27 -!- mode/#webgui [+v Radix-wrk] by ChanServ
05:49 -!- vayde [n=vayde@c-24-118-7-216.hsd1.mn.comcast.net] has quit ["Exit- stage left"]
08:01 <+Radix-wrk> anyone about?
08:03 <+Radix-wrk> I've got an old 6.8.10 webgui setup - can I download the latest wre and upgrade it using that to the latest?
09:01 -!- dapperedodo [n=joeri@77.163.53.25] has joined #webgui
09:42 -!- AMH_bob [n=bob@alphamega-dmp.xs4all.nl] has joined #webgui
09:47 -!- Radix-wrk [n=Jesse@203.161.68.67] has quit ["Leaving"]
10:24 < AMH_bob> Hallo Diego,
10:24 < AMH_bob> Heb je een link naar de opzet gemaakt voor de product catalogue?
10:31 -!- BartJol [n=Administ@host1.procolix.nl] has joined #webgui
10:37 < BartJol> does anyone knows what version of WebGUI is currently on the translation site? we plan to try to keep the Dutch translation up to date, but I can't find the version. Is it always the one most recent?
11:21 -!- besonen_mobile_ [n=besonen_@71-220-198-145.eugn.qwest.net] has joined #webgui
11:23 -!- besonen_mobile [n=besonen_@71-220-198-145.eugn.qwest.net] has quit [Read error: 104 (Connection reset by peer)]
12:20 < xdanger> shouldn't the read_limit be a little higher by default than 64Mb?
12:20 < xdanger> http://www.webgui.org/community-wiki/server-wont-upload-big-files
12:23 < BartJol> maybe someyhing to do with users that randomly uploading files
14:01 -!- SDuensin [n=Scott@12-217-75-184.client.mchsi.com] has quit ["Leaving"]
15:37 -!- BartJol [n=Administ@host1.procolix.nl] has quit [Read error: 110 (Connection timed out)]
15:51 -!- BartJol [n=Administ@host1.procolix.nl] has joined #webgui
15:53 -!- SDuensin [n=Scott@153.sub-75-206-47.myvzw.com] has joined #WebGUI
15:55 < SDuensin> Greetings.
16:47 < BartJol> Same question as this morning (at least morning for me), is the WebGUI version on the translation server always the latest version?
17:05 < nuba> BartJol: i believe whoever is working on any translation for webgui, is using the translation server
17:06 < BartJol> well, no\
17:06 < nuba> since most people would prefer a web based interface than editing the source files, i guess
17:06 < BartJol> Arjan is working on a local machine
17:06 < BartJol> and I'm not sure whether he commited the translation
17:06 < nuba> the translation server also has the convenience of automatically keeping track of new strings as they're added to the releases
17:07 < BartJol> I know
17:07 < nuba> well in the early days you had to mail JT the translation to be imported there
17:07 < BartJol> I remember from my first translation a couple of years ago
17:08 < BartJol> but, is the server always running the latest (stable) version?
17:10 < nuba> dunno
17:11 < BartJol> but probably we will be using the PB translation server in the future
17:11 < nuba> oh sorry i just  read you question again, i got it wrong
17:11 < nuba> i thought you asked if the translations at the translation server were always the latest ones
17:11 < nuba> you were asking about the webgui version tho
17:11 < nuba> my bad
17:11 < BartJol> no sweat, thanks
17:11 < BartJol> is the script I send you working?
17:12 < nuba> i didnt use it yet, im hacking other pressing things atm
17:12 < nuba> but i'll have use for it, thanks for sending :)
17:13 < nuba> just not while this deadlines are breathing in my neck :/
17:13 < BartJol> :)
17:13 < BartJol> off course
17:24 -!- dapperedodo [n=joeri@77.163.53.25] has left #webgui []
18:18 -!- AMH_bob [n=bob@alphamega-dmp.xs4all.nl] has left #webgui []
18:40 -!- BartJol [n=Administ@host1.procolix.nl] has left #webgui []
18:48 -!- perlmonkey2 [n=perlmonk@ip68-97-51-70.ok.ok.cox.net] has joined #webgui
18:51 -!- perlmonkey2 [n=perlmonk@ip68-97-51-70.ok.ok.cox.net] has quit ["Leaving."]
19:40 -!- wgGuest26 [n=wgGuest2@86.121.144.173] has joined #webgui
19:41 < wgGuest26> hello
19:41 < wgGuest26> i have a small problem at installation
19:41 < wgGuest26> ca someone please helpme ?
19:42 -!- wgGuest26 [n=wgGuest2@86.121.144.173] has quit [Client Quit]
19:43 -!- wgGuest82 [n=wgGuest8@86.121.144.173] has joined #webgui
19:43 < wgGuest82> hello
19:44 < wgGuest82> someone can help me with a problem at installation please?
19:50 < nuba> whats going wrong?
20:13 < wgGuest82> hello nuba
20:14 < wgGuest82> when i am runnig wreconsole.pl i am getting this message
20:14 < wgGuest82> [root@localhost sbin]# /usr/bin/perl wreconsole.pl
20:15 < wgGuest82> 'jsonToObj' will be obsoleted. Please use 'from_json' instead. at /usr/lib/perl5/vendor_perl/5.8.8/Config/JSON.pm line 134
20:15 < wgGuest82> 'jsonToObj' will be obsoleted. Please use 'from_json' instead. at wreconsole.pl line 1024
20:15 < wgGuest82> malformed JSON string, neither array, object, number, string or atom, at character offset 0 ["(end of string)"] at 
20:31 <@preaction> why are you using your system's perl when you should be using the WRE's perl?
20:35 < wgGuest82> i did not knew
20:36 <@preaction> best to do this: . /data/wre/sbin/setenvironment.sh <- note the space between the . and the /
20:37 <@preaction> then just "perl wreconsole.pl" or even just "wreconsole.pl"
20:37 <@preaction> the JSON that ships with the WRE should wokr
20:37 < wgGuest82> one moment
20:37 < wgGuest82> i try this now
20:38 <@preaction> but, if you could open up a new terminal, do a perldoc -l JSON, and let me know what the $VERSION is, I would appreciate it
20:38 <@preaction> if they're going to deprecate something, we're going to need to prepare for it
20:40 < wgGuest82> ok
20:40 < wgGuest82> 1 sec please
20:42 < wgGuest82> the version of perl is 5.8.8 and version of webgui si 0.8.1
20:44 <@preaction> i mean what version of the JSON module you were using that gave you that error
20:45 <@preaction> also, just to make sure, you're using the WRE 0.8.1, which probably downloaded WebGUI 7.4.18
20:47 < wgGuest82> yes i am using wre 0.8.1
20:48 < wgGuest82> in the JSON.pm i found a variabile that states version=2.04
20:48 < wgGuest82> i dont have much experience with perl modules
20:50 < wgGuest82> perldoc -l JSON echo the path /usr/lib/perl5/5.8.8/JSON.pm
20:50 < wgGuest82> perldoc -l JSON echo the path /usr/lib/perl5/site_perl/5.8.8/JSON.pm
21:02 <@preaction> now open up that file and look for $VERSION
21:04 < wgGuest82> i found $JSON::VERSION='2.04'
21:04 <@preaction> that works, thanks
21:04 < wgGuest82> $XS_VERSION='2.01'
21:05 < wgGuest82> ok
21:24 -!- perlmonkey2 [n=perlmonk@ip68-97-51-70.ok.ok.cox.net] has joined #webgui
21:48 -!- perlDreamer [n=ckuskie@nat039.mxim.com] has joined #webgui
21:51 < perlDreamer> to answer BartJol's question from this morning, I believe the translation server runs off of SVN HEAD, not the latest stable version.
21:56 < wgGuest82> perldreamer can you help me with some problems at installation please
21:56 < wgGuest82> ?
21:56 < perlDreamer> what kind of problems are you having, wgGuest82?
21:58 < wgGuest82> one moment please
21:59 < wgGuest82> 'jsonToObj' will be obsoleted. Please use 'from_json' instead. at /usr/lib/perl5/vendor_perl/5.8.8/Config/JSON.pm line 134
22:00 < wgGuest82> this is the error i getting
22:00 < perlDreamer> Are you using the WRE?
22:00 < wgGuest82> JSON is version 2.04
22:00 < wgGuest82> perl is 5.8.8
22:00 < wgGuest82> i am running red hat with wre 0.8.1
22:01 < perlDreamer> the WRE supplies its own JSON and perl and apache and mysql
22:01 < perlDreamer> you must not be running it
22:01 < perlDreamer> did you run . setEnvironment from the WRE sbin directory?
22:01 < wgGuest82> i don run apache or mysql
22:01 < wgGuest82> yes
22:01 < perlDreamer> what does perl -v report?
22:02 < perlDreamer> actually, hold on a sec
22:02 < perlDreamer> how can you run WebGUI without using apache or mysql?
22:02 < wgGuest82> This is perl, v5.8.8 built for i386
22:02 < wgGuest82> etc
22:03 < wgGuest82> i am using apache
22:03 < wgGuest82> i just dont have the service started
22:04 < perlDreamer> what are you doing to produce the error message?
22:04 < wgGuest82> perl wreconsole.pl
22:04 < wgGuest82> after that localhost.localdomain:60834
22:04 < perlDreamer> and you continue to get this error message after you do the . setEnvironment ?
22:04 < wgGuest82> yes
22:05 < perlDreamer> which shell are you using?
22:05 < wgGuest82> bash
22:06 < wgGuest82> after i run setenviroment.sh i run echo $? and it returns 0
22:06 < perlDreamer> please do echo $PATH > myPath and paste it using the pastebin site in the title.  webgui.pastebin.com
22:06 < perlDreamer> then paste the URL here
22:07 < perlDreamer> it's not possible that you run the setenvironment script and still point to an operating system perl library.
22:07 < nuba> besides the jsonToObj warnings wgGuest82 is getting this too
22:07 < nuba> wgGuest82> malformed JSON string, neither array, object, number, string or atom, at character offset 0 ["(end of string)"] at
22:07 < nuba> looks like a malformed json file to me
22:07 < wgGuest82> hello nuba
22:08 < wgGuest82> you're back
22:09 < nuba> yup. im always online on irc, even if im not at the computer, i leave the client connected here
22:09 < wgGuest82> http://webgui.pastebin.com/d3e852bf5
22:09 -!- ckotil [n=newtrino@snare.grnoc.iu.edu] has quit [Remote closed the connection]
22:09 -!- ckotil [n=newtrino@snare.grnoc.iu.edu] has joined #webgui
22:09 < nuba> just found out this week's FLOSS is about YUI
22:09 -!- Netsplit orwell.freenode.net <-> irc.freenode.net quits: CIA-20
22:10 < nuba> FLOSS Weekly 23: The Yahoo User Interface Library
22:10 < nuba> http://twit.tv/floss23
22:10 < perlDreamer> wgGuest82: your path is not being set by the setEnvironment script
22:11 < perlDreamer> so it's still pointing to the system perl, not the WRE one
22:11 -!- Netsplit over, joins: CIA-20
22:11 < perlDreamer> and hence your JSON warning about obsolence
22:11 < wgGuest82> you are right
22:11 < wgGuest82> how do i corect this
22:11 < perlDreamer> I _think_ the setenvironment script is in /data/wre/sbin
22:12 < perlDreamer> head over there and see if I'm right
22:12 < perlDreamer> I don't use the WRE myself
22:12 < wgGuest82> yes it is there
22:12 < wgGuest82> and from there i run it
22:13 < nuba> you run it with . /data/wre/sbin/setenvironment
22:13 < perlDreamer> . <space> setenvironment ?
22:13 < nuba> its a dot, a space, then 
22:13 < nuba> /data/wre/sbin/setenvironment
22:13 < perlDreamer> yeah, what nuba said
22:13 < wgGuest82> yes
22:13 < wgGuest82> i try again now
22:14 < nuba> it would be nice havin setenvironment setting a $WRE variable
22:14 < nuba> so you could just ask people to paste here the output of echo $WRE
22:16 < nuba> or maybe echo $SHELL $WRE
22:16 < perlDreamer> I like it.  Make it an RFE and I'll see if I can implement it.
22:16 < wgGuest82> guys thanks for the help
22:17 < wgGuest82> it is working now
22:17 < nuba> cool
22:17 < wgGuest82> i probably messed up something
22:17 < perlDreamer> I'm guessing your earlier problem was not specifying the whole path to setenvironment
22:17 < perlDreamer> but in any case, I'm glad it's working for you now
22:17 < wgGuest82> i run wreconsole.pl with the perl from wre
22:17 < wgGuest82> thks for you help and patience
22:18 < wgGuest82> have a good night
22:18 < wgGuest82> :d
22:18 < nuba> you too
22:18 < nuba> oh also
22:18 < nuba> you can change your nick with /nick mynewnick
22:18 < nuba> so you can hang out here with a proper nick instead of wgGuestxx if you want
22:19 -!- wgGuest82 is now known as Bodanel
22:19 < Bodanel> ok
22:20 < nuba> there you go
22:20 < Bodanel> i dont use irc channels so much
22:20 <@apeiron> Although if you're gonna do that, you may as well get a proper IRC client. :) irc.freenode.net, #webgui
22:20 < nuba> next time a Bodanel joins here we may remember its you :)
22:20 < Bodanel> il do that
22:20 < Bodanel> ok
22:21 < Bodanel> you will see me again
22:21 < Bodanel> now i am experimenting webgui
22:21 < Bodanel> bye everybody
22:21 < perlDreamer> later
22:21 < nuba> have fun
22:21 <@apeiron> Ciao, have fun.
22:37 -!- Bodanel [n=wgGuest8@86.121.144.173] has quit []
23:10 < perlDreamer> He was very persistent.  That's a  good trait in new users
23:10 <@preaction> as long as he learns / listens
23:11 < perlDreamer> "indeed"
23:19 < nuba> yeah
23:20 < nuba> he seemed thankful, and looking forward to come back as he goes experimenting
23:20 < nuba> it would be good if experimenters of today turned channel regulars of tomorrow :)
23:25 <@preaction> i'm starting to think there needs to be a measure of critical installed mass to start upping the population here
23:27 < nuba> worried it could turn all of us into 1st level of customer support ? :)
23:27 < nuba> that can be scary :/
23:29 <@preaction> not really, we kinda are
23:30 <@preaction> for my purpose, in order to keep on top of what's going on in the WebGUI community, i read every board and post when it's appropriate
23:31 <@preaction> usually when i notice something that could be wrong with WebGUI, such as the new JSON API that's going to cause problems unless we switch
23:50 -!- SDuensin [n=Scott@153.sub-75-206-47.myvzw.com] has quit ["Leaving"]
--- Day changed Tue Jan 08 2008
00:32 < Radix__> woohoo.. my source build of the wre completed
00:34 < Radix__> didn't think that was ever going to happen to be honest :)
00:41 <@preaction> :p oh ye of little faith
00:46 < Radix__> it kept on failing at GSSAPI
00:47 < Radix__> it'd try and install Authen::SASL, which had a required module gssapi and it would always fail to install
00:47 < Radix__> I found the right things to apt-get to get it working in the end
00:48 <@preaction> what version of debian/ubuntu?
00:48 < Radix__> still wouldn't build tho.. so ended up doing a full clean and rebuild and went to bed
00:48 < Radix__> debian 4, etch
00:48 <@preaction> might want to post a build log to the dev list with the difficulties you went through
00:48 < CIA-20> WebGUI: doug * r5252 /WebGUI/ (4 files in 4 dirs): Added userDefined fields to GalleryAlbum assets. Modified Gallery::Utility to migrate userDefined fields from Threads to GalleryAlbums. Added tests for the previous.
00:48 < Radix__> the prebuilt one didn't work.. wanted a specific version of glibc
00:49 < Radix__> would it be worth packaging this one up as a binary for debian 4?
00:49 < perlDreamer> preaction: could you please add asset committing to the Gallery Utility test while your'e working on it?
00:50 <@preaction> is that why the purgeRevision is failing?
00:50 < perlDreamer> yes
00:50 <@preaction> better question: is that a bug in purgeRevision?
00:50 < perlDreamer> perhaps, but why get the working tag if you're not going to commit it?
00:51 <@preaction> to isolate my changes and be able to roll them all back at the end
00:51 < perlDreamer> that won't isolate your changes since each addChild will get the current working tag anyway
00:52 <@preaction> it also creates the working tag so that i have it to be able to roll it back without having to go look for it
00:52 < perlDreamer> I didn't know that uncommitted tags could be rolled back
00:53 < perlDreamer> I'll have to study the versioning code some more
00:54 <@preaction> all a rollback does is purge all the assets in the version tag
00:54 <@preaction> rather, all the revisions
00:54 <@preaction> revisions can be pending and still be purged
00:55 < perlDreamer> I see
00:55 < perlDreamer> but then you're still stuck with getParent always failing
00:56 <@preaction> is getParent versioning-aware? that might be a bug too
01:16 -!- ascii [n=ascii@host-64-139-233-19.nctv.com] has joined #webgui
01:33 -!- SDuensin [n=Scott@12-217-75-184.client.mchsi.com] has joined #WebGUI
01:47 < perlDreamer> preaction: my biggest frustration is always trying to figure out what is a bug in code, vs poorly written code.
01:47 < perlDreamer> In other words, who is wrong?
01:48 < perlDreamer> Only rizen can answer that question
01:48 < perlDreamer> since he wrote it
01:49 <@preaction> i tend to think about what I would want, i would want versioning to be a completely-working method to change as much as I want without pushing them to the live version until i'm satisfied
01:49 <@preaction> no matter if i'm using code or actually editing it myself
01:49 <@preaction> so to say, the "spirit" of the code, not the letter ;)
01:50 < perlDreamer> If rizen's cool with changing that, I'm perfectly cool with it too.  I just like passing test suites
01:50 <@preaction> i'm more a fan of failing test suites, it exposes bugs ;)
01:50 <@preaction> correct tests that fail ;)
01:51 < perlDreamer> well, if y'all have a weekly kind of meeting, maybe you could bring it up and talk it through
01:51 < perlDreamer> if the API can change or be altered/fixed, then it should be done
01:55 <@preaction> not sure if those are API issues really, some parts of the Asset class just aren't very aware of versioning, and we're running into them a lot
01:55 <@preaction> i can write something to the dev list about it
01:58 <@preaction> ... and after an hour of migrating content, i begin to worry that the script is just spinning its gears...
01:59 -!- elnino [n=ninow@user-38q49cv.cable.mindspring.com] has joined #webgui
02:01 < elnino> hi.  The help file system is very powerful with the i18n support and all, but my macros are really only useful to us. Is there a way I can write help files within WebGUI, without having to create a corresponding i18n file as well?
02:01 < perlDreamer> no
02:01 < perlDreamer> although the macro help has been pulled out of the core and put into the wiki instead
02:01 < perlDreamer> you could take a similar approach
02:03 -!- Radix-wrk [n=Jesse@203.161.68.67] has joined #webgui
02:03 -!- mode/#webgui [+v Radix-wrk] by ChanServ
02:03 < elnino> so, my custom macros will no longer list automatically in the webgui help?
02:06 < perlDreamer> it's been a while since i've worked in that system
02:06 < perlDreamer> I think they'll list
02:06 < perlDreamer> I don't think they link anymroe
02:08 < perlDreamer> that's as of 7.4, btw
02:08 < elnino> That's too bad, It was certainly convienent to see al the "help" in one place.  But Good to know before I spend all my time writing .pm help files.  Is there 
02:08 < elnino> I'll look at the 7.4 readme/history to see if I can figure it out...
02:08 < elnino> Thanks!
02:13 <@preaction> i'm thinking about re-writing the Settings tab in order to (a) accomodate as much stuff from the .conf file as possible and (b) since there'll be a whole slew of new stuff, adding some search capabilities to make setting things easier
02:13 <@preaction> yea or ney? it's a longer-term project, since the Report asset is first
02:14 <@preaction> something like CompizConfig Settings Manager, with the tabs on the side and a searchbar thingy
02:16 < nuba> two thumbs here for moving things from .conf to Setting!
02:16 < nuba> thumbs up, that is
02:17 <@preaction> in order to start up, though, the DSN, database user, and database password must remain in the config file, but those are unlikely to ever need changing, whereas installing assets requires changing config
02:18 < elnino> The only thing the change log says is that they removed the Table of Contents - which is where the macros help files were linked into. I suspect that there are people out there that have put in a lot of time customizing help, I'm surprised that this was removed? Was this a reqeust?
02:18 <@preaction> it's a performance issue, i18n is loaded into memory, being Perl data structures
02:18 <@preaction> without the i18n for the Help files, we cut something like 12M resident
02:19 <@preaction> iirc. the IRC logs have the true answer
02:19 < elnino> Hmm. That makes sense> I guess I would have had it as an install option. But I'm glad I found out before writing a bunch of help files.
02:20 < elnino> re config: less config files more settings via the application I think is much easier. Sounds good to me.
02:21 < nuba> i like config files having just as little as required to boot
02:21 -!- elnino [n=ninow@user-38q49cv.cable.mindspring.com] has quit ["Leaving"]
02:22 <@preaction> actually i'll have to look. i'm pretty sure preload.perl does some magic with the configured class names. that might have to change to use either Module::Pluggable or WebGUI::Pluggable
02:23 <@preaction> which could increase the resident size, since all the classes in WebGUI::Form, WebGUI::Macro, WebGUI::Asset WebGUI::Workflow::Activity will be loaded at the get-go, no matter if any site is configured to use them
02:25 < nuba> true
02:27 < perlDreamer> preaction: you might want to check the IRC archives for a discussion on that
02:27 < perlDreamer> rizen hasn't been favorable about that in the past
02:28 < perlDreamer> something about the config being loaded on startup, where as settings are hit on every request
02:28 <@preaction> humph... sounds like a challenge to me ;)
02:28 <@preaction> but yes, that would be bad
02:29 < perlDreamer> you know where the online archives are...
02:29 < nuba> are the irclogs archived somewhere?
02:29 < perlDreamer> yes
02:29 < perlDreamer> jukka does it for us
02:29 < perlDreamer> http://mentalhouse.net/irc/logs/webgui/
02:30 < nuba> cool. i've got irssi on auto-logging, but of course that works only when im here :)
02:32 <@preaction> i'll write something to download them and index them later :p
02:35 <@preaction> would it be prudent to make an Asset/api.t test that can be copy/pasted to make sure that all assets conform to the same API? perhaps an additional one for Wobjects?
02:35 <@preaction> god.. give me a week off and just see what explodes out of my head :P
02:36 < perlDreamer> preaction: I think the real answer is Test::Class
02:36 < perlDreamer> copy and paste is bad
02:36 < perlDreamer> very bad
02:36 < nuba> preaction: wget -r to download, grep to search :)
02:36 <@preaction> ooh
02:36 <@preaction> nuba, thanks
02:37 <@preaction> perlDreamer, will investigate before i start writing this api test for my new Report interface
02:37 < nuba> grep -i to make it case insensitive, and -C 10 to give you 10 lines beore and after each grep hit
02:37 < perlDreamer> I'd love to see the Gallery tests refactored to reuse code, rather than copy and paste
02:37 -!- Khaytsus [i=wally@khayts.us] has joined #webgui
02:38 < Khaytsus> I updated my WebGUI from 7.3.18 to 7.3.22, no problem.  Updated 7.3.22 to 7.4.19 and now when I restart httpd I get this:
02:38 < Khaytsus> Error loading WebGUI::Asset::Template::TemplateToolkit! - Can't locate Template.pm
02:38 < Khaytsus> Didn't have any upgrade errors...  And I'm not finding anything on this error.  Any suggestions?  The website sorta shows, but doesn't load much beyond the main page content.
02:39 <@preaction> Khaytsus, is Template installed?
02:39 < perlDreamer> It's probably coming through the config file, with TT as a plug-in
02:39 < Khaytsus> Is it separate?  I'm really not sure
02:39 < Khaytsus> Let me look at config
02:39 <@preaction> Template is perl's Template Toolkit
02:40 <@preaction> TT or TT2
02:40 < Khaytsus> Hmm, let me check that
02:40 < Khaytsus> Template::Toolkit ?
02:40 <@preaction> perlDreamer, i like Test::Class. I like it a lot. it could compartmentalize our tests immensely (each test sub could create its own version tag and rollback after its done)
02:41 <@preaction> Khaytsus, no. it's just Template
02:41 < perlDreamer> yes
02:41 <@preaction> Template.pm
02:41 < perlDreamer> but think of the rewrite
02:41  * perlDreamer shudders
02:41 <@preaction> perlDreamer, admittedly, but if we can use it going forward
02:42 < perlDreamer> I agree, but I won't have time myself to work on it until after the testing book chapter is written
02:42 < perlDreamer> it would need a branch
02:42 < perlDreamer> 'cause we won't make 7.5 either
02:42 <@preaction> and it's an ugly rewrite, but it doesn't have to be a rewrite really, they can co-exist until the Test::Class heirarchy covers enough
02:42 < Khaytsus> preaction: Okay I'm a moron, I'm not sure where to find that then..  cpan I'm coming up with a bunch of template related stuff, but.
02:42 <@preaction> Khaytsus, try cpan Bundle-Template perhaps?
02:42 <@preaction> durno though
02:43 <@preaction> you don't need to install it
02:43 <@preaction> unless you actually use it
02:43  * perlDreamer heads home
02:43 -!- perlDreamer [n=ckuskie@nat039.mxim.com] has quit ["Download Gaim: http://gaim.sourceforge.net/"]
02:43 < Khaytsus> Okay, let me peep in the config now
02:43 < Khaytsus> Aha, I found their website
02:43 < Khaytsus> duh, just "install Template"
02:43 <@preaction> you can just remove the WebGUI::Asset::Template::TemplateToolkit from your Template Plugins section
02:46  * Khaytsus hugs preaction 
02:46 < Khaytsus> Nothing in the config, but installing TT2 did it :D
02:48 < Khaytsus> Thank you!
02:48 < Khaytsus> Do you happen to know one other question?  I had some collaboration systems set to 1 year archive, didn't really mean to..  Is there an *easy* way to find the articles, assuming I can't just search for them?  Ie:  I want to unarchive all of them.  I've searched Assets all over, they stashed anywhere together?
02:50 <@preaction> they're all in the database. they're in the same place, really, just have a different state
02:50 <@preaction> search the wiki for "magic numbers"
02:50 <@preaction> sorry, it's "status"
02:52 < Khaytsus> I saw a reference to that, let me read it closer
02:55 < Khaytsus> So the most direct method is sql queries for status=archived ?
02:55 <@preaction> yeah
02:55 <@preaction> otherwise, though i don't know what other consequences this might have, just update assetData set status=approved where status=archived
02:56 <@preaction> if i'm not mistaken, the only thing that uses the archived status is the collab system
02:56 < Khaytsus> k
03:01 <@preaction> standard disclaimers apply
03:04 < Khaytsus> yep, I'm umm..  attempting a select first..  sql is not a strength :)
03:04 < Khaytsus> aha, got it
03:05 < Khaytsus> select title from assetData where status like 'archived';
03:05 < Khaytsus> Looks safe
03:07 < Khaytsus> hmm, I see dups..  I wonder if it archives old versions
03:08 <@preaction> probably. not sure if the archive system is version-aware
03:09 < Khaytsus> I'm reading the entries to see
03:12 -!- metanil [n=akhadka@137.48.138.72] has joined #webgui
03:13 < Khaytsus> aha yeah I can do selective updates with a title,revisionDate query
03:34 < Khaytsus> Yep, that worked.  I had to edit something else so it would let me commmit, then the 'fixed' threads are showing :)
03:37 <@preaction> oh, since you did a raw DB edit, clear the cache by rm -rf /tmp/Web