| *** didymo has joined #schooltool | 00:24 | |
| *** wgrant has joined #schooltool | 01:08 | |
| *** aelkner_ has quit IRC | 02:43 | |
| *** alga has quit IRC | 03:25 | |
| *** aelkner has joined #schooltool | 04:00 | |
| *** Aiste has joined #schooltool | 08:47 | |
| *** didymo has quit IRC | 09:16 | |
| *** jinty has joined #schooltool | 11:29 | |
| *** jinty has quit IRC | 11:55 | |
| *** jinty has joined #schooltool | 12:00 | |
| *** alga has joined #SchoolTool | 12:46 | |
| *** ignas has joined #schooltool | 14:48 | |
| jinty | ignas: re zope.proxy, looks like the Architecture: tag in the debian/control file is incorrect | 15:12 |
|---|---|---|
| ignas | yep | 15:12 |
| ignas | the question is | 15:12 |
| ignas | what is the correct way of handlinh dch i? | 15:12 |
| ignas | i have like - ubuntu1 on gutsy and ubuntu2 on hardy | 15:12 |
| jinty | just make sure that you apt-get source the latest version | 15:13 |
| jinty | so the next will be ubuntu3 | 15:13 |
| ignas | should i do ubuntu3, then switch it to gutsy, release | 15:13 |
| ignas | then ubuntu4 ,swtich it to hardy, release? | 15:13 |
| jinty | actually, hardy should probably be: | 15:13 |
| jinty | ubuntu1hardy1 | 15:13 |
| ignas | oh | 15:13 |
| jinty | but now we are splitting hairs | 15:13 |
| ignas | i see | 15:14 |
| jinty | i.e. I don't think you should care at all | 15:14 |
| ignas | i care because i have to re release both of them, and would like to do that with less hassle if possible ;) | 15:14 |
| ignas | what do you do? | 15:15 |
| jinty | what happens in a normal repository is that packages get moved between repositories | 15:15 |
| jinty | taking the example of debian unstable and debian etch | 15:15 |
| ignas | is there even a way to apt-get source gutsy version on hardy? | 15:15 |
| jinty | yes, just add both repositories to the sources.list of apt | 15:16 |
| jinty | then you always apt-get source the highest version number | 15:17 |
| jinty | deb-src lines only I think | 15:17 |
| ignas | yeah, but "downgrading" to gutsy before releasing feels a bit backward, hardy always has the newer version, so i'll probably re release a fixed hardy version | 15:18 |
| ignas | though it will seem insane in the change log "switch to hardy" "fix a bug + switch to gutsy" "switch to hardy" | 15:18 |
| jinty | you could choose to always work on the latest release | 15:31 |
| jinty | and then beck-port all your changes | 15:31 |
| jinty | i.e. latest ubuntu release | 15:31 |
| jinty | back-port | 15:31 |
| *** th1a has joined #schooltool | 15:32 | |
| *** aelkner has quit IRC | 16:06 | |
| ignas | jinty: by the way, would it be very evil if i'd add provides: python-zopeinteface to python-zope.interface? | 16:13 |
| jinty | I thought I'd already added it | 16:13 |
| ignas | yes you did | 16:14 |
| ignas | hmm, interesting | 16:14 |
| ignas | why is it conflicting with buildbot then :/ | 16:14 |
| ignas | The following packages have unmet dependencies: | 16:14 |
| ignas | python-twisted-core: Depends: python-zopeinterface (>= 3.2.1-3) | 16:14 |
| ignas | version number? | 16:14 |
| jinty | I think provides doesn't work with versions | 16:15 |
| jinty | another option is an empty python-zopeinterface package that depends on python-zopeinterface | 16:15 |
| jinty | or to change the name of our package | 16:16 |
| jinty | second zopeinterface > zope.interface | 16:16 |
| ignas | hmm | 16:18 |
| th1a | I guess we're going to more or less wait for aelkner here... | 16:30 |
| ignas | i see | 16:30 |
| th1a | ignas: Did you figure out anything about the 64 bit packages? | 16:30 |
| ignas | th1a: i think i have fixed 64 bit packages | 16:30 |
| ignas | can't test them at the moment | 16:30 |
| ignas | because i don't have a 64 bit machine i can wreak havoc on | 16:31 |
| th1a | What was the problem (in theory)? | 16:31 |
| ignas | Architecture was set to All | 16:32 |
| ignas | instead of being set to Any | 16:33 |
| th1a | Ah. | 16:33 |
| ignas | all means that the binary package will work on all platforms, while any means it can be build for any platform | 16:33 |
| ignas | or something like that | 16:33 |
| th1a | Did you push the changes it up to Launchpad? Does this change mostly affect how the packages are built? | 16:34 |
| ignas | pushed it | 16:35 |
| ignas | and new packages have been build already | 16:35 |
| ignas | so apt-get update + apt-get upgrade should fix everything | 16:35 |
| th1a | OK. We can have our guinea pig test them then. | 16:36 |
| ignas | yep | 16:36 |
| ignas | :) | 16:36 |
| ignas | it would be nice if launchpad would provide download statistics ;) | 16:38 |
| th1a | Yes. | 16:38 |
| Lumiere | I am sure that data | 16:39 |
| Lumiere | is being collected | 16:39 |
| th1a | If only we had an inside connection... | 16:39 |
| * th1a goes to get some more coffee. | 16:39 | |
| th1a | Lumiere: So CanDo got some good news. | 16:42 |
| Lumiere | yes it did | 16:43 |
| Lumiere | the state has given us another 40k for summer development | 16:43 |
| Lumiere | and our meeting with them | 16:43 |
| Lumiere | is in a couple weeks. | 16:43 |
| th1a | I think it is going to be important to get CanDo's installation story in line with SchoolTool's. | 16:44 |
| Lumiere | yes | 16:46 |
| Lumiere | that should happen in 2 weeks | 16:46 |
| Lumiere | when eldar and aelkner sit down and sprint in NJ | 16:46 |
| jinty | Lumiere: look at schooltool-common/schooltool-2008, there's space for cando in there | 16:46 |
| Lumiere | that is my entire goal | 16:46 |
| Lumiere | right now I am personally working on cron jobs and start/stop scripts (although I may steal the start/stop from schooltool) | 16:47 |
| th1a | The amount of work packaging has taken has also convinced me of how much it is a strategic advantage for us. | 16:47 |
| jinty | Lumiere: the start-stop in schooltool-common is re-usable, no need to steal | 16:48 |
| Lumiere | jinty: that is what I mean by steal ;) | 16:48 |
| Lumiere | th1a: yes, I have been in agreement on that. I just have to get to the point where I can tell someone to do it | 16:49 |
| th1a | Yes. | 16:49 |
| Lumiere | right now eldar is going into finals and filip/andrew/chris are heading into APs | 16:50 |
| th1a | SUre. | 16:50 |
| Lumiere | and I will not be the one to force them to work hours over getting grades | 16:50 |
| Lumiere | :) | 16:50 |
| ignas | Lumiere: just do the integration on the egg level | 16:52 |
| ignas | and make a nice cando egg | 16:52 |
| ignas | the rest is like 30 minutes of work for me | 16:52 |
| Lumiere | ignas: aelkner should know that by now | 16:53 |
| Lumiere | he's leading that piece | 16:53 |
| ignas | know that? piece? | 16:53 |
| Lumiere | aelkner should know how to make the integration work on the egg level | 16:54 |
| th1a | Perhaps this is something we should discuss when aelkner arrives. | 16:54 |
| Lumiere | yes | 16:54 |
| Lumiere | which should be any time now | 16:54 |
| th1a | Depending on http://maps.google.com/maps?f=q&hl=en&geocode=&q=philadelphia&ie=UTF8&t=h&z=10&iwloc=addr&layer=t | 16:55 |
| wgrant | ignas: Generally used versioning is to upload to the latest release, and append ~hardy1, ~gutsy1, etc. for the previous releases. Not append hardy1 on top of the Gutsy version. | 16:56 |
| wgrant | I think there's a guide on the LP wiki somewhere. | 16:57 |
| ignas | i see | 16:58 |
| wgrant | That also means that upgrades won't fall over and die horribly. | 16:58 |
| ignas | so i apt-get source some-package, then fix it, then dch -i the version, then add ~hardy - upload, change ~hardy to ~gutsy - upload | 16:58 |
| wgrant | Why would you apt-get source it? Don't you have the source already? | 16:59 |
| ignas | not always | 17:00 |
| ignas | sometimes i don't have the most up to date version | 17:00 |
| ignas | especially if it is a package that was auto generated, then fixed by me, then fixed by jinty | 17:00 |
| wgrant | Didn't you upload them in the first place? | 17:00 |
| wgrant | Ah. | 17:00 |
| th1a | jinty: So do we still technically have to wait for Zope 3.4 to be "released" before pushing things into Debian? | 17:02 |
| wgrant | Is it ever going to be released? | 17:02 |
| th1a | I think so. | 17:03 |
| wgrant | Ah. | 17:03 |
| jinty | th1a: I'm starting to get debian/ubuntu people fishing | 17:03 |
| th1a | Actually, I think the Zope 2 release wants a new Zope 3 release, so that's an additional impetus. | 17:03 |
| th1a | jinty: Fishing for? | 17:03 |
| jinty | th1a: for something to upload to debian | 17:04 |
| jinty | the current packaging is painful in quite a few ways | 17:04 |
| th1a | You mean, what's in Debian now? | 17:04 |
| jinty | but there's a lot to do on our packages before they can be uploaded | 17:04 |
| ignas | indeed | 17:04 |
| ignas | python2.5 vs python2.4 stuff would be a nice thing to fix | 17:04 |
| jinty | at least backwards compatibility and copyright would need to be fixed | 17:05 |
| jinty | copyright needs to be manually done for every package | 17:05 |
| th1a | Damn copyright. | 17:05 |
| jinty | which kinda sucks big hairy piles | 17:05 |
| th1a | That's a vivid image. | 17:06 |
| *** aelkner_ has joined #schooltool | 17:07 | |
| th1a | hi aelkner_. | 17:07 |
| aelkner_ | herro :) | 17:07 |
| Lumiere | herring? | 17:07 |
| aelkner_ | did i miss much? | 17:08 |
| Lumiere | not really | 17:08 |
| th1a | Not much relevant to you. | 17:08 |
| aelkner_ | shall i report? | 17:09 |
| th1a | Actually, why don't you start, aelkner_. | 17:09 |
| th1a | Yes. | 17:09 |
| aelkner_ | i'm feeling good about things | 17:09 |
| aelkner_ | i made my final push this weekend | 17:09 |
| aelkner_ | for the meeting today at 12:30 | 17:09 |
| jinty | th1a: anyway any objections to giving others access to make these changes in our PPA | 17:09 |
| jinty | ? | 17:10 |
| aelkner_ | shall i pause my report? | 17:10 |
| th1a | No objections. | 17:10 |
| th1a | Keep going aelkner_. | 17:10 |
| aelkner_ | ok | 17:10 |
| aelkner_ | i will be able to demonstrate the whole system | 17:10 |
| aelkner_ | minus real email | 17:11 |
| aelkner_ | but the fake email will be demostrable as printout to the terminal window | 17:11 |
| aelkner_ | so i can show them that when they add a message, an email is sent | 17:11 |
| aelkner_ | and when they visit the notifyGoals.html page | 17:11 |
| aelkner_ | the emails are sent for all goals that are due today | 17:11 |
| aelkner_ | they can worry about getting an smtp server going here | 17:12 |
| aelkner_ | and then i could hook up to that | 17:12 |
| Lumiere | and I will help aelkner_ and their sysadmin however I can | 17:12 |
| aelkner_ | one last piece that i haven't done | 17:12 |
| aelkner_ | thanks jason | 17:12 |
| Lumiere | with getting the connection made to the email server | 17:12 |
| aelkner_ | is the tire 3 | 17:12 |
| aelkner_ | tier 3 | 17:12 |
| aelkner_ | that's the expulsion record (when all other interventions have failed) | 17:13 |
| th1a | We need to try to do this without setting up a mail server just for SchoolTool. | 17:13 |
| aelkner_ | they can tell me what they want today, and i'll have it ready by Thursday | 17:13 |
| aelkner_ | like i said, when there's no mail server | 17:13 |
| aelkner_ | the mail gets sent to the terminal window | 17:13 |
| aelkner_ | those ar ethe two choices | 17:14 |
| th1a | I mean, SchoolTool just needs a regular mail account on their regular mail server, right? | 17:14 |
| th1a | Or the district's server? | 17:14 |
| aelkner_ | schooltool just needs an smtp server host and port name | 17:14 |
| aelkner_ | in SLA's case, that wold be here | 17:14 |
| ignas | th1a: smtp does not need authentication by default | 17:14 |
| ignas | th1a: as long as the sender is on the same server | 17:15 |
| Lumiere | smtp doesn't need authentication at all | 17:15 |
| Lumiere | unless it is configured by the sysadmin to need it | 17:15 |
| th1a | OK. So anyhow, moving on... | 17:16 |
| aelkner_ | i'd have to did into zope.sendmail to see how to authenticate | 17:16 |
| aelkner_ | anyway | 17:16 |
| aelkner_ | i'll get their reactions tosay | 17:17 |
| aelkner_ | and make any adjustments for Thursday | 17:17 |
| aelkner_ | that would be May 1 | 17:17 |
| aelkner_ | there'll be a number of admin stuff as well | 17:17 |
| aelkner_ | like making the same kind of make file changes i did for the gradebook | 17:18 |
| aelkner_ | or i should say jelkner-buildout | 17:18 |
| th1a | Just do anything possible to avoid setting up a new mailserver. | 17:18 |
| aelkner_ | i'll get the sysadmin folks here to do that | 17:18 |
| * th1a wonders if we could just use a gmail account. | 17:18 | |
| Lumiere | th1a: that requires authentication and tls | 17:18 |
| Lumiere | </sysadmin> | 17:18 |
| aelkner_ | that's all i got | 17:19 |
| th1a | So I'll be in Philly a week from today. | 17:20 |
| aelkner_ | i saced your email | 17:20 |
| aelkner_ | so i'll be expecting you | 17:20 |
| th1a | mmm... Butterscotch Krimpets. | 17:20 |
| aelkner_ | hugh? | 17:21 |
| th1a | ignas? | 17:21 |
| th1a | Did you do anything last week, ignas? | 17:21 |
| th1a | ;-) | 17:21 |
| ignas | eh | 17:21 |
| ignas | released schooltool | 17:21 |
| ignas | ;) | 17:21 |
| th1a | I'm very happy with how it came out. | 17:22 |
| Lumiere | lol | 17:22 |
| th1a | Gearing up for an ongoing documentation effort now. | 17:22 |
| th1a | What's next? | 17:22 |
| th1a | for ignas, that is. | 17:23 |
| ignas | fixing some new bugs | 17:23 |
| ignas | like the error page one | 17:23 |
| ignas | adding a release branch | 17:23 |
| ignas | i missed that while i was integrating vanguardistas.builder with our svn repository | 17:23 |
| ignas | fixing nightlies to build properly for schooltool and schooltoo.gradebook | 17:24 |
| ignas | integrate lyceum with schooltool-2008 properly | 17:24 |
| ignas | add a person management page | 17:24 |
| ignas | that is different from person home page | 17:24 |
| th1a | We do need a person home page. | 17:25 |
| ignas | divide menu items to show up on one of them depending on where it makes sense | 17:25 |
| th1a | Better one. | 17:25 |
| ignas | yep, and we still need something different for the case when someone clicks on a person | 17:25 |
| ignas | so i am thinking of making them 2 pages | 17:25 |
| ignas | instead of 1 page | 17:25 |
| ignas | with action buttons showing up in places they belong to | 17:26 |
| th1a | Yes, it is different if you're looking at your own page or someone else's. | 17:26 |
| th1a | Or at least that's how we're playing it now. | 17:26 |
| ignas | preparing for svn to bzr switch | 17:27 |
| ignas | planning the next 6 months | 17:27 |
| ignas | that's kind of everything that is on my mind at the moment | 17:28 |
| th1a | Do we agree that the big hurdle is terms & time? | 17:28 |
| ignas | yes | 17:28 |
| th1a | Just making sure ;-) | 17:29 |
| aelkner_ | and we're going to mgigrate to z3c.form, right? | 17:29 |
| ignas | aelkner_: slowly, yes | 17:29 |
| ignas | aelkner_: but it's a side task | 17:29 |
| aelkner_ | i agre that it's nto a priority | 17:29 |
| aelkner_ | but | 17:30 |
| aelkner_ | i disagree with it being a slow task | 17:30 |
| Lumiere | slow or side? | 17:30 |
| aelkner_ | i believe it should be done all at once | 17:30 |
| aelkner_ | to avoid wasted overhear of task-switching | 17:30 |
| aelkner_ | everhead | 17:30 |
| aelkner_ | overhead | 17:30 |
| th1a | aelkner is dyslexic this morning. | 17:30 |
| aelkner_ | damn my typing is bad today | 17:30 |
| aelkner_ | we don't have to decide now when to do it | 17:31 |
| ignas | we'll see, z3c.form is just a part of a bigger push | 17:31 |
| aelkner_ | but it should be a focused effort | 17:31 |
| ignas | towards pagelets and layouts | 17:31 |
| aelkner_ | imho | 17:31 |
| aelkner_ | right | 17:31 |
| aelkner_ | is it too early to discuss europython goals | 17:31 |
| aelkner_ | for sprinting that is | 17:31 |
| ignas | don't know, haven't thought about it yet | 17:32 |
| th1a | Probably. | 17:32 |
| aelkner_ | ok | 17:32 |
| aelkner_ | is it too early to order plane tickets :) | 17:32 |
| th1a | No. We should do that. | 17:32 |
| aelkner_ | ok | 17:32 |
| th1a | It is fairly likely that if we're making a push to batch-update something in SchoolTool at EP we'll be updating everything to fit with the new terms and time structure. | 17:33 |
| th1a | If we've got time to think about z3c.forms, I'll be happy. | 17:33 |
| ignas | th1a: i am not sure how far i will be along the way on europython | 17:34 |
| th1a | ignas: True. | 17:34 |
| ignas | which is why i am not making any specific suggestions for europython | 17:34 |
| th1a | Yes. I agree it is too early to say. | 17:35 |
| th1a | One thing we'll be discussing over the next year is what kind of changes are "allowed" between 1.0 beta and 1.0 final. | 17:35 |
| th1a | I'm not sure if changing form engines is allowed. | 17:35 |
| ignas | th1a: users won't see anything | 17:35 |
| th1a | I'd say "yes" though. | 17:36 |
| ignas | th1a: ok, they will see small improvements, like little red markers near required fields for example | 17:36 |
| ignas | th1a: look at the person form, are you aware it is a z3c.form not a formlib form since sprint ;) | 17:37 |
| th1a | Right. | 17:37 |
| th1a | Before we rap up, do we want to return to the discussion about CanDo eggs? | 17:38 |
| th1a | wrap up. | 17:38 |
| aelkner_ | anout cando | 17:38 |
| aelkner_ | as eldar and i are sprinting soon | 17:38 |
| aelkner_ | our goal will be to get rid of the current cando paradigm | 17:39 |
| aelkner_ | of having it use a different skin | 17:39 |
| aelkner_ | and instead provide a tab called 'Competencies' | 17:39 |
| aelkner_ | in this way, it will be identical in conept | 17:39 |
| aelkner_ | to schooltool.gradebook | 17:40 |
| aelkner_ | or schooltool.intervention | 17:40 |
| aelkner_ | so | 17:40 |
| aelkner_ | shouldn't we make sure that whatever way we package it | 17:40 |
| aelkner_ | is the same way we package those other two | 17:40 |
| aelkner_ | btw | 17:40 |
| Lumiere | sure | 17:40 |
| ignas | i am not really sure about that at the moment | 17:41 |
| aelkner_ | could we also call it schooltool.competencies? | 17:41 |
| aelkner_ | to keep things consistent | 17:41 |
| ignas | are you sure cando users will not want any more differences from schooltool | 17:41 |
| aelkner_ | absolutely | 17:41 |
| ignas | i mean - refactoring cando into schooltool.competencies and cando.app | 17:41 |
| aelkner_ | jelkner has commited to it | 17:41 |
| aelkner_ | no | 17:41 |
| ignas | sounds like something that is a little bit more manageable | 17:41 |
| aelkner_ | schooltool.competencies should be just like the other two i mentioned | 17:42 |
| Lumiere | aelkner_: I still think that cando should be a separate app on installation | 17:42 |
| aelkner_ | i don't think so | 17:42 |
| Lumiere | it should include schooltool core | 17:42 |
| ignas | yes, the difference is that there should probably be a package that sets up the right configuration | 17:42 |
| aelkner_ | i think jelkner has the right idea on this one | 17:42 |
| Lumiere | but right now installing cando would break a schooltool instance | 17:42 |
| ignas | ans maybe sometimes adds or changes something | 17:42 |
| aelkner_ | well, all installations have that same issue | 17:43 |
| aelkner_ | like jelkner-buildout | 17:43 |
| ignas | just in case You and I disagree on what is the better way ;) | 17:43 |
| aelkner_ | or sla-buildout | 17:43 |
| aelkner_ | we should solve these things in the SAME way | 17:43 |
| aelkner_ | not two or three different ways | 17:43 |
| aelkner_ | less is more | 17:43 |
| aelkner_ | igas: you'll get me to agree to your way :) | 17:44 |
| aelkner_ | sooner or later | 17:44 |
| aelkner_ | i'll only to wanbt to get you to agree | 17:44 |
| aelkner_ | to having it ONE way | 17:44 |
| ignas | there are these small things, like having a "race" dropdown in an add Person form for example | 17:44 |
| ignas | and there always will be, and you can't solve them one way | 17:44 |
| aelkner_ | person is a whole other discussion | 17:45 |
| ignas | some people need it one way, some need it another way | 17:45 |
| aelkner_ | i have thoughts on that that i've held back | 17:45 |
| ignas | some people need competencies, some don | 17:45 |
| ignas | don't | 17:45 |
| aelkner_ | some poeple need interventions, some not | 17:45 |
| ignas | some people want lyceum.journal to be integrated more closely into their schooltool instance, some don't | 17:45 |
| aelkner_ | some poeple need a gradebook, some not | 17:45 |
| ignas | and have some space for small customizations makes it easier to manage | 17:45 |
| aelkner_ | we have to consider all of these possibilities | 17:45 |
| ignas | because these configurations have "names" | 17:46 |
| ignas | if you say schooltool.sla - i know what you are talking about, and i could write an evolution script that adds schooltool.lyceum.journal to that configuration | 17:46 |
| Lumiere | brb | 17:46 |
| ignas | writing a universal evolution engine that adds/removes any of the gradebook/intervaentions/competencies | 17:46 |
| ignas | to any of the instances is something we just can't do at the moment | 17:47 |
| aelkner_ | maybe i'll be able to do that | 17:47 |
| ignas | not with current architecture of the application | 17:47 |
| aelkner_ | we can discuss that at EP | 17:47 |
| ignas | there is no way to *uninstall* things | 17:47 |
| ignas | in ZODB | 17:47 |
| ignas | there is no way to discover plugins in schooltool | 17:47 |
| aelkner_ | we might be able to come up with an uninstall engine | 17:48 |
| ignas | at the moment | 17:48 |
| th1a | I think for this fall, it is ok if the way you install CanDo is "apt-get install cando" as a separate server and we punt on the pluggability issues for various components. | 17:48 |
| ignas | which is what i am suggesting | 17:48 |
| ignas | because it works | 17:48 |
| th1a | Yes. | 17:48 |
| ignas | and it works now | 17:48 |
| aelkner_ | ok | 17:48 |
| th1a | I mean, we have a longer range pluggability vision, but it is probably post 1.0. | 17:48 |
| ignas | not like I haven't been thinking about plugin architectures for the last 18 months | 17:48 |
| th1a | Yes. | 17:48 |
| th1a | So what does that mean for what aelkner and eldar need to do? | 17:50 |
| aelkner_ | regardless of packaging | 17:51 |
| ignas | well - i'd think they should create schooltool.competencies module that adds the competency part, and move the rest to cando.app | 17:51 |
| ignas | to keep it working mostly the same | 17:51 |
| ignas | and make both of these proper eggs | 17:52 |
| ignas | actually the first step would be | 17:52 |
| th1a | What is cando.app? | 17:52 |
| aelkner_ | i don't see a need for cando.app | 17:52 |
| aelkner_ | schooltool.competencies will cover it | 17:52 |
| th1a | Yes, getting rid of cando.app is the idea, I think. | 17:52 |
| ignas | th1a: emm | 17:52 |
| aelkner_ | we just need a cando-buildout | 17:52 |
| aelkner_ | there can be a cando package | 17:53 |
| aelkner_ | like there is an sla package | 17:53 |
| aelkner_ | and ajelkner2007 package | 17:53 |
| aelkner_ | but | 17:53 |
| aelkner_ | those packages have about 100 lines of code | 17:54 |
| aelkner_ | and config | 17:54 |
| aelkner_ | nothing more | 17:54 |
| aelkner_ | the guts of cando will be in schooltool.icomp... | 17:54 |
| aelkner_ | schooltool.competencies | 17:54 |
| aelkner_ | the purpose of sla or jelkner2007 | 17:54 |
| aelkner_ | was to include the schooltoll padckages of interest | 17:55 |
| aelkner_ | and to override index.html view of app | 17:55 |
| aelkner_ | to redirect to the Manage tab for managers | 17:55 |
| aelkner_ | and the Gradebook tab for all others | 17:55 |
| aelkner_ | maybe cando redirects to the | 17:55 |
| aelkner_ | Competencies tab | 17:55 |
| aelkner_ | but that's it | 17:55 |
| ignas | th1a: you see the small problem is | 17:56 |
| ignas | make-schooltool-instance script | 17:56 |
| ignas | only understands schooltoo.stapp2007 | 17:56 |
| ignas | schooltool.stapp2008spring | 17:56 |
| ignas | lyceum.app | 17:56 |
| ignas | and it can't suddenly realize that you need <include package="schooltoo.competencies" /> | 17:56 |
| ignas | in your site.zcml | 17:56 |
| ignas | th1a: what schooltool-2008 is | 17:56 |
| ignas | it's a meta package that depends on schooltool.stapp2008spring, and relies that on the egg level | 17:56 |
| ignas | there will be an entry point | 17:57 |
| ignas | that gives the right configuration of the application | 17:57 |
| ignas | aelkner_: yes, that's what i am suggesting | 17:57 |
| ignas | aelkner_: which is what i want cando.app to be | 17:57 |
| th1a | Oh, so when you say "cando.app" you don't mean "a big bucket of CanDo-specific code." | 17:57 |
| aelkner_ | can we call it just cando? | 17:57 |
| aelkner_ | like sla and jelkner2007? | 17:57 |
| ignas | you probably can, and sla is schooltool.sla underneath it seems | 17:58 |
| aelkner_ | yes th1a, tha's what ignas means | 17:58 |
| ignas | i am talking about python level namespaces | 17:58 |
| aelkner_ | true | 17:58 |
| ignas | having configuration in cando.app | 17:58 |
| aelkner_ | sla IS schooltool.sla | 17:58 |
| aelkner_ | and jelkner2007 is schooltool.jelkner2007 | 17:58 |
| ignas | yes | 17:59 |
| ignas | so you can do schooltool.cando if you want to | 17:59 |
| ignas | or cando.app | 17:59 |
| ignas | to leave cando viable as a namespace package | 17:59 |
| aelkner_ | i prefer schooltool.cando | 17:59 |
| aelkner_ | less is more | 17:59 |
| Lumiere | I agree with aelkner_ | 17:59 |
| Lumiere | schooltool.cando | 17:59 |
| ignas | ok, hope you can manage the evolution | 17:59 |
| ignas | scooltool.cando makes sense | 17:59 |
| ignas | but still | 18:00 |
| aelkner_ | that will be part of the project | 18:00 |
| Lumiere | although at some point I would like to see schooltool.instances.* or something | 18:00 |
| aelkner_ | Lumiere: ? | 18:00 |
| Lumiere | so that the namespace buildouts all live in a different place from schooltool packages | 18:00 |
| aelkner_ | oh | 18:00 |
| Lumiere | aelkner_: schooltool.gradebook living next to schooltool.sla | 18:00 |
| Lumiere | is confusing | 18:00 |
| ignas | Lumiere: i think you are misusing the term buildout | 18:00 |
| ignas | i'd start from making cando package work the way it is, and then start moving stuff into schooltool.competencies | 18:01 |
| Lumiere | ignas: I am using it to refer to things that would be buildouts | 18:01 |
| ignas | and then when there is nothing left to move, work on renaming cando to schooltool.cando | 18:01 |
| aelkner_ | ignas: i agree | 18:01 |
| ignas | Lumiere: everything is a buildout, every egg has a buildout for itself | 18:01 |
| aelkner_ | eldar and i will get cando to have Competencies tab | 18:01 |
| aelkner_ | before we start changing package name | 18:01 |
| aelkner_ | s | 18:01 |
| aelkner_ | as there are evolution issues at that point | 18:02 |
| ignas | Lumiere: just like with zope | 18:02 |
| ignas | aelkner_: and a lot of them i am afraid | 18:02 |
| aelkner_ | well, that's what we have eldar for | 18:02 |
| ignas | look at the last schooltool evolution script to see how to "reset" the database | 18:02 |
| aelkner_ | i believe he will deal with that after our sprint | 18:02 |
| ignas | to only jave new style references | 18:02 |
| ignas | s/jave/have/ | 18:02 |
| aelkner_ | ignas: we have to learn how to do these things | 18:02 |
| aelkner_ | AND | 18:02 |
| aelkner_ | it has to be possible | 18:03 |
| aelkner_ | and even reasonablly easy to do | 18:03 |
| aelkner_ | or zope sucks | 18:03 |
| ignas | aelkner_: i would not be that sure | 18:03 |
| aelkner_ | and i'm not ready to say that | 18:03 |
| ignas | emm, not zope sucks | 18:03 |
| aelkner_ | i can write a scrpt to patch an SQL database in one day | 18:03 |
| Lumiere | ignas: I was trying to make a distinction between the configuration packages and the code packages I guess | 18:03 |
| aelkner_ | if it takes there with zope, ok | 18:04 |
| ignas | there is no system in the world that allows easy evolution of databases during long spans of time | 18:04 |
| aelkner_ | if not i start bombimg | 18:04 |
| ignas | one off sql scripts | 18:04 |
| ignas | are as good as one off zodb scripts | 18:04 |
| ignas | they only work on 1 deployed instance | 18:04 |
| ignas | and ZODB does make it extra difficult | 18:04 |
| aelkner_ | we only have one deployed instance | 18:04 |
| ignas | very good | 18:04 |
| ignas | that will make it possible | 18:04 |
| ignas | but still difficult | 18:05 |
| aelkner_ | we will have gthis solved by summer's end | 18:05 |
| aelkner_ | and then we may start to have more than one deployed instance | 18:05 |
| aelkner_ | but that's down the road | 18:05 |
| ignas | sounds feasible | 18:05 |
| aelkner_ | but i understand your concern | 18:06 |
| aelkner_ | the theoretical part | 18:06 |
| aelkner_ | of having various deployments in various stages of evolution | 18:06 |
| *** povbot has joined #schooltool | 18:54 | |
| *** ignas has quit IRC | 19:06 | |
| *** Aiste_ has joined #schooltool | 19:07 | |
| *** ignas has joined #schooltool | 19:07 | |
| *** Aiste has quit IRC | 19:07 | |
| *** mgedmin has joined #schooltool | 19:28 | |
| aelkner_ | th1a: why do computers suck? | 19:57 |
| *** jinty has quit IRC | 20:35 | |
| *** ignas has quit IRC | 21:14 | |
| *** aelkner_ has quit IRC | 21:19 | |
| *** mgedmin has quit IRC | 21:25 | |
| *** aelkner has joined #schooltool | 22:34 | |
| Lumiere | aelkner: do we have a method for start/stop on cando instances | 22:40 |
| Lumiere | I cannot find a clean one for the life of me | 22:40 |
| aelkner | you write a python script that accepts start, stop, and restart as command line args | 22:41 |
| aelkner | that way you can install it into init.d | 22:41 |
| aelkner | to stop cando | 22:41 |
| aelkner | you do a ps-aef | grep start-cando | 22:41 |
| aelkner | extract the pid | 22:42 |
| aelkner | and kill it | 22:42 |
| aelkner | i have an example that does this | 22:42 |
| aelkner | i have to look for it | 22:42 |
| aelkner | do you remember your password on alan.elkner.net? | 22:47 |
| aelkner | Lumiere: ? | 22:47 |
| Lumiere | yea | 22:49 |
| aelkner | one sec | 22:49 |
| Lumiere | I can't do the grep start-cando | 22:49 |
| Lumiere | the machine runs multiple cando instances | 22:49 |
| Lumiere | only one of them needs a start/stop | 22:49 |
| Lumiere | and cando doesn't use paste yet or I'd use the paste init.d from schooltool-common | 22:50 |
| aelkner | can't you find a way to distinguish the one that needs starting and stoping from the output of ps -aef? | 22:51 |
| aelkner | i have an idea | 22:52 |
| aelkner | you write a script that calls start-cando | 22:52 |
| aelkner | you call it something else | 22:52 |
| aelkner | and you can then find that in ps -aef | 22:52 |
| aelkner | there's always a way | 22:52 |
| aelkner | so look: | 22:52 |
| aelkner | go to your account on my machine and look at rubycas | 22:53 |
| Lumiere | I can rename it yes | 22:53 |
| aelkner | you can copy that script and change the sopt routine | 22:53 |
| aelkner | stop | 22:53 |
| aelkner | to grep for your script name | 22:54 |
| aelkner | you should be able to install that at init.d | 22:54 |
| Lumiere | k | 22:54 |
| aelkner | there's a delay of three seconds after starting that you might want to change | 22:55 |
| aelkner | maybe make it ten seconds | 22:55 |
| Lumiere | yea, this will do | 22:55 |
| Lumiere | but it is definately hackish | 22:55 |
| aelkner | ah, but it works | 22:55 |
| Lumiere | yes | 22:55 |
| Lumiere | long run I want to do the double fork method somehow | 22:55 |
| aelkner | you could do that | 22:56 |
| aelkner | i was able to put that together in one hour | 22:56 |
| Lumiere | yea | 22:56 |
| aelkner | and that's all i wanted to spend on it | 22:56 |
| Lumiere | I'll worry about the other method later | 22:56 |
| aelkner | but if you have a better way, i could benefit as well | 22:56 |
| Lumiere | because that other method should be how we handle it | 22:56 |
| Lumiere | for everyone | 22:56 |
| aelkner | yes | 22:57 |
| Lumiere | it would involve keeping pid files in var | 23:00 |
| Lumiere | for each instance ;) | 23:00 |
| Lumiere | but this will do | 23:00 |
| Lumiere | aelkner: take a look at cando.initd | 23:15 |
| Lumiere | in my homedir | 23:15 |
| aelkner | so you changed start and stop | 23:15 |
| aelkner | to use your script | 23:15 |
| Lumiere | to use variables | 23:15 |
| aelkner | i see | 23:16 |
| Lumiere | anyways I'll finish the cron scripts tonight | 23:17 |
| Lumiere | and set em up wednesday | 23:17 |
| Lumiere | while welsh is out | 23:17 |
| Lumiere | I want to test em a few times wednesday | 23:17 |
| Lumiere | before I turn em on | 23:18 |
| *** povbot` has joined #schooltool | 23:58 | |
| *** povbot has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!