| *** ignas has quit IRC | 02:56 | |
| *** subir has joined #schooltool | 06:19 | |
| *** pcardune has quit IRC | 06:22 | |
| *** didymo has quit IRC | 07:23 | |
| *** th1a_ has joined #schooltool | 07:47 | |
| *** th1a has quit IRC | 07:48 | |
| *** ignas has joined #schooltool | 09:09 | |
| *** didymo has joined #schooltool | 09:09 | |
| *** aelkner has joined #schooltool | 09:26 | |
| *** subir has quit IRC | 09:37 | |
| *** ignas has quit IRC | 09:41 | |
| *** ignas has joined #schooltool | 10:11 | |
| *** ignas has quit IRC | 10:22 | |
| *** ignas has joined #schooltool | 10:22 | |
| *** didymo has quit IRC | 10:29 | |
| *** lisppaste5 has quit IRC | 13:05 | |
| *** lisppaste5 has joined #schooltool | 13:19 | |
| *** Aiste has quit IRC | 14:04 | |
| *** alga has joined #SchoolTool | 15:19 | |
| *** wbrady has joined #schooltool | 16:13 | |
| *** pcardune has joined #schooltool | 16:26 | |
| *** th1a_ is now known as th1a | 16:29 | |
| th1a | I guess we're going to move the meeting back two hours today. | 16:29 |
|---|---|---|
| th1a | Can anyone not make that? | 16:30 |
| aelkner | i'm ok for then | 16:31 |
| aelkner | th1a: ayt? | 16:32 |
| th1a | I am here. | 16:33 |
| Lumiere | I may not be able to make the first half | 16:33 |
| Lumiere | 11:30 is when I drive to work | 16:33 |
| aelkner | th1a: you didn't receive my provate messages? | 16:33 |
| th1a | I just didn't see it. | 16:34 |
| aelkner | so you did get the messages? | 16:35 |
| th1a | Did you get my response? | 16:35 |
| aelkner | no i didn't | 16:36 |
| * th1a wishes private IRC chats would succeed or fail more explicitly. | 16:36 | |
| th1a | Yes, that's fine. | 16:36 |
| aelkner | Lumiere knows what we need to do | 16:36 |
| aelkner | jason, what did you set up for me so that i would be able to use private chat | 16:36 |
| aelkner | i think tom needs to do the same | 16:36 |
| aelkner | seomthing about registering his nick | 16:37 |
| aelkner | ? | 16:37 |
| th1a | I don't give a shit about registering my nick. | 16:37 |
| th1a | I have too many computers. | 16:38 |
| th1a | I can't keep track of which ones don't work. | 16:38 |
| th1a | There are 30,000 ways of asking me a question electronically. | 16:38 |
| th1a | That's the most annoying one. | 16:38 |
| aelkner | yes, i could always use gmail | 16:39 |
| Lumiere | th1a: PMs fail to registered nicks | 16:39 |
| Lumiere | if your nick is not registered | 16:39 |
| th1a | Lumiere: Yes, I understand. | 16:39 |
| Lumiere | and registering your nick is just a training exercise in typing "/ns identify pass" every time you log in | 16:39 |
| th1a | Yes, thank you, Lumiere. | 16:40 |
| aelkner | that sounds like a pain | 16:40 |
| aelkner | i wouldn't blame him for not wanting to do that each time | 16:40 |
| aelkner | Lumiere: why don't i need to do that? | 16:40 |
| *** wbrady has quit IRC | 16:41 | |
| Lumiere | aelkner: you put it into xchat's login screen | 16:44 |
| Lumiere | so xchat does it for you | 16:44 |
| aelkner | got it | 16:44 |
| *** Aiste has joined #schooltool | 17:22 | |
| *** jinty has joined #schooltool | 17:23 | |
| ignas | Hi | 17:28 |
| aelkner | ignas: hey there | 17:31 |
| aelkner | does this mean you're back? | 17:31 |
| th1a | ignas is looking for help on his test via IRC. | 17:32 |
| ignas | nope '_ | 17:32 |
| th1a | We won't help you cheat, ignas! | 17:32 |
| ignas | ;) | 17:32 |
| * ignas always has a buffer | 17:32 | |
| ignas | aelkner: yes i am back | 17:33 |
| aelkner | what happened to the test? | 17:34 |
| ignas | well - i am done with it | 17:34 |
| ignas | just that - when I have no idea how long is the trip from one place to another + have no idea how long will it last - i "allocate" more time than I need ;) | 17:35 |
| ignas | just in case i won't be able to catch a bus in time and etc. | 17:35 |
| aelkner | th1a: shall we meet then? | 17:36 |
| Lumiere | if we meet now you get me for 30 minutes | 17:36 |
| th1a | I don't care either way. | 17:36 |
| pcardune | howdy | 17:36 |
| th1a | I guess we're all here now. | 17:37 |
| th1a | aelkner: Want to kick things off? | 17:37 |
| pcardune | th1a: i liked your post about teach for america | 17:37 |
| th1a | pcardune: Thanks. | 17:38 |
| aelkner | opk, i finished with the demographics last week | 17:38 |
| aelkner | and started on the narrative report cards | 17:39 |
| aelkner | which cuased me to realize | 17:39 |
| aelkner | that although i can develop them already | 17:39 |
| aelkner | i will need to get sla's section info into schooltool for them to use it | 17:39 |
| aelkner | even though the system integration work is not scheduled to begin until April 15ish | 17:40 |
| aelkner | i will want them using schooltool before then | 17:40 |
| aelkner | that means that I will need to import their sections | 17:40 |
| aelkner | as it is, we have no section import | 17:40 |
| th1a | Yes... that's what I was trying to tell you before. | 17:41 |
| aelkner | too bad i'm a lousy listener :) | 17:41 |
| aelkner | so, just like i have written a test routine | 17:41 |
| aelkner | that i use to get sections into functional tests | 17:41 |
| aelkner | i think we need an xml import for courses, sections, members, instructors | 17:42 |
| aelkner | all in one import | 17:42 |
| aelkner | so the result would be to create the courses | 17:42 |
| aelkner | create the sections (auto-aisning the ids) | 17:42 |
| aelkner | asigning | 17:42 |
| aelkner | then, add the instuctors/students into the sections | 17:43 |
| ignas | something like schooltool import functionality, but without timetables? | 17:43 |
| aelkner | the import could asume that the person import was already done | 17:43 |
| ignas | lyceum import | 17:43 |
| aelkner | you're saying that lycuem has an xml import for sections? | 17:43 |
| ignas | not xml | 17:44 |
| ignas | csv | 17:44 |
| aelkner | but how can you csv multi-level data? | 17:44 |
| ignas | aelkner: it's even integrated with timetables | 17:44 |
| ignas | th1a saw the format I think | 17:44 |
| th1a | I can't say I recall it. | 17:45 |
| ignas | aelkner: 2 parts - 1 csv for student group connections | 17:45 |
| th1a | Generally, schools are more likely to be able to handle CSV. | 17:45 |
| ignas | and 1 csv for group - section - timetable stuff | 17:45 |
| th1a | So if you can pull it off, it is preferable. | 17:45 |
| aelkner | actually, i will be doing the imports myself | 17:45 |
| th1a | Yes, but doing something more easily reused has advantages. | 17:46 |
| aelkner | so it could e xml if that made more sense, but that's the question | 17:46 |
| ignas | xml makes sense | 17:46 |
| th1a | I'd say the first step is to look at ignas's work. | 17:46 |
| aelkner | so what is the more easily reused thing? | 17:46 |
| aelkner | sure | 17:46 |
| th1a | I'm not anti-XML for that. | 17:46 |
| th1a | this. | 17:46 |
| th1a | I'm just saying, all things being equal, there is an advantage to CSV. | 17:47 |
| pcardune | +1 xml | 17:47 |
| th1a | Because otherwise you're eventually asking schools to transform their CSV to XML. | 17:47 |
| aelkner | th1a: what do you see as the advantage to csv? | 17:47 |
| ignas | true, open office can more easily generate CSV's for example | 17:47 |
| th1a | And it is easier for them to transform CSV to CSV. | 17:47 |
| pcardune | +1 csv | 17:48 |
| ignas | aelkner: for example you could do lyceum approach - have an open office spreadsheet with all the markings and explanations | 17:48 |
| aelkner | in my case, there is no open-office file that i know of | 17:48 |
| ignas | aelkner: i know | 17:48 |
| ignas | you asked for CSV upsides | 17:48 |
| aelkner | they have a mysql | 17:48 |
| aelkner | that's fine, i asked, thanks | 17:48 |
| aelkner | anyway, i will have to write some python | 17:48 |
| aelkner | to query the mysql database | 17:48 |
| aelkner | and build the import fild | 17:49 |
| aelkner | file | 17:49 |
| aelkner | that's why i figured that i could just build an xml file | 17:49 |
| aelkner | since i'm the one dong it | 17:49 |
| Lumiere | at the point you're doing querying from a db to make it | 17:49 |
| Lumiere | go xml | 17:49 |
| pcardune | aelkner: in that case you might as well build a python pickle... | 17:49 |
| aelkner | that's an idea | 17:50 |
| Lumiere | I still prefer xml | 17:50 |
| Lumiere | human readable > not human readable | 17:50 |
| aelkner | another good point | 17:50 |
| th1a | Pickle is least desirable in something we'd distribute. | 17:50 |
| Lumiere | I am thinking | 17:51 |
| Lumiere | if the import fails | 17:51 |
| Lumiere | I want to be able to look at the data I was importing | 17:51 |
| Lumiere | I would have to load it into python | 17:51 |
| Lumiere | if it was a pickle | 17:51 |
| aelkner | i agree with the advantage to having an xml file to look at | 17:52 |
| ignas | Lumiere: well - if alekner is doing it for others to reuse - i'd go with CSV, if it's for only him, as in one off script - pickles are fine | 17:52 |
| pcardune | pretty print? | 17:52 |
| ignas | :D | 17:52 |
| aelkner | pcardune: what do you mean? | 17:52 |
| Lumiere | ignas: I agree about the CSV | 17:52 |
| th1a | It is ok for aelkner to do this however is most convenient for him. | 17:53 |
| Lumiere | yes | 17:53 |
| aelkner | i will do what is most convenient for mw, yes, but | 17:53 |
| aelkner | i also happen to see an advantage to an xml | 17:53 |
| aelkner | for importing courses, sections, and teachers/students all in one file | 17:53 |
| pcardune | aelkner: prettyprint lets you print out python data structures in a slightly easier to read way | 17:54 |
| aelkner | i could see that being commonly reused | 17:54 |
| aelkner | pcardune: i know what pretty print does | 17:54 |
| aelkner | but what does that have to do with it? | 17:54 |
| pcardune | aelkner: makes it easier to read your pickle | 17:54 |
| ignas | unpickle + pretty print to get a human readable representation | 17:54 |
| Lumiere | he is saying you can load pickle and print it nicely | 17:54 |
| pcardune | lol | 17:55 |
| aelkner | ok, but i don't see what is wrong with standardizing an xml format | 17:55 |
| Lumiere | just remember with an xml format | 17:55 |
| Lumiere | that you need to order the data | 17:55 |
| aelkner | like Lumiere said, it's human readable | 17:55 |
| Lumiere | oh... dont take xml as instantly human readable | 17:55 |
| aelkner | order the data? | 17:55 |
| Lumiere | :) | 17:55 |
| Lumiere | you need people before courses before sections | 17:56 |
| ignas | could we get on topic please? | 17:56 |
| Lumiere | (and timetables etc) | 17:56 |
| Lumiere | ignas: when did we get off? | 17:56 |
| ignas | this technical discussion should be carried on after the meeting | 17:56 |
| aelkner | fine | 17:56 |
| aelkner | so, that's my report | 17:56 |
| th1a | OK. Thanks aelkner. | 17:57 |
| th1a | ignas? | 17:57 |
| ignas | the week was not too productive :/ | 17:58 |
| ignas | after talking to lyceum i got a couple more tasks for them to do | 17:58 |
| ignas | but to work on journal again | 17:58 |
| ignas | I had to migrate lyceum to schooltool.lyceum.journal | 17:58 |
| ignas | which involves a tricky evolution script | 17:58 |
| ignas | in progress of implementing it (i think it works now) | 17:58 |
| ignas | I also tried adding functional tests for schooltool.lyceum.journal | 17:59 |
| ignas | before that all the tests were in lyceum.ftests | 17:59 |
| ignas | and discovered that there is no way to schedule more than 1 term using our timetabling UI | 17:59 |
| ignas | because of a bug | 17:59 |
| ignas | and found a bug in schooltool breadcrumbs as well | 17:59 |
| ignas | fixed both of them | 18:00 |
| ignas | but have to fix a unit test for breadcrumbs now | 18:00 |
| Lumiere | I wonder if that breadcrumb bug affects cando competency views | 18:00 |
| ignas | as soon as i'll be done with that, i'll finish up schooltool.lyceum.journal functional tests | 18:00 |
| ignas | migrate lyceum to use schooltool.lyceum.journal | 18:01 |
| ignas | make it possible to edit both teachers and public event descriptions from the journal view at the same time | 18:01 |
| ignas | and start working on the overlays | 18:01 |
| ignas | oh, and it seems that CanDo buildout integration branch is working with trunk schooltool quite fine | 18:02 |
| ignas | but ccarey still can't commit to the cando repository | 18:02 |
| ignas | which is slowing us down | 18:02 |
| ignas | because he has to send me all the patches | 18:02 |
| Lumiere | ignas: ask jinty or someone at PoV to add him | 18:03 |
| th1a | jinty does it. | 18:03 |
| ignas | i thought so too | 18:03 |
| th1a | He just hasn't gotten to it? | 18:03 |
| jinty | I thought I'd given him access | 18:03 |
| th1a | Hi jinty! | 18:03 |
| Lumiere | eldar did it at one point (and probably still can) | 18:03 |
| jinty | more than a week ago | 18:03 |
| Lumiere | but he's still at school | 18:03 |
| ignas | jinty: he thought he got it, but he still can't commit i think | 18:03 |
| ignas | he tried entering his username + password | 18:04 |
| ignas | but got permission denied | 18:04 |
| ignas | that's all i have to say | 18:05 |
| ignas | i'll try to be more focused this week :/ | 18:05 |
| th1a | So to make State of the Nation up to date, it should include schooltool2008, right? | 18:06 |
| th1a | Which would otherwise be like schooltool2007? | 18:06 |
| ignas | hmm, i think it should | 18:08 |
| ignas | I will try and find some time to update all the docs | 18:08 |
| th1a | Yes, I need to update more user oriented docs, and I feel more comfortable doing that looking at your docs. | 18:09 |
| ignas | Ok | 18:10 |
| ignas | i'll even try to package up a schooltool2008 tarball | 18:10 |
| ignas | 2008spring that is | 18:10 |
| th1a | Ah. Right. | 18:10 |
| th1a | Lumiere, pcardune, jinty: Anything you'd like to throw in? | 18:11 |
| pcardune | not me | 18:12 |
| jinty | hmm, not really | 18:12 |
| th1a | jinty: What's the state of Zope packages on LP? | 18:12 |
| jinty | I need to speak to ignas about the builder I've installed on schooltool.org | 18:13 |
| th1a | I'm on a clean hardy box now, so I can work through our install process. | 18:13 |
| jinty | get that building/releasing schooltool packages | 18:13 |
| th1a | jinty, ignas: Now's a good time. | 18:13 |
| th1a | For me, at least ;-) | 18:14 |
| jinty | and perhaps try mass converting the KGS to packages | 18:14 |
| Lumiere | nope | 18:14 |
| ignas | ooohh | 18:14 |
| jinty | ignas: want to take it for a ride? | 18:14 |
| Lumiere | other then a cando meeting tonight 8pm EST (sorry ignas/jinty) | 18:14 |
| Lumiere | I have nothing to add | 18:14 |
| ignas | would be nice | 18:15 |
| jinty | ignas: did you get the mail I sent the other day | 18:15 |
| ignas | hmm | 18:15 |
| ignas | which day? | 18:15 |
| jinty | Auto building of packages/tarballs for eggs (Feb 03) | 18:16 |
| ignas | oh, yes I saw that | 18:16 |
| ignas | hmm | 18:17 |
| * jinty commits something to schooltool.buildconfig | 18:17 | |
| ignas | hmm | 18:18 |
| * ignas broke it | 18:18 | |
| ignas | http://localhost:11082/projects/schooltool.gradebook | 18:19 |
| jinty | yes, I can fix that if you answer me a question: | 18:20 |
| jinty | what is the correct thing to do when a project doesn't have a tags directory in svn? | 18:20 |
| ignas | hmm with schooltool repository | 18:20 |
| ignas | i'd use the tags that schooltool is using | 18:20 |
| ignas | i think we had both schoolbell and schooltool | 18:21 |
| ignas | use the same tags directory | 18:21 |
| Lumiere | I am out, be back later from the office | 18:21 |
| ignas | Lumiere: see you | 18:21 |
| jinty | 'file:///svn/schooltool/branches/schooltool.gradebook' non-existent in that revision | 18:21 |
| jinty | is the directory it's trying to use | 18:21 |
| th1a | bye Lumiere. | 18:21 |
| jinty | so the schema would be /branches/schooltool.gradebook/branch1 | 18:23 |
| ignas | i'd go with /branches/branch1 | 18:23 |
| ignas | actually | 18:23 |
| ignas | not to nest our repository even deeper | 18:23 |
| jinty | then how do you tell which branch is from which project? | 18:23 |
| ignas | and not to make bzr export scripts more confused | 18:24 |
| ignas | shouldnb | 18:24 |
| ignas | hmm | 18:24 |
| ignas | prefixes? | 18:24 |
| ignas | i mean schooltool-2007 | 18:24 |
| ignas | schooltool.gradebook-0.1 | 18:24 |
| ignas | schooltool.gradebook-0.1.xx | 18:24 |
| ignas | bzr export scripts are relying on such directory structure in the repository ... | 18:24 |
| ignas | i wouldn't want to go and fix them :/ as I am not the author | 18:25 |
| jinty | yeah, we can do that, it's just a matter of sub-classing the backend and changing some functions | 18:25 |
| ignas | http://source.schooltool.org/trac/browser/tags | 18:25 |
| ignas | has schooltool, schooltool.ldap and schoolbell already | 18:25 |
| ignas | anyway | 18:25 |
| ignas | I am a lot more worried about bzr branches though | 18:26 |
| jinty | yeah, I only got a backend for subversion | 18:26 |
| ignas | have you seen how the test/release cycle works with bzr in buildbot? | 18:26 |
| ignas | in bzr branches | 18:26 |
| jinty | no, not yet | 18:27 |
| jinty | you mean you use buildbot to make releases? | 18:27 |
| ignas | nightlies | 18:27 |
| ignas | make release target | 18:27 |
| ignas | that should get renamed to make nightly I guess | 18:27 |
| ignas | but yes - buildbot is releasing all the bzr eggs every now and then | 18:27 |
| ignas | after someone commits a working schooltool change set | 18:28 |
| ignas | to trigger the chain reaction | 18:28 |
| jinty | yes, but it's a little more difficult to have buildbot release a package, I think | 18:28 |
| ignas | i mean - i'd rather use your system to make actual releases | 18:29 |
| ignas | the ones that get tagged branched | 18:29 |
| ignas | and packaged into debs | 18:29 |
| ignas | buildbot is for nightlies only | 18:29 |
| jinty | ok, you want to try releasing schooltool.buildconfig then? | 18:29 |
| jinty | see how that goes? | 18:30 |
| ignas | i guess | 18:30 |
| ignas | what's involved ? clicking "release" ? | 18:30 |
| jinty | yeah | 18:31 |
| jinty | then wait | 18:31 |
| ignas | done | 18:31 |
| jinty | eventually a new egg should appear: http://ftp.schooltool.org/schooltool/eggs/ | 18:31 |
| jinty | and a debian package at: /var/www/buildresult/ | 18:31 |
| jinty | (I don't upload automatically to LP because the packages need a gpg signature) | 18:32 |
| jinty | but you can trivially sign that package and upload it to launchpad | 18:32 |
| * ignas is releasing schooltool + KGS 3.4 packages into http://ftp.schooltool.org/schooltool/eggs/3.4/ | 18:32 | |
| ignas | i see | 18:33 |
| ignas | hmm, where should I start to add bzr support? | 18:33 |
| jinty | http://localhost:11082/tarball_import.html | 18:33 |
| jinty | you can import single egg tarballs | 18:33 |
| ignas | upload the builder ssh key to launchpad so it could make branches i guess | 18:34 |
| jinty | you could make a schooltool.bzrsupport package for bzr support | 18:34 |
| ignas | http://localhost:11082/tarball_import.html fails if i try linking to: http://ftp.schooltool.org/schooltool/eggs/3.4/schooltool.devtools-0.4.tar.gz | 18:35 |
| jinty | what's the error? | 18:35 |
| ignas | something about reading from None | 18:35 |
| ignas | TypeError: 'NoneType' object is not callable<br /> | 18:35 |
| ignas | t.write(tarball.read()) was the line | 18:35 |
| jinty | wonderful ;)! | 18:36 |
| jinty | as I said, alpha quality | 18:36 |
| ignas | :) | 18:37 |
| jinty | schooltool.svn.interfaces.IProject | 18:37 |
| jinty | no.... | 18:37 |
| jinty | vanguardistas.svn.interfaces.IProject | 18:37 |
| ignas | what is the proper way to develop+test a the bzr support for this thing by the way? | 18:37 |
| jinty | is the interface your bzr backend needs to support | 18:37 |
| * jinty thinks it should be moved to a different package | 18:38 | |
| ignas | :) | 18:39 |
| lisppaste5 | jinty pasted "IProject" at http://paste.lisp.org/display/55739 | 18:39 |
| jinty | you see any problems implementing that for bzr? | 18:40 |
| ignas | hmm, i'd have to look at example outputs of these | 18:41 |
| ignas | get_branches + get_released are the ones i am not sure how to implement if we'd keep the stuff in launchpad | 18:42 |
| ignas | though - we can keep releases on schooltool.org ;) | 18:42 |
| ignas | and only have upstream in launchpad | 18:42 |
| lisppaste5 | jinty pasted "Current Tests for vanguardistas.svn.Project" at http://paste.lisp.org/display/55740 | 18:42 |
| jinty | ignas: those are the current tests for the svn backend | 18:43 |
| ignas | i see | 18:44 |
| jinty | currently the svn backend only can cope with version numbers like 1.2.3 | 18:44 |
| *** lhuynh has joined #schooltool | 18:44 | |
| jinty | but one day I hope to write more robust code | 18:45 |
| ignas | yeah, and i have them 0.1 0.2 at the moment | 18:45 |
| jinty | ... | 18:45 |
| ignas | so i guess your scripts might be failing because of that too | 18:45 |
| jinty | I think that might work, but it's not tested | 18:45 |
| ignas | what command do you use to build eggs? | 18:45 |
| jinty | sdist | 18:45 |
| ignas | buildout sandboxes do "bin/buildout setup sdist" | 18:45 |
| jinty | I never bother with binary eggs at all | 18:46 |
| th1a | mgedin's N810 just arrived. | 18:46 |
| ignas | the biggest problem is setting this thing up so I could test it locally :/ | 18:46 |
| ignas | th1a: i'll tell mgedmin as soon as he'll come to the office | 18:46 |
| jinty | the problem is that it's so intertwined with the system that setup is a bit difficult | 18:47 |
| ignas | I see | 18:47 |
| ignas | so how do you develop/test it? | 18:47 |
| ignas | when you want to add a feature or fix a bug | 18:48 |
| * jinty patches the thing on a server, then copies the bugfix locally and commits it ;) | 18:48 | |
| jinty | but for the backends, I generally try to have tests | 18:48 |
| jinty | it's just the upper layers that are _very_ difficult to test | 18:49 |
| jinty | I mean the current apt integration requires root | 18:49 |
| ignas | is it possible to run tests for the svn backend without everything else being available? | 18:49 |
| jinty | yes, definitely | 18:49 |
| jinty | it's split up over a few packages | 18:49 |
| ignas | what's the url of the svn backend repository? | 18:50 |
| jinty | vanguardistas.svn, vanguardistas.buildtools, vanguardistas.builder and schooltool.buildconfig | 18:50 |
| ignas | so i could try and set it up while you are available | 18:50 |
| * jinty is still wondering where to host the thing | 18:52 | |
| jinty | http://svn.vanguardistas.net/public/ | 18:52 |
| jinty | is temporary | 18:52 |
| * ignas found a checkout | 18:53 | |
| ignas | was in "src/STACK/..." ;) | 18:54 |
| jinty | you should be able to run the tests for buildtools and buildconfig without any setup | 18:54 |
| jinty | (I think) | 18:54 |
| jinty | Any suggestions where we should host the code? | 18:55 |
| jinty | (gotta be in svn) | 18:55 |
| jinty | so that I can give you commit acces? | 18:55 |
| ignas | hmm | 18:55 |
| ignas | no idea | 18:55 |
| ignas | i mean - I use darcs or bzr whe I want to host something ;) | 18:56 |
| * jinty was thinking of code.google.com or svn.zope.org | 18:56 | |
| * ignas is getting test failures in vanguardistas.svn | 18:57 | |
| jinty | ? | 18:57 |
| lisppaste5 | ignas pasted "failures" at http://paste.lisp.org/display/55743 | 18:58 |
| jinty | hmm, the tests run for me | 19:02 |
| jinty | I guess it's a different svn or setuptools version | 19:02 |
| * jinty is using whatever came with etch | 19:02 | |
| ignas | the ignas vs jinty | 19:03 |
| ignas | in tests :D | 19:03 |
| *** mgedmin has joined #schooltool | 19:03 | |
| jinty | ok, so one test bug fixed :) | 19:05 |
| jinty | but the others are still unexplained | 19:05 |
| ignas | hmm setuptools 0.6c6 here | 19:06 |
| ignas | what's yours ? | 19:06 |
| jinty | 0.6c3 | 19:07 |
| jinty | could also be python revision | 19:07 |
| * jinty sees that he needs to use the more robust method for introspecting eggs here | 19:08 | |
| jinty | Never use: from distutils.core import run_setup | 19:08 |
| ignas | :) | 19:08 |
| jinty | it is buggy as all hell | 19:08 |
| * jinty considers having vanguardistas.svn depend on vanguardistas.buildtools | 19:11 | |
| * ignas is trying to set up vanguardistas.svn in a buildout | 19:11 | |
| ignas | to test with an up to date Zope3.4 + up to date setuptools | 19:12 |
| ignas | hmm | 19:14 |
| ignas | your dependencies in vanguardistas.svn are not set up properly | 19:14 |
| ignas | ok setuptools 0.6c7 (the latest and greatest) are failing the same way | 19:17 |
| ignas | and vanguardistas.svn depends on zope.interface + zope.schema | 19:17 |
| ignas | while is not stating that in setup.py | 19:17 |
| *** nitromaster has joined #schooltool | 19:21 | |
| jinty | ignas: I'd prefer to keep those commented out until we have a python-zope.interface and python-zope.schema | 19:23 |
| jinty | debian package dependencies are computed from setup.py dependencies | 19:24 |
| ignas | oh | 19:24 |
| ignas | hmm | 19:24 |
| ignas | might be a problem for schooltool | 19:24 |
| jinty | it's a very difficult problem | 19:25 |
| ignas | unless you do the KGS import first | 19:27 |
| * jinty swears profusely at setuptools.sandbox for printing to stdout | 19:27 | |
| jinty | yes, but I havn't had time to think about that yet... | 19:27 |
| ignas | most of my eggs that are using bzr depend on schooltool | 19:31 |
| ignas | or on other eggs in bzr ;) | 19:31 |
| jinty | ignas: could you svn up and test? | 19:32 |
| jinty | I've made the setuptools introspection more robust | 19:32 |
| jinty | all tests still apss here | 19:32 |
| ignas | still failing the same way | 19:33 |
| ignas | you should try virtualpython + easy_install zope.interface + easy_install zope.schema + vp/bin/python test.py | 19:34 |
| ignas | to test it with an up to date setuptools | 19:34 |
| ignas | i can give you precise instructions if you have the time | 19:34 |
| ignas | wget http://svn.colorstudy.com/virtualenv/trunk/virtualenv.py | 19:35 |
| ignas | python2.4 virtalenv.py my_sandbox | 19:35 |
| ignas | cd my_sandbox | 19:35 |
| ignas | bin/easy_install zope.testing zope.interface zope.schema | 19:36 |
| ignas | bin/python PATH_TO_YOUR_TEST.py | 19:36 |
| jinty | but I want to test it with what comes in etch;) | 19:36 |
| jinty | until I want to upgrade to lenny | 19:37 |
| ignas | well - I want you to test it with both | 19:37 |
| ignas | for future proofness | 19:37 |
| jinty | ok ok ;) | 19:37 |
| jinty | all tests pass with your virtual env | 19:39 |
| *** lhuynh has quit IRC | 19:41 | |
| ignas | hax! | 19:41 |
| jinty | including with python2.5 | 19:42 |
| * jinty wonders what kind of code he's been writing | 19:42 | |
| ignas | wow | 19:43 |
| ignas | tests suddenly passed :/ | 19:43 |
| jinty | hrumpf | 19:44 |
| jinty | but I did find a bug | 19:44 |
| jinty | it seems when making a release from a branch that is not trunk | 19:45 |
| jinty | the release is made from the trunk anyway | 19:45 |
| ignas | :D | 19:45 |
| jinty | fixed... | 19:57 |
| jinty | ignas: ok, so the KGS + debian packages problem | 19:58 |
| th1a | mgedmin: So... can I play with your 810? | 19:58 |
| jinty | ignas: I'd like to use the packaged KGS for my company as well | 19:58 |
| mgedmin | th1a: sure | 19:58 |
| mgedmin | just don't break it :) | 19:58 |
| jinty | ignas: so is it a problem if we host/create one? | 19:59 |
| th1a | mgedmin: Can you restore these to factory settings? | 19:59 |
| mgedmin | yes | 19:59 |
| th1a | OK. I won't install any software. | 19:59 |
| jinty | then we upload it to LP | 19:59 |
| mgedmin | feel free | 19:59 |
| th1a | I really just want to compare it to the XO. | 19:59 |
| mgedmin | you can always reflash the whole OS image to get to the factory state | 19:59 |
| ignas | jinty: don't think so | 19:59 |
| ignas | jinty: i mean - who is the Zope3 release manager for debian? | 20:00 |
| mgedmin | except it doesn't change anything on the "internal memory card" which, as far as I've heard, contains GPS maps | 20:00 |
| mgedmin | (and isn't really a removable memory card, just 2 gigs of flash that pretend to be one) | 20:00 |
| jinty | er, I think the way to do this is to build one | 20:00 |
| ignas | jinty: precisely | 20:00 |
| jinty | ignas: show that it works | 20:00 |
| jinty | then get debian to upload the thing | 20:00 |
| ignas | indeed | 20:01 |
| jinty | in the meantime we can server the KGS from schooltool's PPA | 20:01 |
| jinty | (for ubuntu releases) | 20:01 |
| * ignas wants to get schooltool.stapp2008spring into a working + releasable deb format asap | 20:01 | |
| th1a | mgedmin: Oooh... shiny. | 20:01 |
| ignas | jinty: will we have enough space? | 20:01 |
| jinty | dunno, are there limits? | 20:01 |
| ignas | something around a gig or two I think | 20:02 |
| jinty | should be enough | 20:02 |
| th1a | On LP? | 20:02 |
| ignas | yes | 20:02 |
| ignas | how much space do we have there? | 20:02 |
| th1a | That might become an issue but I'm sure we can get more. | 20:02 |
| th1a | I have connections. | 20:02 |
| jinty | I think 1 gig is definitely enough | 20:03 |
| jinty | all the package files are compressed | 20:03 |
| ignas | jinty: what is missing to get a schooltool release with full dependencies into PPA? | 20:09 |
| * ignas even has a tarball that has all of the eggs | 20:09 | |
| ignas | for schooltool2007 | 20:09 |
| ignas | jinty: http://ftp.schooltool.org/schooltool/schooltool2007-buildout.tgz | 20:09 |
| ignas | with an extra tarball of libxml ;) | 20:09 |
| jinty | hmm, how about we try making zope.interface, zope.schema and zope.testing pacakges? | 20:10 |
| ignas | we can, anything i can help you with in doing that? | 20:11 |
| jinty | let me see how the thing goes | 20:11 |
| * jinty wants a Tarball import page capable of spidering a KGS | 20:11 | |
| ignas | spidering might be a bad idea, there are all the eggs from pypi | 20:13 |
| ignas | only eggs that are managed by Zope3 comunity are in packages.cfg or controled-versions.cfg though | 20:13 |
| ignas | I think | 20:13 |
| ignas | so you might want to read the config file, and generate a list of eggs that you want to package from it | 20:14 |
| jinty | that's an idea | 20:14 |
| jinty | better than uncontrolled spidering | 20:14 |
| ignas | if you read Zope3-dev there was stephan's email that linked to the config files | 20:15 |
| jinty | but how to get the tarballs, once you have the eggs | 20:15 |
| ignas | hmm, you are not building debian packages from source eggs, are you? | 20:16 |
| jinty | yes | 20:18 |
| jinty | you get the tarball | 20:18 |
| jinty | unpack it | 20:19 |
| jinty | get the current debian packaging, apply it to the new tarball | 20:19 |
| jinty | and release | 20:19 |
| Lumiere | I am betting that LP would love to have someone abuse PPA hard | 20:24 |
| Lumiere | and make sure it can take punishment | 20:24 |
| *** alga has quit IRC | 20:24 | |
| th1a | I'm sure Canonical is pretty good at torturing LP themselves. | 20:25 |
| jinty | ignas: what would be a good set of zope packages to start releasing? zope.interface, zope.testing... | 20:27 |
| ignas | i think these would do | 20:28 |
| ignas | zc.recipe.egg | 20:28 |
| ignas | zc.buildout | 20:28 |
| ignas | would be good candidates too | 20:28 |
| ignas | j1m would be so glad :D knowing how much he dislikes system pythons in general ;) | 20:30 |
| mgedmin | jim is going to love the flood of ubuntu bug reports for zc.buildout :) | 20:30 |
| Lumiere | lol | 20:32 |
| jinty | ignas: ok, I've uploaded python-zope.interface and python-zope.testing to the PPA | 20:34 |
| jinty | you want to do one? | 20:35 |
| ignas | http://download.zope.org/zope3.4/controlled-packages.cfg - the file I have talked about (i think ;) | 20:36 |
| jinty | erp, need API to translate these specifications into tarball URLS | 20:37 |
| ignas | http://pypi.python.org/packages/source/z/zc.recipe.cmmi/zc.recipe.cmmi-1.1.0.tar.gz#md5=39b215ad57710d8564cc5fab6138dc4d failed :/ | 20:37 |
| * jinty swears brutally at the code that he wrote hiding errors | 20:38 | |
| jinty | I think that might have to be manually packaged first | 20:38 |
| jinty | the default "make package from nothing" script is quite simplistic | 20:39 |
| jinty | how did zc.buildout go? | 20:39 |
| ignas | don't know ;) didn't say anything ;) | 20:41 |
| ignas | just clicked import | 20:41 |
| ignas | and it refreshed without warnings | 20:41 |
| jinty | ok, so then it worked :) | 20:41 |
| * jinty needs status messages as well | 20:42 | |
| ignas | :) | 20:42 |
| lisppaste5 | jinty pasted "Procedure to sign tarballs and upload them to the PPA" at http://paste.lisp.org/display/55747 | 20:43 |
| jinty | ignas: does that make sense? | 20:43 |
| ignas | looks like it does | 20:44 |
| jinty | after that, hopefully the PPA will build it properly | 20:44 |
| jinty | then anyone wanting to modify the packaging can download/upload it to the PPA | 20:45 |
| ignas | what about existing zope.interface packages in Ubuntu ? | 20:45 |
| jinty | a tarball upgrade will then use the latest packaging from the PPA | 20:45 |
| jinty | there is only a python-zopeinterface | 20:46 |
| jinty | honestly, I'm not sure | 20:46 |
| ignas | won't we conflict with it? and what happens if we will overwrite their files ... | 20:46 |
| jinty | I think we should just conflict against zope3 and python-zopeinterface | 20:46 |
| ignas | hope so ;) | 20:47 |
| jinty | I'm not aware of another solution that is safe | 20:47 |
| ignas | will you upload the files to PPA | 20:47 |
| ignas | or should I try doing that | 20:47 |
| jinty | you go for it | 20:47 |
| ignas | .dput.cf ? not .dput.cfg | 20:48 |
| jinty | the whole point of me building out this infrastructure is that I wouldn't be a critical blockage | 20:48 |
| jinty | yep .cf | 20:48 |
| jinty | no g | 20:48 |
| jinty | th1a: what's the status of the old zope3 tarball based release? | 20:52 |
| th1a | Release of Zope3? | 20:52 |
| jinty | because by diving into packages now, we are risking breaking that | 20:52 |
| jinty | I mean the schooltool release that depends on the zope3 package | 20:52 |
| jinty | that's all in out PPA | 20:52 |
| ignas | Successfully uploaded packages. | 20:52 |
| ignas | Not running dinstall. | 20:52 |
| jinty | ignas, that's cool | 20:53 |
| jinty | https://edge.launchpad.net/~schooltool-owners/+archive | 20:53 |
| th1a | I'm not following you, jinty. | 20:53 |
| jinty | I mean python-zope.interface and the old zope3 package are incompatible | 20:54 |
| jinty | which is file | 20:54 |
| jinty | fine | 20:54 |
| th1a | I'm not worried about the stuff we previously put in PPA, if that's what you mean. | 20:54 |
| jinty | as we can build parallel dependency trees | 20:54 |
| jinty | yes, that's what I mean | 20:54 |
| jinty | If we carry on putting stuff in the PPA we'll eventually break all that | 20:55 |
| th1a | That's ok. | 20:55 |
| th1a | It never really worked anyhow. | 20:55 |
| jinty | ok, so then I'm going to delete the old zope3 package | 20:56 |
| jinty | and things like python-zope.ucol | 20:56 |
| ignas | isnt python-zope.ucol | 20:56 |
| ignas | a proper egg-deb | 20:57 |
| ignas | hmm | 20:57 |
| ignas | or is it having wrong dependencies? | 20:57 |
| jinty | yes, but it depends on zope3 | 20:57 |
| jinty | or we can just update the dependencies... | 20:57 |
| jinty | probably the best option actually | 20:57 |
| jinty | but the monolithic zope3 monster definitely has to go | 20:57 |
| * jinty get's afraid thinking of ZConfig | 20:58 | |
| jinty | best that we have a resonable zope dependency tree before doing irreversible things | 20:59 |
| ignas | true | 20:59 |
| jinty | Ok, we've gotta wait for the PPA to build the binary packages, then test what we've built | 21:00 |
| * jinty takes a break | 21:01 | |
| ignas | :) | 21:03 |
| jinty | ok, zope-interface built | 21:23 |
| jinty | but the setup.py introspection broke again | 21:24 |
| jinty | but we should see if we can install the thing | 21:24 |
| *** Aiste has quit IRC | 21:24 | |
| *** Aiste has joined #schooltool | 21:25 | |
| * jinty watches the zc.buildout tests fail mostly horribly and wonders if buildout conflicts with the system python a bit | 21:34 | |
| jinty | Ran 279 tests with 3 failures and 43 errors | 21:34 |
| ignas | no idea | 21:35 |
| ignas | where can I look at these failures? | 21:35 |
| ignas | jinty: hmm, it seems that buildout works | 22:03 |
| ignas | jinty: at least on one of my sandboxes ;) | 22:03 |
| ignas | not to launch a nightly script that would do this for all the packages in KGS ;) | 22:04 |
| ignas | and print a user friendly list to make upload to PPA easier to do | 22:04 |
| *** ignas has quit IRC | 22:11 | |
| *** grishahun has joined #schooltool | 22:13 | |
| grishahun | can anyone help me with installing schooltool on ubuntu 7.10 Gusty? I've got som dependency problems | 22:14 |
| Lumiere | grishahun: what are you installing from? | 22:16 |
| grishahun | i'm installing schooltools from synaptic | 22:17 |
| Lumiere | th1a: you here? | 22:17 |
| Lumiere | installing from packages is a bad idea right now | 22:17 |
| Lumiere | they're wildly out of date | 22:17 |
| grishahun | thx | 22:18 |
| grishahun | I realized | 22:18 |
| Lumiere | we are working toward a release in the next version or two | 22:18 |
| grishahun | can you help me to install it any other way | 22:19 |
| Lumiere | yea sec | 22:19 |
| grishahun | thank you | 22:20 |
| Lumiere | http://www.schooltool.org/documentation/setting-up-a-development-server is a start | 22:21 |
| *** mgedmin has quit IRC | 22:21 | |
| Lumiere | not sure how up to date it is | 22:21 |
| Lumiere | th1a: is looking to update all these soon | 22:21 |
| grishahun | all i need is the offline functions of schooltool. tracking my students' test results, missings, etc | 22:24 |
| * Lumiere stabs th1a | 22:24 | |
| *** didymo has joined #schooltool | 22:40 | |
| *** grishahun has left #schooltool | 22:49 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!