| *** aelkner has quit IRC | 02:44 | |
| *** aelkner has joined #schooltool | 02:49 | |
| *** lisppaste5 has quit IRC | 03:21 | |
| *** aelkner has quit IRC | 03:25 | |
| *** lisppaste5 has joined #schooltool | 03:32 | |
| *** aelkner has joined #schooltool | 03:50 | |
| *** fsufitch has quit IRC | 07:05 | |
| *** didymo has quit IRC | 07:13 | |
| *** aelkner_ has joined #schooltool | 07:21 | |
| *** aelkner has quit IRC | 07:22 | |
| *** aelkner_ has quit IRC | 07:27 | |
| *** aelkner has joined #schooltool | 07:27 | |
| *** didymo has joined #schooltool | 09:59 | |
| *** jinty has joined #schooltool | 11:49 | |
| *** didymo has quit IRC | 11:57 | |
| *** jinty_ has joined #schooltool | 11:59 | |
| *** jinty has quit IRC | 12:07 | |
| *** jinty has joined #schooltool | 12:08 | |
| *** jinty_ has quit IRC | 12:16 | |
| *** jinty_ has joined #schooltool | 12:18 | |
| *** jinty__ has joined #schooltool | 12:27 | |
| *** jinty has quit IRC | 12:33 | |
| *** jinty_ has quit IRC | 12:35 | |
| *** jinty has joined #schooltool | 12:36 | |
| *** jinty__ has quit IRC | 12:44 | |
| *** jinty_ has joined #schooltool | 12:46 | |
| *** jinty has quit IRC | 12:54 | |
| *** jinty_ has quit IRC | 12:54 | |
| *** jinty has joined #schooltool | 12:55 | |
| *** jinty_ has joined #schooltool | 13:04 | |
| *** jinty has quit IRC | 13:13 | |
| *** jinty_ has quit IRC | 13:13 | |
| *** jinty has joined #schooltool | 13:14 | |
| *** jinty_ has joined #schooltool | 13:23 | |
| *** jinty has quit IRC | 13:31 | |
| *** jinty_ has quit IRC | 13:32 | |
| *** jinty has joined #schooltool | 13:32 | |
| *** jinty_ has joined #schooltool | 13:42 | |
| *** jinty__ has joined #schooltool | 13:48 | |
| *** jinty has quit IRC | 13:49 | |
| *** jinty has joined #schooltool | 13:59 | |
| *** jinty_ has quit IRC | 13:59 | |
| *** jinty__ has quit IRC | 14:06 | |
| *** jinty_ has joined #schooltool | 14:08 | |
| *** jinty has quit IRC | 14:17 | |
| *** jinty has joined #schooltool | 14:18 | |
| *** jinty_ has quit IRC | 14:26 | |
| *** ignas has joined #schooltool | 14:30 | |
| *** jinty has quit IRC | 14:34 | |
| *** jinty has joined #schooltool | 14:36 | |
| *** jinty has quit IRC | 14:45 | |
| *** jinty has joined #schooltool | 14:46 | |
| *** alga has joined #SchoolTool | 14:55 | |
| *** jinty has quit IRC | 15:04 | |
| *** jinty has joined #schooltool | 15:07 | |
| *** jinty has quit IRC | 15:18 | |
| *** jinty has joined #schooltool | 15:19 | |
| *** jinty_ has joined #schooltool | 15:28 | |
| *** alga has quit IRC | 15:31 | |
| *** jinty has quit IRC | 15:35 | |
| *** jinty has joined #schooltool | 15:38 | |
| *** jinty_ has quit IRC | 15:46 | |
| *** jinty_ has joined #schooltool | 15:47 | |
| *** jinty has quit IRC | 15:56 | |
| *** jinty has joined #schooltool | 15:57 | |
| *** jinty_ has quit IRC | 16:05 | |
| *** jinty has quit IRC | 16:18 | |
| *** wbrady has joined #schooltool | 16:27 | |
| wbrady | aelkner: hey, are pretty busy right now? | 16:28 |
|---|---|---|
| aelkner | about to have a meeting | 16:29 |
| aelkner | where are you, jeff's classroom? | 16:29 |
| wbrady | yea 2nd period | 16:29 |
| wbrady | i was just wondering, if I had a question about why CanDo does or has something who would it be best to ask? | 16:29 |
| aelkner | you could ask me | 16:30 |
| wbrady | ok cool | 16:30 |
| aelkner | hav any question at the mement? | 16:30 |
| wbrady | i put together a list of some questions but it might be better to go over them at a different time when you have a little more time | 16:31 |
| *** test123_ has joined #schooltool | 16:31 | |
| aelkner | why don't you email me the list, and we could discuss it later | 16:32 |
| th1a | test123_: Hi Ian. | 16:33 |
| ignas | hi everyone | 16:33 |
| test123_ | hi | 16:33 |
| wbrady | aelkner: ok its some very unofficial notes but I'll send it to you | 16:34 |
| aelkner | that fine | 16:34 |
| th1a | Since test123_ is here, we should probably focus first on getting Ian's project over the finish line. | 16:34 |
| test123_ | sounds good to me | 16:34 |
| th1a | There's some discussion needed with ignas on the final direction. | 16:34 |
| test123_ | Do you want to lead on this Tom? | 16:34 |
| th1a | If we get through with that there are some bugs to discuss with ignas. | 16:35 |
| wbrady | aelkner: actually allen I don't have your email address | 16:35 |
| th1a | Well, I just read test123_'s email. | 16:35 |
| th1a | So we can go through that. | 16:35 |
| th1a | Hm... I guess I should forward it to ignas & aelkner. | 16:36 |
| th1a | OK. Sent. | 16:36 |
| *** jstraw has joined #schooltool | 16:37 | |
| aelkner | got it | 16:37 |
| *** jstraw is now known as Lumiere | 16:37 | |
| th1a | Staring at the end... yes, we have switched to a dependency on lxml. | 16:37 |
| th1a | Instead of the libxml2 python bindings. | 16:37 |
| ignas | oh, yeah, I have forgot to update tizard buildout | 16:37 |
| test123_ | ok does that mean that the cando doct needs to be updated? | 16:37 |
| ignas | to not build libxml2 anymore | 16:37 |
| ignas | CanDo? | 16:38 |
| th1a | Actually, everything needs to be updated, I guess. | 16:38 |
| th1a | Documentation-wise. | 16:38 |
| test123_ | http://cando.schooltool.org/installation/installation | 16:38 |
| ignas | cando is using a bit different installation procedures than the ones that schooltool is using I think | 16:39 |
| th1a | I'm a little unclear on exactly what the effects are on Mac OS X in terms of the underlying libxml dependency. | 16:39 |
| * ignas does not have any macs around | 16:39 | |
| ignas | but I could help someone with enough time | 16:39 |
| ignas | to make schooltool run on their machine | 16:40 |
| Lumiere | the cando docs are wildly out of date to current trunk | 16:40 |
| ignas | remotely | 16:40 |
| th1a | lxml still uses the libxml C library, and I'm not sure how daintily it has to match versions. | 16:40 |
| th1a | That's been a real headache in the past with the regular libxml2 bindings. | 16:40 |
| test123_ | OK,, that's helpful Tom.. I am just keen not to bundle C libraries we don't need | 16:40 |
| th1a | Well, Mac OS X should ship with libxml2. | 16:41 |
| test123_ | and establish a least effort way of keeping the mac and ubuntu builds in sync | 16:41 |
| test123_ | which I think may best be achieved by agreeing a ''contract'' with ignas in the form of | 16:41 |
| test123_ | a ''.deb'' package dependency.. | 16:42 |
| ignas | well - with us migrating from libxml2 we don't need to build libxml2 anymore | 16:42 |
| ignas | so i'll just update tizard | 16:42 |
| ignas | to take advantage of these changes | 16:42 |
| test123_ | while OSX ships with 2.5 and we are using 2.4 it may be best not to use their library? | 16:43 |
| ignas | and the rest of the dependencies are eggs | 16:43 |
| th1a | Well, one thing I'd say is that after the initial release you need for your immediate use, you should only update when we do a release every six months. | 16:43 |
| test123_ | do we plan to advance to 2.5? | 16:43 |
| th1a | libxml 2.5? | 16:43 |
| test123_ | python 2.5 | 16:43 |
| th1a | Oh. That's really a Zope question. | 16:44 |
| th1a | I presume eventually, but it doesn't seem to be a priority. | 16:44 |
| th1a | It is not coming in the near future. | 16:45 |
| th1a | OK Leopard ships with libxml 2.6.16 | 16:47 |
| test123_ | So we need a policy to deal with the fact that we have a non-standard OSX python. We are copying it into the self-contained schooltool-tizard folder and at present as in ubuntu we symlink the /usr/local (which we need to buildout ) at the risk of interfering with other stuff that is there? or do we | 16:48 |
| test123_ | copy all the assets we need into the self contained OSX folder? | 16:48 |
| th1a | From the lxml instructions: On MacOS-X 10.4, you can try to use the installed system libraries when you build lxml yourself. However, the library versions on this system are older than the required versions, so you may encounter certain differences in behaviour or even crashes. A number of users reported success with updated libraries (e.g. using fink), but needed to set the environment variable DYLD_LIBRARY_PATH to the directory where fink keeps the libra | 16:49 |
| th1a | ries. | 16:49 |
| ignas | th1a: hmm, we could try depending on an older lxml if it's available on OSX | 16:49 |
| th1a | The Mac 10.5 libraries are newer, I'd say we should just try it with themand see what happens. | 16:49 |
| ignas | th1a: but I can't create a proper workaround witohut a mac :( | 16:50 |
| th1a | Yes, that's probably true, ignas, but I bet it just works. | 16:50 |
| ignas | and at the moment - libxml2 will break the build | 16:50 |
| test123_ | what about icu? | 16:50 |
| ignas | it's not compatible with lxml egg | 16:50 |
| ignas | test123_: what specific error message are you getting with icu? | 16:51 |
| ignas | because you shouldn't need any symlinks to build it i'd assume | 16:52 |
| test123_ | Its not an problem to install in usr/local but I am not aware if it is bundled with the system? | 16:52 |
| th1a | test123_: It does not appear to be. | 16:53 |
| *** mgedmin has joined #schooltool | 16:53 | |
| ignas | :( ICU support seems to be a significant problem on all the systems | 16:53 |
| * ignas would have had windows build working if not for ICU | 16:53 | |
| ignas | and both python libraries for ICU need C compiler and libraries available | 16:54 |
| test123_ | my gut instinct is to create a self-contained archive with all the assets we agree we need as binaries on the Mac side | 16:54 |
| th1a | test123_: Yes. | 16:54 |
| th1a | Disc space is cheap. | 16:55 |
| test123_ | and to work with an informal ``deb'' system on ubuntu | 16:55 |
| *** wbrady has quit IRC | 16:55 | |
| test123_ | This also has the advantage of forcing us to be transparent on the dependencies | 16:55 |
| test123_ | which if I understand you correctly will only change at most every six months | 16:56 |
| th1a | Once you aren't tracking our trunk, yes, it will be a twice a year update. | 16:57 |
| test123_ | The downside is that a bit more testing will be needed to relocate assets in the OSX 10.5.1 system we | 16:57 |
| test123_ | have hacked together | 16:57 |
| test123_ | I would appreciate sign-off from Ignas once we have that done | 16:57 |
| th1a | I think we have to figure out what we'd mean by "sign-off." | 16:58 |
| ignas | as for the package thingie | 17:00 |
| ignas | i'll need some testers after the meeting | 17:00 |
| th1a | I can certainly try out the installer on my Mac. | 17:00 |
| ignas | i'll need ubuntu testers too | 17:01 |
| ignas | i am trying to package a schooltool2007 egg release | 17:01 |
| ignas | that would be a huge tarball | 17:01 |
| ignas | with everything needed (except for all the libicu, libxml, gcc stuff) | 17:01 |
| test123_ | The ''.deb package'' dependencies -- what they are and where they are located -- plus any environment variables, symlinks or path settings that need to be hardwired should be documented by us and agreed by Ignas as necessary and sufficient. This gets round the problem that our reverse engineering may throw up assets that have been deprecated (eg the libmxl2 python bindings) and that we may be passing the functional tests | 17:02 |
| test123_ | == def sign off | 17:03 |
| test123_ | ? | 17:03 |
| th1a | Does that make sense to you ignas? | 17:03 |
| ignas | I can try, but without an actual system to poke around it is a bit difficult ... most of the dependencies are listed in build-deps of the schooltool debian package, so as soon as we'd come up with schooltool2008 debian package | 17:04 |
| ignas | we'd have the full list of both - build and install dependencies | 17:04 |
| th1a | It sounds like that's what test123_ needs. | 17:04 |
| test123_ | Not sure if we need the build dependencies. I am not proposing that we do any compilation on the Mac (except for icu). | 17:05 |
| ignas | well - icu and lxml both need to compile some things | 17:06 |
| ignas | ZODB needs to do some compiling too | 17:06 |
| ignas | unless you can precompile everything | 17:06 |
| ignas | because it's a Mac and it just works | 17:06 |
| th1a | I would try to precompile everything. | 17:06 |
| test123_ | sorry ... my intention is to pre-compile -- including ICU. | 17:06 |
| ignas | that makes sense | 17:07 |
| test123_ | My assumption is that we get hardware independence (PPC and Intel) from the universal binaries | 17:07 |
| th1a | Oh yeah... I think you're on your own with that one. | 17:08 |
| test123_ | OK | 17:08 |
| th1a | Hopefully it will just work ;-) | 17:08 |
| test123_ | thanks.. I think I have what we need. I will post progress at ../germanbight. | 17:09 |
| th1a | OK. | 17:09 |
| Lumiere | hi th1a | 17:09 |
| th1a | hi Lumiere. | 17:10 |
| th1a | What's up? | 17:10 |
| Lumiere | for those of you who read the topic... there isn't a CanDo dev meeting tonight... but starting next week Monday evenings will be CanDo dev meetings | 17:10 |
| th1a | OK. | 17:11 |
| th1a | If there is nothing else from test123_... | 17:11 |
| th1a | can we look at https://bugs.edge.launchpad.net/schooltool/+bug/187772 | 17:11 |
| th1a | ? | 17:11 |
| th1a | I like ignas's suggestion. | 17:12 |
| th1a | That one overlays calendars by finding them and clicking an "Add this calendar" button. | 17:13 |
| Lumiere | so it only checks for conflicts with calendars you have selected? | 17:13 |
| th1a | Conflicts? | 17:13 |
| ignas | nothing related to conflicts | 17:13 |
| th1a | It is just to select which calendars you can overlay visually on yours. | 17:13 |
| ignas | these are related to the ability to see the events of other person in your calendar | 17:14 |
| Lumiere | ah | 17:14 |
| Lumiere | then yes ignas's suggestion seems good | 17:14 |
| th1a | Then we could just make it easy to remove obsolete ones with a click. | 17:14 |
| th1a | Which solves the "too many obsolete section calendars in my legend" bug. | 17:15 |
| th1a | Well, not perfectly, but good enough. | 17:15 |
| Lumiere | anything outside of term shouldn't display anyways | 17:15 |
| th1a | Well, ideally, if you went back to that date you'd still see the events. | 17:15 |
| ignas | we'd just move the checkboxes to the "more" view | 17:16 |
| th1a | It is really more of a problem with cluttering up the list of overlaid calendars. | 17:16 |
| th1a | Checkboxes? | 17:16 |
| ignas | the "enabled" "disabled" "tt" | 17:16 |
| th1a | I was thinking get rid of the "More" view. | 17:16 |
| ignas | so the "more view" would allow you to "remove" "hide" "set color maybe" | 17:16 |
| th1a | Hm. | 17:17 |
| ignas | we still need a way to "manage" overlays, and I always wanted to have "color picking" ;) | 17:17 |
| th1a | We need color picking. | 17:17 |
| th1a | You may be right, ignas. | 17:17 |
| Lumiere | allowing color setting is nice... but I think that getting a good set of colors is a first step | 17:18 |
| ignas | we have a good set of colors ;) | 17:18 |
| th1a | I'll send an email to the main list. | 17:18 |
| th1a | We have good colors. | 17:18 |
| th1a | I don't know if we have enough. | 17:18 |
| th1a | How many do we have? | 17:18 |
| ignas | around 8-10 i think | 17:18 |
| th1a | Any other thoughts about this. | 17:20 |
| th1a | I think I can write up a proposal at this point. | 17:21 |
| ignas | not really, i'd rather try out what we are thinking about first | 17:21 |
| ignas | and then see how we can improve on it | 17:21 |
| ignas | you can start downloading http://ftp.schooltool.org/schooltool/schooltool2007-buildout.tgz | 17:21 |
| ignas | and if python2.4 install.py + bin/make-schooltool-instance instance + bin/start-schooltool-instance instance/ works | 17:23 |
| ignas | i'll update installation instructions for users | 17:23 |
| ignas | and maybe even package such a thing for trunk schooltool | 17:24 |
| th1a | Trying it... | 17:25 |
| th1a | jinty was working on Zope debs in Launchpad last weekend. | 17:27 |
| ignas | cool | 17:27 |
| th1a | I don't know how it came out yet... the fix we needed should have been committed. | 17:27 |
| th1a | aelkner: What are your plans for the week? | 17:29 |
| aelkner | last week, franky and i worked a little on drupal cas | 17:30 |
| aelkner | we have it logging into cas, but logging out is more problematic | 17:30 |
| aelkner | because drupal uses something called "drupal views" | 17:31 |
| aelkner | and currently there is no view for logout | 17:31 |
| aelkner | so hitting the logout button does nothing | 17:31 |
| th1a | eh? | 17:31 |
| th1a | You can't logout? | 17:31 |
| aelkner | nope | 17:31 |
| aelkner | one wonders how they ever logged out | 17:31 |
| th1a | This is the problem with interoperability. | 17:32 |
| aelkner | even without cas in the equation | 17:32 |
| th1a | You have to interoperate with such screwed up stuff. | 17:32 |
| aelkner | that's life | 17:32 |
| aelkner | one just has to deal with the screwed up thing which we are | 17:32 |
| aelkner | anyway, we only spent a couple of half-days on it | 17:32 |
| aelkner | so progress has slowed some | 17:33 |
| aelkner | otherwise | 17:33 |
| aelkner | i created a Demographics obvject as per ignas' suggestion | 17:33 |
| aelkner | that lives in its own container off of the applicationh root | 17:33 |
| aelkner | with __name__ the same as corresponding BasicPerson | 17:34 |
| aelkner | then i created an edit view that incorporates the Demographics with the BasicPerson fields | 17:34 |
| th1a | That sounds good. | 17:34 |
| aelkner | then I created my own import class | 17:34 |
| aelkner | to import the students from the csv output of sla's student database | 17:35 |
| aelkner | the edit view and the import are working save for two things | 17:35 |
| aelkner | 1) i need to work out how to deal with sla's need for two advisors | 17:35 |
| aelkner | 2) i need functional tests | 17:36 |
| aelkner | sometime in the next two days, i should have this all done | 17:36 |
| Lumiere | are the advisors under the same title? | 17:36 |
| aelkner | ? | 17:36 |
| Lumiere | are they both "faculty advisor" | 17:36 |
| aelkner | oh, in the csv you mean | 17:37 |
| Lumiere | just in general | 17:37 |
| aelkner | they just have an advisor field that has up to two advisors | 17:37 |
| Lumiere | ok | 17:37 |
| aelkner | they just use last names with a '/' between them if there are two | 17:37 |
| aelkner | anyway | 17:37 |
| aelkner | after i'm done with that, i'll move on to the narrative report cards | 17:38 |
| th1a | OK. Don't get hung up on CAS at this point. | 17:38 |
| aelkner | not getting hung up, but an working slowly to move it along | 17:39 |
| aelkner | itnot being a priority works well with franky not being available a lot | 17:39 |
| th1a | OK. | 17:40 |
| th1a | Good. | 17:40 |
| aelkner | i'd like to talk advisors with ignas following the dropping of the gravel | 17:40 |
| th1a | I think that's it for this meeting. | 17:40 |
| * th1a drops the bag of gravel. | 17:40 | |
| th1a | ignas: Why do I need libxml2 for this buildout? | 17:41 |
| th1a | Oh... it is for ST2007? | 17:41 |
| ignas | aelkner: hmm, advisor relationship has no limitations for the ammount of possible advisors | 17:45 |
| ignas | so i'd look at it's implementation in BasicPerson | 17:45 |
| ignas | maybe hide the "advisor" attribute in the UI | 17:45 |
| ignas | and add IStudent(person).advisors to manage them | 17:45 |
| ignas | like we have ILearner(person).sections | 17:46 |
| ignas | or IAdvisee() | 17:46 |
| ignas | you will have to do the limmiting to 2 yourself though | 17:46 |
| ignas | so you might have advisor1 and advisor2 attributes | 17:46 |
| ignas | and a read only attribute advisors that returns a list of all the advisors that are not None | 17:47 |
| aelkner | right now there are two things keeping a student from having two advisors | 17:47 |
| aelkner | one is in BasicPerson, the other the IStudent(person) adapter | 17:48 |
| aelkner | for the first case, not prob | 17:48 |
| aelkner | I could duplicate the property in my Demographics object | 17:48 |
| aelkner | to assign a second advisor to the person | 17:49 |
| ignas | yes | 17:49 |
| aelkner | the problem with IStudent(person) is that it limits the number of advisors to one | 17:49 |
| aelkner | and i don't know how much code depends on that being the case | 17:50 |
| ignas | none actually | 17:50 |
| ignas | advisor is more of an inverse relationship | 17:50 |
| ignas | student can see who is his advisor and that's it | 17:50 |
| ignas | while advisor can access persons information | 17:51 |
| ignas | or his grades | 17:51 |
| aelkner | BasicPerson has an advisor property that accesses the IStudent interface of the object | 17:53 |
| aelkner | and gets its advisor (singular) | 17:53 |
| aelkner | so there is code the depends on the fact that IStudent has only one advisor | 17:53 |
| ignas | yes, but you can "not use" the advisor property | 17:54 |
| ignas | you should look for all the places that use that property | 17:54 |
| ignas | and see whether they matter | 17:54 |
| ignas | and just add 2 advisors to the IDemographics | 17:55 |
| ignas | i think IStudent interface would just return the first advisor | 17:55 |
| aelkner | it asserts that there is only one | 17:55 |
| aelkner | so having a second will cause a crash | 17:55 |
| ignas | then you'll have to make sure no one uses advisor property | 17:56 |
| ignas | on the IBasicPerson | 17:56 |
| ignas | maybe even rewrite it so | 17:56 |
| ignas | that IBasicPerson would not have advisor in it | 17:56 |
| ignas | rewrite BasicPerson Add and Edit Views | 17:56 |
| ignas | to pull that property from IStudent interface | 17:57 |
| ignas | (original views) | 17:57 |
| aelkner | you're talking changing BasicPerson, right? | 17:57 |
| ignas | refactoring, yes | 17:57 |
| aelkner | and its views | 17:57 |
| ignas | a little bit | 17:57 |
| ignas | to make 1 "advisor" a property of the IStudent interface | 17:57 |
| ignas | rather than a property of IBasicPerson | 17:57 |
| ignas | property in a non-python sense | 17:58 |
| aelkner | advisor is already a propety of IStudent | 17:58 |
| ignas | property in a non-python sense | 17:58 |
| aelkner | how so? | 17:58 |
| ignas | "attribute" | 17:58 |
| ignas | to make IStudent interface the part that only allows 1 advisor | 17:58 |
| ignas | instead of having it ingrained in the BasicPerson code | 17:59 |
| aelkner | but if IStudent allows only one advisor, it won't allow me to have two | 17:59 |
| ignas | you will have to make sure that IStudent is only used in the Original Add/Edit person view | 18:00 |
| ignas | as you are overriding it - no one will be using IStudent in your system | 18:00 |
| ignas | because - well you can't make the system not use BasicPerson ;) but you can make your system not use IStudent | 18:01 |
| ignas | so IStudent will be a part of the presentation rather than the data model | 18:01 |
| aelkner | not sure what you mean, but i want to point out a few things | 18:02 |
| aelkner | advisor is in the schema for BasicPerson | 18:02 |
| aelkner | but my edit view could omit it | 18:02 |
| aelkner | and I could add two advisors to the Demographics | 18:02 |
| aelkner | and that in turn could avoid using IStudent | 18:03 |
| ignas | that is one way to do it, yes | 18:03 |
| aelkner | and access the relationships directly | 18:03 |
| ignas | it still leaves the attribute "advisor" on BasicPerson though | 18:03 |
| ignas | and that attribute is broken if there is more than 1 advisor | 18:04 |
| ignas | but if you are feeling safe enough to do it that way | 18:04 |
| ignas | it is an option too | 18:04 |
| aelkner | is there anyone in the field depending on the presence of the advisor attribute? | 18:05 |
| aelkner | lyceum for instance? | 18:05 |
| aelkner | remember that the edit view does depend on it | 18:06 |
| aelkner | so if anyone at lyceum uses that edit view to change the advisor | 18:06 |
| ignas | lyceum only uses basicperson through it's add/edit views so | 18:06 |
| aelkner | then I can't just remove it | 18:06 |
| ignas | so if you will update add/edit views | 18:06 |
| ignas | appropriatelly | 18:06 |
| ignas | and use IStudent.advisor as one of the schema fields | 18:07 |
| ignas | all the tests should pass | 18:07 |
| ignas | and everything should work | 18:07 |
| aelkner | IStudent.advisor is a read-only property at the moment | 18:08 |
| aelkner | it | 18:09 |
| aelkner | it's the advisor property in BasicPerson that handles adding, removing the advisor | 18:09 |
| ignas | you can change all that ;) | 18:09 |
| ignas | as long as you'll make sure everything keeps working | 18:10 |
| ignas | you can even add some functional tests in basicperson that check whether advisor still works properly | 18:10 |
| aelkner | so you're suggesting the following steps: | 18:11 |
| aelkner | 1) remove advisor from BasicPerson interace and implementation | 18:11 |
| aelkner | 2) change BasicPerson add and edit views to use IStudent.advisor | 18:12 |
| aelkner | 3) make IStudent.advisor a write as well as read property | 18:12 |
| aelkner | 4) add functional tests to make sure the views still work | 18:12 |
| ignas | 4) -> 0) | 18:13 |
| ignas | 4) make sure IStudent is not used anywhere else except for Add/Edit Person views | 18:13 |
| aelkner | IStudent is only used in BasicPerson | 18:14 |
| aelkner | for the advisor property | 18:14 |
| aelkner | this i already grepped | 18:15 |
| ignas | good | 18:15 |
| aelkner | and the tests of course | 18:15 |
| aelkner | so i think i know what i need to do now | 18:16 |
| ignas | good | 18:17 |
| ignas | the whole meaning of this thing is to make "advisor" property/relationship external to the BasicPerson instead of it being internal | 18:18 |
| ignas | so people who will want only 1 advisor will always be able to adapt to IStudent without blocking others from having multiple advisors | 18:19 |
| aelkner | but after i make the change to remove the advisor property from BasicPerson | 18:20 |
| aelkner | the list of people whoo only want one advisor will be zero, right? | 18:20 |
| aelkner | for the moment at least | 18:21 |
| ignas | nope, all the people who will not be overriding Add/Edit person views | 18:21 |
| ignas | will still get 1 advisor | 18:21 |
| ignas | so it will be a *default* instead of a *limitation* | 18:22 |
| aelkner | oh, right, because my change to IStudent and its use in the add/edit views will make that the default behaviour | 18:22 |
| ignas | yes | 18:22 |
| aelkner | ok, sounds good | 18:22 |
| Lumiere | bbl | 18:34 |
| *** Lumiere has quit IRC | 18:35 | |
| *** th1a_ has joined #schooltool | 18:53 | |
| *** Evannn has joined #schooltool | 19:08 | |
| *** alga has joined #SchoolTool | 19:51 | |
| Evannn | hey, is anyone active in here? | 19:53 |
| Evannn | i had a quick question and was hoping someone in here might have the answer. | 19:53 |
| ignas | yes | 19:54 |
| Evannn | it's regarding tinyzis, specifically.. is that okay? | 19:55 |
| *** wbrady has joined #schooltool | 19:56 | |
| Evannn | (i was searching google for a solution and the logs of this channel came up, from when one of the developers had my same problem trying to get tinyzis working.. but i couldn't find a solution.) | 19:56 |
| Evannn | basically, i'm having a problem with zope i think.. i'm getting ImportError: No Module named transaction | 19:59 |
| Evannn | when i run: python test.py | 20:00 |
| th1a | Hi Evannn. | 20:00 |
| th1a | Did you install the zodb? | 20:01 |
| Evannn | hey th1a... i'm pretty sure i installed the zodb. i was following the directions on http://www.uvasoftware.com/wiki/TinyZISInstall-RH | 20:02 |
| Evannn | is there a way to test if zodb was successfully installed? | 20:04 |
| th1a | Well, if you can't import transaction something's wrong. | 20:05 |
| th1a | And it rings a bell, but I'm having trouble remembering what causes this. | 20:05 |
| th1a | Do you have more than one Python installed? | 20:05 |
| Evannn | there should only be one installed.. is there any way to check for that? it had python installed when i got it. | 20:07 |
| Evannn | by got it, i mean.. got access to the box. it's a vps we're using as a test ZIS.. it's redhat with plesk stuff. | 20:08 |
| th1a | I can find several examples of me having this conversation over the past two years, but I can never remember the answer. | 20:09 |
| th1a | *sigh* | 20:09 |
| Evannn | haha yeah, that's what i noticed when looking through the logs... i was thinking it'd be a good idea to come here and get a solution so anyone with this problem in the future will be able to find it quickly via google when these logs come up. | 20:10 |
| th1a | Can you just start an interactive session in Python and see what happens if you do "import transaction" | 20:11 |
| Evannn | same exact error | 20:12 |
| Evannn | most recent call is: File "<stdin>", line 1, in ? | 20:12 |
| th1a | Yes. | 20:15 |
| Evannn | well, that wasn't a question actually.. the question mark was part of the response. | 20:16 |
| th1a | I know ;-) | 20:16 |
| Evannn | haha alright. | 20:16 |
| th1a | otoh, I'm drawing a complete freaking blank on this. | 20:16 |
| th1a | can you import zodb? | 20:17 |
| Evannn | "import zodb" at the python prompt? | 20:17 |
| th1a | Yeah. | 20:17 |
| Evannn | ahh... | 20:17 |
| Evannn | no module named zodb | 20:17 |
| Evannn | must not have installed right. | 20:18 |
| Evannn | is zodb in yum? | 20:18 |
| th1a | Probably not. I'm not sure. | 20:19 |
| th1a | Looks like not. | 20:19 |
| th1a | Perhaps try the tarball here: http://pypi.python.org/pypi/ZODB3 | 20:19 |
| Evannn | i was just about to wget that. | 20:20 |
| *** wbrady has quit IRC | 20:20 | |
| Evannn | im assuming i just run setup.py? | 20:22 |
| th1a | yes.. python setup.py install (as root) | 20:24 |
| Evannn | oops.. need gcc | 20:25 |
| th1a | Hmmm... | 20:26 |
| th1a | What happened when you did easy_install the first time? | 20:26 |
| *** Lumiere has joined #schooltool | 20:26 | |
| Evannn | src/BTrees/MergeTemplate.c:88: error: expected â=â, â,â, â;â, âasmâ or â__attribute__â before â*â token | 20:26 |
| Evannn | lots of errors like that | 20:26 |
| Evannn | same thing actually, with the ZODB3 setup.py | 20:27 |
| Evannn | at the end it gives me: error: command 'gcc' failed with exit status 1 | 20:27 |
| th1a | Oh... you got errors for easy_install? | 20:27 |
| Evannn | but i have gcc installed. | 20:27 |
| th1a | Perhaps you need python-devel. | 20:27 |
| Evannn | doh! | 20:28 |
| Evannn | i totally read that in the docs too... i must have just assumed i had it. stupid me. | 20:28 |
| * th1a won't mention how many times he's done this. | 20:29 | |
| Evannn | haha. | 20:29 |
| Evannn | looks like i need lxml now. | 20:30 |
| th1a | Probably the same issue. | 20:31 |
| Evannn | yeah... any particular version that plays nice with tinyzis? there seems to be quite a few.. http://pypi.python.org/pypi/lxml/ | 20:31 |
| th1a | It shouldn't matter. | 20:33 |
| Evannn | src/lxml/etree_defs.h:44:31: error: libxml/xmlversion.h: No such file or directory | 20:37 |
| Evannn | i get that now followed by 1000+ errors when running lxml's setup.py | 20:37 |
| Evannn | (i have libxml2 installed.. i double checked.) | 20:37 |
| Evannn | seems to be a path issue.. | 20:38 |
| mgedmin | no, you need libxml-dev | 20:39 |
| Evannn | ohh. | 20:39 |
| Evannn | hmm i did yum install libxml2-devel ... seemed to install find but i'm still getting the same errors. | 20:41 |
| Evannn | (with "error: command 'gcc' failed with exit status 1" at the very end) | 20:41 |
| Evannn | i have... /usr/include/libxml2/libxml/xmlversion.h | 20:45 |
| Lumiere | I would check yum for python-lxml | 20:46 |
| Lumiere | the more you can pull from repository the better imo | 20:46 |
| Evannn | i agree.. unfortunately it doesn't have lxml :( | 20:47 |
| Evannn | perhaps this will help | 20:52 |
| Evannn | http://dag.wieers.com/rpm/packages/python-lxml/ | 20:52 |
| *** alga has quit IRC | 20:55 | |
| Evannn | alright.. i got python-lxml installed... now the tinyzis test.py gets a lot further, but still throws back some errors... "registered new agen RamseySIS..." | 21:02 |
| th1a | Those aren't errors. | 21:02 |
| Evannn | the first error i get is error: (98, 'Address already in use') | 21:02 |
| Evannn | ohh | 21:03 |
| th1a | Most of them are just logging to the console. | 21:03 |
| th1a | So you know something's happening. | 21:03 |
| th1a | That can be turned off in the configuration. | 21:03 |
| Evannn | ahh, alright. | 21:03 |
| Evannn | so is it safe to run the tinyzis setup.py now? | 21:03 |
| Evannn | or do i even need to run that? can i just run the tz-run.py right out of the directory? | 21:05 |
| th1a | You can just run it from there. | 21:06 |
| Evannn | i still get socket.error: (98, 'Address already in use') | 21:07 |
| Evannn | does that mean the port is in use or something? | 21:07 |
| th1a | Yes. | 21:07 |
| Lumiere | which may mean that it was already started | 21:07 |
| Evannn | oohhh.. duh. stupid plesk. | 21:08 |
| Lumiere | ah | 21:08 |
| Evannn | naw, it's plesk. | 21:08 |
| Lumiere | I hate all those CPs | 21:08 |
| Lumiere | I run webmin occasionally for setting up samba | 21:08 |
| th1a | You should be able to change that in a config file. | 21:08 |
| Evannn | ha, same here. | 21:08 |
| Lumiere | (it does a far better job then I can) | 21:08 |
| Evannn | i just installed webmin on my new dedicated server i got last night. | 21:08 |
| Lumiere | but even then I have it set not to start | 21:08 |
| Lumiere | I start it manually then connect to it... do what I need then cut it off again | 21:09 |
| Evannn | unfortunately this vps i'm on now belongs to work and it has sites all set up with plesk so i can't really touch it. | 21:09 |
| Evannn | you only run samba when you need it? | 21:09 |
| Evannn | or webmin | 21:09 |
| Lumiere | webmin | 21:10 |
| Lumiere | and I only really use webmin to configure samba | 21:11 |
| Evannn | ah, yeah okay... all this tinyzis stuff is slowing my brain down. | 21:11 |
| Evannn | sweet! tinyzis is up and running! thank you so much everyone. | 21:12 |
| Evannn | i really appriciate the help. | 21:12 |
| th1a | No problem. | 21:17 |
| *** mgedmin has quit IRC | 21:17 | |
| *** ignas has quit IRC | 21:17 | |
| *** Ninno has joined #schooltool | 22:44 | |
| *** test123_ has quit IRC | 23:22 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!