| *** menesis has joined #schooltool | 01:45 | |
| *** menesis has quit IRC | 02:37 | |
| *** aelkner has quit IRC | 08:48 | |
| *** aelkner has joined #schooltool | 08:54 | |
| *** menesis has joined #schooltool | 09:39 | |
| *** yvl has joined #schooltool | 10:19 | |
| *** mgedmin has joined #schooltool | 10:26 | |
| mgedmin | menesis, does schooltool still have buildbot slaves on muskatas? | 10:27 |
|---|---|---|
| mgedmin | if so, some maintenance may be needed to recreate all virtualenvs | 10:27 |
| mgedmin | because they tend to break after a distro python2.6 package upgrade | 10:27 |
| mgedmin | with errors like ImportError: cannot import name urandom | 10:27 |
| mgedmin | anyway, if you get those errors, now you know why | 10:36 |
| *** mgedmin has left #schooltool | 10:36 | |
| menesis | I have fixed that already | 10:51 |
| *** aks has joined #schooltool | 13:05 | |
| *** aks has quit IRC | 14:32 | |
| *** replaceafill has joined #schooltool | 16:21 | |
| *** th1a has joined #schooltool | 16:26 | |
| th1a | hi yvl, menesis, aelkner, replaceafill. | 16:31 |
| yvl | good morning :) | 16:31 |
| replaceafill | good morning/afternoon | 16:31 |
| aelkner | morning | 16:31 |
| th1a | hm... menesis is who we really need today... | 16:33 |
| th1a | yvl: Know where he is? | 16:35 |
| yvl | nope | 16:35 |
| * th1a sighs. | 16:35 | |
| yvl | seems like he worked from home | 16:35 |
| th1a | Sure. | 16:36 |
| th1a | You're not his keeper. ;-) | 16:36 |
| th1a | aelkner, would you like to start? | 16:36 |
| aelkner | ok, i fixed the gradebook average when minimum is set bug | 16:37 |
| aelkner | there are tests to prove it works now | 16:37 |
| aelkner | also, i figured out how to fix the pesky broken gradebook tests die to evolve1 | 16:38 |
| aelkner | due to | 16:38 |
| aelkner | yvl, did you see what i did there? | 16:38 |
| yvl | no | 16:38 |
| aelkner | didn't my note to the developer's list get through? | 16:38 |
| replaceafill | it did | 16:39 |
| aelkner | anyway, here's the revision: http://bazaar.launchpad.net/~schooltool-owners/schooltool.gradebook/flourish/revision/502 | 16:39 |
| aelkner | there's no reason to add the scoresystems utilities to the container when we're not also creating it at the same time | 16:39 |
| yvl | aelkner, sorry, got lost in all the mail from LP | 16:39 |
| aelkner | the thing was that tests fire AppInit before database init | 16:40 |
| aelkner | start-server does it the other way around | 16:40 |
| aelkner | anyway, the tests pass now, and i think the evolve1 change is safe | 16:41 |
| yvl | why the utilities were there in the first place? | 16:41 |
| th1a | There was a bit of an email flood. ;-) | 16:41 |
| yvl | in the tests | 16:41 |
| yvl | evolve 1 is supposed to find the registered scoresystem utilities, if there are any, and *move* them to the container | 16:42 |
| yvl | fresh databases should not have those utilities | 16:42 |
| yvl | why there were utilities found in tests? | 16:42 |
| yvl | is it a faulty (outdated) setup? | 16:42 |
| yvl | or is there a more serious problem? | 16:43 |
| yvl | feel free to keep it as a workaround, but I'd like those questions answered before merging that fix to trunk | 16:43 |
| yvl | and I'm not convinced it's a right fix, sorry :/ | 16:44 |
| th1a | Well, perhaps yvl can answer them soonish. | 16:44 |
| yvl | we can of course sweep it under the rug, merge to trunk, and assume everything is fine for now | 16:45 |
| aelkner | as i said, AppInit got fired before database init | 16:45 |
| yvl | yes | 16:45 |
| yvl | that is intentional | 16:45 |
| aelkner | AppIiit adds three scoresystems if it finds and empty container | 16:45 |
| yvl | yes! | 16:46 |
| yvl | that is also intentional | 16:46 |
| aelkner | then evolve1 runs, and it adds the same scoresstems again! | 16:46 |
| yvl | yes! | 16:46 |
| yvl | but! | 16:46 |
| yvl | only if it finds utilities! | 16:46 |
| yvl | should they be there? | 16:46 |
| yvl | no. | 16:46 |
| yvl | anyway, it's just a faulty test set up probably | 16:48 |
| yvl | not a very big deal | 16:48 |
| aelkner | look at test_evolve1 is you want the answer | 16:48 |
| aelkner | it adds the utiltiies | 16:48 |
| aelkner | the faulty part is that AppInit runs and adds the scoresystems container first | 16:49 |
| yvl | ??? | 16:50 |
| *** menesis has quit IRC | 16:51 | |
| aelkner | yvl, look, i fixed the broken tests, i didn't even break them in the first place | 16:51 |
| yvl | and I said that is a hack not a fix | 16:51 |
| aelkner | if you have a better fix, by all means, i'm all for it | 16:51 |
| yvl | what is so difficult to understand? | 16:51 |
| yvl | I thought th1a mentioned that I will have to fix those eventually | 16:51 |
| yvl | and I said that the fix is just hiding the problem, not fixing it | 16:52 |
| yvl | and I'm reluctant to merge that fix because it should expose broken databases | 16:52 |
| th1a | It is just a little pissing contest. | 16:52 |
| th1a | I'm not taking sides. | 16:53 |
| yvl | :) | 16:53 |
| th1a | Can we move on? | 16:53 |
| yvl | sure | 16:53 |
| aelkner | i'm not in any contest, i made my report, that is all | 16:53 |
| th1a | OK, so aelkner and I are going to do a little review of imports the next few days, right aelkner? | 16:54 |
| aelkner | sure | 16:54 |
| * th1a moves post-its around. | 16:55 | |
| th1a | OK, lets go over that later today. | 16:55 |
| th1a | 3:00? | 16:56 |
| th1a | 2:00? | 16:56 |
| th1a | When do you get up? | 16:56 |
| aelkner | :) | 16:56 |
| aelkner | well, i'm up now | 16:56 |
| aelkner | but we could meet in the afternoon | 16:56 |
| aelkner | name the time | 16:56 |
| th1a | 2:00. | 16:57 |
| aelkner | ok | 16:57 |
| th1a | Thanks aelkner. | 16:57 |
| th1a | replaceafill? | 16:57 |
| replaceafill | ok | 16:58 |
| replaceafill | i finished the projects gradebook update | 16:58 |
| replaceafill | you can see it at http://69.164.203.135:6660/calendar | 16:58 |
| replaceafill | eunita.winkey@apsva.us:teacher | 16:58 |
| replaceafill | http://69.164.203.135:6660/schoolyears/2012-2013/SY/sections/13106-45-1-1/gradebook-projects | 16:58 |
| replaceafill | adding skills is a little bit hacky | 16:59 |
| replaceafill | since you can't really search for a skillset/skill in the beginning | 16:59 |
| replaceafill | i also finished the update equivalent skills functionality | 16:59 |
| replaceafill | so if you add a skill that's also in the section skills gradebook | 16:59 |
| replaceafill | the project updates that | 17:00 |
| th1a | Hacky in the sense that we should at least get rid of the checkboxes that lie? | 17:00 |
| replaceafill | yes | 17:00 |
| replaceafill | as we said last week, it's the same filter we use in /nodes | 17:00 |
| replaceafill | so we have that problem there too | 17:00 |
| replaceafill | i don't know how to "detect" the right level :( | 17:01 |
| replaceafill | (to stop showing) | 17:01 |
| replaceafill | is it like "show only layers with nodes"? | 17:01 |
| * th1a doesn't know if searching for everything isn't almost easier than removing the boxes. | 17:01 | |
| th1a | aelkner could do that right now actually. | 17:01 |
| th1a | Make these searches find skillsets and skills. | 17:01 |
| replaceafill | during the weekend i checked this one: https://bugs.launchpad.net/schooltool/+bug/1062868 | 17:02 |
| replaceafill | so i thought i could also merge the style change to Done links i have in a branch | 17:03 |
| replaceafill | i'll do that today | 17:03 |
| th1a | Oh, God yes. | 17:03 |
| th1a | That's definitely one of those "You haven't merged WHAT?" bugs. | 17:03 |
| replaceafill | th1a, i realized the task involves moving some Done links out of view page templates | 17:04 |
| replaceafill | now that we have the page content viewlet manager approach | 17:04 |
| th1a | Does this include adding them above and below lists? | 17:04 |
| replaceafill | i think we should always stick to it | 17:04 |
| replaceafill | and leave content_templates out when possible | 17:05 |
| replaceafill | th1a, correct | 17:05 |
| replaceafill | when using viewlets that's really easy | 17:05 |
| th1a | OK, yes, clean that stuff up. | 17:05 |
| replaceafill | we also have the after="*" functionality with viewlets | 17:05 |
| replaceafill | so we can put the Done at the bottom | 17:06 |
| replaceafill | etc | 17:06 |
| replaceafill | kk, i'll work on that today | 17:06 |
| replaceafill | and then i'll move back to completion reports | 17:06 |
| replaceafill | now that we have the section skills fix | 17:06 |
| replaceafill | dwelsh is also requesting help on updating the social skills server | 17:06 |
| replaceafill | i'll fwd you some emails he just sent | 17:07 |
| * replaceafill done | 17:07 | |
| th1a | Yes... don't let Welsh bypass me completely. ;-) | 17:08 |
| th1a | Thanks replaceafill. | 17:08 |
| th1a | yvl? | 17:08 |
| yvl | I spent the day looking over the assigned bugs | 17:08 |
| yvl | and how to fix them :) | 17:09 |
| yvl | I have two things that are not clear | 17:09 |
| yvl | one is https://bugs.launchpad.net/schooltool/+bug/1047266 | 17:09 |
| yvl | it's a good question what's the best way to solve it | 17:09 |
| *** menesis has joined #schooltool | 17:10 | |
| yvl | another is unfinished business with school logo | 17:10 |
| th1a | Probably an "Assign to all linked sections" option. | 17:10 |
| yvl | hmm | 17:10 |
| th1a | Definitely move the title above the logo on the login page. | 17:10 |
| th1a | That's driving me crazy. | 17:10 |
| yvl | hmm | 17:11 |
| yvl | the problem with that is that css is not that friendly there :) | 17:11 |
| yvl | so it's not 2 minute change | 17:11 |
| yvl | but I will move that | 17:11 |
| yvl | I realized just now it's a 10 minute change | 17:12 |
| yvl | as far as timetable assignment goes, you'd like a button "apply and assign to linked sections" ? | 17:13 |
| yvl | that simply puts copies in linked sections with start/end dates matching the sections' terms | 17:13 |
| th1a | Probably more like "Apply schedule to all terms for this section." | 17:13 |
| yvl | you can assign start/end dates | 17:14 |
| yvl | and they can be outside of a section | 17:14 |
| th1a | Apply schedule to all terms (linked sections). | 17:14 |
| yvl | so | 17:15 |
| yvl | if the user enters a start/end date | 17:15 |
| yvl | that spans two terms | 17:15 |
| yvl | since he can do that | 17:15 |
| yvl | and then clicks next | 17:15 |
| yvl | and hits a button to apply the change to all four linked terms that year | 17:15 |
| yvl | what should happen? | 17:15 |
| th1a | I guess if you're using start and end dates you don't need the buttons. | 17:16 |
| yvl | Maybe a separate page to "add for all linked sections" then? | 17:16 |
| yvl | as in "add timetable" link | 17:16 |
| th1a | You'd need the initial form to cue the user to the span of linked sections. | 17:17 |
| th1a | I thought you were saying that if you have the date span you could just infer that if it spanned multiple terms it would have to be applied that way. | 17:17 |
| yvl | that too | 17:17 |
| yvl | but now I think that it would be best | 17:17 |
| yvl | if we just had a separate form: | 17:18 |
| yvl | one for assigning the timetable | 17:18 |
| yvl | another for assigning timetable for all linked sections | 17:18 |
| yvl | the latter would not allow you to pick dates, | 17:18 |
| yvl | but would apply to all linked sections using their start/end dates | 17:19 |
| yvl | I think that would work | 17:19 |
| aelkner | that would be easier for the user, too | 17:20 |
| aelkner | not having to put dates in that one already put in previously is a friendly ting to do | 17:20 |
| th1a | I just don't like the user having to figure out which link is the right one for him. | 17:21 |
| th1a | If both could be on the same page, that might be best. | 17:21 |
| yvl | then we can use the magic of third-nav | 17:21 |
| th1a | That would be ok. | 17:21 |
| yvl | kk | 17:21 |
| th1a | I'd make whole terms the default. | 17:21 |
| aelkner | yes | 17:22 |
| yvl | +1 | 17:22 |
| yvl | thanks guys | 17:22 |
| yvl | so another question is a bit more techy | 17:22 |
| yvl | I didn't resize the school logo on upload | 17:23 |
| yvl | I think we need different resolutions anyway | 17:23 |
| yvl | (high res for reports) | 17:23 |
| yvl | but serving the big one to users will kill the server | 17:23 |
| yvl | well, at least for some people ;) | 17:24 |
| yvl | so I | 17:24 |
| yvl | so I'd like to do thumbnails, | 17:24 |
| yvl | in a somewhat similar way google plus does | 17:24 |
| th1a | Ah, yes. We do need to make sure they'll look nice printed. | 17:24 |
| yvl | right | 17:24 |
| yvl | so I thought about views that automatically resize the photo | 17:25 |
| yvl | like manager/photo/w200 | 17:25 |
| yvl | (width 200, height auto) | 17:25 |
| yvl | manager/photo/320x240 | 17:25 |
| yvl | manager/photo/h400 | 17:25 |
| yvl | (height 400, width auto) | 17:26 |
| yvl | but the question remains, when to cache? | 17:26 |
| yvl | i.e. when to save to HDD | 17:27 |
| yvl | (for faster access) | 17:27 |
| aelkner | is that a jquery question? | 17:27 |
| yvl | it's a... well architecture question | 17:27 |
| aelkner | or is it a http header that controls browser activity? | 17:27 |
| yvl | no | 17:28 |
| yvl | say, we upload a photo | 17:28 |
| yvl | and we want to store some thumbnails in HDD | 17:28 |
| yvl | where would you like to specify which thumbnails to store? | 17:28 |
| yvl | should they be specified in interfaces.py ? | 17:29 |
| th1a | The high res ones only need to be requested by reports, so they can be a little more expensive than just a web request. | 17:29 |
| yvl | yes | 17:29 |
| yvl | but when doing low res, we have to specify the resolution(s) *somewhere* | 17:30 |
| yvl | options: | 17:30 |
| yvl | 1) interfaces.py | 17:30 |
| yvl | 2) in zcml | 17:30 |
| yvl | 3) in tal | 17:31 |
| yvl | 4) on-demand - always store whatever resolution somebody requested | 17:31 |
| aelkner | 4 sounds practical | 17:32 |
| yvl | yes | 17:32 |
| yvl | but it | 17:32 |
| yvl | it's also dangerous | 17:32 |
| yvl | if somebody runs a script to look at person's photo with 100000 different resolutions... | 17:32 |
| yvl | bye bye free space on HDD | 17:32 |
| aelkner | attack vulnerability | 17:32 |
| yvl | right | 17:32 |
| aelkner | there cold be an algrith that protects against that | 17:33 |
| aelkner | i mean, the file names can show the resolution | 17:33 |
| yvl | yes | 17:33 |
| aelkner | the method that thumbnails can check the hdd | 17:33 |
| yvl | we also could specify "max cache size" in MB | 17:33 |
| yvl | or "max amount of thumbnails" | 17:34 |
| aelkner | i think, the savings in speed comes when the image is already there | 17:34 |
| aelkner | so simply doing an expensive check for too many files of the given type | 17:34 |
| th1a | The logo and person images are different. | 17:34 |
| aelkner | would not be haprmful | 17:34 |
| th1a | The best person image has to be good enough for a passable id card image. | 17:35 |
| th1a | The best report card letterhead logo image needs to be pretty sharp with potentially small lines. | 17:35 |
| th1a | But all we really need is the regular web resolution and the good one. | 17:36 |
| th1a | We don't need to keep high quality person images, and we don't need a bunch of logo images. | 17:36 |
| yvl | well | 17:36 |
| yvl | if you want a good image on the web | 17:36 |
| yvl | it's best to scale down from a very high resolution than nearly-matching resolution | 17:37 |
| yvl | scaling from 140 to 144 pixels is... not the best way to go | 17:37 |
| yvl | but in general I do agree with th1a | 17:38 |
| yvl | ok, that helped! | 17:38 |
| yvl | thanks | 17:38 |
| th1a | OK. | 17:40 |
| th1a | Yes, we don't need to switch to a generalized rescale on demand system. | 17:40 |
| yvl | umm | 17:40 |
| yvl | I will do a semi-generalized resacle on-demand | 17:41 |
| yvl | simply because it's easier | 17:41 |
| th1a | OK. We don't need anything comlicated. ;-) | 17:41 |
| yvl | right! | 17:41 |
| aelkner | yes! | 17:41 |
| yvl | ok, /me done | 17:42 |
| th1a | Thanks yvl. | 17:42 |
| th1a | menesis? | 17:42 |
| menesis | hi | 17:42 |
| menesis | sorry for missing the Friday meeting | 17:43 |
| menesis | on Friday and today | 17:43 |
| menesis | I am making the releases | 17:43 |
| menesis | happy to see many last-minute bugfixes | 17:44 |
| menesis | hope that you have all of your bugfixes merged? | 17:44 |
| th1a | replaceafill unfortunately reminded me that we hadn't merged and finished one of the most important. | 17:44 |
| th1a | The universal Done button improvements. | 17:45 |
| th1a | I'd say that's important enough to wait for. | 17:45 |
| replaceafill | i'll have that "Done" today | 17:45 |
| th1a | :-D | 17:45 |
| replaceafill | :D | 17:45 |
| yvl | :DDD | 17:45 |
| yvl | +! | 17:45 |
| menesis | :) | 17:45 |
| th1a | I think that's thebig one. | 17:46 |
| menesis | will gradebook depend on changes in core? | 17:46 |
| menesis | for the Done button? | 17:46 |
| replaceafill | i think the gradebook has some Done links | 17:47 |
| * replaceafill goes to check | 17:47 | |
| replaceafill | yes | 17:47 |
| th1a | It is pretty universal. | 17:48 |
| replaceafill | menesis, i'd say yes, it will depend on a change to the core css | 17:48 |
| replaceafill | (i mean the gradebook will) :) | 17:48 |
| menesis | :-/ | 17:49 |
| menesis | okay, will see | 17:49 |
| menesis | if only css then little problem | 17:49 |
| menesis | I am halway dealing with translations and updating changelogs | 17:50 |
| th1a | I will need the release notes to make an announcement. | 17:50 |
| menesis | will wait for those Done changes before I hit the "release" button | 17:50 |
| th1a | OK. | 17:51 |
| th1a | Thanks. | 17:51 |
| menesis | buildbot is green again, thanks aelkner for gradebook evolve1 fix | 17:52 |
| th1a | Let's not get into that. | 17:52 |
| yvl | :D | 17:52 |
| aelkner | it's ok, th1a, i agree that the test setup needs to improve | 17:52 |
| aelkner | i just thought it was a priority to get tests to pass | 17:52 |
| th1a | But perhaps we can leave that green until yvl does the real fix. | 17:52 |
| yvl | yes | 17:52 |
| aelkner | exactly | 17:53 |
| th1a | Lovely. | 17:53 |
| th1a | OK. I actually started writing CanDo docs last week. Starting is the hardest part. | 17:53 |
| th1a | Also: | 17:54 |
| th1a | * invoices! | 17:54 |
| th1a | Release notes, menesis. | 17:55 |
| th1a | OK. | 17:55 |
| th1a | See you Wednesday, gentlemen. | 17:55 |
| menesis | yvl, make it so that evolution scripts do not run in test layer setup? | 17:55 |
| replaceafill | th1a, will send mine tonight | 17:55 |
| menesis | thank you for all your fixes again | 17:55 |
| yvl | well, install generations are run on fresh databases in real life | 17:56 |
| yvl | so I'd rather have them on ftests also | 17:56 |
| menesis | install... true. | 17:56 |
| menesis | in unpredictable order | 17:56 |
| menesis | good | 17:57 |
| yvl | where's the gravel? | 17:58 |
| * th1a drops the bag of gravel. | 17:58 | |
| yvl | :D | 17:58 |
| replaceafill | thanks everybody | 17:58 |
| * yvl may have mild OCD | 17:59 | |
| aelkner | thank guys, cya wed | 17:59 |
| yvl | thanks guys | 17:59 |
| *** yvl has quit IRC | 18:04 | |
| replaceafill | is it pypi down? i can't bootstrap a fresh branch :( | 18:27 |
| *** veloutin has quit IRC | 18:29 | |
| *** veloutin has joined #schooltool | 18:29 | |
| th1a | aelkner, replaceafill: I forwarded a question from Lee. | 19:09 |
| replaceafill | th1a, iirc, the skills don't have an order attribute or something and we save them as they're spit by their api | 19:11 |
| th1a | What do we use as labels? | 19:12 |
| replaceafill | consecutive numbers | 19:12 |
| replaceafill | 001, 002 | 19:12 |
| th1a | Aren't they in order? | 19:12 |
| th1a | Or don't we sort by label? | 19:12 |
| replaceafill | i'm not sure if Lee is talking skills gradebook here | 19:13 |
| th1a | How are they sorted there? By label? | 19:13 |
| th1a | There is a "sort skills by label first" bug I filed. | 19:13 |
| * replaceafill is checking | 19:13 | |
| replaceafill | :( | 19:17 |
| replaceafill | .values() | 19:17 |
| replaceafill | they're not sorted | 19:18 |
| replaceafill | in the gradebook at least | 19:18 |
| th1a | Well, that's a bug. | 19:19 |
| th1a | I can have aelkner do those today. | 19:19 |
| replaceafill | kk | 19:19 |
| replaceafill | on a first glance i can't see anything "sorting" the skills in the gradebook | 19:19 |
| replaceafill | http://69.164.203.135:6660/skills | 19:21 |
| replaceafill | http://69.164.203.135:6660/skills/8088-2011-6 | 19:22 |
| replaceafill | ggrr damn time out! | 19:29 |
| replaceafill | how do i know what server is timing out!?!?! | 19:29 |
| *** aks has joined #schooltool | 19:36 | |
| *** aks has joined #schooltool | 19:36 | |
| replaceafill | menesis, you around? | 20:01 |
| *** aks has quit IRC | 20:02 | |
| menesis | replaceafill, I am | 20:15 |
| replaceafill | menesis, i'm having trouble downloading zc.buildout :( | 20:15 |
| replaceafill | when running bootstrap | 20:15 |
| replaceafill | http://pypi.python.org/simple/zc.buildout/ is timing out for me | 20:16 |
| replaceafill | using virtualenv | 20:16 |
| replaceafill | but i can access the page in my browser | 20:16 |
| replaceafill | any clue? | 20:16 |
| *** aks has joined #schooltool | 20:16 | |
| *** aks has joined #schooltool | 20:16 | |
| replaceafill | i thought maybe using the --download-base will let me use my egg cache or something, but i can't make it work :( | 20:18 |
| replaceafill | i don't know who to blame: virtualenv, setuptools, distribute, pypi, my IPS :D | 20:18 |
| replaceafill | ISP* | 20:19 |
| th1a | Vint Cerf | 20:19 |
| replaceafill | :D | 20:20 |
| replaceafill | hhmm | 20:25 |
| * replaceafill starts blaming his ISP | 20:25 | |
| replaceafill | bootstraping works in our VPS | 20:25 |
| * replaceafill goes to "borrow" some wireless ;) | 20:26 | |
| *** replaceafill has left #schooltool | 20:26 | |
| *** replaceafill has joined #schooltool | 20:31 | |
| replaceafill | it's my damn ISP >:( | 20:32 |
| replaceafill | at least i was able to bootstrap | 20:32 |
| replaceafill | i just don't have to touch my environment anymore | 20:32 |
| *** aks_ has joined #schooltool | 21:00 | |
| *** aks_ has joined #schooltool | 21:00 | |
| *** aks has quit IRC | 21:00 | |
| *** aks_ has quit IRC | 21:00 | |
| th1a | hi aelkner. | 21:03 |
| aelkner | hey th1a | 21:04 |
| th1a | OK, so we have a few things which came up. | 21:05 |
| aelkner | shoot | 21:05 |
| th1a | This sorting issue in CanDo, which is also I think a general problem. | 21:05 |
| th1a | You got an email. | 21:05 |
| th1a | It also goes with this: https://bugs.launchpad.net/cando/+bug/1049992 | 21:06 |
| th1a | Which should be applied everywhere skills are listed. | 21:06 |
| th1a | Or, perhaps they should also be sorted by parent node then label. | 21:07 |
| th1a | Sorted by parent node, then skillset then skill, all by label. | 21:07 |
| aelkner | well, the email is about get_verso.py and the xls being out of order | 21:12 |
| aelkner | if we are scanning or otherwise processing the data from their api in the wrong order | 21:13 |
| th1a | There is that. | 21:13 |
| aelkner | then the labels that we generate would be in the wrong order anyway | 21:13 |
| th1a | But also replaceafill doesn't think we're sorting them in the gradebook, and I know we aren't sorting them other places. | 21:13 |
| aelkner | so sorting in the app by label wouldn't solve that problem | 21:13 |
| th1a | So we probably have a problem with our belt AND our suspenders. | 21:13 |
| aelkner | but, yes, we could definitely make sure sorting is being done by label in the app | 21:13 |
| aelkner | :) | 21:13 |
| aelkner | i can definitely look into both issues | 21:14 |
| replaceafill | we use .values() for generating the list of activities, which you can't rely on | 21:14 |
| replaceafill | and the gradebook views methods don't sort the activities in any way | 21:14 |
| aelkner | .values() returns them in the order they are added | 21:15 |
| aelkner | if get_verso adds them in the right order, then we'd be ok there | 21:15 |
| replaceafill | in a sorted container that is true | 21:15 |
| aelkner | i was just saying that they values() returns in order added, not sorted | 21:15 |
| replaceafill | also true :) | 21:16 |
| aelkner | thing about get_verso is that it builds this huge dictionary as it queries their api | 21:16 |
| aelkner | and dictionaries don't have a reliable sort order, they go by the random hash | 21:17 |
| aelkner | so we would need to have a field in the objects that is sortable and use that | 21:17 |
| aelkner | i'd have to look over the code a bit | 21:17 |
| th1a | That sounds easy enough. | 21:17 |
| replaceafill | aelkner, if you're sorting by label, title you need the request | 21:18 |
| replaceafill | ie you need to do it in views | 21:19 |
| replaceafill | (for using collation) | 21:19 |
| replaceafill | is collation a word? :D | 21:19 |
| aelkner | yes :) | 21:19 |
| replaceafill | phew | 21:19 |
| aelkner | get_verso.py doesn't have a request, but we should be using the order that they deliver in the api anyway | 21:20 |
| replaceafill | do relationships work like dictionaries? | 21:21 |
| replaceafill | sorting-wise | 21:21 |
| aelkner | yes | 21:22 |
| aelkner | but the solution is to get the get_verso sort right | 21:22 |
| aelkner | the result would be to have labels that wold sort in the same order | 21:23 |
| aelkner | these labels are not anything that would be internationalization-sensitive | 21:23 |
| aelkner | they are things like 01 02 ... | 21:23 |
| aelkner | they would sort right in any country | 21:23 |
| replaceafill | but they're strings... | 21:24 |
| aelkner | the trick is to get them into the xls file in the right order | 21:24 |
| aelkner | strings of digits sort the same everywhere, right? | 21:24 |
| replaceafill | for aps case | 21:24 |
| replaceafill | not general | 21:24 |
| aelkner | well, i was just referring to get_verso | 21:25 |
| aelkner | but i agree that user-defined labels will be international-sensitive | 21:25 |
| aelkner | and the collator would need to work based on that | 21:25 |
| replaceafill | thanks | 21:25 |
| * replaceafill appreciate any effort on improving i18n :) | 21:25 | |
| aelkner | point taken :) | 21:26 |
| aelkner | th1a, anyway, i'll look into Lee's claim that the skills are out of order in the xls file | 21:34 |
| aelkner | and i'll let you know what i find when i do | 21:34 |
| th1a | But also fix the sorts in the app too. | 21:35 |
| aelkner | yes, that, too | 21:35 |
| th1a | Also... I'd like the search forms for the nodes to search skillsets and skills too, if that is practical. | 21:35 |
| aelkner | yeah, i saw that one, that would be a departure from the way we have always presented data | 21:46 |
| aelkner | mixed data types in container views has never been done | 21:46 |
| aelkner | that doesn't mean it's not possible, just strange, i'd have to look into what that would mean | 21:46 |
| th1a | Look into it. We either need to do that or just remove skillsets and skills from the the choices there, since you can't actually search for them at all there. | 21:47 |
| aelkner | you mean remove the checkboxes? | 21:47 |
| aelkner | by there, you mean the Search tab? | 21:48 |
| aelkner | when you say 'remove skillsets and skills from the the choices there' | 21:48 |
| aelkner | in the node container view, the Search tab, there is no skillsets or skills choices, so i don't understand | 21:49 |
| aelkner | there are checkboxes for layers, but that's it | 21:49 |
| aelkner | oh, assigning skillsets to courses, replaceafill could tell you about that one | 21:51 |
| aelkner | that's the accordion view, is that what you mean? | 21:51 |
| replaceafill | aelkner, http://69.164.203.135:6660/nodes | 21:51 |
| replaceafill | the filter used in that view | 21:52 |
| replaceafill | shows two layers Skillsets and Skills which don't have nodes | 21:52 |
| aelkner | has no mention of skillsets and skills, so what about it? | 21:52 |
| aelkner | oh, the layers called Skillsets and Skills | 21:52 |
| replaceafill | :) | 21:52 |
| aelkner | th1a, you know we created those two layers as part of the document view | 21:53 |
| aelkner | they are pretend layers to allow the user to rename SkillSet and Skill to DutyArea and Competency (or whatever) | 21:54 |
| aelkner | only in the document tree views | 21:54 |
| aelkner | there are actually no nodes that have those layers | 21:55 |
| aelkner | it's just a way for us to have the document tree view for adding a SkillSet be: | 21:55 |
| th1a | Well... I don't care why the view is lying. | 21:55 |
| aelkner | Adding DutyArea | 21:56 |
| th1a | It IS lying, right? | 21:56 |
| aelkner | i guess we can have the logic to exlude the two lowest layers on the hierarchy | 21:56 |
| aelkner | so they are not included | 21:56 |
| aelkner | or | 21:57 |
| th1a | Or we could include them in the search. | 21:57 |
| aelkner | i was just about to say :) | 21:57 |
| aelkner | like i said, it's weird, but possible in theory | 21:57 |
| aelkner | if we were to include skillsets and skills in the Search tab, would we want to re-remove the SkillSets tab? | 21:59 |
| aelkner | i mean, is there any reason to have it at that point? | 21:59 |
| th1a | Probably not. | 22:00 |
| aelkner | ok, i'll look into including skillsets and skills in the search | 22:01 |
| * replaceafill goes to get lunch, bb in ~45 | 22:05 | |
| replaceafill | th1a, zyt? | 23:42 |
| th1a | yes replaceafill. | 23:42 |
| replaceafill | quick question | 23:42 |
| replaceafill | do you want a Done button under EVERY listing table? | 23:42 |
| replaceafill | some of them are really short | 23:42 |
| replaceafill | like Documents in cando for example | 23:42 |
| th1a | The question would be do I want it OVER every table. | 23:43 |
| replaceafill | ah | 23:43 |
| th1a | The default for a short list would be under. | 23:43 |
| th1a | The answer is no. | 23:43 |
| replaceafill | kk | 23:43 |
| replaceafill | for longer tables, like /person, we'll use two right? | 23:43 |
| replaceafill | above and below | 23:43 |
| replaceafill | i mean /persons | 23:43 |
| replaceafill | hhhmm didn't we put one next to the filter buttons? | 23:44 |
| * replaceafill goes to see | 23:44 | |
| th1a | Yes. | 23:44 |
| replaceafill | kk | 23:45 |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!