| *** th1a has quit IRC | 00:16 | |
| *** replaceafill has quit IRC | 01:46 | |
| *** menesis has joined #schooltool | 11:28 | |
| *** alga has joined #schooltool | 11:37 | |
| *** menesis has quit IRC | 12:08 | |
| *** menesis has joined #schooltool | 12:53 | |
| *** yvl has quit IRC | 13:29 | |
| *** th1a has joined #schooltool | 15:01 | |
| *** th1a has quit IRC | 15:20 | |
| *** jelkner has joined #schooltool | 15:39 | |
| *** th1a has joined #schooltool | 16:05 | |
| *** yvl has joined #schooltool | 16:09 | |
| *** replaceafill has joined #schooltool | 16:19 | |
| *** jelkner has quit IRC | 16:29 | |
| th1a | hi replaceafill, aelkner_, yvl, menesis. | 16:32 |
|---|---|---|
| replaceafill | good morning/afternoon | 16:32 |
| menesis | hi | 16:32 |
| aelkner_ | morning | 16:32 |
| th1a | OK, let's discuss release plan B. | 16:32 |
| yvl | morning guys | 16:33 |
| th1a | I have a feeling we're not going to have flourish in Oneiric. | 16:33 |
| replaceafill | :( | 16:33 |
| th1a | And let me state from the beginning I'm not blaming menesis for this, and I wouldn't have done anything differently. | 16:33 |
| yvl | hmm | 16:34 |
| yvl | so plan B is to have a bleeding edge PPA | 16:34 |
| th1a | menesis: Just in terms of getting packages into Ubuntu, do you think it is still possible to get flourish into Oneiric? | 16:34 |
| yvl | and a stable version in Oneiric? | 16:34 |
| th1a | Plan B is to have a stable version in Oneiric and a stable version in the PPA, essentially. | 16:34 |
| th1a | And "release" in November. | 16:35 |
| menesis | I was going to write an email explaining why we don't want flourish in oneiric and how we work next | 16:35 |
| menesis | but as th1a said | 16:35 |
| menesis | we have a stable version in oneiric | 16:36 |
| menesis | it is there. | 16:36 |
| menesis | and we put our flourish package in the ppa | 16:37 |
| menesis | several frequent releases | 16:37 |
| th1a | One question I have now is how much work would it be to put flourish in a separate PPA for the last several Ubuntu releases? | 16:37 |
| th1a | That might make it a lot easier to get testing. | 16:37 |
| menesis | and polish until we are ready to call this 2.0 | 16:38 |
| menesis | and then we have it in the next LTS tested and translated | 16:38 |
| th1a | We're not going to sit around polishing it though. | 16:38 |
| th1a | It should go into Oneiric+1 universe in November. | 16:39 |
| menesis | of course | 16:39 |
| th1a | We'll release and roll out the new website & docs in November. | 16:40 |
| menesis | one of my points why a 2.0 is not ready is website and docs | 16:40 |
| menesis | but I think it is possible to do this by November | 16:41 |
| th1a | Yes. | 16:41 |
| menesis | to answer your earlier questions, | 16:41 |
| menesis | 1. no, it is way too late (1.5 months) to push such a big change to Oneiric. | 16:42 |
| menesis | 2. it is possible to put flourish packages in a separate PPA | 16:42 |
| menesis | at least to natty, it is easy | 16:43 |
| menesis | don't know what's missing from lucid, but I can put additional zope packages in the PPA as well | 16:44 |
| menesis | the /dev ppa | 16:44 |
| th1a | OK. I'd like to have flourish for natty available in a PPA by Oct. 13. | 16:44 |
| th1a | And then in November, our website will be encouraging people to use the PPA (flourish) version. | 16:45 |
| menesis | remember that servers running schooltool are still on lucid, and they get everything from the ppa | 16:46 |
| menesis | even with natty, everything is in ubuntu. but updates are only in the ppa | 16:46 |
| th1a | OK. | 16:47 |
| th1a | Any other questions about this right now. | 16:49 |
| th1a | ? | 16:49 |
| menesis | not from me. I am ready for plan B. | 16:50 |
| replaceafill | just want to say, and maybe i'm not aware of the real issues, but for some reason this makes me feel sad :( | 16:50 |
| replaceafill | i won't say more though :) | 16:50 |
| th1a | One thing... if we're six weeks too late on getting this in, perhaps you could have told me that six weeks ago. ;-) | 16:50 |
| th1a | Realistically, we should probably always release a month *after* Ubuntu to ensure getting in the next one. | 16:51 |
| th1a | The thing is, schools weren't really going to be upgrading to Oneiric in the middle of the school year anyhow. | 16:52 |
| menesis | true | 16:53 |
| th1a | So PRACTICALLY it isn't a big deal. | 16:53 |
| menesis | I lost hope only two weeks ago. not six. | 16:53 |
| th1a | I need to talk to Critical Links now actually. That's probably the biggest point. | 16:53 |
| * th1a needs menesis to be more productively pessimistic. | 16:54 | |
| yvl | "Students in Australia attend school for 200 days a year. Their school year lasts from late January to mid December." | 16:54 |
| yvl | just FYI | 16:54 |
| th1a | So November is still fine for them! | 16:54 |
| th1a | ;-) | 16:54 |
| menesis | :) | 16:54 |
| yvl | nad Brazil I think | 16:55 |
| th1a | October (and April) are spring and fall everywhere, right? | 16:55 |
| yvl | *and | 16:55 |
| th1a | So most schools are in session for both. | 16:55 |
| th1a | Which is good. | 16:56 |
| menesis | not all regions have spring or fall | 16:56 |
| menesis | they have rain and dry seasons :) | 16:56 |
| replaceafill | we call it winter and summer :) | 16:56 |
| th1a | OK. Moving on... | 16:57 |
| th1a | menesis: What have you been up to? | 16:57 |
| yvl | (umm, in Australia it does start in Jan, I think) | 16:57 |
| menesis | I have worked tirelessly on Monday, so yesterday/today was slow | 16:58 |
| *** jelkner has joined #schooltool | 16:58 | |
| menesis | I was looking at the mentioned intervention failure, but only managed to do a small workaround | 16:59 |
| menesis | and I continue to hunt for non-translatable strings | 16:59 |
| th1a | Should aelkner_ be doing something else on that intervention bug? | 17:00 |
| menesis | also made pages have a <title>, but this was written last week | 17:00 |
| aelkner_ | i was going to look at that, menesis | 17:00 |
| menesis | I don't plan to look more at that intervention widget | 17:00 |
| aelkner_ | the intervention bug, that s | 17:00 |
| menesis | but as I written somewhere, it works bad when there are errors in the form | 17:01 |
| aelkner_ | i saw your suggestion, so ill look into it | 17:01 |
| menesis | it is a mix of formlib and z3c widgets | 17:01 |
| aelkner_ | and those areas are always way more trouble than they are worth! | 17:02 |
| menesis | would be better to port old skin views to z3c.form and only have one proper widget | 17:02 |
| aelkner_ | ah | 17:02 |
| th1a | For the most part I don't want menesis banging his head on these kinds of bugs -- they need to be handed off to other developers. | 17:02 |
| aelkner_ | yes, i would expect that | 17:02 |
| aelkner_ | but thanks for your analysis up to now menesis | 17:03 |
| menesis | I looked into it only because others could not reproduce the bug | 17:03 |
| th1a | So far today I've encouraged menesis to be more pessimistic and do less work. | 17:03 |
| menesis | but I am fixing bugs as well! ;) | 17:04 |
| th1a | I know menesis! | 17:04 |
| th1a | How's the translation situation look menesis/replaceafill? | 17:06 |
| replaceafill | on my part: | 17:07 |
| replaceafill | i finished the es-sv translations for the plugins | 17:07 |
| replaceafill | set them up at http://69.164.203.135:6660 | 17:07 |
| menesis | some things in left sidebar are not translatable. | 17:08 |
| replaceafill | and noticed some issues in the gradebook/journal | 17:08 |
| replaceafill | right | 17:08 |
| replaceafill | i noticed the same | 17:08 |
| replaceafill | also, tertiary navigation for add activity | 17:08 |
| menesis | there are many templates that do not have i18n:domain, but those are easy to fix, was doing that before the meeting | 17:09 |
| replaceafill | menesis i have a question for you | 17:09 |
| replaceafill | would it be ok to translate the .po files directly in the locales directories | 17:09 |
| replaceafill | so i don't need to go to rossetta? | 17:09 |
| menesis | yes | 17:10 |
| replaceafill | cool, thanks | 17:10 |
| menesis | in rosetta it is explained that you can download, translate, and upload the po file | 17:11 |
| menesis | but we can edit the .po in place | 17:11 |
| menesis | that's what I was doing | 17:11 |
| replaceafill | yes, it's easier for debugging | 17:11 |
| replaceafill | i wasnt :( | 17:11 |
| replaceafill | i was downloading every change from rosetta :D | 17:11 |
| replaceafill | kind of slow process, getting the mail, putting the file, etc | 17:11 |
| menesis | find problematic place, fix it, make extract-translations; msgmerge lt.po schooltool.pot; msgfmt lt.po; restart, see changes | 17:12 |
| replaceafill | ah ok, will follow that pattern | 17:13 |
| menesis | but of course you have to commit the updated file | 17:13 |
| menesis | it will be imported into rosetta | 17:13 |
| menesis | automatically | 17:13 |
| replaceafill | and of course i had the typical issues when translating | 17:14 |
| replaceafill | gender issues | 17:14 |
| replaceafill | strings with ${deploy-sheet} in the middle! | 17:14 |
| replaceafill | where ${deploy-sheet} is a link :D | 17:15 |
| th1a | What's the better way to do that? | 17:15 |
| th1a | Change the interface? | 17:15 |
| replaceafill | i thought *that* was the better way | 17:16 |
| replaceafill | did you notice those menesis? | 17:16 |
| replaceafill | there are like 2 or 3 in the gradebook | 17:16 |
| replaceafill | from the "What is This?" sections | 17:16 |
| menesis | what's wrong with ${deploy-sheet}? | 17:16 |
| replaceafill | nothing, it just feels funny | 17:17 |
| replaceafill | th1a i think i'll finish translating core today | 17:18 |
| th1a | replaceafill: OK. | 17:18 |
| menesis | replaceafill: we should write down problematic strings somewhere | 17:18 |
| replaceafill | good idea | 17:19 |
| menesis | because I am having problems, too. | 17:20 |
| replaceafill | like "'Minimum' passing score" | 17:20 |
| replaceafill | gave me trouble in spanish | 17:20 |
| menesis | for example with the recent "${person}'s" sidebar title | 17:20 |
| replaceafill | ah! true! | 17:20 |
| th1a | Do we need something different there entirely? | 17:21 |
| menesis | or even better with sections: Math (1)'s | 17:21 |
| menesis | yes | 17:22 |
| th1a | What's the more common way to do possessives? | 17:22 |
| th1a | Outside English... | 17:22 |
| aelkner_ | use of 'of' | 17:23 |
| aelkner_ | and reversing the word order | 17:23 |
| th1a | Should we just put the name or title there and hope the possessive meaning can be inferred? | 17:24 |
| replaceafill | th1a that's what i did in the spanish version | 17:24 |
| replaceafill | since "De ${section}" doesnt make too much sense | 17:25 |
| th1a | It would be nice to keep it in English. Can we do that without confusing translators? | 17:25 |
| yvl | there are many options, though not sure if they're any good | 17:25 |
| yvl | like "Section ${section}" | 17:25 |
| yvl | or "${section} : " | 17:25 |
| yvl | just my two cents | 17:26 |
| menesis | I don't like the title in the menu heading | 17:26 |
| yvl | (or "Section:" without the ${section}) | 17:26 |
| th1a | We need to make it clear that, for example, the link is to the section's calendar, not the user's calendar. | 17:27 |
| yvl | or even two-liner: | 17:28 |
| yvl | black, bold "Section" | 17:28 |
| yvl | gray, thing "${section}" | 17:28 |
| th1a | I'm not following that yvl. | 17:29 |
| th1a | For example? | 17:29 |
| th1a | Well, we'd better move on... | 17:30 |
| th1a | Thanks menesis. | 17:31 |
| menesis | To see my point, look at this screenshot of section page http://i.imgur.com/BibPN.png | 17:31 |
| th1a | Anything else replaceafill? | 17:31 |
| replaceafill | th1a no, i'm done | 17:31 |
| menesis | but the it is the title repeated in the content div are less needed than the one in sidebar | 17:32 |
| th1a | menesis: Yes. | 17:32 |
| th1a | Thanks replaceafill. | 17:33 |
| th1a | yvl? | 17:33 |
| yvl | I got a bit stuck these two days | 17:33 |
| yvl | most of the problems are solved | 17:34 |
| yvl | the one I'm working now is supporting chromium / firefox / html_unit | 17:34 |
| th1a | Perhaps you and replaceafill could meet tomorrow to get him going on this? | 17:35 |
| yvl | they kind of give different outputs in some cases | 17:35 |
| yvl | here's the thing | 17:35 |
| yvl | three drivers vary somewhat | 17:35 |
| yvl | that either: | 17:35 |
| yvl | needs to be solved | 17:35 |
| yvl | or one driver should be picked | 17:36 |
| yvl | and others forgotten | 17:36 |
| th1a | (I'm only excluding aelkner_ because I suspect a bleeding edge functional testing framework will drive him crazy(ier)) | 17:36 |
| aelkner_ | :) | 17:36 |
| yvl | ( I assumed that much :) ) | 17:36 |
| th1a | Probably pick one. | 17:36 |
| yvl | I knew you'd say that | 17:36 |
| aelkner_ | and i appreciate it | 17:36 |
| yvl | why? | 17:37 |
| yvl | and wich one? | 17:37 |
| yvl | and why that one? | 17:37 |
| * yvl is not attacking th1a or anything - it's just the questions that bother me a lot | 17:37 | |
| th1a | It seems like less work to pick one. | 17:37 |
| yvl | maybe | 17:37 |
| yvl | firefox; works and ships with Ubuntu | 17:37 |
| yvl | but is slow as hell | 17:38 |
| yvl | chromium - fast, but not default in ubuntu | 17:38 |
| replaceafill | slower than our old ftests? ;) | 17:38 |
| yvl | firefox?... you bet! | 17:38 |
| replaceafill | ah?!?! | 17:38 |
| replaceafill | really?!? | 17:38 |
| aelkner_ | i only use firefox if that means anything | 17:38 |
| replaceafill | i thought anything would be faster than our old ftests :D | 17:38 |
| yvl | nope | 17:38 |
| th1a | Does the differences in drivers reflect actual differences in the browsers as a user would see, or just how they work with selenium? | 17:39 |
| yvl | don't get your hopes up with tests running faster | 17:39 |
| yvl | th1a - both | 17:39 |
| yvl | oh, and also the output we compare in tests | 17:39 |
| yvl | thing is, the differences are quite little | 17:39 |
| yvl | I think I can solve it | 17:39 |
| *** jelkner has quit IRC | 17:39 | |
| yvl | but can't give an estimate | 17:40 |
| yvl | there's just a list of things to try out | 17:40 |
| yvl | (btw, html_unit is fastest, but runs on emulated browser and is not visual) | 17:40 |
| th1a | I'd pick Chromium or Firefox. | 17:41 |
| th1a | And personally I've not used Firefox in years... | 17:41 |
| th1a | I'd pick one and if we really want to we can support the second later on, right? | 17:42 |
| yvl | but people with stock Ubuntu may do that | 17:42 |
| yvl | I'd like to ask some time to thing and try working this out | 17:42 |
| aelkner_ | firefox has firebug, so i wouldn't use anything else | 17:42 |
| th1a | Yes... my personal taste shouldn't be the deciding factor. | 17:42 |
| yvl | and resort to a single browser when the rest of it fails | 17:42 |
| yvl | aelkner_, chromium has some nice built-in stuff too | 17:42 |
| yvl | but your point is valid | 17:43 |
| th1a | Realistically, it should be Firefox unless it is so much slower it messes with our productivity. | 17:43 |
| yvl | if I had to pick, I'd pick chromium because it's almost twice as fast on my machine | 17:43 |
| th1a | That's a lot. | 17:43 |
| yvl | that's why I want to try to make them all work | 17:44 |
| yvl | on the same tests | 17:44 |
| replaceafill | i'd pick firefox because that's what ubuntu users have by default | 17:44 |
| replaceafill | right? | 17:44 |
| th1a | All things being equal, I'd take firefox, but half as fast is a lot. | 17:44 |
| replaceafill | but users dont run tests | 17:44 |
| yvl | I'm glad you see why I'm puzzled | 17:45 |
| th1a | A lot of OUR time! | 17:45 |
| replaceafill | and we can pick packages and specific tests | 17:45 |
| replaceafill | th1a but we can select a specific test to run | 17:45 |
| replaceafill | like bin/test -t foo.txt | 17:45 |
| th1a | So... 30 seconds vs. 1 minute? | 17:45 |
| replaceafill | we don't have to run the whole suit | 17:45 |
| th1a | Times... thirty times a day? | 17:46 |
| replaceafill | :| | 17:46 |
| aelkner_ | suit and suite are very different things :) | 17:46 |
| replaceafill | aelkner_ sorry | 17:46 |
| replaceafill | missed the e ;) | 17:46 |
| yvl | :) | 17:46 |
| aelkner_ | and i got SO confused! | 17:46 |
| replaceafill | ¬¬ | 17:46 |
| th1a | We're used to it aelkner_. | 17:47 |
| replaceafill | anyway, that's just my 2 cents | 17:47 |
| aelkner_ | but i'd like to try on a test suit on of these days, perhaps with a matching hat | 17:47 |
| yvl | frankly, I'd really like to keep hammering. | 17:47 |
| replaceafill | aelkner_ :)) | 17:47 |
| yvl | after the annoyances I had to go through to make them run | 17:47 |
| aelkner_ | i vote for yvl getting firefox to work better :) | 17:47 |
| yvl | ... | 17:47 |
| yvl | it just seems not that difficult now | 17:48 |
| yvl | then again - I'm working on this few weeks now | 17:48 |
| th1a | OK. | 17:48 |
| yvl | but you've no idea how crazy these people are | 17:48 |
| th1a | My concern is that not only will getting both running take longer now, it may be costly in the long run. | 17:48 |
| yvl | a very valid concern! | 17:49 |
| th1a | Constant fiddling. | 17:49 |
| yvl | right | 17:49 |
| yvl | then again - it's a bet | 17:49 |
| th1a | The speed issue is not trivial. | 17:49 |
| yvl | I bet that we'll save more time while writing tests than fiddling for two browser support | 17:49 |
| th1a | Well, use your judgement. | 17:50 |
| yvl | ok | 17:50 |
| th1a | Are we done then? | 17:50 |
| yvl | yep | 17:50 |
| th1a | Oh, replaceafill & yvl, want to meet tomorrow? | 17:50 |
| yvl | hmm | 17:51 |
| yvl | well, the code is out there | 17:51 |
| yvl | I can finish form filler tomorrow.... | 17:51 |
| yvl | we may meet, replaceafill, if you want | 17:52 |
| replaceafill | sure | 17:52 |
| yvl | cool | 17:52 |
| replaceafill | maybe i should take a current test and port it? | 17:52 |
| yvl | hmm | 17:53 |
| yvl | if you find a good one | 17:53 |
| replaceafill | like gradebook_management.txt!!! | 17:53 |
| replaceafill | :) | 17:53 |
| yvl | a good one :D | 17:53 |
| yvl | but it has a lot of dependencies | 17:53 |
| yvl | might run slow | 17:53 |
| yvl | then again - a good stress-test | 17:53 |
| replaceafill | i think so | 17:54 |
| yvl | +1 | 17:54 |
| replaceafill | i'll give it a try today if i can and let you know tomorrow | 17:54 |
| th1a | OK. Sounds good. | 17:54 |
| yvl | okay | 17:54 |
| th1a | Thanks guys! | 17:55 |
| * th1a drops the bag of gravel. | 17:55 | |
| yvl | thanks guys! | 17:55 |
| replaceafill | thanks everybody | 17:55 |
| yvl | oh, replaceafill - a small thing to keep in mind | 17:55 |
| replaceafill | yes? | 17:56 |
| yvl | Selenium does not wait for "action" to complete | 17:56 |
| yvl | instead it waits an implicit amount of time to finish a query | 17:56 |
| yvl | like xpath and such | 17:56 |
| yvl | so if you click on "Add" in one browser | 17:56 |
| yvl | and immediately after that open a window in another browser... you get the idea | 17:57 |
| yvl | as in manager.query.button("add").click() | 17:57 |
| yvl | bob.open("/stuff") | 17:57 |
| yvl | on stuff | 17:57 |
| replaceafill | ah | 17:57 |
| yvl | * no stuff | 17:57 |
| yvl | but manager.query.button("add").click() | 17:57 |
| yvl | print manager.query.xpath("title") | 17:57 |
| yvl | and there you go | 17:57 |
| yvl | also | 17:58 |
| yvl | in buildout.cfg | 17:58 |
| yvl | don't remember if I wrote about this | 17:58 |
| yvl | you can set | 17:58 |
| yvl | selenium.implicit_wait = 5 | 17:58 |
| yvl | or something | 17:58 |
| yvl | need to fine-tune to your machine | 17:58 |
| replaceafill | ah, no i think you didnt | 17:58 |
| replaceafill | ah ok | 17:58 |
| replaceafill | yvl question | 17:58 |
| replaceafill | where can i find the buildout options for the selenium part? | 17:59 |
| replaceafill | are they the same as the google-code driver page | 17:59 |
| yvl | they're ours and undocumented | 17:59 |
| yvl | sorry | 17:59 |
| replaceafill | ah ok | 17:59 |
| replaceafill | in the recipe? | 18:00 |
| yvl | you can try looking at the source of schooltool.devtools | 18:00 |
| yvl | selenium_recipe.py | 18:00 |
| replaceafill | ah ok | 18:00 |
| yvl | but it's not very... clear | 18:00 |
| replaceafill | will do | 18:00 |
| replaceafill | :) | 18:00 |
| replaceafill | np, i was just wondering about that | 18:00 |
| yvl | if it was documented it would be really easy to use ;) | 18:02 |
| yvl | well... good luck to you! | 18:02 |
| replaceafill | thanks | 18:02 |
| yvl | see you tomorrow ;) | 18:02 |
| replaceafill | bye yvl | 18:02 |
| *** yvl has quit IRC | 18:07 | |
| replaceafill | th1a do you mind if i start with the selenium task today? and put translations aside for a moment? | 18:22 |
| *** ignas has joined #schooltool | 18:22 | |
| replaceafill | i'd like to change to a new activity | 18:22 |
| th1a | OK. | 18:22 |
| replaceafill | thanks | 18:23 |
| *** ignas has quit IRC | 18:32 | |
| replaceafill | yay! | 19:42 |
| * replaceafill just ported his first ftest to stest | 19:42 | |
| th1a | How's it look? | 19:42 |
| replaceafill | it's a small one though, i'm going to try the chrome driver | 19:43 |
| replaceafill | to see it twice faster ;) | 19:43 |
| th1a | Yeah... I wonder if that's mostly startup time. | 19:44 |
| *** menesis has quit IRC | 19:49 | |
| *** menesis has joined #schooltool | 19:49 | |
| *** menesis has quit IRC | 19:56 | |
| *** menesis has joined #schooltool | 19:57 | |
| *** ignas has joined #schooltool | 19:58 | |
| *** menesis has quit IRC | 20:13 | |
| th1a | aelkner_: ayt? | 21:05 |
| replaceafill | lol, i was running chromium :D | 21:16 |
| replaceafill | ah :( the firefox driver is broken | 21:16 |
| th1a | Oh, chromium not chrome? | 21:28 |
| replaceafill | it's called chromium in ubuntu | 21:30 |
| th1a | replaceafill: I'm confused. Does it work? | 21:47 |
| replaceafill | yes, but only with the chromium driver | 21:48 |
| th1a | Oh, there are different drivers. I see. | 21:51 |
| replaceafill | yes | 21:51 |
| replaceafill | chromium, firefox and html_unit | 21:51 |
| replaceafill | yvl's current branch works well with chromium | 21:51 |
| replaceafill | it seems like firefox sends the keys in a wrong way | 21:52 |
| replaceafill | i'm downloading selenium-server to test html_unit | 21:52 |
| *** th1a has quit IRC | 22:52 | |
| *** th1a has joined #schooltool | 23:16 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!
