IRC log of #schooltool for Monday, 2013-03-18

*** menesis has quit IRC01:27
*** yvl has joined #schooltool10:23
*** ignas has joined #schooltool14:38
*** menesis has joined #schooltool14:59
th1ahi yvl, menesis.15:31
yvlgood morning th1a15:31
th1areplaceafill is at PyCon... I guess it is 6:30 AM there.15:31
*** replaceafill has joined #schooltool15:32
replaceafillgood morning/afternoon15:32
yvlhey replaceafill  :)15:32
replaceafill(sorry i'm late, it's 6:30 am here :D)15:32
th1aThat's what I just said.  ;-)15:32
th1aHow's PyCon, replaceafill?15:33
replaceafillreally cool15:33
replaceafilllearned a couple of things ;)15:33
replaceafilland got a free raspberry pi!15:33
yvlwhoa15:33
yvlcongrats! :)15:33
replaceafill:D15:33
th1aThat's like finding $35 on the sidewalk!15:34
replaceafill:))15:34
th1aSpeaking of which, replaceafill, I need your invoice.15:35
replaceafillth1a, ah ok, will send it later today15:35
replaceafillis somebody already doing his report?15:35
replaceafilli have a couple of questions15:36
yvlnobody started yet15:36
th1aGo ahead replaceafill.15:36
replaceafillah ok15:36
replaceafillthanks15:36
replaceafillyvl, first question:15:37
th1aYou got here at 9:31.  We aren't that fast.15:37
replaceafillglenda requested the two completion reports to be empty at the beginning15:37
replaceafillso i've used a fromPublication check in my table filter15:37
replaceafillbut yesterday i used the acc data, and noticed that the report was taking a while to load even then15:38
replaceafillthe reason is that i'm getting all the information needed from the section in the items() method of the table15:38
replaceafillso the report always goes through every sections, gets data for it, and then filters15:39
replaceafilli was thinking of putting the part to extract data from the section at the end of filtering15:39
replaceafill(haven't tried yet)15:40
replaceafillbut i think that will really speed the report up15:40
replaceafillcan you recommend a better approach?15:40
yvlhmm15:40
replaceafilli'm not sure if there's an intermediate step between table's items and the filter's filter method15:41
yvlyou probably can override .items() in table itself15:41
yvlbasically there are two steps15:41
yvlsetting up15:41
yvland rendering15:41
yvlfiltering happens in the first part15:42
yvlbut right before that it obtains the items to filter15:42
yvlif you look at schooltool/table/table.py15:42
yvlyou can see def items():15:43
yvlreturn self.source.values()15:43
replaceafillwhen do you recommend extract information from the items to use in the table?15:43
replaceafillbecause that part is what takes time15:43
yvlumm15:43
replaceafillhttp://bazaar.launchpad.net/~schooltool-owners/schooltool.virginia/trunk/view/head:/src/schooltool/virginia/browser/report.py#L117915:44
yvlif I understand correctly, you want to skip the item extraction entirely when fromPublication15:44
replaceafillcorrect15:44
replaceafillwhen not fromPublication :)15:44
yvl:)15:44
yvlso... just add return[] and be done with it?15:45
replaceafillah, you mean in the table formatter!!!!15:45
yvlyes15:45
replaceafillnot in the filter15:45
yvlyes15:45
replaceafillDUH MOMENT!15:45
yvl:D15:45
replaceafillthe damn table is the manager of the filter15:45
* replaceafill slaps his face15:45
yvlyes :D15:45
replaceafillok, glad that's not that difficult ;)15:46
replaceafillthanks15:46
replaceafillsecond question15:46
yvlglad to help :)15:46
replaceafilli remember you added a view to render pdf from a form15:46
replaceafillcould you point me to that?15:46
replaceafillis it in the gradebook?15:46
yvlskin.flourish.report.PDFForm ?15:47
replaceafilli think the view landed somewhere because i can see the machinery in flourish15:47
replaceafillyes, but i mean, where it's used15:48
replaceafillwell, i think i can grep that now15:48
yvlbasicperson/browser/person.py:class ProfileGeneralPart(flourish.report.PDFForm)15:48
replaceafillgot it15:48
replaceafillit's the profile pdf15:49
replaceafillthanks, i was looking like crazy in the gradebook :D15:49
yvl:)15:49
replaceafillth1a, so i finished the scr for the section15:49
replaceafilli just have the changes that glenda requested on friday left15:50
th1aOK.15:50
replaceafilladding a filter for course15:50
replaceafilland did she answer your last email?15:50
replaceafillspecifically your total row question15:50
th1aLet's leave it as is for now.15:51
replaceafill"I'd rather only include that one if so, because the others are just a confusing distraction."15:51
replaceafillok, i'll turn the totals for % columns into averages then15:51
replaceafilli should be done with these today15:52
replaceafilland yesterday we started the quiz sprint15:52
replaceafilldavid and paul are working on updating cando grades from the quiz data15:52
replaceafilli think we made good progress for one afternoon15:53
replaceafillthat's it from me15:53
th1aGlenda said: "The most important totals are the total number of students and the average percentage of those students hitting or exceeding the 80% attainment mark. Total # skills is irrelevant, but I'd leave the column in for now."15:54
th1aThanks replaceafill.15:54
replaceafillah ok15:55
replaceafill(i don't think i received that one...)15:55
th1aI just forwarded it.15:55
replaceafillah thanks15:55
th1aThanks replaceafill.15:57
th1ayvl?15:57
yvlone sec.15:57
yvlok15:58
yvlI added a table of reports in person homepage15:58
yvlthen I tried quickly adding more reports15:59
yvlso, it's not a home run :/15:59
yvlnow I have better tracebacks, which is good for debugging and user bug reports15:59
yvlrunning stuff remotely can cause cascade of fail instead of a single fail16:00
th1aYes, that's the scary part.16:00
th1aDo you know what's going wrong?16:00
yvlnow, yes16:01
yvland I'm glad stuff broke16:01
yvlif we had user reports in form of "failed to commit to DB"16:01
yvlwithout explanation of what and where and why16:01
yvlit would be... well, tough16:01
yvlI also had to deal with security, which I though I could mostly avoid16:01
yvl(security caused fails)16:02
yvland for last 1.5 days I'm at a stage where16:02
yvl"well, I'll fix this one thing and everything will run now!"16:02
yvland then it uncovers another small issue16:02
th1aWell, that's understandable.16:03
yvltwo minutes ago I ran the last experiment, and again a small thing - missing URL in "fake" request (because we're not using traversal)16:03
yvla line of code to fix :)16:03
yvlprobably will uncover something16:03
yvlor just work16:03
yvlI'm very annoyed by the fact I can't give an estimate at the moment :/16:04
yvloh, and there were some small issues with file downloading, resolved now16:04
th1aOK.  Finding a bunch of small bugs now is not a problem.16:05
th1aFinding an "Oh, shit, this is fundamentally broken" would be.16:05
yvltrue16:05
yvlSo - I'm digging.16:06
yvlfull speed :)16:06
yvloh16:07
yvland I did not look into journal bug :/16:07
yvlyet16:07
th1aOK.16:07
th1aBetter to work on the reports for now.16:07
yvlat the moment it looks like the recent bug somebody reported is manifestation of one in Jeffs instance16:08
th1aWe can always do bugfix releases.16:08
yvlright16:08
yvlwell, I'll notify as soon as I have it working16:09
th1aOK.  Thanks yvl.16:10
th1amenesis, replaceafill will let you know when we're ready for a CanDo update, hopefully later today.16:10
th1aOf course it may be too late for you today, depending on your schedule.16:10
th1aCan you poke menesis, yvl?16:13
yvlumm16:14
yvlhe's working from home16:14
* yvl can poke by phone though!16:14
th1aOK.16:14
yvlpoked16:15
menesishello16:17
th1ahi menesis.16:17
menesisok I will be ready to upload the update if it's not too late16:18
th1aOK.  Thanks.16:21
th1aAnything else to add menesis?16:21
menesisno, I got a cold again since friday :( not too bad but hard to work with a headache16:24
th1aOK.  No problem.16:24
menesisI was running tests with various z3c.form versions16:25
replaceafilli noticed the change to version.cfg and removed my pin down :)16:25
replaceafilland cando works16:25
menesisbut a lot of later releases have one problem or another16:26
menesisI pinpointed a change that caused test failure in virginia16:27
menesisbut stopped there, didn't investigate16:29
menesisother than that, no good news16:31
menesisbut now that I'm out of bed, will work today16:32
th1aOK.  Glad you're feeling better.16:32
th1aAll right.  Thanks guys.16:34
th1aI'm almost done with the basic CanDo docs.16:34
th1aJust need to do reports.16:34
th1aAnd I might as well wait for those to be updated.16:34
th1aI generated a million tiny bug reports for replaceafill in the process.16:34
replaceafilli noticed :)16:35
th1aOK.  Thanks guys.16:36
th1aHave a good week.  See you Wednesday.16:36
th1aEnjoy California, Douglas.16:36
th1aWhere are you exactly?16:36
replaceafillhaven't seen much, but will try :)16:37
replaceafillsanta clara16:37
replaceafillwe're in the not-so-fun part of town :(16:37
th1aThe real Real America?16:38
replaceafillyes!16:38
replaceafillit's a 2.2 mile walk from here to the convention center16:39
replaceafillbut hey, i just ran 5k on saturday ;)16:39
replaceafillhttps://lap.io/event/2013-pycon-5k-charity-fun-run/results16:39
replaceafilllook at 137 :D16:39
replaceafilli almost die trying ;)16:39
th1aDid you beat jelkner?16:40
replaceafill:D16:40
replaceafilljelkner was working at that time :)16:40
th1aSure he was.16:42
replaceafillok i'll go get ready for leaving16:43
replaceafillsee you later guys16:43
yvlhave fun replaceafill! ;)16:43
replaceafill:P16:43
*** replaceafill has quit IRC16:43
yvla bit of good new, th1a16:44
yvljust generated person profile report16:44
th1aExcellent.16:45
yvlsection roster keeps generating empty pdfs though16:45
yvlwill look into that next16:45
th1aOK.16:45
th1aAnd yes, time invested in good error messages at this point is important.16:45
yvloh yes.16:46
yvlok, see you Wed16:46
yvlwith some good news I hope :)16:46
th1aThanks yvl.16:48
th1aOh...16:48
* th1a drops the bag of gravel.16:49
th1a;-)16:49
yvl;)16:49
*** replaceafill has joined #schooltool18:22
replaceafillth1a, you around?19:36
*** replaceafill has quit IRC20:09
*** replaceafill has joined #schooltool20:13
th1areplaceafill:  hi.20:39
*** replaceafill has quit IRC21:26
*** replaceafill has joined #schooltool22:23
replaceafillth1a, just shared a doc with you22:24
replaceafilli found a bug in the progress report22:24
th1aWhat is it?22:25
replaceafillsee the last line of the google spreadsheet22:25
replaceafillwell, actually, this worksheet is the *right* way :)22:25
th1aProgress report totals?22:26
replaceafillnot the totals, the actual rows for the sections22:26
replaceafillthe last column22:26
replaceafillAvg. % Required Skills Attained22:26
replaceafilli want to share this with glenda and have the values double check22:27
replaceafillthat ok?22:27
th1aSo... is there a bug?22:27
replaceafillbefore we get the "this number is wrong" :(22:27
replaceafillyes22:27
replaceafilland since we have no test for that report, i thought i should use *this* data and create one22:27
replaceafillnow that i know the right numbers to watch22:27
replaceafilli don't have glenda's gmail address22:28
th1areplaceafill, I have not idea what the problem is.22:28
replaceafillth1a, np, i'll just fix it22:28
th1aUh...22:28
replaceafilldo you have glenda's gmail address?22:28
replaceafilli don't know if someone can access the formulas if i just send her the link22:29
th1aYou really don't want to explain the problem to me?22:29
replaceafillah sure22:32
replaceafillwe basically copied the VA reports from the old cando, right?22:33
th1aOK.22:33
replaceafillwhen calculating the averages, that logic is using skills attained / skills evaluated22:34
replaceafillwhen it should use (imho) skillls attained / total skills for the section22:34
replaceafillat some point i remember glenda saying: "i wish i understand these numbers better"22:34
replaceafill(or something like that)22:34
replaceafillso i just set this spreadsheet to confirm the values are correct22:35
replaceafilli haven't finished my test but i'm sure when i run it with this data, that last column won't be 54.16%22:36
replaceafillas it should be22:36
th1aThis is an average we didn't have in the report before?22:36
replaceafillwe did22:37
replaceafillit's just wrong22:37
th1aWelsh didn't think it was wrong?22:37
replaceafillno22:38
replaceafilli think he never really did the math on it22:38
th1aOh...22:38
th1aCan you show me the actual report?22:38
replaceafillsure22:38
replaceafillis a screenshot ok?22:38
th1aYes.22:38
replaceafillour wireless sucks right now22:38
replaceafillok, hold on22:39
* replaceafill is connecting his external hd again22:42
*** replaceafill_ has joined #schooltool22:50
*** replaceafill_ has joined #schooltool22:50
*** replaceafill_ has joined #schooltool22:50
*** replaceafill has quit IRC22:53
*** replaceafill_ has quit IRC22:56
*** replaceafill has joined #schooltool23:02
th1aAh, conference wifi.23:02
replaceafill>:(23:02
replaceafilli'm even trying webchat!23:02
replaceafillwhatever gets my typing first23:02
replaceafillok23:02
replaceafillhttp://imagebin.org/25078223:02
replaceafillthat's acc data23:03
th1aIs this the new version?23:03
replaceafillyes23:03
th1aThis is the SIMPLER version?23:03
replaceafillyes23:03
replaceafillbut this is the progress report23:03
replaceafillthe completion one is simpler23:03
replaceafill(va has two "completion" reports...)23:03
th1aOK, so what's the question?23:04
th1aThis report is bullshit.23:04
replaceafill:D23:05
replaceafillsee? that's why i wanted to send it directly to glenda23:05
replaceafillbecause i knew you would hate it :(23:05
th1aWell, you're going to have to write this down anyhow.23:05
replaceafilli just want somebody to double check my spreadsheet values are ok23:06
th1aIt's just taking a long time to get to telling me what the problem is.23:06
replaceafillbecause i think my spreadsheet is correct23:07
replaceafilland the report is not23:07
th1aWHAT ARE WE TALKING ABOUT?  WHAT IS THE PROBLEM?23:07
replaceafilli know i'm never clear enough23:07
th1aYou're usually more clear.  ;-)23:08
th1aLet's put it this way... if we were sitting next to each other, you'd be pointing at something.  What?23:08
replaceafillthe division in the report!23:10
replaceafillthe last column23:10
th1aAverage percentage of required skills attained?23:10
th1aAh.  Dave wanted it that way.23:11
th1aSee, that was easy.23:11
th1aLiving online!23:12
replaceafillmy point is:23:12
replaceafillhow can you tell this is wrong from this single line, you need skills and grades, right?23:12
replaceafillthat's why i created the spreadsheet23:12
th1aIs this the question:23:13
th1a> when calculating the averages, that logic is using skills attained / skills evaluated23:13
th1a when it should use (imho) skillls attained / total skills for the section23:13
th1aIf so, Welsh wanted the first way.23:13
th1ahello, replaceafill?23:16
*** replaceafill has quit IRC23:16
*** replaceafill has joined #schooltool23:47

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!