| *** replaceafill has quit IRC | 02:20 | |
| *** menesis has joined #schooltool | 08:05 | |
| *** ignas has joined #schooltool | 12:02 | |
| *** ignas has quit IRC | 13:36 | |
| *** menesis has quit IRC | 13:41 | |
| *** menesis has joined #schooltool | 14:58 | |
| *** replaceafill has joined #schooltool | 15:59 | |
| th1a | menesis: The dns handoff is creeping forward. | 16:30 |
|---|---|---|
| th1a | hi menesis, replaceafill, yvl. | 16:31 |
| replaceafill | good morning/afternoon | 16:31 |
| menesis | hi | 16:32 |
| menesis | th1a: good news | 16:32 |
| yvl | ? | 16:37 |
| th1a | I was waiting for yvl. | 16:37 |
| th1a | Were you lurking? | 16:37 |
| yvl | yes | 16:37 |
| yvl | hi :) | 16:37 |
| th1a | You usually say "hi." | 16:37 |
| th1a | OK... | 16:37 |
| yvl | sorry | 16:37 |
| th1a | Probably replaceafill should start with some questions for yvl. | 16:38 |
| yvl | sure | 16:38 |
| replaceafill | yes | 16:38 |
| replaceafill | ok | 16:38 |
| replaceafill | so, my task (no pun intended) was simple | 16:38 |
| replaceafill | convert the scr report in cando into a remote report | 16:38 |
| replaceafill | (simple in theory) ;) | 16:39 |
| replaceafill | so this is my first attempt: | 16:39 |
| replaceafill | http://bazaar.launchpad.net/~replaceafill/schooltool.cando/remote_scr/revision/203 | 16:39 |
| replaceafill | getting the dialog was almost automatic | 16:39 |
| replaceafill | using the existing code | 16:39 |
| replaceafill | but then i realized that ReportTasks use intids | 16:39 |
| replaceafill | on the context of the report | 16:40 |
| replaceafill | currently we have two remote reports: | 16:40 |
| replaceafill | section roster and person profile | 16:40 |
| replaceafill | (at least in core) | 16:40 |
| replaceafill | both of those pdf work on the context itself (section and basicperson) | 16:40 |
| replaceafill | the scr pdf works on a studentgradebook adapter | 16:40 |
| replaceafill | so when the reporttask tried to do the int id look up | 16:41 |
| replaceafill | it failed | 16:41 |
| replaceafill | because the adapter has no intid | 16:41 |
| replaceafill | yvl, so i created my own report task | 16:41 |
| replaceafill | and set the task_factory on the download dialog | 16:41 |
| yvl | makes sense so far | 16:41 |
| replaceafill | thank God :D | 16:41 |
| replaceafill | yes | 16:41 |
| replaceafill | this had some little issues | 16:42 |
| replaceafill | like properties swallowing the actual error :( | 16:42 |
| replaceafill | but i kind of figured it out | 16:42 |
| replaceafill | ah | 16:42 |
| replaceafill | the base_filename was funny | 16:42 |
| replaceafill | the original method in pdfpage | 16:42 |
| replaceafill | assumes the view has a __name__ attribute | 16:42 |
| replaceafill | but in the remote context this doesn't happen | 16:43 |
| replaceafill | so that required some pdb'ing | 16:43 |
| replaceafill | the part i'm stuck right now is | 16:43 |
| replaceafill | i can get the task added to the container, the message set | 16:43 |
| replaceafill | but when the task tries to execute i get this error: | 16:44 |
| replaceafill | http://imagebin.org/253458 | 16:44 |
| replaceafill | interpreting the failures is still not clear to me | 16:44 |
| replaceafill | but i assume all those tracebacks go in order, right? | 16:44 |
| yvl | yes | 16:44 |
| replaceafill | i mean, the first error is the execute part | 16:44 |
| replaceafill | and here i'm stuck because i don't know what's wrong | 16:45 |
| replaceafill | i mean | 16:45 |
| replaceafill | the reporttask has its own execute | 16:45 |
| replaceafill | and gradebookreporttask inherits that | 16:45 |
| replaceafill | i was looking for some security zcml that i'm missing | 16:46 |
| yvl | hmm | 16:46 |
| replaceafill | but couldn't find any | 16:46 |
| yvl | maybe it tries to obtain context | 16:46 |
| yvl | and attribute error fires there? | 16:46 |
| yvl | and by accident, propagates? | 16:46 |
| yvl | no, that makes no sense | 16:47 |
| replaceafill | yvl, i tried to pdb the execute method in reporttask | 16:47 |
| replaceafill | but i guess celery calls that? | 16:47 |
| replaceafill | and at that point the schooltool process (foreground) has nothing to do with it | 16:47 |
| replaceafill | i get a pdbquit exception in the celery log instead | 16:48 |
| replaceafill | the separated logs are great btw | 16:48 |
| yvl | oh | 16:48 |
| yvl | did you restart celery? | 16:48 |
| yvl | make stop | 16:48 |
| yvl | make run | 16:48 |
| replaceafill | yes | 16:48 |
| replaceafill | actually | 16:48 |
| replaceafill | i had several issues trying to use an existing database | 16:49 |
| replaceafill | so i decided to start everything from scratch | 16:49 |
| replaceafill | using a sample xls and importing everything | 16:49 |
| replaceafill | another thing worth mentioning is that we have a node catalog in cando | 16:49 |
| yvl | hmm | 16:50 |
| replaceafill | check my branch and you'll see i had to comment that | 16:50 |
| replaceafill | because i was getting an INode problem just with the profile pdf | 16:50 |
| replaceafill | related to catalog | 16:50 |
| replaceafill | ah hold on | 16:50 |
| yvl | could you quickly look at... | 16:50 |
| yvl | bin/celeryd | 16:50 |
| replaceafill | i think i got a screenshot... | 16:50 |
| yvl | and check what ST eggs are included? | 16:50 |
| replaceafill | http://imagebin.org/253461 | 16:51 |
| replaceafill | that is the one | 16:51 |
| replaceafill | (happens when cando catalog is enabled) | 16:51 |
| replaceafill | and you try to run the profile pdf | 16:51 |
| replaceafill | without any other modification of mine | 16:51 |
| replaceafill | ah ok, checking | 16:52 |
| replaceafill | yvl, ah | 16:52 |
| replaceafill | just your celery branch | 16:52 |
| replaceafill | /home/replaceafill/sandboxes/remote_reports/schooltool_celery/src | 16:52 |
| yvl | right | 16:52 |
| replaceafill | i added the gradebook and cando in my buildout | 16:52 |
| yvl | base.cfg needs an update | 16:52 |
| *** ignas has joined #schooltool | 16:53 | |
| replaceafill | adding virginia actually to [package] | 16:53 |
| menesis | "'IFoo' has no attribute 'providedBy'" error means that IFoo interface is not in pythonpath | 16:53 |
| replaceafill | (since it depends on the other two) | 16:53 |
| yvl | fixed pushed | 16:53 |
| yvl | (my branch) | 16:53 |
| replaceafill | ah got it | 16:54 |
| replaceafill | i need to run buildout again, right? | 16:55 |
| replaceafill | bin/celeryd has schooltool.virginia now | 16:55 |
| replaceafill | and cando and the gradebook | 16:55 |
| yvl | great | 16:56 |
| yvl | apologies | 16:56 |
| replaceafill | np | 16:56 |
| replaceafill | so, let me uncomment the catalog | 16:56 |
| replaceafill | great! progress | 16:58 |
| replaceafill | new error :D | 16:58 |
| replaceafill | 'Section' object has no attribute 'values' | 16:58 |
| replaceafill | this i can debug! | 16:58 |
| replaceafill | :D | 16:58 |
| yvl | :D | 16:59 |
| replaceafill | all in all i'm getting most of it | 16:59 |
| replaceafill | i mean, tasks, messages, etc | 16:59 |
| replaceafill | and kind of how they interact | 16:59 |
| replaceafill | the transaction manager is still above my head | 16:59 |
| yvl | cool | 16:59 |
| replaceafill | but i think i'm making progress | 17:00 |
| yvl | tm is also somewhat simple | 17:00 |
| replaceafill | even though i can't render the report yet ;) | 17:00 |
| * yvl has to push some tweaks on error handling | 17:00 | |
| replaceafill | ah this got my attention | 17:00 |
| yvl | failures to fail were not handled well | 17:00 |
| replaceafill | some flooding: | 17:00 |
| replaceafill | title = format_message( | 17:01 |
| replaceafill | _(u"Requested report ${filename}."), | 17:01 |
| replaceafill | mapping={'filename': view.filename}) | 17:01 |
| replaceafill | why is that format_message necessary? | 17:01 |
| replaceafill | why not just: _(u'...', mapping={...}) | 17:01 |
| replaceafill | i don't get exactly what format_message does | 17:01 |
| replaceafill | (not related to the remote stuff per se) | 17:01 |
| *** ignas has quit IRC | 17:02 | |
| yvl | you're right | 17:02 |
| yvl | format_message is not necessary there | 17:02 |
| yvl | I think | 17:02 |
| replaceafill | if i understand correctly, format_message is necessary when you don't control the message creation, right? | 17:03 |
| replaceafill | or when you want to modify the existing mapping | 17:03 |
| *** ignas has joined #schooltool | 17:03 | |
| replaceafill | but no biggie :) | 17:04 |
| replaceafill | th1a, ok, i can keep debugging now | 17:04 |
| replaceafill | again, i think i'm moving forward | 17:04 |
| th1a | OK... would you be able to work late today or tomorrow yvl, or just be around? | 17:05 |
| replaceafill | yvl, in theory, pdf reports don't need to change in order to become remote, right? | 17:05 |
| th1a | Or perhaps replaceafill can get up super early... | 17:05 |
| replaceafill | sure, i can get up early | 17:06 |
| yvl | I can stay around later tomorrow | 17:06 |
| replaceafill | i was going to ping the mailing list last night | 17:06 |
| replaceafill | but i realized that it was a security declaration missing | 17:06 |
| yvl | as for pdf reports - in theory - yes | 17:06 |
| replaceafill | and properties usually send me in another direction | 17:06 |
| th1a | OK, so let's plan on that. | 17:07 |
| th1a | For tomorrow. | 17:07 |
| replaceafill | kk | 17:07 |
| replaceafill | yvl, maybe with the base.cfg change, my existing database will work | 17:07 |
| yvl | it should now | 17:07 |
| replaceafill | and i don't have to start from scratch | 17:07 |
| replaceafill | great, i'll test that too | 17:07 |
| * replaceafill done asking | 17:08 | |
| th1a | OK, thanks replaceafill. | 17:08 |
| th1a | yvl? | 17:08 |
| yvl | I was adding report failure messages | 17:09 |
| yvl | then made bugs, and noticed that failing of failure notifications are not handled well | 17:09 |
| yvl | also, today ran into same thing with base.cfg | 17:10 |
| yvl | so failures are reported better now, need to finish last dialog | 17:10 |
| yvl | also, need to think about reporting failures of failures directly to sysadmin | 17:11 |
| yvl | successful task with failing notification gets missing (but can be found in the logs) | 17:12 |
| yvl | so I'll commit part of it... today I think | 17:12 |
| yvl | then - show number of unread messages near login | 17:13 |
| yvl | visiting Home will reset this | 17:13 |
| yvl | as you can see messages there | 17:13 |
| yvl | that's that | 17:14 |
| th1a | OK. Keep pushing. | 17:14 |
| th1a | Thanks yvl. | 17:14 |
| th1a | menesis? | 17:14 |
| menesis | I fear my status is the same as last time. | 17:15 |
| menesis | was refactoring scripts and writing commands | 17:16 |
| th1a | Well, that needs to be done. | 17:16 |
| menesis | I know :( | 17:17 |
| th1a | Nothing to do but keep at it. | 17:18 |
| menesis | I am, no distractions | 17:18 |
| th1a | That's all I can ask. | 17:20 |
| th1a | Unless I'm going to fly over and sit at your shoulder. | 17:20 |
| th1a | Thanks menesis. | 17:20 |
| th1a | yvl will have to stop by and pat you on the back or kick your butt as necessary. | 17:20 |
| th1a | I exchanged a few interesting emails with Glenda. | 17:22 |
| th1a | It seems possible that VA CTE will get out of the business of directly supporting CanDo more quickly than we might have thought, which may be good because then SIELibre can step in. | 17:23 |
| replaceafill | any chance the counties stop using it? :( | 17:24 |
| *** th1a has left #schooltool | 17:24 | |
| *** th1a has joined #schooltool | 17:25 | |
| th1a | Well... I guess that would be counties that don't like it or would never pay for support anyhow. | 17:26 |
| th1a | OK, I think that's it. | 17:27 |
| th1a | Have a good week/end gentlemen. | 17:27 |
| th1a | yvl and replaceafill: talk tomorrow. | 17:27 |
| yvl | sure | 17:28 |
| th1a | replaceafill: Send me your invoice! | 17:28 |
| replaceafill | will do | 17:28 |
| replaceafill | th1a, ah ok | 17:28 |
| replaceafill | will do too | 17:28 |
| th1a | I may be at my accountant's tomorrow morning. | 17:28 |
| th1a | Tax time. | 17:28 |
| yvl | replaceafill - email me if you hit a wall | 17:28 |
| * th1a drops the bag of gravel. | 17:28 | |
| yvl | thanks guys | 17:28 |
| replaceafill | yvl, will do :) | 17:28 |
| replaceafill | thanks everybody | 17:28 |
| replaceafill | yay, i passed the section issue | 17:33 |
| replaceafill | yvl, i've seen this one before: | 17:33 |
| replaceafill | newInteraction called while another interaction is active. | 17:33 |
| replaceafill | don't understand it though | 17:33 |
| replaceafill | something competing with something? | 17:34 |
| yvl | just mishandled interactions | 17:35 |
| yvl | need to look into this, but not fatal I think | 17:35 |
| yvl | seems to happen when failing, but tracebacks are propagated properly | 17:35 |
| yvl | IIRC it just needs interaction start/end to be refactored out to... | 17:36 |
| yvl | the task itself I think | 17:36 |
| yvl | well, one level up | 17:36 |
| yvl | pushed improved error handling btw | 17:36 |
| replaceafill | nothing i can do in my gradebook task level? | 17:37 |
| replaceafill | ah ok, i'll pull | 17:37 |
| yvl | nothing you need to do :) | 17:37 |
| replaceafill | so, at some point the report will run? | 17:38 |
| replaceafill | maybe i should try a simpler report | 17:38 |
| replaceafill | although there's no more simpler reports :D | 17:38 |
| replaceafill | "You can download this report at a later time from home page." :) | 17:40 |
| replaceafill | nice | 17:40 |
| replaceafill | YAY!!! | 17:40 |
| replaceafill | remote scr pdf generated | 17:40 |
| yvl | congrats! | 17:40 |
| yvl | \o/ | 17:40 |
| replaceafill | ¬¬ | 17:40 |
| replaceafill | come on! it's huge for me :P | 17:41 |
| * yvl is sincerely happy! | 17:41 | |
| replaceafill | :D | 17:41 |
| replaceafill | excellent! | 17:41 |
| * replaceafill starts remoting everything | 17:41 | |
| replaceafill | yvl, last question and i let you go | 17:41 |
| yvl | :DDD | 17:41 |
| yvl | sure | 17:42 |
| replaceafill | so, the custom report task for the gradebook is a valid approach? | 17:42 |
| replaceafill | i mean, the hacking of the context | 17:42 |
| yvl | I would say - yes | 17:42 |
| yvl | and it's not a hack :) | 17:42 |
| replaceafill | right | 17:42 |
| replaceafill | it's meant to be :D | 17:42 |
| replaceafill | cool, i think i have enough to try the scr for the section | 17:43 |
| yvl | great | 17:43 |
| replaceafill | thank you yvl! | 17:43 |
| yvl | thank you! | 17:43 |
| yvl | also, if you try remoting old style reports, there may be stuff missing | 17:44 |
| * yvl can't tell offhand what exactly | 17:44 | |
| replaceafill | ah, np, i'll stick to new style ones | 17:44 |
| replaceafill | unless you want me to try old style ones | 17:45 |
| replaceafill | but i think most of them are in the gradebook | 17:45 |
| yvl | you can stick to new style ones :) | 17:45 |
| replaceafill | kk | 17:45 |
| yvl | and/or port old style ones to new style :> | 17:45 |
| replaceafill | ah! good point | 17:46 |
| replaceafill | th1a, should i keep "remoting" cando reports? | 17:46 |
| replaceafill | th1a, i was thinking of doing the section scr | 17:46 |
| th1a | Well... | 17:46 |
| replaceafill | that one takes like 1 second per student, so i think it's a good candidate | 17:46 |
| replaceafill | (i know, it shouldn't take that long...) :( | 17:47 |
| th1a | These are dialog reports. | 17:47 |
| th1a | Right. | 17:47 |
| replaceafill | yes | 17:47 |
| th1a | I guess at this point yes, until you run into a wall. | 17:47 |
| replaceafill | kk | 17:47 |
| yvl | also, if you have spare time... | 17:47 |
| yvl | and it's a very long running report... | 17:47 |
| th1a | You want to find the next sticking point but not spend the next three hours banging your head against it. | 17:47 |
| yvl | you can look at schooltool.export.importer | 17:48 |
| yvl | for progress bars | 17:48 |
| replaceafill | yvl, ah, i did notice the remote tasks there | 17:48 |
| replaceafill | yvl, will do | 17:48 |
| yvl | but this is more of a bonus task :) | 17:48 |
| replaceafill | yvl, so in your branch, importers are actually using the remote stuff, right? | 17:48 |
| yvl | disabled | 17:48 |
| replaceafill | ah | 17:48 |
| yvl | they could :) | 17:49 |
| yvl | at least mega importer | 17:49 |
| yvl | ok | 17:49 |
| yvl | time for me to go | 17:49 |
| yvl | best of luck! | 17:49 |
| replaceafill | thanks, see you yvl | 17:49 |
| *** menesis has quit IRC | 19:47 | |
| replaceafill | th1a, i was able to make the section scr remote | 19:51 |
| replaceafill | i kind of cheated, but it worked | 19:52 |
| replaceafill | i'm going to get rid of the cheating by cleaning that report | 19:52 |
| replaceafill | i wonder if i should dare to make the completion report pdf remote... | 19:55 |
| replaceafill | nah, i'll check progress bars instead | 19:55 |
| th1a | replaceafill: Good. | 20:16 |
| th1a | ! | 20:16 |
| *** replaceafill has quit IRC | 20:18 | |
| *** menesis has joined #schooltool | 20:26 | |
| *** replaceafill has joined #schooltool | 20:30 | |
| *** replaceafill has joined #schooltool | 20:30 | |
| *** replaceafill has quit IRC | 20:37 | |
| *** replaceafill has joined #schooltool | 20:37 | |
| *** menesis has quit IRC | 20:37 | |
| *** menesis has joined #schooltool | 20:40 | |
| *** ignas has quit IRC | 21:14 | |
| *** menesis has quit IRC | 23:00 | |
| *** menesis has joined #schooltool | 23:27 | |
| *** replaceafill has quit IRC | 23:45 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!