Perl 6 - the future is here, just unevenly distributed

IRC log for #webwork, 2013-04-09

| Channels | #webwork index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:00 mgage_ yes -- we are far away from that -- but that is the direction that peter and david's code is going
00:00 mgage_ and that seems to be in line with what I have been hearing about modern web apps
00:01 goehle I thought about doing that, but that will break math3/math2
00:01 goehle and i'm concerned about changing lots of stuff when you are trying to polish for release
00:01 goehle we could take your twofactor approach
00:01 mgage_ me too -- but how big a break is it?  -- adding two lines of code to each of the system templates?
00:01 goehle I could do the workaround for 2.7
00:02 goehle and the better fix for develop
00:02 goehle umm, In theory its not too bad.  But I would be commenting out code in some pretty major files, and its not clear to me whta the ramificaitons are
00:02 mgage_ ok -- but develop might not even need the fix -- given how much it is changing
00:02 goehle it will certainly break custom themes
00:03 mgage_ in general though I'm happy with fixing things in two ways -- a short term fix which hopefully clears the way for a longer term fix later.
00:03 goehle develop will need the fix one way or the other
00:03 mgage_ yes -- but there  are not very many of those (as far as i know)
00:03 goehle maybe not as we have described it, but you have the same probelm
00:03 mgage_ ok
00:04 mgage_ we should get peter and david to chime in on that one
00:04 goehle yeah, I"m sure they have a plan for how they will deal with that.  Honestly its probably just to comment out the jquery calls in the perl files
00:05 goehle assume that going forward all themes will have at least jquery
00:05 goehle and bootstrap
00:06 mgage_ that would be my guess.   -- one could create a safety.  comment out the files in the individual modules and have them call SUPER to add the query files -- once that is working turn off the query lines in just one place in the SUPER output_JS in ContentGenerator.pm
00:07 mgage_ just in doing that you have refactored the existing code so that the bootstrap and jquery general calls are only made from one place.
00:07 mgage_ that's kind of worth doing all on it's own
00:08 mgage_ the output_JS subroutines have been added one at a time as they were needed so it's probably time to refactor anyway.
00:09 mgage_ whether it would be better to have the bootstrap/jquery calls in templates or in the perl modules I can't tell right at the moment, but with the refactoring one could make that choice pretty easily at the end -- and switch back and forth if necessary.
00:10 mgage_ also -- this is interesting -- if I  do the same experiment on hosted2  the tabs work correctly and no error messages.
00:10 mgage_ you could do this:
00:10 mgage_ git checkout myrelease2.7
00:10 goehle on hosted 2 you dont have bootstrap tabs
00:11 goehle they are tabber tabs there
00:11 mgage_ I see
00:11 mgage_ so I don't have the latest bootstrap.js to contend with.
00:11 goehle you dont have the latest math4
00:11 Paul_Pearson joined #webwork
00:12 mgage_ got it.
00:12 goehle which does some js trickery to turn tabber tabs into bootstrap tabs
00:13 mgage_ ok -- next question -- is there some fundamental rewrite that would be better than trying to change tabbed tabs to bootstrap tabs? -- something that would start out with bootstrap tabs?
00:13 goehle most certainly
00:13 goehle but you asked that I not modify the perl code when implementing bootstrap tabs for 2.7
00:14 goehle (thats why I'm hesitant to start now with this jquery thing)
00:14 mgage_ true enough.
00:14 goehle I have an idea
00:14 goehle and it would be a nice project
00:14 goehle but it probably collides with what david and peter are doing
00:15 mgage_ It's always a balancing act.  The version on hosted2 doesn't look too bad -- so perhaps it's better to stabilize it there and save the true fix for develop
00:15 goehle I recommend we go with my cludge
00:15 goehle its not elegent
00:15 goehle but its pretty safe
00:15 goehle and really doesn't impact performance much
00:15 mgage_ ok -- I'm game
00:15 goehle and will leave math2 completely intact
00:15 goehle (and other themes)
00:15 mgage_ so describe it once more
00:15 goehle I load jquery bootstrap in math4/system.template
00:16 goehle then call js_output
00:16 goehle then load jquery bootstrap again
00:16 mgage_ ok -- let's do that.
00:16 mgage_ that sounds stable -- and people can start refactoring from there -- they'll have something stable to start from
00:17 goehle actually
00:18 goehle I just have to load bootstrap again
00:18 goehle so there is only one duplication
00:18 mgage_ kk
00:18 goehle my idea was to try and move all of the formatting/script/whatever from webwork and into math4 without breaking anything
00:19 goehle so that the perl is just outputing divs with generic tags
00:19 goehle and then math4 adds the appropriate classes to make things work
00:20 mgage_ that's a good principle -- but sometimes it's actually worth fixing the original problem in the perl rather than jump through a ridiculous number of hoops. (Something I have to keep telling Davide -- who insists in working around some fundamental flaws in PG -- that would not be that hard to just fix. :-)  )
00:21 mgage_ yes -- and moving the perl code so that it just outputs content and divs seems to me to be an excellent idea.
00:21 goehle this was more to seperate out the formatting from the perl
00:21 mgage_ yes
00:21 mgage_ I don't think that is very different from what Peter and David are working on
00:21 goehle however, this would most likely overlap with the current gui overhall
00:22 goehle yeah
00:22 mgage_ they are using a framework called backbone which has a similar separation model
00:22 mgage_ it might even take over a fair percentage of the perl in webwork2 (although not the perl in the PG section) --
00:23 goehle its a big project
00:23 mgage_ node.js might also serve the gui model
00:23 mgage_ yes -- it's a major overhaul -- which is why it's not out yet :-)
00:23 mgage_ but it's getting closer.
00:24 goehle is this on develop or somewhere else?
00:24 mgage_ it's in develop --hang on a minute
00:25 mgage_ https://devel1.webwork.rochester.edu/webwork2/maa102_english/
00:25 mgage_ there are some timing issues.
00:25 mgage_ you sometimes have to wait 20 seconds for all of the javaScript to load
00:28 mgage_ see if profa/profa  login work now on the link above?
00:29 goehle login works
00:29 goehle some of the pages look radically different but seem a little broken
00:30 mgage_ that's about right -- and the slowness is probably also due to some bug -- it's not ready for prime time yet -- but you can hack on it and see where it is going.
00:30 mgage_ telephone brb
01:30 mgage_ joined #webwork
01:41 goehle looks like the upgrade worked
01:42 goehle or merge rather
01:42 mgage_ so far yes -- I just brought it down to my lap top.  -- Give me a few minutes to test and
01:42 mgage_ I'll push it up to my github and to hosted2
01:54 mgage_ joined #webwork
01:55 mgage_ ok -- hosted2 has all of your changes now (and a couple of mine)
02:01 goehle i'll take a look
02:01 goehle I also figured out what I can do that might be a nice project
02:04 goehle looks good on hosted
02:04 goehle too tired to really see if there are bugs
02:05 mgage_ ok -- looks good for the moment -- I'll play with it a little before I go to bed.  I had a plan to add the timedate picker to the ProblemSetDetails page.
02:05 goehle yeah
02:05 goehle you mentioed that too me
02:05 goehle I looked at it but it seems a little more complicated
02:05 goehle than I was willing to do at the time
02:05 mgage_ Mostly just taking the script written in Homeworksets editor2 and putting it in ProblemSetDetail
02:06 goehle it shouldn't really be that bad though.  I would just have to figure out how the js works
02:06 goehle yeah.  I seem to remember homeworkseteditor2 not being clear to me.  But this was at the end of the code camp so I might have not been firing on all cylinders
02:06 mgage_ I think I have an idea of how to make it work cleanly, although not elegantly -- the js is pretty clear -- it's just a matter of adjusting the names of the input text boxes
02:07 mgage_ it's actually simpler than the main homeworkseteditor2 because there is only one homework set available at a time
02:08 goehle anyway, my idea was, once david and peter get the perl code set.  I might try to recreate the current math4 as closely as possible using their perl backend and just js/css/backbone
02:08 goehle itll be a good test to see if the setup they have is flexible.  And some people will probably want to stick with a more classic look, so it may even be used
02:08 mgage_ sounds good -- you can help with comments on the API if you want -- I have to do that myself -- its on the openwebwork github
02:09 mgage_ true enough -- there are some difficulties with moving a large group of users on to new things ---- even when they are "clearly" better
02:09 mgage_ :-)
02:10 goehle I'll poke around a bit.  However, I should probably let them finish some of the perl first.  Just so we aren't stepping on each others toes
02:10 mgage_ the API is just a document -- so it should be the tools that you wish had available to accomplish your js/css/backbone front end
02:11 goehle that would be good to look at now then
02:11 goehle just to make sure its possible
02:11 goehle but not now now :)  I'm off to bed
02:11 goehle ttyl
02:11 mgage_ ttyl
03:09 rbeezer joined #webwork
03:16 mgage_ rbeezer:hi
03:16 rbeezer mgage: I've got some problems with hosted2, are you available for a minute?
03:16 rbeezer crossing in the ether!
03:17 mgage_ yep -- what are you seeing?  I just made some tweaks (sent in by Geoff)
03:17 rbeezer Individual problems do not want to render, for me, nor for students
03:17 rbeezer I can provoke the error, gimme a sec
03:17 mgage_ which course 210 or 220?
03:17 rbeezer 210
03:17 mgage_ k -- give me a minute
03:18 mgage_ I'm in your course -- which homework set?
03:19 rbeezer Call Stack:  http://pastebin.com/B3rN4rUk
03:19 rbeezer Chapter 9 is the current one causing problems, any problem
03:19 mgage_ I see- hang on one sec.
03:20 mgage_ I didn't update that course with the database change
03:21 rbeezer aah...
03:21 mgage_ one thing that is going to be iffy is that you may loose some past answer access
03:21 mgage_ the data is there -- but in a table that won't be accessible any longer
03:21 rbeezer OK, go for it
03:22 mgage_ ok -- should work now
03:22 rbeezer Yep!  Thanks!
03:22 mgage_ if you have 5 minutes try switching the theme to math4 in the course configuration
03:23 rbeezer I do
03:23 rbeezer Very modern  ;-)
03:24 rbeezer The Sage notebook could use a similar facelift
03:24 rbeezer Now I know what all the chatter has been about here on IRC
03:24 mgage_ it's hard to keep up -- the web interface is moving very quickly right at the moment
03:25 rbeezer HW Editor2: I like the different color buttons (lt/dk blue) - it helps you see what is important (Take Action!)
03:25 mgage_ pick one of the older sets, look at a problem and then click editor 3 -- that's one of the more impressive changes from my point of view
03:25 mgage_ choose "view" tab and then take action
03:26 mgage_ much less confusing than switching browser windows
03:27 rbeezer Very nice! Except I have no scroll bar, no resizing on the overlay?  I like no new-tab in my browser.
03:28 mgage_ I get a vertical scroll bar on chrome
03:29 mgage_ I'm using a huge screen, so I might not notice this
03:29 rbeezer Firefox/Ubuntu/Huge screen, no scroll bar  :-(
03:30 mgage_ yep -- I see that --
03:30 mgage_ on firefox/macos
03:30 rbeezer thought maybe it was "hidden" - no action if I try in the right places
03:30 mgage_ it's using twitter bootstrap css  and query -- but I guess they are not perfect across browsers
03:31 rbeezer Welcome to cross-browser testing  :-(
03:31 rbeezer Great idea to have the overlay, that sort of thing will help a lot
03:32 mgage_ I'm reporting the bug as we speak :-)
03:32 rbeezer I'm emailing the students as we speak, WW is up!  Then calling it a night
03:33 mgage_ kk -- sorry for the outage -- I thought I had upgraded all of the classes
03:33 mgage_ scroll bar works in safari  -- so I guess web-kit is ok
03:34 rbeezer no problem, Mike - I appreciate the quick help
03:35 rbeezer I sent email to everybody, so maybe you will see it, just ignore
03:35 mgage_ np
03:35 rbeezer off-duty
03:35 mgage_ it's nice to hear what's going on in the background   ttyl
04:08 mgage_ joined #webwork
11:15 mgage_ joined #webwork
12:21 mgage_ joined #webwork
12:26 goehle joined #webwork
12:41 mgage_ joined #webwork
12:59 goehle joined #webwork
13:02 goehle joined #webwork
13:06 mgage_ joined #webwork
14:16 mgage_ joined #webwork
14:27 aubreyja joined #webwork
15:37 Pbelley joined #webwork
15:37 Pbelley hi
15:37 mgage_ hi
15:37 Pbelley how are you today?
15:37 mgage_ good -- and you?
15:37 Pbelley good
15:38 mgage_ how are things going with your math editor?
15:38 Pbelley I have question about localization
15:38 mgage_ ok
15:38 Pbelley the math editor is released
15:38 Pbelley I have done a pull request as saspatou
15:38 mgage_ where did you send the pull revuest to?
15:39 Pbelley SourceForge
15:39 mgage_ ok -- that's why I didn't see it.  would it be possible to send it to   github.com/openwebwork/webwork2   branch develop?
15:40 Pbelley oups
15:40 Pbelley mine mistake
15:40 Pbelley I've pull it to github
15:40 Pbelley It's another project who I use SourceForge
15:41 Pbelley It's suppose to be here
15:41 mgage_ https://github.com/openwebwork/webwork2/pulls
15:41 Pbelley as release candidat?
15:42 mgage_ I suggest that you request that it be pulled to devel -- at least for the moment
15:42 mgage_ the branch devel
15:42 mgage_ (notice that webwork2-dev is not being used any more (long story)
15:43 Pbelley ok
15:43 Pbelley I'm set to it
15:44 Pbelley then if you want I may zip mine local server and send it to you via skype?
15:46 mgage_ I haven't received large files via Skype before.    -- but I'm not sure we are talking about the same thing
15:47 mgage_ do you have your math equation editor available on git hub?
15:47 Pbelley not yet
15:48 mgage_ ok -- I see
15:48 mgage_ sure go ahead and try to send me a zip file via Skype and I'll see what happens -- if it's not too large you can probably email it to me -- my mailbox will take fairly large files.
15:48 mgage_ anyway -- you had a question about localization
15:49 Pbelley I will merge to version 2.7 of webwork and I'll add equation editor and I'll do pull request
15:49 Pbelley It's okay like that?
15:50 Pbelley For de localisation, using poedit. wich path I have to add to path list to have every String?
15:50 mgage_ I think that it would be better for this first submit to merge to develop, add the equation editor and then issue a pull request -- I'm trying not to add features to 2.7 right at the moment so that I can finish it and release it
15:51 Pbelley ok
15:51 Pbelley now I'm in webwork2-dev server
15:51 Pbelley in localhost
15:52 Pbelley the pull request is suppose to be here
15:52 mgage_ ok -- it's better if you could use  openwebwork/webwork2   and pull the develop branch from there
15:52 Pbelley ok
15:53 mgage_ are you using poedit to find all of the strings in the webwork repository? or to create the translation dictionary
15:54 Pbelley to find all string because i'm translating set makers part
15:54 Pbelley I've add all needed maketext
15:54 Pbelley but I need to localise them to merge with webwork2.pot
15:55 mgage_ ok -- I'll have to look that up and send you an email -- it has been a little while since I've used poedit for that -- there were some perl scripts that I think worked better for finding all of the strings -- but I will have to go review my student's notes to see what they were.
15:56 mgage_ I understand. I'll try to send you the right settings this evening.
15:57 Pbelley thanks you so much !
15:57 mgage_ merci bien pour votre efforts
15:58 Paul_Pearson joined #webwork
15:58 Pbelley merci :)
17:25 Paul_Pearson1 joined #webwork
17:28 goehle hey mgage
17:32 mgage_ joined #webwork
17:37 rwerw joined #webwork
17:38 rwerw how to write No solution for system solving equation
17:43 goehle hey mgage_
17:43 mgage_ hi
17:43 goehle have you looked at math4 using ie?
17:44 goehle it leaves soemthing to be desired
17:44 mgage_ no -- 'fraud not -- I don't have that on Macos
17:44 mgage_ afraid not
17:44 mgage_ that doesn't surprise me -- everyone has trouble with the older versions of i.e.
17:45 mgage_ which version are you using -- is it 9 ? that is supposed to be reasonable.
17:45 goehle I was using whatever version they have on our classromo machines
17:45 goehle I"m trying to get my virtual machine working now to check it out
18:52 Pbelley joined #webwork
18:59 Paul_Pearson joined #webwork
19:06 BelleyPa joined #webwork
20:37 mgage_ joined #webwork
20:59 Pbelley joined #webwork
21:10 Paul_Pearson joined #webwork
21:59 mgage_ joined #webwork
22:31 aubreyja joined #webwork
22:41 mgage_ joined #webwork
23:55 Paul_Pearson joined #webwork

| Channels | #webwork index | Today | | Search | Google Search | Plain-Text | summary