# IRC log for #webwork, 2013-06-05

All times shown according to UTC.

Time Nick Message
00:11 whytheplatypus \exit
13:47 ChanServ joined #webwork
13:55 mgage joined #webwork
14:28 mgage_ joined #webwork
16:50 goehle joined #webwork
16:50 goehle hey mgage
16:50 mgage hi
16:50 goehle i'm trying to test out the lastest relase/2.7
16:50 mgage kk
16:51 mgage you saw that Gavin is having trouble with it -- I can't figure out why since I don't see the same effects he sees.
16:51 goehle I did see your email
16:51 goehle but it was unclear to me what his trouble was
16:52 goehle I can't even get my servers tarted ;(
16:52 mgage by the way you can follow along instructions at http://webwork.maa.org/wiki/Github (using other branches of WW topic ) to see if I've left anything out.
16:53 mgage what does the error message look like?  -- I can send you some of Gavin's picks in a minute.
16:53 mgage pics
16:53 goehle Number found where operator expected at /opt/webwork/webwork2/lib/WeBWorK/ContentGenerator.pm line 2019, near "# FIXME -- this gives the correct locations for release/2.7"
16:53 goehle
16:53 goehle but I've gone over the code and cant find any issues
16:53 goehle and its not somethign that was changed in the latest pull anyway, as far as aI can see
16:54 goehle I mean that *should* be a comment
16:54 mgage this kind of error shouldn't prevent start up
16:54 mgage usually this is just an annoying warning message.
16:54 goehle Global symbol "$site_url" requires explicit package name at /opt/webwork/webwork2/lib/WeBWorK/ContentGenerator.pm line 2016.\nGlobal symbol "$MathJax" requires explicit package name at /opt/webwork/webwork2/lib/WeBWorK/ContentGenerator.pm line 2016.\nsyntax error at /opt/webwork/webwork2/lib/WeBWorK/ContentGenerator.pm line 2019, near "# FIXME -- this gives the correct locations for release/2.7"\nCompilation failed in require at (eval 1095) line 3.
16:54 mgage when you restart the apache server to you just get the usual information report about the server? or do you get this warning as well.
16:55 goehle its attached to a bunch of compillation errors
16:55 goehle it looks like its trying to parse the comments as perl code
16:55 goehle which I dont understand
16:55 mgage kk.
16:55 mgage let me look at this -- looks like a typo near line 2016 in CotentGenerator.pm
16:56 goehle https://github.com/openwebwork/webwork2/blame/release/2.7/lib/WeBWorK/ContentGenerator.pm
16:57 goehle I wasn't able to see one
16:57 goehle it uses the new // operator
16:58 mgage that could be it -- perhaps using that is premature.
16:58 mgage type    perl --version into your command line and see which version of perl is being used
16:58 goehle aaaah
16:58 goehle that woudl explain it
16:58 goehle 5.8.8
16:58 goehle for me
16:58 mgage this can be a little tricky because you want to know which version the apache server is using
16:59 mgage yep that would do it.  I have 5.12 --- the current version is I believe 5.18
16:59 mgage the push to "modern perl" seems to have started with about 5.10
17:00 mgage I wonder if that could have anything to do with Gavin's problem although that seemed purely CSS to me.  I'll send the pics.
17:08 goehle joined #webwork
17:08 goehle CDR@%7ut
17:09 mgage ???
17:09 goehle nothing :P
17:35 goehle hmm
17:35 goehle so whats going on with his css?
17:37 mgage I really couldn't tell at least from a distance.  I asked him the usual debugging questions.  It's possible he doesn't have the latest version of something although he just pulled from the standard distribution.  Having a link hardcoded in the wrong place might cause something like this but I asked him to check for links -- as you see.  That was an issue for Peter Staab who had a bunch of local links that ended up causing problems.
17:38 mgage in one of the pics the size of the info box stretches across the page (hence the yellow color on the left hand side)
17:40 goehle is this hapening for all of them, but you cant see it because of the coloring
17:40 goehle or just sometimes?
17:41 mgage I don't know.
17:42 mgage if you have some time you can ask Gavin that.  The other thing I thought of which I haven't emailed him about is to remove the conf/templates directory (or move it out of the way) -- it's possible something in there is being read that interferes with the real css in the themes directory
17:43 mgage i just figured the css discussion might be getting out of my league and you might have a better chance helping him out.
17:43 goehle yeah
17:43 goehle its hard to tell if its always happening
17:44 goehle or just happening sometimes
17:44 mgage by the way math4 theme worked ok for him -- but if there are going to be issues in general for people switching to 2.7 release I'd like to know about it sooner rather than later.
17:44 goehle math3 hasn't gotten the kind of love math4 has gotten
17:45 goehle so its quite possible there are math3 specific  issues
17:45 goehle side question
17:45 goehle how do I reset a password for someone in the admin course (me!) using the command line?
17:45 mgage that's ok if math3 is not really fancy but I want it to be acceptable.
17:46 mgage there is a script for that in webwork2/bin  called addprofa
17:46 mgage (no instructions of course :-) )
17:47 mgage what you do is, in   webwork2/bin type   wash (for WeBWorK shell) and then the course name -- in this case admin
17:47 mgage then to the wwsh command line type    source  addprofa   and profa will be added.
17:48 mgage you need to copy profa first and enter the appropriate information.  my version is called addGage and has my credentials hard coded into it.  Then I can simply add myself to any course to gain access when I'm hosting a whole bunch of courses on a server.
17:49 mgage wash should be  wwsh  by the way in the previous posts (stupid auto spell)
17:50 mgage yes -- but you need to make a copy of addprofa to a file called addgoehle   and then add your credentials into that file
17:50 goehle wait
17:51 goehle right
17:51 goehle rigth
17:54 mgage but I expect you will find having an addgoehle script useful
17:56 goehle that is useful
17:56 goehle although I ended up just using putPassword
17:57 goehle Gavin's issue does look like css
17:57 goehle but I"m not having that problem on my (newly upgraded) machine
17:58 mgage me either -- perhaps we can get a link so that we can view it directly --even as students -- and get some idea of what the problem is.
17:58 mgage using firebug like toolks
17:58 mgage tools
17:59 mgage by the way if the user already exists you may have to change the "addXXX" commands to "putXXX" commands if the record already exists.
18:07 goehle right
18:07 goehle do you have an install of the latest 2.7 available
18:07 mgage yes -- I have it running on my laptop
18:07 goehle does the footer on the problem list page for an individual homework look broken to you
18:12 mgage I don't think so -- not sure if we have the same page in mind.  the information about theme, ww version and pg version is all there.
18:12 goehle on my page its intersecting with the navigation bar really badly
18:12 goehle I cant figure out why
18:13 mgage on mine it is not centering -- looks like there might be a local override from that page.
18:16 goehle hmm
18:16 goehle this is in math3?
18:19 mgage I can't get it to fail anymore.
18:19 mgage all the math4 footers are centered.
18:19 mgage all the math3 footers are now left justified
18:19 goehle I sent you a couple of screenshots
18:19 mgage either is fine -- I recommend leaving it like this
18:20 mgage mine look like the first picture
18:21 goehle the wierd bit is that I'm getting both behaviours
18:21 mgage oops -- no they look like the second one in the example I just tried.
18:25 mgage I'm seeing both as well.  It's a "float" issue and it depends on how long the side bars on the left and right are.  I think what is needed is a clear:both div just above the footer.  Does that seem right to you?
18:26 goehle I think its the preamble/postamble issue I fixed in develop
18:27 goehle the reason its different on the Problems list page
18:27 goehle and the homeworks list page
18:27 goehle is that the hproblems page generates a problem for the info panel
18:27 goehle which we know produces a stray end div tag
18:28 mgage I see.  We should fix that here as well I think.
18:28 goehle just to make sure I"m right
18:29 goehle does commenting out the problem Postamble line in defaults.conf fix the problem for you
18:29 goehle ?
18:34 goehle one side note
18:34 goehle you changed the default editor to pgproblemeditor3
18:34 goehle but that doesn't work in math3
18:39 mgage good point I guess. I also made math4 theme the default -- my idea is to make sure that people use it at least once -- I think it's a big improvement.  However you have a good point -- and at the very least I'll need to document this fact very carefully.
18:39 goehle yeah
18:39 goehle I'm confused now though
18:39 goehle didn't I add basic jquery functionality to math3
18:39 goehle or was that only in develop?
18:40 goehle I think that was only in develop
18:40 mgage I think so.
18:40 goehle ok so should I change the default editor to PGProblemEditor2?
18:40 goehle or do youthink math4 adoption will be good enough that it doesnt matter
18:41 goehle and do you want me to change the default theme to math4
18:41 goehle (its still math3 currently)
18:41 mgage actually check pgproblemeditor2 carefully to make sure all of its buttons work -- it's not much different from pgeditor1
18:41 goehle right
18:41 goehle I remember a bunch of those were broken
18:41 mgage I thought I'd made math4 the default theme, but if it isn't change it.
18:42 mgage we could expose both editor1 and editor3 as defaults.
18:42 mgage that might be best.
18:42 goehle yeah
18:42 goehle math3 still has trouble with small screens
18:42 mgage For the css tweaks that really should be back ported to 2.7
18:42 goehle because of the way modal works
18:42 goehle what should be backported
18:42 goehle the preamble/postamble fix?
18:43 mgage I think that is ok.  I regard that as a feature and small screens can upgrade to math4 theme
18:43 mgage I think preamble/postamble fix, and perhaps the minimal jquery support for math3?
18:45 mgage one way to do the cherry pick (if you have time) is to pull a fresh copy of 2.7,  manually add the changes to the math3.css files and whatever we did for preamble/postamble and then issue a pull request.
18:45 mgage it might cause a small conflict when we merge back into develop but I don't think it will be too bad.
18:45 goehle there isn't really much of a way around it
18:45 goehle pgproblem editor 2
18:45 goehle works ok, actually, although the append function is broken
18:46 mgage what did we end up doing to fix preamble/postamble?  I remember why the problem is there but not which fix we decided on .
18:46 goehle we decided to remove preamble/postamble completely
18:46 goehle and leave it up to the content generators to put the problem body in div tags
18:47 mgage kk -- so that is a simple change to default.config and math3.css
18:47 goehle yeah
18:47 goehle and the issue with PGProblemEditor2 was a typo
18:47 goehle since its also got the tabs
18:47 goehle I recommend we go with
18:47 goehle problem editor 2 and problem editor 3
18:47 mgage that's what I expect.
18:48 goehle so the look will be consistent, and people will have a fallback if the modal stuff doesn't work
18:49 mgage kk then let's go with editor 2 and 3.    If you can fix the defaults.config/localOverrides so that math4 theme is the default, those two editors are defaults,  the postamble/preamble is empty and so that the right css is being used in both math3 and math4
18:49 goehle right
18:49 goehle I should be able to do that
18:49 mgage that would be great.  once you send the pull request I'll look it over quickly and pull it in.
18:50 mgage there is (or should be) a lot of this last minute detail fixing just before a release.
18:50 mgage it's hard for one person to catch everything.
18:50 goehle yeah
18:50 mgage I appreciate the help :-)
18:51 goehle no worries
18:51 goehle half of it is indirectly my fault anywya :)
19:02 goehle hmm
19:03 goehle check out the library browser in math3
19:03 mgage kk
19:04 mgage there is an empty box at the bottom on the left.  That has been there for a while.
19:04 mgage for some reason I couldn't get it to disappear entirely when there were no siblings.
19:05 mgage I think something similar happens for logout and perhaps login
19:05 mgage I think we can live with that.
19:05 mgage unless the fix is super easy
19:05 goehle i'm seeing worse things
19:06 mgage this is library browser 1-- right?
19:06 goehle yeah
19:06 mgage there are 4 of them around
19:06 mgage what are you seeing -- I haven't noticed anything yet.
19:06 goehle I'm sending you a screenshot
19:07 aubreyja joined #webwork
19:08 mgage aubreyja: hi
19:12 aubreyja Hi - how's it going?
19:15 aubreyja mgage
19:17 goehle hey mgage
19:17 goehle the library browser isn't working for me ;(
19:17 mgage aubreyja: it's going ok -- geoff and I are working on last minute details for release/2.7
19:17 aubreyja ah, need help?
19:17 mgage I think I have the chinese.po file taken care of for release/2.7
19:18 goehle right now I'm trying to find out why the library browser doesn't add when you click add
19:18 aubreyja great
19:18 mgage let me check it out.  but the screen shot you sent me of library browser 1 used math4 theme and it seemed to look fine
19:18 goehle I suppose it was
19:19 goehle it took me by suprise since i"m used to the older library browser with the giant table
19:19 mgage ok  -- that's a bug and I suspect that is the subject of this commit to develop https://github.com/openwebwork/webwork2/commit/b48766d8ae8f57b25ac4fc6fe4fe55706540e7b3
19:20 mgage by jj.
19:20 goehle ok
19:20 goehle I"ll backport that too
19:22 goehle hmm
19:22 goehle that doesnt seem to fix it
19:25 aubreyja we need to be sure HTML::Scrubber ends up in check_modules.pl
19:27 aubreyja we're still using defined(@array) where we should just be using @array (causes deprecation warning in perl 16.1+
19:27 aubreyja 5.16.1+
19:28 aubreyja also, 2.7 gives me this warning when I start it: WARNING: database.conf.dist is newer than database.conf: UPDATE database.conf!
19:28 aubreyja but thats no longer true right? It entirely runs off of database.conf.dist now?
19:28 mgage jj upgraded the librarybrowser1 so that it uses some ajax and such like but still looks largely the same as the old library browser.  For the moment it's more usable than the the other options which made more ambitious changes.
19:29 goehle hmm
19:29 goehle I still can't get it to add problems
19:30 aubreyja goehle - it won't add problems to sets with math4 or other themes too?
19:30 goehle in general
19:30 goehle its something about the webservice I think
19:31 goehle I get soem console log warnings when I click on stuff
19:32 goehle it does work on develop though
19:32 goehle hmm
19:32 goehle so that commit must fix it
19:32 goehle so I must not have backported it corredtly
19:32 aubreyja also, the 'Your modelCourse in the courses directory is out of date…' warning - this is expected now or did I screw up?
19:33 mgage aubreyja:  that's actually useful -- you should move database.conf out of the way -- look at the http://webwork.maa.org/wiki/Github while you are upgrading -- you can help debug that also.  it mentions what to do with database.conf
19:34 mgage the bug with the add button was fixed by jj in develop but not in release/2.7 (according to his notes at least)
19:34 goehle yeah
19:35 goehle unfortunately some things have changed  in SetActions since 2.7
19:35 goehle so I can't just coyp paste his changes
19:35 mgage kk -- I'll add HTML::Scrubber right now
19:35 mgage yeah -- I was afraid of that
19:36 mgage we may have to just commit a different kind of fix for add for release/2.7
19:36 aubreyja yup, e.g. this ain't right from that page: cp -RPpi /opt/webwork/courses/modelCourse /opt/webwork/courses/modelCourse
19:37 mgage oops -- should be from webwork2/courses.dist
19:38 aubreyja fixed
19:39 aubreyja also, I'd prefer that when working under webwork/ we use relative paths - I'm doing this in my home dir on my mac (for example) and would just like to copy and paste. I'll fix that too.
19:39 mgage HTML::Scrubber is in check_modules in release/2.7 as of a few days ago -- do a git pull to update
19:39 aubreyja ok, I didn't actually run check_modules.pl - just noticed it when I tried to start it up :)
19:40 aubreyja Also, maybe a button to upgrade the modelCourse? Or at least instructions in the error message….
19:41 mgage ok -- I actually used complete paths for the cd  to make sure that people were really sure about where to work -- you can change it to relative paths if you wish if you think it will work better.  I was worried that people might get lost.
19:42 mgage I was wondering about changing the instructions so that instead of copying courses.dist/modelCourse to courses/....
19:43 mgage we could create a link.  Once that is done the modelCourse upgrades are automatic.  What do you think?   We could do that in the instructions for this release if you think it's a good idea.
19:43 aubreyja That's a good idea.  I'd say do it
19:43 mgage ok -- you try it also to make sure it works on your machine.
19:43 aubreyja sure - doing it now
19:44 goehle hmm
19:45 aubreyja yep, that error message goes away, so the admin course is seeing the modelCourse through the link ok
19:46 aubreyja Uh oh, the upgrade script is going to try to upgrade modelCourse.old and fail since there are no db tables of that name...
19:47 goehle I think I fixed it
19:48 mgage ok -- perhaps we should have modelCourse moved  to ../modelCourse.old so that it doesn't cause that problem.
19:48 aubreyja oh, no, it did upgrade the modelCourse.old, presumably by creating the tables.  Maybe we can just have it deleted so we don't have an unnecessary set of tables in the db?
19:49 mgage at this point you can delete modelCourses.old from the web using the admin course -- that will work unless there are permissions issues.
19:49 mgage but for instructions we'll have modelCourse moved and saved somewhere outside the courses directory to avoid this issue
19:49 aubreyja right, but following the current instructions users will have a modelCourse.old that needs upgrading or deleting.
19:49 aubreyja ok right - that's my point
19:50 mgage goehle: you fixed the add button?
19:50 goehle yeah
19:50 mgage cool
19:50 aubreyja brb
19:50 goehle looking at the library browser
19:50 goehle is there supposed to be boxes around the problems?
19:50 aubreyja joined #webwork
19:51 mgage let me check
19:52 aubreyja in math3 I get  the page footer in the middle of the main menu and the main menu overlaps it where they intersect
19:52 aubreyja I see this when looking at a set with 1 problem in it
19:52 goehle right
19:52 goehle but only on the problems page
19:52 goehle but not on the other pages
19:52 goehle like the problem set list page
19:52 goehle or a single homewokr page
19:52 aubreyja yes,
19:52 goehle yup
19:52 goehle fix inc for that
19:53 aubreyja kk
19:53 mgage no -- I don't think the library boxes had boxes around them.
19:53 goehle ok
19:54 goehle hmm
19:54 goehle well
19:56 mgage I think that jj had disabled the preamble/postamble variables to display things in browsers. so our change doesn't make any difference in the library.   -- I've checked a couple of older installations and I don't see any boxes in the library browser.
19:56 goehle ok
20:01 mgage aubreyja: could you pull the chinese changes into release/2.7.   Are those the most recent versions of the .po files?  I created the .mo file and added it.
20:01 mgage brb
20:02 aubreyja Those are the most recent changes - I'll do the pull
20:02 mgage kk ty
20:12 mgage back
20:12 goehle i'm going to submit a pull request for the 2.7 backports
20:12 mgage kk  -- I'm looking at your pull request to develop now
20:13 goehle thats a minor bugfix
20:13 goehle hmm
20:13 goehle so my pull request to 2.7 is including my changes to site_info.txt
20:13 goehle I want to keep those changes for my site
20:13 goehle but I shouldn't pull them
20:15 mgage this is a bit of hack but do this:  move site_info.txt to site_info.save;   git checkout site_info.txt ;(get's the old one)   git add site_info.txt and then git commit -m"fixed site_info"
20:15 mgage then you can move site_info.save back to site_info.txt
20:15 goehle sure
20:15 goehle I was wondering if there was a better way :)
20:16 mgage git stash is supposed to do this kind of thing, but I'm a little unsure about it.   git stash; git checkout site_info.txt; git add and commit  and then  git stash pop to get site_info back
20:18 mgage I'll wait until you've fixed the site_info file before doing the pull
20:18 goehle done
20:23 goehle Can't use an undefined value as a HASH reference at /opt/webwork/webwork2/lib/WeBWorK/PG.pm line 263.
20:23 goehle I'm getting a lot of these
20:23 goehle nm
20:23 goehle my bad, I forgot to restart apache
20:24 mgage in your pull request there is a change in SetActions.pm -- is that your change?
20:24 goehle yah
20:24 goehle at least that was part of that fix in the other commit
20:25 mgage kk
20:25 goehle one thing
20:25 goehle the edit links all point to PGProblemEditor1
20:26 mgage the edit links in the Homeworks editor all do that --- the links from a problem itself gives you a choice.
20:26 goehle right
20:26 mgage I think we can leave it like that
20:26 goehle ok
20:26 mgage when they start complaining that they want editor3 we can do a "bug fix"
20:27 mgage by the way you can work around this quickly by looking at the url and adding 3 after the PGProblemEditor node. -- I often do this
20:27 goehle sure
20:28 mgage on the other hand I seldom edit from the homework editor page -- usually from the problem itself -- not sure if that is true of everyone
20:29 mgage ok -- I'm merging the pull request.  your changes fixed the problem with "add" in the problem library browser? I didn't see any changes in setActions that would have affected that.
20:30 goehle it was working for me
20:30 mgage kk I'll have it in a minute
20:30 mgage and I'll test it
20:32 goehle although now I'm having trouble getting the library browser to show problems
20:34 mgage add was actually working ok in my version of release/2.7 library browser before this update -- not sure why
20:34 mgage I'm pulling the latest we'll see what happens.
20:34 goehle ok
20:39 mgage the library works fine for me -- using math4 theme -- I'lll try math3 theme
20:40 mgage yeah -- it's working there for me also
20:41 goehle great
20:42 mgage did the library browser show problems for you?
20:42 goehle no, but i think its because I moved over to a class that had lots of stuff in it so I could see everything populated
20:42 goehle and my links to the natinal problem library point to the wrong place now
20:43 goehle yeah
20:43 goehle it works now
20:43 mgage quite likely -- for the next release we should put something in the upgrade courses script of the admin page that checks the library links
20:43 mgage I don't think I want to do that just yet
20:44 goehle IT shouldnt matter for me
20:44 goehle new courses will have the right links
20:45 mgage if you have updated modelCourse -- yes.
20:50 mgage aubreyja:   about using relative links   - I realize it makes copy paste a little harder if you have a different base for webwork I think having the complete path when I do cd will help reduce mistakes.
20:50 aubreyja np, it's not a big deal - we can go with that if you prefer
20:51 goehle hmm
20:51 goehle another bug
20:52 goehle so the "select all" buttons
20:52 goehle on basically all of the pages
20:52 goehle are broken
20:52 mgage I think for the moment.  I tend to like to cd to the directory where I'm working.  the alternative could be to make all commands as if you sat in the /opt/webwork directory -- which is easy if you do copy paste somewhat automatically but not as easy if you are checking things out as you update.
20:53 mgage most likely this is a javaScript that isn't being found -- the location of javaScripts is quite different in release/2.7 and in develop
20:56 goehle I tracked it down
20:56 goehle the name of some of the forms has changed
20:57 mgage kk
20:58 mgage aubreyja: I updated the directions for the modelCourse in the wiki github document.
20:59 mgage we'll probably have to duplicate some of that information in other places as well.  this is mostly directed at those updating from some version prior to 2.6
20:59 mgage but there are some issues that apply to those upgrading from 2.6 as well.
21:00 aubreyja looks good. any reason not to just delete the old model course?
21:00 goehle Hey
21:00 goehle I submitted a new pull request for the submit buttonthing
21:01 goehle I'm going to head off for a bit
21:01 goehle feel free to leave that open
21:01 goehle and when I get back I can help out fixing any other things that pop up
21:04 mgage kk -- here is something up your ally.  If I view a list of problems in library browser 1 and then click the randomize icon -- then when the problem comes back the width changes.
21:06 mgage the update doesn't seem to know how big the page is
21:06 mgage talk to you later.  thanks for the help.
21:38 mgage joined #webwork
22:46 goehle back mgage
22:48 mgage hi
22:48 goehle taking a look now
22:49 goehle I also have something you might want to look at
22:49 mgage math4 -- but it might be true for math3 as well
22:50 goehle In math3 the datepickers on the ProblemSetDetail page produce a lot of javascript errors
22:55 mgage I'm seeing something slightly different .  For homework editor2  the date pickers work on both the page that lists the entire homework set and also on the page where only the dates are allowed to be changed.
22:55 goehle this would be the Set Detail page
22:56 mgage for homework1 it works on the page with the entire homework but doesn't work at all on the page where only the dates are allowed to be changed.
22:57 mgage I don't see any error messages however in chorme
22:57 mgage chrome
22:57 goehle huh
22:57 mgage it's my work so I can fix some of it.
22:58 mgage (actually I was fixing work done by Grant He, he didn't quite get a chance to finish it.)
22:58 goehle ah nm
22:58 goehle I'm in math4
22:59 goehle or rather
22:59 goehle math4 on the develop branch
22:59 goehle Uncaught TypeError: Cannot set property 'parseDateTime' of undefined jquery-ui-timepicker-addon.js:1010
22:59 goehle Uncaught TypeError: Cannot call method '_newInst' of undefined
22:59 mgage that might not be up to date yet.  -- in particular it probably doesn't load that jquery file
23:00 goehle I did fix the jquery load
23:00 mgage make sure you can see all the icons in the date picker.  there was a time when some of those were missing
23:00 goehle I cant see the time picker at all
23:00 goehle and on the homework sets editor page I can see the time picker but cant move the time sliders
23:00 mgage I think I have the icons fixed but it might just be that I have them cached.
23:01 goehle Uncaught TypeError: Cannot call method 'addClass' of undefined
23:06 mgage kk.  somehow I thought I had fixed all of this.  --
23:06 mgage here is what I get:  for homeworkset2 -- the date picker works
23:08 mgage for homeworkset1 the date picker works for the page with all the homework but not for the page with only dates -- I can fix that.
23:08 mgage I'll need more detail about the example where the time slider doesn't work
23:09 goehle for me on homework sets 2
23:09 goehle when I use edit selected sets
23:09 goehle I can open the date picker thing
23:10 goehle and I can select dates
23:10 goehle but if I try to use the time slider I get
23:10 goehle Uncaught TypeError: Cannot call method 'addClass' of undefined
23:10 goehle which means that some jquery is coming up empty when it shouldn't be
23:11 mgage and math4 theme?
23:11 mgage using chrome?
23:11 goehle yup
23:12 goehle The page at https://webwork2.wcu.edu/webwork2/TestCourse/instructor/sets2/ ran insecure content from http://code.jquery.com/ui/1.10.2/themes/smoothness/jquery-ui.css.
23:12 goehle I also get this error
23:12 goehle which I have been inoring but maybe I shouldn't
23:12 mgage ok -- I haven't been able to reproduce this.  about the only difference is operating system and perhaps which version of chrome
23:13 mgage likely that is an error that would cause trouble.
23:13 goehle I'll try to get the css file into the server then
23:13 goehle I also wasn't able to reproduce the width changing bug you were talking about
23:15 mgage but I'm not getting that error on my laptop -- the security might be different on your test server -- but that could be tricky since there are lots of calls to jquery.com -- can you make that call https?
23:15 goehle shouldnt it be alocal call anyway?
23:16 goehle since you cant tell which servers will be running http/https?
23:16 mgage it might have been generated from within the date picker java script
23:17 goehle no
23:17 goehle it was in the .pm
23:17 goehle I tried to get that file and make it local
23:17 goehle but its just made htings worse
23:17 goehle now there are lots of.png files which are missing
23:17 goehle and it didn't fix the sliders
23:18 mgage yeah -- the .png files were one of my issues.
23:19 goehle hmm
23:19 goehle well lets go back to the library browser
23:19 goehle so you were saying when you click the randomize button you get different sizes
23:19 goehle that doesn't happen for me
23:21 mgage ok -- it's not happening for me on hosted2 either.
23:24 goehle hmm
23:27 ionparticle wow, marathon debugging session?
23:28 goehle a bit
23:28 goehle we are getting 2.7 ready to be merged into the master branch
23:28 goehle btw
23:28 goehle a bunch of changes to math4 just got loaded into the develop branch
23:28 goehle I tried not to clobber any of your stuff but its possible some things got overwritten
23:28 mgage ok -- I think the date picker needs a bit more work.  I'll try to pull down all the relevant query files and host them locally.
23:29 mgage ionparticle:  I also cherry picked some of your changes to put in 2.7 since they seemed like really nice additions.
23:29 ionparticle awesome
23:30 ionparticle <goehle>  The page at https://webwork2.wcu.edu/webwork2/TestCourse/instructor/sets2/ ran insecure content from http://code.jquery.com/ui/1.10.2/themes/smoothness/jquery-ui.css.
23:30 mgage if you recognize some behavior I missed during the cherry picking you can add it and send a pull request to release/2.7
23:30 goehle yeah
23:30 goehle chrome doesn't like cross stite scripting
23:30 goehle I'll be back I have to go for a bit
23:30 mgage you can view what it oohs like at  https://hosted2.webwork.rochester.edu/webwork2/maa105
23:31 ionparticle just need to change the include to remove http:
23:31 ionparticle it just wants everything to be https if you're on a https page
23:31 mgage with https  it seemed to loose the link to the images.
23:31 ionparticle hmm
23:32 mgage I'm thinking maybe I'll just host the code locally?
23:32 mgage then I can be pretty sure no browser will get upset about domain changes.
23:33 ionparticle which images does it have a problem with?
23:33 mgage the little left and right images that appear in the date picker window.
23:34 ionparticle joined #webwork
23:35 mgage ionparticle: the images might be a separate issue.  my laptop may have  some stuff cached.
23:36 mgage you can see the behavior at https://hosted2.webwork.rochester.edu/webwork2/maa105/instructor/sets2/ if you edit the dates on the homework set using chrome with the web inspector turned on
23:36 mgage [blocked] The page at https://hosted2.webwork.rochester.edu/webwork2/maa105/instructor/sets2/?eff…age&theme=&key=cZlMByogyzKNJNYXueJwroXq4rvgPWh2&editMode=1&visible_sets=hi ran insecure content from http://code.jquery.com/ui/1.10.2/themes/smoothness/jquery-ui.css.
23:37 ionparticle ok, trying to recreate it on my dev server
23:38 mgage the images I think I can fix, just by including them on the server.
23:42 ionparticle haha
23:42 ionparticle giant list of errors trying to start the server
23:42 goehle_ joined #webwork
23:43 mgage does the procedure at http://webwork.maa.org/wiki/Github help?  That needs to be debugged by more people.
23:43 ionparticle uhh
23:43 ionparticle uh-oh
23:43 ionparticle my server is still on perl 5.8.8
23:44 mgage aaah.  It could be we're getting a little aggressive on that -- I thought going to perl 5.10 would be safe :-(
23:45 ionparticle it's fine, we can deal with it
23:45 ionparticle RedHat is just amazingly slow at upgrading their repos
23:45 mgage I think there are only one or two places where this breaks (I used the   a//b  symbol for  (defined(a) and a)? a : b  )
23:45 mgage it's so much nicer.
23:46 goehle_ yeah, I was having trouble with this earlier
23:46 ionparticle did you get it working with 5.8.8?
23:46 mgage ok -- there are red hat users however so we'll need to warn them.  -- I thought apple was slow but they are on 5.12
23:47 ionparticle RHEL5 is on 5.8.8, that's my dev server
23:47 ionparticle RHEL6 is on 5.10.1
23:47 goehle_ I did not.  It was an older virtual machine and I just abandonded it
23:47 goehle_ I moved over to our newer maching which is RHEL6
23:49 goehle_ oh mgage
23:49 goehle_ it occurs to me we may have been working cross purposes
23:49 mgage here
23:49 goehle_ I just realized I was still on develop.  So all of those datePicker bugs may not actually be in 2.7
23:50 Paul_Pearson joined #webwork
23:51 mgage ok -- but at least one of them is a real bug -- in ProblemSetDetail1   -- it seems to me it would be safer if we hosted all the jquery code on the webwork server to avoid the risk of security issues.
23:52 mgage how big a group will be using RHEL5 instead of RHEL6?  will the IT departments scream if they are asked to upgrade?  I think I could back out of the few features of 5.10 that I used.
23:57 goehle_ A friend of mine says its pretty common
23:59 goehle_ although I'm told you can upgrade past 5.8 manually
23:59 goehle_ they would need to get an rpm for it though