Perl 6 - the future is here, just unevenly distributed

IRC log for #webwork, 2012-09-09

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

All times shown according to UTC.

Time Nick Message
01:12 JohnTravis joined #webwork
01:39 JohnTravis mgage:  You home?
01:42 mgage_ johnTravis: hi
01:42 JohnTravis Hey Mike.  Sorry to wake you!  :)
01:43 mgage_ nope -- not a sleep just went down to the kitchen to get some wine
01:43 mgage_ what's up?
01:43 JohnTravis As you know, the PREP committee requested we pursue a proposal for next summer
01:44 mgage_ yes -- I saw that in your email -- haven't focused on it yet because I've been squashing bugs most of the day.
01:44 mgage_ do you want help with the writing?
01:44 JohnTravis The committee for some reason must need some other proposals
01:44 JohnTravis I got another email back today asking us to pursue both proposals.
01:45 JohnTravis Briefly, #1 is problem authoring only.
01:45 mgage_ hang on -- let me read and catch up
01:45 JohnTravis #2 is model course development only
01:45 JohnTravis ok
01:46 mgage_ I'm having trouble getting to the google doc -- did you send it to michaelegage@gmail.com?
01:47 JohnTravis checking
01:47 mgage_ if you sent it to the other email address could you send it to that one? michaelgage5 isn't connected to google drive yet
01:47 JohnTravis I did 5.  Will do michaelgage
01:47 mgage_ ty
01:48 mgage_ never mind -- I finally figured out how to get in as 5
01:49 JohnTravis You can now edit using whichever persona you desire.
01:49 mgage_ thanks
01:50 JohnTravis I had given up on the idea since we obviously missed the deadline at the beginning of August.  I was too tired to pursue it any more at the time.
01:50 mgage_ I understand -- you were doing enormous amount this summer -- I wasn't ready to deal with it either  and I wasn't involved in as many projects.
01:51 JohnTravis There is benefit to going the PREP route...I would seem that since this program is advertised broadly that we would be able to bring in a set of people beyond our normal reach.
01:52 mgage_ I personally like the second project a bit better in some ways -- it  is a bit more focused but the first one is a pretty good outreach for getting people who are slightly interested  more involved.
01:52 JohnTravis Roughly, the first workshop would progress through perl in stages...from editing the "pi problem" to focusing carefully on various aspects of problem writin.
01:53 JohnTravis This is the way I generally learn anyway...steal and augment.  My next thing to learn is answer checkers and making them more insightful.
01:53 mgage_ There are at least two other proposals we should keep track off -- a possible one with Beezer, et. al. for AIM combining webwork and sage and UTMOST
01:53 JohnTravis In my mind, separating the two allows for the second to be more productive.
01:54 JohnTravis Yes.  I though about pinging Beezer about this.
01:54 JohnTravis He might be watching it appears.
01:55 mgage_ that is  a very good point -- that was a difficulty in the one that we did two years ago.  and the learning progression is pretty much the same for everyone.
01:55 JohnTravis My main question at this juncture is whether to bring in others and give them the project of to keep doing it myself.
01:55 JohnTravis In particular, whose name goes on the proposal.
01:55 mgage_ I'm trying to remember what the second proposal was --- the google summer of code thing is coming up again -- but it seems to me there was something else as well.
01:56 JohnTravis <Brain Freeze>
01:56 mgage_ well -- to a large extent it's up to you --
01:56 mgage_ yep
01:57 JohnTravis Paul Pearson came to mind
01:57 JohnTravis and Anneka and Alina for the second.
01:57 JohnTravis Paul for the first.
01:57 mgage_ I kind of like getting other people to put their names on proposals, even when I'm doing a fair amount of work -- it just helps broaden the pool.
01:57 mgage_ Paul
01:57 mgage_ is going to be on the job market for next fall I think -- that would be my only reservation.
01:58 JohnTravis Do you see any problem with bringing them in on these projects...presuming Paul can handle it with the job search.
01:59 mgage_ Certainly bring them in on the project.  perhaps not heading it.  you know that Alina is already point person for a code camp to be held at UNCC some time in the spring?   that might be enough organizing for the spring-summer.
01:59 mgage_ both alina and anneke are good choices
02:00 JohnTravis I'll add them to the document and see how things settle down.  Budget may force some choices.
02:00 mgage_ so the first workshop is completely online?
02:00 JohnTravis Yes.
02:00 JohnTravis Only problem authoring.
02:01 mgage_ well as you said the official organizing committee might get slimmed down
02:01 JohnTravis Hope is to perhaps record the sessions and edit them down for posting into an online tutorial
02:01 mgage_ yes
02:02 JohnTravis I haven't heard from Jason and was hoping to catch him on here.  How much does he want to be involved...
02:02 mgage_ I like the idea of people signing up for the web conference in teams of some kind -- I think it makes it more effective.  Either they are physically in the same place or they know each other and have worked together before.
02:03 JohnTravis Great idea.  I suppose a requirement is too much but a suggestion would be great...even perhaps a price break with two for the price of 1.5.
02:04 mgage_ I expect he will be glad to help out -- same as me -- as much as is needed and that we can fit in. We also won't be offended if our places are taken by newcomers.  :-)  As I said it's important to get new people involved and that doesn't always happen if the old ones insist on hanging around for everything.
02:04 mgage_ price break is a good idea -- hadn't thought of that.
02:05 mgage_ jason and djun were both on earlier but I think they have quit for the day.
02:06 JohnTravis Ok.  I'll wait till Monday late before bothering them again.   I added Djun's name to one of the presenter's list but need to add him to the doc as well.
02:07 JohnTravis Thanks for being a sounding board.  I didn't want to keep moving forward on this without having some feedback.
02:09 mgage_ Sure I think he would be interested -- and very good at it.  He has been helping out with getting webwork going in local community colleges in the Vancouver area .
02:09 JohnTravis Ok.  I've added your groups idea with my reduced registration idea for the 1st workshop.
02:09 mgage_ np -- sorrry I haven't jumped on it sooner.  But I'm trying to make what is available in the github repositories as stable as possible.
02:10 JohnTravis That is a better use of your time.  I can do this.
02:10 mgage_ I can't guarantee no bugs (not anywhere near enough testers) but I can promise to get them squashed as quickly as possible -- usually in a couple of hours -- at most a couple of days.
02:11 JohnTravis Wish I had time to convert our new (unused) server here to the git repo.  I NEED the achievements for the secondary course I'm working on.
02:11 mgage_ next to work more on the documentation -- at the moment ww2.5.1.1 is pretty stable
02:11 JohnTravis Soon, hopefully I'll be able to test as well.
02:11 mgage_ you don't need to convert to git if you don't want to
02:11 mgage_ for now
02:12 JohnTravis Yes I do.  We have a new server set up now with an updated OS and I want to move to it asap.  However, I don't want to convert midstream.
02:12 mgage_ just cd /opt/newwebwork and then type    svn checkout ---hang on I'll get the exact url
02:13 mgage_ you can set them up in parallel.
02:13 JohnTravis You are very good at this?
02:13 JohnTravis !
02:14 mgage_ http://webwork.maa.org/wiki/Github#Download_WeBWorK_via_svn_to_an_svn_repository
02:15 JohnTravis Got it.  Ok, I'll give it a try.
02:16 mgage_ I believe what you want  is svn checkout https://github.com/openwebwork/webwork2/trunk
02:16 JohnTravis From /opt/webwork or from /opt/webwork/webwork2
02:17 JohnTravis Just checking to be certain.
02:19 mgage_ it get's placed in a directory called trunk -- so you could do it from /opt/webwork
02:19 mgage_ then you rename trunk to    webwork2_github and your old webwork2 to  webwork2_svn and make a link      ln -s webwork2_github webwork2
02:19 mgage_ then to switch back do  ln -s webwork2_svn webwork2   -- actually do that first just to make sure everything is still working.
02:19 JohnTravis Gotcha.  Switch back and forth via a ln
02:20 JohnTravis Same for pg?
02:20 mgage_ yep -- but make sure the old system continues to work first even with the link -- then you can configure the new system -- and if you've broken something you switch back in a minute and work on the configuration again.
02:20 mgage_ yes -- same thing for pg
02:20 mgage_ brb
02:21 mgage_ I'm needed for about 5 minutes
02:32 JohnTravis Apparently I have to fix something in the conf files.
02:33 Paul_Pearson joined #webwork
02:33 mgage_ back
02:33 JohnTravis Hey Paul Pearson
02:34 JohnTravis mgage...need to change new directories to wwadmin and not root.
02:35 mgage_ yes -- it will take you a while to get the site.conf and localOverrides.conf files set up -- it's the same information as in global.conf.dist but just in different places.
02:35 mgage_ yep that too.  just make sure the ln to the old version of webwork2 allows you to run -- then you can take your time about getting the new version on line
02:36 JohnTravis Well neither is "officially" online at the moment since this one is still new.  The old server is still running admirably.
02:36 mgage_ Paul -- I've pushed some updates to openwebwork/webwork2-dev --- I think they are pretty stable
02:37 mgage_ and I'll push them to openwebwork/webwork2 later tonight or tomorrow.
02:37 mgage_ which version are you guys running at Hope?
02:37 mgage_ Paul_Pearson: hi
02:39 Paul_Pearson Hi Mike,
02:40 Paul_Pearson The three Hope courses are all calc 1 and all on https://test.webwork.maa.org/webwork2/ which is running 2.5.1.1 right now
02:41 mgage_ oh right -- ok that reminds me to do updates on that server.  I haven't yet.  Are you having any issues with the courses?
02:49 Paul_Pearson Nothing major.
02:49 Paul_Pearson The livegraphics3d problems I wrote won't print the graph in pdf hardcopy.
02:50 mgage_ I've fixed several minor things.
02:50 mgage_ Did they ever do that? I thought java apps didn't do well with printing. or do you  mean the static pictures?
02:50 Paul_Pearson Etc.  I think I've submitted pretty much everything I've noticed via bugzilla over the past couple of weeks.
02:50 Paul_Pearson I mean the static jpg or png's.
02:51 Paul_Pearson It's not a high priority.
02:51 mgage_ I think I might have gotten that at the end of the code camp at winona -- can you tell me a course and a homework set/rproblem that causes trouble?
02:51 Paul_Pearson My multivariable calc students are OK with things the way they are.
02:52 Paul_Pearson https://test.webwork.maa.org/webwork2/hope_math_232/hardcopy/09.1_Graphs_images_level_sets/
02:52 Paul_Pearson so hope_math_232
02:53 Paul_Pearson homework 09.1_Graphs_images_level_sets
02:53 mgage_ I think it was a typo in PGalias -- i handled .gif files but not .png  -- but your problem could be something different
02:53 mgage_ k -- I'll look at it.
02:54 Paul_Pearson For example, in problem 1, there should be static png images included in the hardcopy, but they're not.
03:04 mgage_ I'll take a closer look -- I think it might be because of something tricky that Davide was doing that doesn't work anymore
03:04 mgage_ this one might take until tomorrow.
03:05 mgage_ I've updated on test.webwork  and haven't seen any major screams yet.
03:05 mgage_ :-)
03:10 Paul_Pearson So you updated it 2 minutes ago and nobody has screamed yet?  That's a good start :)
03:11 mgage_ there have been several dozen hits already
03:11 mgage_ from hope and from lowell high school
15:25 aubreyja joined #webwork
15:26 mgage_ hi
15:26 aubreyja Hi
15:26 aubreyja Were you able to get OPL-update running by adding the missing files?
15:26 aubreyja (In other words, should I continue bug hunting?)
15:27 mgage_ yes -- by the time I figured out what was wrong enough to explain it I had also figured out how to fix it -- probably --
15:27 mgage_ you should check my work -- but I've had it running for a few minutes on both hosted2 and test and no obvious errors.
15:28 aubreyja ok, having a look: openwebwork/webwork2-dev?
15:28 mgage_ yes -- that should be uptodate -- I changed "include" to "require" in the script -- include is only defined by CourseEnvironment
15:29 aubreyja i.e. is that where you changed the script - ok
15:29 mgage_ most of the changes were to add the files in webwork-open-problem-library
15:29 aubreyja You called him 'iron particle' :)
15:29 mgage_ also what should be done about localOverrides
15:30 mgage_ yeah -- dyslexic I guess -- hope he is not offended :-)
15:30 mgage_ and the definition of problemLibrary{version} -- is that new? is it still needed for backwards compatibility?
15:31 aubreyja no, that was in NPL-update so it's not new
15:31 aubreyja but in SetMaker.pm there is a switch I had to change
15:31 mgage_ right -- but we should add some comments.
15:31 aubreyja previously the only two allowed cases were == 1 and ==2, and
15:31 aubreyja ok, yes
15:32 aubreyja djun suggested that we key off of the VERSION file in the top level opl directory
15:32 mgage_ Right - that needs to be set to 2.5 rather than 2 to get OPL-update to choose the OPL_* tables rather than the NPL- tables.    -- I don't believe this is true.  OPL-updates is keying off the VERSION file
15:32 mgage_ kk -- just should add some comments to localOverride .dist   so that people don't get confused.
15:33 aubreyja is it? Maybe someone changed that - when I wrote it in winona it was looking for the problem library version - just a sec, let me refresh my memory
15:33 mgage_ probably means that one can even leave the problemLibrary{version}=2 without problems if we key off the VERSION file
15:34 aubreyja ah, ok - I remember now what happened:
15:34 mgage_ git log -p -- OPL-update  has been useful to me for recalling what happened (particularly if I wasn't there)
15:35 mgage_ the stuff after -- restricts the files that are reported on   -- another variant is to add --full-history -- this catches things that have been added and then removed subsequently -- otherwise those changes aren't reported.
15:36 aubreyja originally I think we had it in webwork-open-problem-library/ i.e. above the actual library, and I was using $problemLibrary{version}, but then we moved it below OpenProblemLibrary/ so we could more easily get it from $libraryRoot
15:37 aubreyja I'm taking the easy way out and looking over https://github.com/openwebwork/webwork2-dev/blame/master/bin/OPL-update
15:38 aubreyja I agree we should add the table definitions to database.conf.
15:38 aubreyja Right now the exact choice between NPL and OPL update is also in DBListing.pm
15:38 mgage_ that's handy too -- thanks
15:38 aubreyja so if we can factor it out and make the choice once that would be better
15:39 mgage_ kk
15:40 mgage_ where is the link to the "blame" files from openwebwork/webwork2-dev?
15:41 aubreyja if you look at a specific file the link will be at the top
15:41 mgage_ got it -- thanks -- this is quicker and less error prone than using git directly
15:42 mgage_ OTOH you can use git on your local repository
15:43 mgage_ hmm -- where are my changes of include to require?
15:44 mgage_ doesn't look like that made it into this version -- I suspect if you try to run this script it won't work -- claiming main::include not defined
15:45 aubreyja unrelated - do you know how to turn of the mac voice over? My son somehow turned it on on my wife's computer and now we can't turn it off!
15:46 mgage_ yeah -- I think it's command 5 or command 8 but you can also turn it off from the systems config
15:46 mgage_ systems preferences
15:46 aubreyja thanks - brb
15:47 mgage_ under universal access -- the command is command then function F5
15:47 aubreyja ok- problem solved - thanks
15:48 aubreyja where's the require supposed to be?
15:48 aubreyja ah, I found it include($libraryRoot/VERSION");
15:49 mgage_ where include is now -- line 102.   Could you check and make sure it fails before you change it? this worked for me but I was puzzled as to how include stayed in there at all
15:49 mgage_ it should have caused errors the first time it ran -- unless there is something more mysterious going on
15:49 aubreyja just a sec - checking
15:50 mgage_ it just might be one of those things that got done in a hurry at the end of Winona
15:51 aubreyja it was definitely one of things that got done in a hurry at the end of Winona - I was doing it :)
15:52 aubreyja actually finishing it during the final presentations before I had to leave
15:52 aubreyja Aaron's been sending me the video by the way, so I'm going to post it on our youtube channel
15:53 mgage_ if you can check it out and verify that changing include to require is the right thing then push it to openwebwork/webwork2-dev and I'll update my various local copies on hosted2, test, and my machines
15:54 aubreyja hmm…I can run this in ww2.5.2 and it seems to be working...
15:54 aubreyja my line 102 is include(…)
15:55 mgage_ ok -- I have been testing in ww2.5.1.1 -- that's why i wanted you to look.  I might have made the wrong fix.
15:55 mgage_ Robin first reported this bug
15:56 aubreyja ok, let me switch to 2.5.1.1 and check it there
15:56 mgage_ I think the only place is include is defined is in running CourseEnvironment  but something clever might have been done there.
15:58 aubreyja working for me in 2.5.1.1 too, using same defaults.config
15:59 aubreyja isn't' inlcude() aliased to main:: in CourseEnvironment.pm...
16:00 mgage_ what I see on github is that the include() to require() commit was made to ww2.5.1.1 and to ww2.5.2 but not to master
16:01 aubreyja oh, but I haven't updated and have include in my OPL-update...
16:01 aubreyja and I see that include is shared witht he safe compartment in which the config files are read
16:01 aubreyja so it *shouldn't* be available for OPL-update, right?
16:02 mgage_ that's what I thought
16:02 aubreyja ah, but there's a loop starting at line 188 in CourseEnvironment
16:03 aubreyja I think it takes all of the vars out of the safe compartment and puts them into main
16:04 aubreyja maybe not - I went through this carefully at one point, and now I'm remembering it in fragments - let me look for a minute
16:04 aubreyja better yet, I can print %main...
16:05 mgage_ I can do the following -- update to mgage/ww2.5.2 -- change require to include -- run OPL I get error  main::include not defined
16:06 aubreyja ok, I found the error
16:07 aubreyja it's working when using NPL-tables but not when using OPL-tables
16:07 mgage_ interesting
16:07 aubreyja remove your VERSIOn file in the OPL, it should get NPL tables and work
16:09 aubreyja see there it never needs include because -e $libraryRoot/VERSION fails
16:09 mgage_ yep
16:09 aubreyja so require it is then
16:10 mgage_ ok -- glad we got that figured out -- it really takes at least two people in these final versions
16:10 aubreyja well…let's see - should we put it in an eval block to catch any errors and then default to NPL-tables?
16:11 aubreyja or, actually, let's do that in localOverrides.conf the include($libraryRoot/VERSION)
16:11 mgage_ what makes listDB look at OPL tables instead of NPL tables?  when I didn't have VERSION there I was getting a database error   webwork.OPL_DBsubject not found.
16:11 aubreyja then we can get $ce{problemLibrary}{version} from $ce
16:12 mgage_ I like that better I believe we were specifying the version two different places -- and that's always reisky
16:12 mgage_ risky
16:12 aubreyja lemme check
16:13 mgage_ it's in getTables line 90 of listDB
16:13 mgage_ brb
16:13 aubreyja yes, there it's definitely usiing the problemLibrary version defined in $ce
16:15 aubreyja here's a nice pro tip I picked up on the inter webs: http://pastebin.com/Rr3kqZ3C
16:16 aubreyja that function in your .bashrc or .bash_profile will have your prompt always tell you what git branch you're on
16:20 aubreyja ok, I'll make the changes to the config files and OPL-update and push it and make a pull request. I also want to add an NPL-update alias for people with that….
16:20 mgage_ sounds good
16:22 aubreyja I want to test it with the remotes tricks you sent me yesterday
16:26 aubreyja wait - how do I set a push remote separate from a tracking remote?
16:27 mgage_ not sure of the shortest method but   git branch --upstream url sets the pull url
16:28 mgage_ almost   git branch --help reveals its git branch --set-upstream   url
16:28 aubreyja also, is there a newer version of the dmg or is the one at webwork.maa.org/ww-downloads/WW_dev_small2.dmg the current version?
16:30 mgage_ I have newer versions on my machine -- getting those finalized and pushed out is a project I've put off for a bit. -- you can use the small2.dmg one -- it will work pretty well and you can update the webwork aspects of it using git
16:30 aubreyja pesky docs
16:30 aubreyja kk
16:30 mgage_ I think it is missing git however which means you need that command on your own machine
16:32 mgage_ hmmm -- I'm not sure -- it may be that set-upstream resets both push and pull.  It would be nice
16:33 mgage_ there were a way to set them differently.
16:37 mgage_ more than you want to know:  http://longair.net/blog/2011/02/27/an-asymmetry-between-git-pull-and-git-push/
16:50 aubreyja when I submit the pull request, should I submit the request to all three of openwebwork-dev/master ww2.5.1.1 and ww2.5.2 - is there an easy way to submit a pull request to all 'relevant' braches?
16:52 mgage_ I haven't found it yet -- you can push to all of them using git but I prefer to have these final commits done over github for the moment so the public record is clearer.
16:52 mgage_ not sure if that is absolutely necessary
16:52 mgage_ but lets do it that way for now.
16:53 mgage_ you can skip master if you want -- my plan is for the people accepting pull requests on openwebwork to remember to update master to whatever the latest stable branch is -- for now ww2.5.1.1
16:54 mgage_ master is what gets checked out by default in a clone request or when using svn and trunk
16:55 aubreyja sure - sounds good
16:56 mgage_ you can issue a pull request from one branch to another in the same repo
17:10 aubreyja I'm having a problem viewing images - somewhere there is a double forward slash…in the url path…any ideas where this might be from?
17:11 mgage_ ok -- this has been reported as well -- which images? --- although are you sure the double forward slash is causing troubles? -- it usually doesn't -- there might be a deeper problem
17:12 mgage_ give me an example of a url of an image that is not working
17:13 aubreyja so webwork 2.5.1.1, pg-dev, setDemo/limits.pg (Problem 2) in the Demo set
17:18 aubreyja left #webwork
17:18 aubreyja_ joined #webwork
17:18 mgage_ works for me https://test.webwork.maa.org/webwork2/gage_demo/Demo/2/?effectiveUser=gage&amp;displayMode=images&amp;user=gage&amp;key=NKFPbNERFnQD59yCTUuLIKk412BxnIda
17:19 aubreyja_ sorry - for some reason I couldn't type and had to close and rejoin the room
17:19 aubreyja_ hmm...
17:19 mgage_ there was a change to PGalias that fixed a problem like this
17:20 mgage_ send me the actual url that is causing the problem (copied from the browser)
17:20 mgage_ that is the url of the image
17:21 aubreyja_ it's on my local machine: localhost:9000/webwork2_course_files/Math_01/tmp//gif/admin-1240-setDemoprob2image1.png
17:22 mgage_ try typing that url alone into your browser.  then removing the double // and seeing if it works in that case
17:22 mgage_ I suspect the .png is not there at the location you expect
17:25 aubreyja yeah, nothing…lemme check
17:26 aubreyja the image is indeed in Math_01/html/tmp/gif
17:27 aubreyja does your image url also  have a double forward slash in it?
17:27 mgage_ does removing the double // fix it
17:28 mgage_ yes
17:29 mgage_ it's probably coming from within PGalias -- or perhaps the definition of the tmp file in defaults.config
17:30 aubreyja hmm..no..doesn't seem to which makes me think it's a local configuration issue
17:39 aubreyja where does directoryFromPath come from in PGalias?
17:47 mgage_ not sure -- likely PGcore
17:56 aubreyja it's in IO.pm
17:59 aubreyja I think it might be safer to use functions in File::Spec (a core perl module) than things like our fileFromPath, directoryFromPath, etc...
17:59 mgage_ probably true -- those were created for perl 4 and webwork 1.9
18:00 aubreyja for example, the getDirDelim is supposed to be a hook for doing this portably, but File::Spec already builds that in
18:00 aubreyja while we only return / right now
18:00 mgage_ not sure if you wan to do it this instant but go ahead and make those changes
18:00 aubreyja sure - I'll start thinking about this - seems like pretty low level stuff, so I'll want to be sure to be careful about it :)
18:01 mgage_ right -- and we gave up on getDirDelim in any case -- I'm not convinced it's used consistently
18:01 aubreyja I'll probably just have our existing functions return the results of applying the File::Spec stuff (so replace our r.e.'s with File::Spec calls)
18:02 mgage_ exactly -- you could slip it into ww2.5.2 at the moment however --that's being lightly tested -- just be careful to monitor error logs on say hosted2 after you've made that change.
18:02 aubreyja kk
18:02 mgage_ that seems wise -- I don't see any real harm in that
18:02 Paul_Pearson joined #webwork
18:02 mgage_ you can send my github a pull request and then I'll pull it into hosted2 and we'll alert Rob to be aware something might be strange
18:58 JohnTravis joined #webwork
19:01 aubreyja well, in searching for the "//" source I discovered that in fact my image has 0 bytes
19:02 aubreyja I think it may be the fedora/RH/CentOS texmf bug...
19:10 aubreyja yep, that was the problem - texconfig
19:11 mgage_ :-)   did you also find where the // was coming from?
19:12 aubreyja no, I would would still like to find where the double forward slash is coming from.  I think it works only because of mod_rewrite
19:12 aubreyja in fact...
19:25 aubreyja well, my suspicion right now is that the reason it works is that the AliasMatch in webwork.apache2.config has a generous regular expression
19:43 aubreyja ok, well, everybody (web server and file system) seem to agree to ignore the //
19:44 mgage_ I think that is common in unix systems (and web urls) -- still annoying
19:44 aubreyja the r.e. in webwork-apache2.config returns a single slashed alias if passed a single slashed url and returns a double slashed alias if passed a single slashed alias
19:44 aubreyja …if passed a double slashed alias
19:44 aubreyja I still want to know where it's coming from!!!! :)
19:45 mgage_ did you check defaults.config and the definition of the temp directories? does one of them have a slash at the end?
19:48 aubreyja nope, checked
19:52 aubreyja How many licks does it take to get to the Tootsie Roll center of a Tootsie Pop?
19:52 aubreyja The world may never know.
20:27 JohnTravis http://www.tootsie.com/gal_machine.php
20:31 aubreyja thx :)
21:03 JohnTravis Paul Pearson...?

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