Camelia, the Perl 6 bug

IRC log for #bioclipse, 2008-08-20

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

All times shown according to UTC.

Time Nick Message
03:41 zarah joined #bioclipse
03:41 ilogger2 joined #bioclipse
03:42 CIA-50 joined #bioclipse
03:42 shk3 joined #bioclipse
03:42 stain joined #bioclipse
05:36 Gpox joined #bioclipse
07:05 masak joined #bioclipse
07:31 olas joined #bioclipse
07:31 olas hi
07:31 zarah hi olas
07:44 jonalv joined #bioclipse
07:52 CIA-50 bioclipse: ospjuth * r6653 /bioclipse2/trunk/plugins/net.bioclipse.qsar/ (17 files in 7 dirs): Added a QsarManager for business logic.
07:59 masak ooh, CIA-50 is alive today
07:59 jonalv CIA-50++
08:03 CIA-50 bioclipse: carl_masak * r6654 /bioclipse2/trunk/ (codetemplates.xml tabless.xml):
08:03 CIA-50 bioclipse: [codetemplates.xml, tabless.xml]
08:03 CIA-50 bioclipse: * added to repository
08:03 CIA-50 bioclipse: * it is possible that these should be moved elsewhere. be bold.
08:10 maclean joined #bioclipse
08:15 EskilA joined #bioclipse
08:27 jonalv olas, what's up with pcore anyway? Why doesn't it compile?
08:32 maclean oh don't say things like that when I'm re-checking out... :(
08:33 masak maclean: just don't include it in you run config and you'll be fine
08:34 masak s/you/your/
08:38 maclean oh, wait. "pcore" - I read "core". never mind :)
08:40 masak what a difference a letter makes
08:46 jonalv yea :)
08:51 egonw joined #bioclipse
08:52 olas jonalv: pcore compiles here
08:52 olas what error msg do you get?
08:53 jonalv The method addBlacklistFilter(List<IResource>) is undefined for the type WSFileDialog
08:54 jonalv in PcorePage
08:54 jonalv row 274
08:56 olas hmm
08:57 olas oh
08:57 olas a missing commit
08:57 * olas has too many workspaces
08:57 jonalv oki
08:59 CIA-50 bioclipse: ospjuth * r6655 /bioclipse2/trunk/plugins/ (4 files in 3 dirs): A missing commit for blacklistFilter casting.
08:59 olas jonalv: did you solve the java1.5 problem with the system props?
08:59 jonalv uhm system props?
08:59 jonalv that was a mac problem I think
09:00 jonalv or are we talking about the ame problem?
09:00 jonalv *same
09:00 olas oh
09:00 olas sry
09:00 olas nvm
09:00 olas did you solve it?
09:00 jonalv sort of
09:01 olas lol
09:01 olas how?
09:02 jonalv uhm why does the exported version of Bioclispe 2 alpha 2 say : DEBUG - ## -noUpdate detected, do not check for updates.
09:02 jonalv That can't be right....
09:02 jonalv that is: the version I just downloaded from sourceforge...
09:02 olas that can't be right
09:02 olas damn
09:02 olas :-(
09:02 olas :-(
09:03 olas seems it's broken
09:03 olas need to fix that
09:03 jonalv new version will come this week I guess then :)
09:03 olas the good thing is: we have no updates that installs this way :-)
09:03 olas friday possibly
09:03 olas a lot of other things to do
09:03 olas and besides, until we want to publich new features...
09:04 olas this is not an issue
09:04 jonalv Would it check for Brunn updaets too?
09:04 jonalv it would wouldn't it?
09:04 olas not by default I think
09:04 jonalv it checks all update site, right?
09:04 olas hmm
09:04 olas need to check that
09:05 olas actually donät think so
09:05 jonalv doesn't the autoupdate check all update sites?
09:05 jonalv If not, could you make it so?
09:05 olas please file a bug
09:05 olas I will fix it later
09:05 olas assign to me
09:05 jonalv I don't even know if it doesn't...
09:06 olas ok, I'll check
09:06 olas hold on
09:06 olas no, it should check all registered update sites
09:06 jonalv good
09:07 olas but this needs more testing
09:07 jonalv I am trying to get Brun to start but it isn't working... :(
09:07 olas I have a feeling it is not perfect
09:07 olas why not?
09:07 jonalv it simply doesn't start
09:07 * olas is writing tests for the QSARManager
09:07 jonalv no error message or anything
09:07 olas error log?
09:07 jonalv nope
09:08 olas start from console?
09:08 jonalv I just ran it with -conoleLog
09:08 jonalv nothing strange in the log
09:08 olas bc runs, right?
09:08 jonalv yea
09:08 olas but not brunn?
09:08 jonalv but the Brunn features doesn't start
09:08 olas do you get logging for bc?
09:08 jonalv yes
09:08 olas bot no logging for brunn?
09:09 egonw zarah: do you have a delicious account?
09:09 jonalv no
09:09 zarah egonw: I did not understand that
09:09 olas did you install from update site?
09:09 jonalv nothing from Brunn at all
09:09 jonalv olas, yea from update site
09:09 egonw zarah: you should... and bookmark all URLs posted which you abbreviate
09:09 zarah egonw: I'm just a bot, you expect me to understand that?
09:09 egonw zarah: and then you can tag them with the name of the channel on which you found them
09:09 zarah egonw: I'm just a bot, you expect me to understand that?
09:09 olas jonalv: check the dir so that all is copid in the bc folder
09:09 egonw zarah: yes, actually, I do :)
09:09 zarah egonw: I see.
09:09 olas all plugins, all features
09:10 olas and make sure all features have correct size
09:10 olas hmm, all plugins
09:10 olas it could be wrong on your export, right?
09:10 jonalv I guess
09:10 olas no error logs on the build update site?
09:10 jonalv actually I don't seem to have the brunn feature
09:11 jonalv nope no error logs from it either
09:11 olas no wonder it won't start if you don't have it installed!
09:11 jonalv olas, I installed it
09:12 jonalv but it obviously didn't work...
09:21 olas it didnät install correctly
09:21 olas since you donät have the feature
09:21 olas ä='
09:22 jonalv
09:23 masak .oO( mm, donuts )
09:34 CIA-50 bioclipse: shk3 * r6656 /bioclipse2/trunk/plugins/net.bioclipse.spectrum/ (5 files in 4 dirs): added the peak picking action
09:46 maclean jonalv: I was re-reading the irc logs from yesterday, and I thought I would say that I'm working on windows at the moment.
09:46 maclean (and osx, but that's not so rare :)
09:50 maclean Hmmm. and, having said that, what once worked (javacc task in cdk) now doesn't.
09:56 thomas_ku joined #bioclipse
09:57 maclean Aha! I found the problem. Or, rather, I found a hack-solution...
09:58 maclean I commented out the shellscript task in the cdk build.xml. It was trying to invoke bash, which windows doesn't have...
10:05 edrin joined #bioclipse
10:07 edrin hi
10:07 zarah hi edrin
10:08 shk3 hi
10:08 zarah hi shk3
10:08 zarah shk3: You have new messages. Write '@messages' to read them.
10:08 shk3 @messages
10:08 zarah shk3: masak said please don't use tab characters in bc2 source. you could reconfigure Eclipse to never insert tabs in your code. see my bioclipse-devel mail from 2008-05-07 for instructions.
10:13 CIA-50 bioclipse: shk3 * r6657 /bioclipse2/trunk/plugins/net.bioclipse.spectrum/ (6 files in 4 dirs): added the integrate action
10:15 edrin what is the problem with tab characters?
10:15 maclean spaces are more consistent between editors.
10:16 maclean http://www.jwz.org/doc/tabs-vs-spaces.html
10:16 zarah maclean's link is also http://tinyurl.com/2az9zn
10:45 masak edrin: tab characters make it harder to maintain a good coding standard in the source tree
10:45 masak banning them altogether makes it much easier
10:46 masak with the added benefit that the space character suddenly becomes the only allowed character looking like a blank horizontal space :)
10:54 edrin left #bioclipse
11:04 jonalv maclean, so just commenting that tag out made it work?
11:04 jonalv I wonder what that tag does...
11:04 maclean Yup.
11:05 jonalv egonw, can you comment?
11:05 maclean I /think/ it was there to check the version.
11:05 egonw no, not at this moment... mail cdk-devel about it
11:05 egonw seems to be a commit I missed
11:06 egonw this was the patch by Rajarshi, right?
11:06 maclean yup.
11:06 egonw to check the CDK version
11:06 maclean yup.
11:06 egonw ok, email him that he broke compilation on windows
11:06 maclean ok.
11:06 egonw better; file a bug report
11:06 malick joined #bioclipse
11:06 egonw thanx guys, for finding the source of the problem!
11:06 jonalv maclean++
11:07 egonw maclean++ indeed
11:07 egonw karma maclean
11:07 zarah maclean has a karma of 2
11:07 maclean yay!
11:07 egonw ok, one has to start somewhere :)
11:10 CIA-50 bioclipse: gilleain * r6658 /bioclipse2/trunk/plugins/net.bioclipse.spectrum/​src/net/bioclipse/spectrum/editor/ChartPage.java: Fix for Layout problems in the ChartPage
11:26 edrin joined #bioclipse
12:41 CIA-50 bioclipse: carl_masak * r6659 /bot/trunk/zarah.pl:
12:41 CIA-50 bioclipse: [zarah] large refactoring of utterances
12:41 CIA-50 bioclipse: zarah now logs her own utterances, and counts them as repetitions in triples
12:42 zarah joined #bioclipse
12:43 CIA-50 bioclipse: ospjuth * r6660 /bioclipse2/trunk/plugins/net.bioclipse.qsar/ (5 files in 4 dirs): Implemented manager methods to populate model from Extension Point.
13:01 CIA-50 bioclipse: carl_masak * r6661 /bot/trunk/zarah.pl: [zarah] tried to fix repetition sensor by moving it up a bit in the loop
13:19 olas I have a question regarding exceptions in managers
13:19 olas jonalv, masak
13:19 masak listening.
13:20 olas if I am to locate something by ID, and the object is not found, what kind of exception should I throw?
13:20 olas so that console outputs my "payload message"
13:20 masak NotFoundException? :)
13:20 olas ok, if a calculation fails?
13:20 masak why does it fail?
13:20 olas it is not found
13:21 olas use has inputted an ID that is not valid
13:21 olas user
13:21 masak hm
13:21 masak in ordinary program flow, I would expect a check to be made first
13:21 olas this is the check :-)
13:21 masak just so that getting an exception is not the only way to find out
13:22 olas please explain
13:22 masak exceptions are supposed to be exceptional, remember :)
13:22 masak i.e. they should occur when something not expected occurs
13:22 olas ok, so how do I report to the user so that he is informed in the console (if command run from there)
13:22 masak like loss of a connection, calculation error or out of resources
13:22 olas "the descriptor could not be found in regstry"
13:23 olas "please enter a valid ID"
13:23 masak olas: maybe one manager method for checking, and one for actually indexing
13:23 olas indexing?
13:23 masak accessing
13:23 olas please explain
13:23 masak then you get the luxury of being _more_ severe in the accessing method
13:23 olas please explain
13:24 olas in detail
13:24 olas what you mean
13:24 masak because now there _is_ a non-exceptional way to find out if there's an element
13:24 masak olas: ok, starting over
13:24 masak 1. provide an elemExists(elem) method (or whatever)
13:24 masak it returns a boolean
13:25 masak so that the user can check before jumping in
13:25 olas sure
13:25 olas oh no
13:25 olas I donät like that
13:25 masak oh no?
13:25 masak you don't have to like it
13:25 masak it's the masak way :)
13:25 olas user says: getDescriptor("AlogP")
13:25 olas if not found, it should say "Please enter a valid descriptor id"
13:25 masak what? no...
13:26 olas why not?
13:26 masak because maybe it wasn't the user who entered the descriptor id
13:26 masak a method can be called in many circumstances
13:26 olas so?
13:26 olas sure
13:26 masak so the error message is evil
13:27 masak it assumes that the user is at the console
13:27 olas that's why I want to throw an exception with a message
13:27 olas that COULD be written in console
13:27 masak ok, let's back up a bit :)
13:27 olas or acted upon if called by code
13:28 masak olas: I think you should provide an exception, yes. use the one that the collection classes use for that purpose, whatever that is.
13:28 masak an exception is far better than returning null
13:28 olas this was my proposal from the beginning
13:29 masak olas: BUT, there has to be a way to guarantee that code will not encounter an exception, regardless of the parameters/inputs
13:29 masak for that, you need a boolean method for checking
13:29 olas why not just catch the exception?
13:29 masak olas: because that's direct abuse of the exception system
13:29 * jonalv is backlogging
13:30 masak jonalv: help me!
13:30 * olas likes to abuse the exception system
13:30 olas masak: I get your point
13:30 masak olas: but you don't agree?
13:30 olas hmm, I guess so
13:30 olas but still
13:30 masak I'm fine with that. I'm not necessarily right.
13:30 olas ehat exception should I throw?
13:31 masak olas: have you checked the collection classes?
13:31 masak or should I?
13:31 olas I guess I can
13:31 masak I'm checking too :)
13:31 masak mother.
13:33 olas illegalargumentexception?
13:33 olas unsupportedoperation?
13:33 jonalv masak, help you with what?
13:33 masak http://java.sun.com/j2se/1.5.0/docs/api​/java/util/NoSuchElementException.html
13:33 zarah masak's link is also http://tinyurl.com/f9qnf
13:33 masak jonalv: convincing olas that exceptions are not for control flow
13:34 * egonw mumbles that throwing exception is slow in Java
13:34 masak s/control flow/program flow control/
13:34 masak egonw: that's just one of the reasons not to use them for fun
13:34 egonw yeah, but might be the one to convince olas not to use them :)
13:35 masak they are slow because they are not overly optimized, because they should only be used when something has gone horribly wrong anyway
13:35 masak olas: exceptions are s ... l ... o ... w
13:35 olas right
13:35 * olas is coding fast
13:35 olas donät want to be slowed down
13:36 jonalv Just to be clear here. I totally think that olas method can throw an exception but  Masak I think you are right. Both "check-method" and exception is a good way I think
13:36 masak glad we agree.
13:36 egonw olas: exceptions are slow on runtime
13:36 egonw they really kill performance
13:36 jonalv This doesn't mean that I always followthis though... :)
13:36 egonw even more than event listeners
13:38 jonalv was event listeners what slowed down bioclipse 1?
13:40 egonw yeah, because events got duplicated or so...
13:41 masak that could slow down just about anything...
13:42 edrin left #bioclipse
13:44 * olas coding is not slowed down by events, only by lengthy irc conversations
13:44 olas :-)
13:45 jonalv olas, what has slowing down of coding got to do with anything?
13:46 masak jonalv: I think he's keeding,,,
13:46 masak in other news, have y'all seen this? http://www.bryceharrington.org/drupal/node/18
13:46 zarah masak's link is also http://tinyurl.com/569ujh
13:47 masak don't know if I got it from this channel or from somewhere else, so I thought I'd post it here
13:47 masak it's relevant for us since we still use Sourceforge's RT system
13:48 jonalv masak, "keeding"?
13:49 jonalv :( // pan is using firefox 2
13:53 masak jonalv: http://www.urbandictionary.​com/define.php?term=I+keed
13:53 zarah masak's link is also http://tinyurl.com/6nd98r
13:54 jonalv uhm that joke about michael jackson was a bit too much I think...
13:55 masak yes, sorry about that.
13:55 masak didn't really check the page before sending it.
13:56 masak unfortunately, a lot of immature content gets put into urban dictionary
13:56 masak apart from that, it's a good idea
13:56 olas another question
13:56 olas if an operation fails, what to throw?
13:57 olas OperationFailedException
13:57 olas (does not exist)
13:57 masak olas: every exception is due to an operation failing...
13:57 olas I send info to a calculation and it returns null
13:57 * olas is not good at exceptions
13:57 masak that sounds like NullPointerException to me
13:58 olas not really
13:58 masak no?
13:58 olas emptyresultexception
13:58 olas rather
13:58 olas zeroresultsexception
13:58 masak an empty result is not an exception
13:58 masak sometimes, that's in order
13:59 masak just return an empty list
13:59 * jonalv feels forced to point out that the javascript console for example only prints the message in the exception ot the name of the exceptions class
13:59 masak jonalv: true.
13:59 egonw olas: you might want to contribute toL
13:59 egonw to:
13:59 egonw http://nrgwiki.nature.com/cyberinfr​astructureforbiology/show/HomePage
13:59 zarah egonw's link is also http://tinyurl.com/59p5hp
14:00 egonw at some point
14:00 egonw not now
14:00 olas ack
14:01 egonw as I see parallels with the Bioclipse platform goals
14:01 maclean "biological cyberinferstructure" sounds like a defence project.
14:01 maclean :)
14:01 olas I like it
14:02 * masak will start using that
14:03 masak hello, I'm a biological cyberinfrastructurologist
14:03 jonalv masak, start using what?
14:03 jonalv aha
14:05 masak brb &
15:17 CIA-50 bioclipse: ospjuth * r6662 /bioclipse2/trunk/plugins/net.bioc​lipse.qsar/src/net/bioclipse/qsar/ (6 files in 3 dirs): Implemented descriptorCalculations.
15:18 CIA-50 bioclipse: ospjuth * r6663 /bioclipse2/trunk/plugins/​net.bioclipse.qsar.test/: Initial import.
15:19 CIA-50 bioclipse: ospjuth * r6664 /bioclipse2/trunk/plugins/net.bioclipse.qsar.test/ (17 files in 9 dirs): Tests for QSAR features with dummycalculator.
15:21 CIA-50 bioclipse: ospjuth * r6665 /bioclipse2/trunk/plugins/net.bioclipse.cdk.qsar/ (2 files in 2 dirs): Adapted to new EP syntax.
16:21 olas left #bioclipse
17:01 malick left #bioclipse
18:38 egonw joined #bioclipse

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