Camelia, the Perl 6 bug

IRC log for #bioclipse, 2008-07-31

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

All times shown according to UTC.

Time Nick Message
00:30 bruno_ joined #bioclipse
04:41 masak joined #bioclipse
04:47 zarah joined #bioclipse
06:43 Gpox joined #bioclipse
07:06 olas joined #bioclipse
07:16 malick joined #bioclipse
07:28 ilogger2 joined #bioclipse
07:36 jonalv joined #bioclipse
07:36 jonalv moin
07:36 zarah hi jonalv
07:43 rojasm joined #bioclipse
07:46 jonalv olas:
07:49 olas yes
07:49 olas jonalv:
07:50 jonalv What do we do now?
07:50 jonalv no sourceforge...
07:50 olas is it down?
07:50 jonalv yup
07:50 jonalv didn't you see the mail?
07:50 olas really down?
07:50 olas no
07:50 olas oh
07:50 olas I do now
07:51 * olas always ignores these mails
07:51 jonalv for at maximum four more hours
07:51 jonalv I think
07:51 olas we do what we can
07:51 olas let's hope SVN is faster after this upgrade
07:52 jonalv I think I might have a late version but I can't build... :(
07:52 olas jonalv: di you test the binary I sent?
07:52 jonalv (I was thinking we could build anoterh test version)
07:52 olas please test on linux, win and mac
07:52 egonw joined #bioclipse
07:52 olas hi egonw
07:52 jonalv olas: only the mac one. I found some things I had to test and gott distracted :)
07:53 jonalv s/test/fix/
07:53 jonalv hi egonw
07:53 egonw hi olas
07:53 egonw hi all
07:53 zarah hi egonw
07:53 olas zarah: shut up
07:53 zarah olas: :)
07:53 olas zarah: are you in a better mood today?
07:53 zarah olas: eh... wha'?
07:54 olas sorry
07:54 olas just asking
08:01 thomas_ku joined #bioclipse
08:02 zarah joined #bioclipse
08:09 olas egonw: around?
08:09 olas in molfiles, how do you tell if it has 3D coordinates?
08:09 egonw really tricky...
08:09 olas is it fixed to read line 5?
08:09 egonw no
08:09 egonw you need to read all atoms, and see of all z coords are 0.0
08:09 egonw but even then, the molecule might be 3D...
08:10 egonw e.g. benzene
08:10 olas I will just read the 3 first and see if z is null
08:10 egonw which unconveniently has been rotated into the XY plane :(
08:10 egonw make it 10
08:10 olas that's ok
08:10 olas how read the first 10 atoms?
08:10 egonw up to line 15?
08:10 olas so 5 is fixed then?
08:10 olas atoms tart on line 5?
08:10 egonw yes
08:11 olas ok
08:11 olas will do
08:11 egonw in MDL molfile V2000 at least :)
08:11 masak "atoms tart"?
08:11 olas :-)
08:11 egonw you funny atoms tart
08:11 olas and I meant it, every word
08:11 egonw mmm... maybe zarah can reply that when someone tells him to shut up ?
08:12 olas @slap egonw
08:12 * zarah clobbers egonw with an untyped language
08:12 masak "him"?
08:12 masak is "zarah" a boy name in the Netherlands?
08:13 olas these dutch people are crazy sometimes...
08:13 olas they only have a slice of bread for lunch
08:13 egonw sorry, must think of zarah as a PhD student to barge aruond to do the dirty work, and most PhDs I know are boys ;)
08:14 olas and drown everything in majo
08:14 jonalv egonw++
08:14 masak (more PhD girls)++
08:14 * olas is a PhD student...
08:14 jonalv no wait I am gonna be a phd student...
08:14 olas @slap egonw hard
08:14 * zarah karate-chops egonw hard into two equally sized halves
08:14 jonalv egonw--
08:15 egonw olas: yeah, good point... might as well start calling you zarah too :) hahahaha
08:15 egonw oh oh...
08:15 olas egonw--
08:15 jonalv karma egonw
08:15 zarah egonw has a karma of 4
08:15 masak @slap olas
08:15 * zarah clobbers olas with an untyped language
08:15 masak this is turning into a brawl :)
08:15 egonw zarah: please add this to your vocabulary (just don't use it on me): kick him in the fork
08:15 zarah egonw: I'm just a bot, you expect me to understand that?
08:15 masak egonw: will do
08:16 masak @update # now added
08:16 zarah joined #bioclipse
08:17 egonw @sla olas in the fork
08:17 egonw hahaha
08:17 egonw @slap olas in the fork
08:17 * zarah secretly deletes olas in the fork's source code
08:17 egonw ok, close enough :)
08:17 jonalv haha
08:17 masak egonw: I think I misunderstood you
08:17 jonalv no that\s not the way to do it :)
08:18 egonw that's sort of time cause/consequence contradictions
08:19 egonw (never mind :)
08:19 masak :)
08:20 egonw anyway, good to see that zarah can handle DNA sequences
08:22 edrin joined #bioclipse
08:22 olas hi edrin
08:22 egonw olas: was the XML pull parser code snippet useful?
08:22 jonalv I can report that the javascript console doesn't start in windows in the test build
08:22 edrin hi olas
08:22 edrin hi all
08:22 zarah hi edrin
08:22 egonw olas: did you see my email on the Blue Obelisk mailing list? Can you see bioclipse contributions?
08:22 egonw hi edrin
08:22 edrin hi egon
08:23 shk3 joined #bioclipse
08:24 olas hi shk3
08:25 olas egonw: in an MDL file, are the spaces fixed as well as separators?
08:26 egonw MDL uses columns
08:26 egonw google for CTFile.pdf
08:26 zarah egonw: I can't do that; complain with masak and jonalv
08:26 masak that is an oft-asked-for function
08:26 egonw ic
08:27 egonw likewise 'spider for ethanol' ?
08:27 egonw zarah: spider for ethanol
08:27 zarah egonw: excuse me?
08:27 masak Spider for president!
08:27 egonw is that the active ingredient of President Camambert ?
08:28 masak must be
08:29 * egonw wonders if he missed a day... is it already Friday?
08:30 masak egonw: nope
08:34 shk3 hi olas
08:35 shk3 I have problems in Bioclipse
08:35 shk3 jonalv has this on linux as well.
08:35 shk3 could you have a look?
08:35 shk3 I can't work on the bugs right now, so it would be great to resolve this.
08:37 masak shk3: I'm fixing it right now
08:37 jonalv shk3: I just found the same problem in the windows version of our test build
08:38 masak shk3: but the problem is that SF's svn is down, so I can't share the solution...
08:38 masak I could paste a patch, though
08:38 jonalv olas: There is something wrong with: http://www.bioclipse.net/dl/20a/biocli​pse2.0-alpha01-pre01.linux.gtk.x86.zip
08:38 zarah jonalv's link is also http://tinyurl.com/5hn7w6
08:38 jonalv That files eeems corrupt
08:38 olas oh no
08:38 olas win32?
08:38 jonalv olas: what do you mean?
08:39 olas did you test the win version?
08:39 olas please do
08:39 jonalv olas: yes and it can't run the javascript console view
08:39 olas tomorrow's the day
08:39 olas ?
08:39 jonalv I just said that a couple of minutes ago
08:39 olas why not?
08:39 * olas will backlog
08:39 jonalv same error as shk3 has
08:40 jonalv something about jruby
08:40 olas ok
08:40 jonalv masak is on it I think
08:40 olas as soon as SVN is up, bugs fixed, I make a new rel pre02
08:40 * olas needs coffee
08:40 jonalv olas: Could you build me a new linux test version from your local (It might be a bit old but better than nothing...)
08:41 egonw http://sourceforge.net/community/forum​/topic.php?id=2993&page#post-6904
08:41 zarah egonw's link is also http://tinyurl.com/6lm8cl
08:41 masak olas: the prereleases are strictly internal to the bioclipse-dev team, I suppose?
08:41 egonw about the SVN downtime
08:41 masak egonw: yes.
08:43 olas masak: not really
08:44 olas anyone feel free to use them
08:44 masak oki
08:44 olas just didn't want to spam th ML
08:44 masak olas: I think that's a good thing
08:47 egonw well, I'd say it was an imporant event
08:47 egonw that could have deserved an email
08:47 olas what event?
08:47 egonw the first BC2 alpha releae
08:47 olas it was not
08:47 olas that will be tomorrow
08:47 egonw oh...
08:47 olas it was to see if it worked on win and mac so far
08:48 olas and as it didn't, I'm happy I did not send an email out
08:50 egonw :)
08:51 masak olas: http://paste2.org/p/53823
08:51 masak olas: there's a Unix tool called 'patch' to help you apply those diffs to your wc
08:52 masak don't know if it comes with Mac OS X per default, but it's definitely available via Macports
08:54 masak (in case you want to export a version of bc2 that probably fixes the problem shk3++ and jonalv++ reported)
08:56 shk3 masak: great you work on that
08:56 masak shk3: the patch is just a simple surgical procedure, removing the ruby env/console. it should fix the problem you experience
08:57 shk3 that's the one at http://paste2.org/p/53823
08:57 shk3 ?
08:57 olas masak: you a hobbu surgeon?
08:57 shk3 how do I apply this?
08:57 masak shk3: indeed.
08:57 shk3 (sorry, never used patches a lot)
08:57 masak shk3: check out 'patch'
08:57 shk3 not at all,  really
08:57 masak the unix command
08:57 shk3 shame on me
08:57 masak olas: I like to play with knives, if that's what you mean
08:58 shk3 installing patch...
08:58 masak olas: not saying I'm particularly _good_ at it...
08:58 masak shk3++ # patch is a useful tool for devs
08:58 olas masak: your are really building peoples karma today
08:59 olas masak++
08:59 masak :)
08:59 masak shk3: meanwhile, I will revert my changes and try to use patch the patch myself
08:59 masak s/use //
08:59 shk3 btw, what about categories for the bioclipse bugs?
09:00 shk3 want to file some jcp problems.
09:00 shk3 there is bc_jchempaint, but that sounds like bc1
09:00 masak it does indeed
09:00 masak is there a reason to keep the "bc_" prefixes in the categories?
09:01 masak are there other types of bugs than "bc_" bugs, I mean?
09:01 shk3 no, i would say
09:01 shk3 in groups there is v2.0 and 2.0-alpha
09:01 shk3 which one are we working on?
09:02 olas alpha
09:02 olas for tomorrow
09:02 olas after that, it's 2.0
09:03 jonalv until something like alpha/2 turns up I guess :)
09:03 masak then it's v2.0-alpha2
09:03 masak not much of a problem
09:04 jonalv nope
09:04 masak as long as the sf-bug groups uniquely identify a real bc version
09:04 shk3 do you really want to release tomorrow?
09:04 shk3 I get the feeling there are lots of bugs
09:04 masak olas: btw, remember to create a tag tomorrow when you release
09:04 masak shk3: it's just an alpha
09:04 shk3 and also architecture problems
09:04 shk3 yeah, ok
09:04 masak shk3: love you too
09:05 jonalv shk3: architecture problem?
09:05 shk3 well, sort of
09:05 jonalv shk3: are you thinking of something special?
09:05 shk3 like file type definitions
09:05 shk3 but olas got the describers working
09:06 shk3 that's a progress
09:06 shk3 so yes, we are making progress
09:06 shk3 about the patch:
09:06 shk3 do file need to be in the trunk/... directory structure?
09:06 masak paste2-- # what did they think when they willfully broke copy-pasting?
09:07 jonalv masak: how did they break something?
09:07 shk3 ah, i need to use -p ?
09:07 masak jonalv: try copying something from a paste and you'll see
09:08 masak it's nothing vim can't fix, but still, it's pointless
09:08 jonalv masak: what? It works perfect for me
09:09 malick joined #bioclipse
09:09 jonalv masak: okey now I see
09:10 shk3 patch runs forever for me
09:10 shk3 is this normal?
09:11 jonalv masak: http://paste2.org/help.xhtml
09:11 jonalv "a minor problem" LOL
09:11 masak yeah
09:12 shk3 does the patch work for you?
09:13 masak shk3: still trying...
09:13 shk3 for me, the command seems to be stuck
09:13 masak shk3: but you might have trouble because you probably have a different directory structure
09:14 masak shk3: ah did you do `patch file.patch`?
09:14 masak shk3: check `man patch`
09:14 egonw patch < file.patch
09:14 masak egonw: right
09:15 shk3 I did patch -p2 file.patch
09:15 egonw make that:
09:15 egonw patch -p2 < file.patch
09:15 egonw patch expects the patch from STDIN
09:15 shk3 ah, something happening
09:15 shk3 done
09:15 shk3 trying in eclipse
09:17 shk3 thanks!
09:17 shk3 works
09:17 masak :)
09:17 shk3 masak++
09:17 * masak happy
09:17 shk3 need to do some admin stuff
09:18 shk3 but hope to finish the coordinates stuff soon
09:18 egonw shk3: I'm doing a NMRShiftDB commit
09:19 shk3 great
09:19 egonw been a while again already
09:19 egonw a single one
09:19 shk3 btw, email from mainz are still not working
09:19 shk3 i am working on that
09:20 egonw ack
09:20 egonw there will be a EBI node too, right?
09:20 jonalv yey sourceforge svn seems to be back up
09:20 egonw ah, cool
09:24 olas nice
09:24 shk3 yes, will do one at the ebi once mainz is upgraded
09:26 egonw too bad Jarl's group doesn't do NMR
09:27 CIA-49 bioclipse: ospjuth * r6357 /bioclipse2/trunk/plugins/net.bioclipse.ui/ (2 files in 2 dirs): Added contenttypes to distinguish between mdlmolfiles with 2D and 3D coordinates.
09:27 olas egonw: why is that bad?
09:27 olas we do other stuff
09:27 olas can't do it all :-)
09:27 CIA-49 bioclipse: ospjuth * r6358 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk10.jchempaint/plugin.xml: Use new MDL contenttypes.
09:28 olas ok folks
09:28 olas We're back on track, keep those commits coming!
09:30 olas shk3: did you see this? http://wiki.bioclipse.net/index.ph​p?title=Contenttypes_and_describer​s#Editors_should_use_contenttypes
09:30 zarah olas's link is also http://tinyurl.com/5b4bcf
09:30 olas I have now done contenttypes for MDL 2D, MDL3D, CML2d, CML3D, PDB
09:31 olas that's what we support in bc1-alpha
09:31 olas oh, I will do SDF as well
09:32 shk3 great
09:32 shk3 cml-multi-molecule should be easy since it is handled the same way as sdf is
09:34 CIA-49 bioclipse: ospjuth * r6359 /bioclipse2/trunk/plugins/n​et.bioclipse.ui/plugin.xml: Added net.bioclipse.contenttypes.sdf content type.
09:37 olas shk3: we donät support CML with mutliple mols yet
09:37 olas so not in alpha
09:37 olas file a bug if you like
09:37 CIA-49 bioclipse: ospjuth * r6360 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk10.sdfeditor/plugin.xml: Use SDF contenttype.
09:38 olas but set group to 2.0
09:38 * olas will do jmol toolbar/menu now
09:38 olas jonalv: export after lunch?
09:39 jonalv olas: sounds great
09:39 shk3 olas: no problem, just thought it would be easy
09:39 jonalv I am sort of without anything to do right now. But I will grab masak to lunch now
09:39 shk3 but no need for it in alpha
09:40 shk3 without anything to do? we can change that ;-)
09:40 * jonalv -> lunch :)
09:40 olas jonalv: there's SO much todo
09:41 olas update help and intro if you like
09:41 olas test releases
09:41 * olas will prepare some for jonalv after lunch
09:49 shk3 does anybody know how to get an IFile from a path string?
09:57 egonw shk3: "Thank you for your contribution! There was a delay, but your data will soon be handled by the administrator."
09:57 egonw which you might not have received because of the email issue
10:00 egonw shk3: btw, my personal page lists a long list of commits which apparently have not reached a reviewer
10:00 egonw same problem, I guess
10:00 olas shk3: there's a convenience class masak and Gpox wrote
10:00 olas ask them
10:02 shk3 yes, sorry
10:02 shk3 I check the problem
10:03 shk3 I hope to updgrade the server and solve the email problem soon
10:03 shk3 then i will send reviewer emails for past submits
10:03 shk3 it's messy right now
10:03 egonw shk3: no rush, just a ping
10:04 shk3 i found the wsfiledialog
10:04 shk3 i need something like this with the possibility to enter a new file name.
10:05 shk3 do we have that?
10:05 olas for what?
10:05 shk3 or is there an eclipse "save as dialog"?
10:05 olas yes
10:05 shk3 for the 2d generation
10:05 shk3 I want to offer the user to write to existing file
10:05 shk3 or create a new one.
10:05 olas mom
10:05 shk3 (thought that could be usefull)
10:07 olas shk3: I'd like Save as... implemented in CMLEditor and MDLEditor as well
10:07 olas that's a key feature
10:07 shk3 so it is not there?
10:07 olas which needs to add the Formats notion
10:07 olas no
10:08 shk3 but that should be an eclipse standard feature for editors?
10:08 shk3 so we shouldn't need to do a dialog?
10:09 olas mom
10:09 olas yes
10:09 olas SaveAsDialog dialog=new SaveAsDialog(getEditorSite().getShell());
10:09 olas see line 332 in StructureTablePage, in sdfeditor plugin
10:09 shk3 ah, will try
10:12 shk3 grrr, how to get an IFile from an IPath?
10:13 shk3 found that
10:16 olas grrr?
10:23 shk3 fuck
10:23 shk3 how can I create a file when I get the IPath from the dialog?
10:23 shk3 i found a method looking for an IFile, but the file must exist for this
10:42 edrin olas: grrr means the noisy of for example a dog when he threatens someone
10:47 olas ResourcesPlugin,xx
10:48 olas (that was for shk3)
10:49 Gpox shk3: ResourcesPlugin.getWorkspace​().getRoot().getFile(IPath)
10:54 * jonalv is back :)
10:54 * jonalv feels threatened by a dog for some reason
10:56 olas jonalv: shk3 is no dog
10:56 olas (just to reassure you)
10:56 jonalv yet again my feelings fail me
10:57 * olas is barbeque'ing lamb sausages for lunch
10:57 olas or grilling?
10:58 jonalv I think I wil lsee if I can help masak with that JS script running thing
10:58 jonalv masak?
10:59 masak \o/
11:03 shk3 btw, I just found the del button is not deleting files in the resource browser
11:03 shk3 shouldn't we have that?
11:03 shk3 with a confirm?
11:05 masak shk3: yes. it's a known bug
11:05 masak hopefully with an SF bug tracking ticket
11:06 CIA-49 bioclipse: shk3 * r6361 /bioclipse2/trunk/plugins/net.bioclipse.cdk.busine​ss/src/net/bioclipse/cdk/business/CDKManager.java: the create2dcoordinates is working (some fine tuning to follow)
11:06 CIA-49 bioclipse: shk3 * r6362 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk.ui/src/net/bioclipse/cdk​/ui/handlers/Create2dHandler.java: the create2dcoordinates is working (some fine tuning to follow)
11:11 jonalv olas: uhm not sure I have seen that ticket...
11:15 malick joined #bioclipse
11:17 malick left #bioclipse
11:22 olas please add it
11:22 * olas is eating
11:23 * jonalv will add it then
11:28 olas back
11:28 jonalv shk3: I seem to be able to generate 2d coordinates for a file already generated 2d coordinates for. IS that the idea? It feels confusing to me...
11:29 egonw_ joined #bioclipse
11:31 olas not good
11:32 * jonalv is confused. What's the point in these 2d coords generation and what's it done from? From the 3d coords? If so why can't jchempaint open a 3d coordinated file with genreated 2d coords but only jmol?
11:44 CIA-49 bioclipse: jonalv * r6363 /bioclipse2/trunk/plugins/net.bioclipse​.jmol/src/net/bioclipse/jmol/business/ (IJmolManager.java JmolManager.java): added jmol.spinOn() and jmol.spinOff()
11:48 masak general question: if an import in package A cannot find a class in package B, what could be wrong? I've already done "Update classpath" / "Project->Clean", and package A imports package B, and package B exports itself.
11:48 masak what else is there that could go wrong?
11:50 rojasm joined #bioclipse
11:54 masak anyway, this error adequately explains why shk3++ had problems before
11:54 masak this seems to have the same cause
11:57 olas strange
11:57 olas target platform?
12:01 masak olas: jonalv++ fixed it for me
12:02 jonalv olas: hwat should I spend my time doing?
12:06 malick joined #bioclipse
12:06 malick left #bioclipse
12:11 olas jonalv: install new Eclipse 3.4
12:11 olas install delta pack
12:11 olas try to export
12:11 olas I will need you to manage that as well
12:11 olas as I must run away in an hour
12:11 olas to meet my carpenter...
12:13 CIA-49 bioclipse: carl_masak * r6364 /bioclipse2/trunk/plugins/net.bioclip​se.scripting.ui/META-INF/MANIFEST.MF:
12:13 CIA-49 bioclipse: [net.bioclipse.scripting.ui] added missing dependency
12:13 CIA-49 bioclipse: to net.bioclipse.scripting
12:15 CIA-49 bioclipse: ospjuth * r6365 /bioclipse2/trunk/plugins/net.bioclipse.jmol/ (8 files in 5 dirs): Added operational menu and toolbar for Jmol with new org.eclipse.ui.menus Extension Point, commands, and handlers with parameters. Solves 1934186.
12:15 * jonalv is fiddling with the jmol manager for the moment. Addings some easy to add features... :)
12:16 egonw_ jonalv: I assume there already is a method to run any random Jmol script?
12:16 olas yes
12:16 olas jmol.run("select all")
12:16 jonalv egonw_: a random? _)
12:16 olas I just added A LOT of menus
12:16 egonw_ yeah, in particular those that lead to root access :)
12:16 olas menu options I mean
12:16 egonw_ it's important you know...
12:17 egonw_ that way we create an industry of Bioclipse virus scanners etc...
12:17 jonalv I want to get the feedback from jmol back somehow
12:17 egonw_ will yields thousands of EU jobs
12:17 jonalv but it seems ccomplicated...
12:17 egonw_ jonalv: register a feed back listener
12:17 egonw_ BC1 has code for that
12:17 egonw_ for routing Jmol feedback to the console
12:18 jonalv exacty, I should look at bc1 then?
12:19 egonw_ yes, mom
12:20 egonw_ I'll get you a pointer
12:20 jonalv only probelm is that this stupid operating system forbids me to run two instances of a program at the same time. So I can't have one Eclipse runnign the bioclipse 2 workspace and one running the Bioclispe 1 workspace... :(
12:20 egonw_ huh?\
12:20 olas egonw, jmol: we already have a listener in bc2
12:20 CIA-49 bioclipse: ospjuth * r6366 /bioclipse2/trunk/plugins/net.bioclipse.​data.sampledata/data/2DStructures/NIST/: Removed NIST 2D structures as some were errornous.
12:20 olas look in bc2 code
12:20 egonw_ I've done that on several occassions
12:21 egonw_ oh...
12:21 * egonw_ stops browsing his BC1 source tree
12:22 CIA-49 bioclipse: ospjuth * r6367 /bioclipse2/trunk/plugins/​net.sourceforge.jseditor/ (META-INF/MANIFEST.MF build.properties): Updated for exports to complete: export .. Strange that this is needed.
12:22 olas masak: have you attacked "run script yet"?
12:22 masak "run script yet"? :)
12:23 olas otherwise I can create the toolbar item for you
12:23 masak olas: looking at it
12:23 masak olas: that would be nice
12:23 olas just did that for jmol
12:23 olas ok
12:23 masak olas++
12:23 olas will be in jseditor plugin
12:23 masak yes
12:25 olas would you like a mnu as well?
12:25 olas menu?
12:25 olas or just in toolbar?
12:25 olas hmm
12:25 olas maybe there is a JSEditor menu already...
12:26 masak olas: I think there is
12:26 masak but the menu item is less critical
12:26 jonalv olas: where is that jmol error listener in Bioclipse 2?
12:26 olas there is a listener
12:26 olas mom
12:27 olas net.bc.jmol.views.StatusListener
12:27 jonalv olas: dz
12:29 jonalv what does the export as image button in the jmol editor do?
12:30 egonw save the jmol view as PNG/GIF/...
12:31 olas it should
12:31 olas not impl yet
12:31 jonalv I don't get any dialog or anything...
12:31 olas there is a bug, assigned to me
12:31 jonalv oh okey :)
12:32 jonalv I realised that what I thought should be easy wasn't exactly easy...
12:32 masak in the meantime, I'll go look at how bc1 implemented the run button...
12:32 jonalv dealing with commands that can vome at anytimne is never fun... :(
12:33 masak jonalv: excuse me?
12:33 olas masak: I'm done
12:33 jonalv is there one jmol instance in bioclipse or one for each jmol editor, view?
12:33 olas commit is coming up
12:33 masak olas++ # that's fast
12:33 jonalv masak: you are excused. (I don't know why though...)
12:33 olas jonalv: not much works with the view (yet)
12:34 olas we will have to sacrifice some things, and that is one
12:34 CIA-49 bioclipse: ospjuth * r6368 /bioclipse2/trunk/plugins/​net.sourceforge.jseditor/ (4 files in 3 dirs): Added toolbar button and handler for running JS scripts. Handler not implemented yet.
12:34 olas I have just tied the menus to the editor for now
12:34 masak jonalv: you said "vome"
12:34 * masak updates
12:34 * olas has to leave for some time
12:34 jonalv masak: I daid vame?
12:34 olas :-(
12:34 masak jonalv: tes
12:34 olas see you later
12:35 jonalv olas: cya
12:35 masak olas: good luck
12:39 CIA-49 bioclipse: carl_masak * r6369 /bioclipse2/trunk/plugins/net.bioclip​se.scripting.ui/src/net/bioclipse/scr​ipting/ui/views/JsConsoleView.java: [JsConsoleView] removed unused imports
12:40 masak gitk++
12:40 masak ...by the way.
12:51 CIA-49 bioclipse: shk3 * r6370 /bioclipse2/trunk/plugins/net.bioclipse.cdk.busine​ss/src/net/bioclipse/cdk/business/CDKManager.java: ask if overwrite existing file
12:52 CIA-49 bioclipse: shk3 * r6371 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk.ui/src/net/bioclipse/cdk​/ui/handlers/Create2dHandler.java: ask if overwrite existing file
12:59 jonalv_ joined #bioclipse
12:59 jonalv_ left #bioclipse
13:10 masak RhinoEditor extends TextEditor, uh huh
13:14 masak ah, yes: RunRhinoScriptAsJobAction
13:45 egonw_ joined #bioclipse
13:57 CIA-49 bioclipse: carl_masak * r6372 /bioclipse2/trunk/plugins/net.sourcef​orge.jseditor/src/net/sourceforge/jse​ditor/handlers/RunScriptHandler.java: [RunScriptHandler] indentation, text messages
14:00 egonw_ shk3: not sure
14:00 egonw_ Uppsala: is there a manager for cdk10?
14:01 jonalv egonw_: no
14:01 shk3 so I create one?
14:01 egonw_ yes\
14:02 shk3 or shall I have a look at the bugs in the modelbuilder and then do it in cdk if possible?
14:02 shk3 perhaps it's not too complicated to fix them.
14:02 egonw_ fix in CDK trunk would be perfect
14:03 jonalv shk3: no
14:03 shk3 will try to do so
14:03 jonalv The latest I heard was that we shouldn't have a cdk10 manager
14:04 egonw_ shk3: ping me if I can be of help
14:04 egonw_ jonalv: ah, ok, thanx for backing up the fix-CDK-trunk choice :)
14:04 egonw_ jonalv++
14:04 egonw_ :)
14:04 jonalv egonw_: huh? lol
14:04 jonalv I just repeated what I was told :)
14:05 egonw_ and I appreciate that :)
14:05 shk3 where is the template library be found?
14:05 shk3 in cdk?
14:06 shk3 looks like templates are not found
14:06 shk3 want to compare to cdk1.0
14:06 egonw_ shk3: ok, will check if templates should be findable...
14:06 egonw_ I think I did check already if the templates are in the right jar...
14:06 egonw_ hang on
14:06 shk3 some are found
14:06 shk3 so the files must be found
14:06 egonw_ ok, good
14:07 shk3 perhaps the file is outdated or so
14:07 shk3 where is the file in?
14:07 egonw_ mom, telephone
14:07 shk3 hang on ;-)
14:10 egonw shk3:  how could the file be outdated?
14:10 egonw you mean the fingerprint index?
14:10 egonw shk3: you could calculate the fingerprint for all tested files with cdk10
14:10 shk3 something like this
14:10 egonw and add tests based on that reference prints for trunk
14:10 egonw then we should say some failing, some working too...
14:11 egonw similar to the lookup
14:11 egonw which would confirm the outdated fingerprint index
14:12 shk3 hm, will try something like this
14:20 CIA-49 bioclipse: carl_masak * r6373 /bioclipse2/trunk/plugins/net.bioclipse.scripti​ng.ui/src/net/bioclipse/scripting/ui/business/ (IJsConsoleManager.java JsConsoleManager.java):
14:20 CIA-49 bioclipse: [IJsConsoleManager]
14:20 CIA-49 bioclipse: [JsConsoleManager]
14:20 CIA-49 bioclipse: added 'eval' method, for running js code
14:28 CIA-49 bioclipse: carl_masak * r6374 /bioclipse2/trunk/plugins/net.bioclipse​.scripting.ui/src/net/bioclipse/scripti​ng/ui/business/JsConsoleManager.java: [JsConsoleManager] implemented eval
14:41 egonw olas:
14:41 egonw final String mapping = "org/openscience/cdk/dict/da​ta/cdk-sybyl-mappings.owl";
14:41 egonw AtomTypeMapper mapper = AtomTypeMapper.getInstance(mapping);
14:41 egonw Assert.assertNotNull(mapper);
14:41 egonw Assert.assertEquals("C.3", mapper.mapAtomType("C.sp3"));
14:41 egonw olas: do you like that?
14:41 egonw I guess he will :)
14:48 CIA-49 bioclipse: carl_masak * r6375 /bioclipse2/trunk/plugins/net.bioclipse.scripti​ng.ui/src/net/bioclipse/scripting/ui/business/ (IJsConsoleManager.java JsConsoleManager.java):
14:48 CIA-49 bioclipse: [IJsConsoleManager]
14:48 CIA-49 bioclipse: [JsConsoleManager]
14:48 CIA-49 bioclipse: added executeFile(IFile) and executeFile(String)
14:49 CIA-49 bioclipse: carl_masak * r6376 /bioclipse2/trunk/plugins/net.bioclipse.scripti​ng.ui/src/net/bioclipse/scripting/ui/business/ (IJsConsoleManager.java JsConsoleManager.java):
14:49 CIA-49 bioclipse: [IJsConsoleManager]
14:49 CIA-49 bioclipse: [JsConsoleManager]
14:49 CIA-49 bioclipse: corrected return type
14:54 malick joined #bioclipse
14:55 malick joined #bioclipse
15:01 shk3 ok, cdk1.0 finds 562 templates with the fingerprints, cdk trunk 15 (and none with exact match)
15:01 shk3 try to find the difference
15:07 CIA-49 bioclipse: carl_masak * r6377 /bioclipse2/trunk/plugins/net.bioclipse.scripti​ng.ui/src/net/bioclipse/scripting/ui/business/ (IJsConsoleManager.java JsConsoleManager.java):
15:07 CIA-49 bioclipse: [IJsConsoleManager]
15:07 CIA-49 bioclipse: [JsConsoleManager]
15:07 CIA-49 bioclipse: implemented executeFile
15:14 shk3 egonw?
15:15 egonw yes
15:17 shk3 where is the code for regeneration of the fingerprints for the modelbuilder3d?
15:21 shk3 ah, seems to be templateextractor
15:22 masak @tell olas 'run script' works. jonalv++
15:22 zarah Consider it noted.
15:23 CIA-49 bioclipse: jonalv * r6378 /bioclipse2/trunk/plugins/ (5 files in 5 dirs): It is now possible to run js files with the play button from the js editor
15:35 egonw shk3: yes, int he builder3dtools module
15:36 shk3 i get totally different fingerprints
15:36 shk3 the fingerprint files are just copied to trunk
15:36 shk3 so I suppose this is the reason
15:37 shk3 but I can't regenerate the fingerprints
15:37 CIA-49 bioclipse: jonalv * r6379 /bioclipse2/trunk/plugins/net.bioclipse​.scripting.ui/src/net/bioclipse/scripti​ng/ui/business/JsConsoleManager.java: Running js scripts from the editor now gives feedback if they fail
15:37 shk3 the templateextractor does not work
15:39 egonw I though steinbeck had fixed that...
15:39 egonw ok, this is why those unit tests for the fingerprints are useful...
15:39 shk3 well, not sure whats the problem
15:40 shk3 the iteratingmdlreader does not iterate a single time
15:40 egonw fingerprints should not change, unless deliberately changed
15:40 egonw now they almost seems to change randomly every week :(
15:40 shk3 well, no idea why they are different
15:41 malick left #bioclipse
15:41 shk3 can you try the fingerprint generation?
15:42 shk3 perhaps I do something wrong
15:42 egonw just run TemplateExtractor.main() ?
15:42 shk3 needs some parameters
15:42 shk3 the input file
15:42 shk3 the output file
15:42 shk3 and two booleans I have no idea about
15:42 egonw might you please pass me the full command line run?
15:43 egonw hahaha :)
15:44 shk3 #/home/shk3/workspace/org.openscience.cdk​/src/main/org/openscience/cdk/modeling/bu​ilder3d/data/ringTemplateFingerprints.txt /home/shk3/testfinger true true
15:44 shk3 without the #
15:44 egonw ack
15:44 shk3 is what I did
15:44 egonw ok, had a look at the two bools
15:44 egonw anyAtom ad anyAtomAnyBond
15:45 egonw IIRC, this is about how to match a ring system
15:45 shk3 yes, but what are we doing?
15:45 egonw right :)
15:45 egonw hang on
15:45 egonw about the explain that :)
15:45 egonw given this ring: C1CCCC1
15:45 egonw five ring
15:45 egonw it does then not match C1OCCC1
15:46 egonw however, if anyAtom == true, then it does
15:46 shk3 ok
15:46 egonw this is to reduce the problem of having combinatorial explosion of possible ring systems
15:46 egonw and address the limited input Christian used
15:46 egonw sort of, for providing back ups
15:47 egonw anyAtomAnyBond likely adds the same for bond orders
15:47 egonw and you can imagine that using C1=CCCC1 as template for C1CCCC1 has more consequences than
15:47 egonw using C1OCCCC1 for C1CCCC1
15:47 shk3 ok, so true true is used?
15:47 egonw no idea
15:48 egonw I really don't
15:48 shk3 as I said, for me the iterating reader doesn't read
15:48 egonw didn't Christian not write a CDK News about it?
15:48 shk3 perhaps you just try with true true
15:48 egonw yes
15:48 shk3 might be
15:48 shk3 I will look for it
15:48 shk3 if it works at all
15:48 * egonw is running 'ant dist-large'
15:48 egonw to allow running the tool
15:49 egonw there should be a unit test for the IteratingMDLReader...
15:49 egonw do those tests fail for you?
15:49 shk3 i did it in eclipse
15:49 shk3 perhaps that was the problem?
15:49 shk3 but the files were found
15:49 shk3 mom
15:52 shk3 they seem to work
15:59 shk3 same problem on command line
16:00 egonw ok that command line has one principle flaw
16:00 egonw the file is gzipped
16:00 shk3 i ungzipped it before
16:01 egonw ok, the tool seems to require that indeed
16:02 egonw ok, nothing much imprssive happens
16:02 egonw Fingerprints:0 are written...ready
16:03 shk3 same for me
16:03 shk3 and this is because the iteratingmdlreader does not iterate
16:03 egonw ok, second problem in tat command line you gave
16:03 shk3 but no idea why
16:03 egonw that .txt is for the output
16:03 shk3 oh
16:04 egonw ringTemplateStructures.sdf.gz is what you need
16:04 * egonw is checking if that works
16:04 egonw that looks much better :)
16:04 egonw Generated Fingerprints: 1400  /
16:05 egonw Fingerprints:10751 are written...ready
16:05 egonw Computing time statistics:
16:05 egonw {3=16, 2=56, 1=258, 0=10404, 6=2, 5=5, 19=1, 4=6, 22=1, 8=1, 50=1}
16:05 shk3 looks better
16:05 egonw no idea what those stats are about :)
16:05 egonw ok, got to buy some food now
16:05 egonw bbl
16:05 shk3 will try with the new fingerprints
16:06 egonw great
16:06 egonw please take the first 10 structures from the SDF file
16:06 egonw and the newly generated FPs
16:06 egonw and make a (single) test for it
16:06 shk3 will do so
16:06 egonw to ensure we can detect when it FPs change again
16:06 egonw thanx
16:52 rojasm joined #bioclipse
17:08 edrin left #bioclipse
19:40 olas @msg
19:40 zarah olas: masak said 'run script' works. jonalv++
19:41 olas left #bioclipse

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