Camelia, the Perl 6 bug

IRC log for #bioclipse, 2010-02-15

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

All times shown according to UTC.

Time Nick Message
06:20 sneumann joined #bioclipse
07:57 sneumann joined #bioclipse
08:01 Gpox joined #bioclipse
08:03 olass joined #bioclipse
08:16 mgerlich joined #bioclipse
08:16 jonalv joined #bioclipse
08:27 masak joined #bioclipse
08:41 s_wolf joined #bioclipse
08:53 egonw joined #bioclipse
08:54 egonw moin
08:54 zarah privet egonw
09:01 egonw Gpox: ping
09:02 Gpox egonw: pong
09:02 egonw I'm updating bioclipse.cheminformatics for the new CDK / JCP
09:02 egonw is the 's' variable in PeriodicTableView.java
09:02 egonw line 192
09:02 egonw refering a symbol?
09:04 Gpox yes
09:04 egonw ack
09:07 annzi joined #bioclipse
09:23 egonw Gpox: ping
09:25 Gpox egonw: pong
09:25 egonw JChemPaintWidget... line 125
09:25 egonw AtomNumberGenerator(new Vector2d(7,-7))
09:25 egonw there is only a constructure with zero arguments
09:25 egonw not sure what to do with this vector
09:32 Gpox missing patch from cdk-externals
09:33 Gpox ignoring it makes the numbers apear right on top of the symbols
09:37 egonw Gpox: ok, please backport that patch
09:38 egonw SADI-interested: http://plindenbaum.blogspot.com/2010/0​2/semantic-web-services-with-sadi.html
09:38 zarah egonw's link is also http://tinyurl.com/y9np79r
09:40 edrin joined #bioclipse
09:40 edrin hi
09:40 zarah hello edrin, you fantastic person you
09:51 masak hello edrin, you fantastic person you
09:56 egonw joined #bioclipse
10:09 egonw Gpox: another thingy...
10:09 egonw I would appreciate if you could look at the RenderModel.getRenderingParameter(Class) ...
10:10 egonw Gpox: d'you think we could morph that method depdending to return T, following the T in the pass IRenderingParameter<T>?
10:10 egonw I guess not...
10:11 Gpox I'll have a look
10:11 Gpox atomnumber_offset backport up on pele
10:11 egonw OK, tack ...
10:12 egonw Gpox: just realizing that that getRenderingParameter should just return the IRenderingParameter...
10:12 egonw and not the value ...
10:12 egonw otherwise, you cannot 'set' it either...
10:13 egonw OK, going offline first now
10:14 annzi left #bioclipse
10:14 annzi joined #bioclipse
11:38 egonw joined #bioclipse
12:30 egonw Gpox: here goes... are you ready?
12:30 egonw everyone: big CDK 1.3.2 update in bioclipse.cheminformatics
12:31 egonw 1. extracting the CDK plugins into a new org.openscience.cdk repository (my github account)
12:31 egonw 2. removal of those plugins from bioclipse.cheminformatics
12:31 egonw that includes the 3rd party deps
12:31 egonw 3. updates in bioclipse.cheminformatics for API changes
12:36 annzi egonw: about chebi id's …when I look into molecule properties I find <http://rdf.farmbio.uu.se/chembl/snorql/?desc​ribe=http%3A%2F%2Fbio2rdf.org%2Febi%3A514447> but for it to correspond to bio2rdf shouldn't it say chebi:514447 instead?
12:36 zarah annzi's link is also http://tinyurl.com/yabxtjv
12:38 egonw annzi++
12:39 egonw please file another issue at http://github.com/egonw/chembl.rdf
12:39 zarah egonw's link is also http://tinyurl.com/ygu5bqw
12:39 annzi will do
12:47 egonw annzi: tack
12:48 annzi np
12:49 egonw Gpox: patches commiteed
12:49 egonw -e
13:02 Gpox possible problem and collition with  org.openscience.cdk.annotation and iofomats in core/externals
13:05 egonw Gpox: indeed ...
13:05 egonw olass: ping
13:05 olass egonw: pong
13:05 egonw regarding those two cdk plugins...
13:05 egonw what do you prefer...?
13:05 egonw keep separate copies in bioclipse.core
13:06 olass ?
13:06 egonw or use those from the new repository?
13:06 olass please explain
13:06 olass write an email to bioclipse-devel?
13:06 olass why do we need copies of the plugins?
13:06 olass deps in core?
13:06 olass Gpox: is there a way around having deps in core?
13:07 egonw yes, for ioformats
13:07 egonw which is used for format detection
13:07 egonw and we wanted to do that, without depending on bioclipse.cheminformatics
13:07 egonw ioformats has no deps on the CDK
13:07 egonw but is developed as part of the CDK
13:10 olass a copy sounds fair with me then
13:11 egonw ok, well, it is the current solution
13:11 olass egonw: please email bioclipse-devel about this transition and how to update our git checkouts accordingly
13:11 olass however
13:12 olass with the new bioclipse git repo coming up, is this really a good solution?
13:12 olass I would much rather prefer to have all Bioclipse-developed stuff in this repo
13:12 olass like JCP and SDFEditor
13:12 olass since these clearly are bioclipse-developments rather than CDK developments
13:12 olass so we should factor these out IMO
13:13 olass JCP-SWT is a bioclipse-project
13:13 olass and so is MolTable
13:13 egonw sorry, please explain
13:13 olass extract moltable and jcp-swt into a new repo separate from the new cdk repo
13:13 egonw JCP-SWT still is in a bioclipse repos
13:13 olass oh, thought that was in bioclipse.cheminformatics?
13:13 egonw there is no such talk about extracting moltable and jcp-swt into a separate repos
13:14 egonw not at this moment...
13:14 egonw well...
13:14 olass well, I just brought it up
13:14 egonw besides factoring out the CDK plugins...
13:14 egonw so that e.g. KNIME can use the CDK plugins easily
13:14 egonw too
13:14 egonw ...
13:14 * olass does not care about KNIME
13:14 egonw we have also been talking about making separate *plugins* for SWT wrappers...
13:15 egonw but don't think we wanted those plugins outside Bioclipse repositories ...
13:15 olass what do you have against moving out the JCP and the moltable out into a separate bioclipse.cdk repo?
13:15 egonw nothing
13:15 olass settled then
13:16 * egonw is confused
13:16 * olass did not understand that this was so high prio
13:16 egonw what is?
13:16 olass what you are doing, factoring out plugins
13:17 olass into separate repos
13:17 egonw you mean, just updating to CDK 1.3.2 ?
13:17 egonw that's not refactoring...
13:17 olass no, factoring out plugins into separate repos
13:17 egonw and certainly not the bioclipse.sdk you just requested
13:17 olass did you not say a new repo?
13:17 egonw bioclipse.cdk you just requested
13:17 * olass is confused
13:17 egonw that makes two of us
13:17 olass skype call?
13:18 egonw sure
13:18 olass IRC is so limited sometimes...
13:18 Gpox copies in bioclipse.core and a note that thees exist in two places, to simplify transiont to the new layout. The rest is another discussion
13:20 Gpox having them in core saves the need to have org.opensience.cdk.git checked out when not needing the checmoinformatics feature
13:25 * olass now understands egonw :)
13:29 egonw Gpox, olass: actually, just the version numbers change :)
13:29 olass ack
13:29 egonw and a bit of JavaDoc, but no changes in cdk.annotation, cdk.ioformats otherwise
13:29 Gpox a version changed when nothing has changed? :)
13:32 egonw yeah, following upstream versioning :)
13:52 egonw mmm... no CIA bot today?
14:11 egonw Gpox: ping
14:12 Gpox egonw: pong
14:12 egonw in our JCP branch...
14:12 egonw is there still a separation between
14:12 egonw scale-to.fit
14:12 egonw and zoom?
14:12 egonw those should be two separate rendering parameters
14:13 egonw zoom by default being 1.0
14:13 egonw scale to fit a boolean making sure that when zoom=1.0 the scene fits the drawing area
14:13 egonw e.g. zoom=1.0 should be the case in the mol table
14:14 egonw there is no zoom there, so it will stay 1.0 always
14:14 egonw the scale-to-fit ensures that the whole molecule still fits into the table cell
14:14 Gpox I don't know, have to check
14:14 egonw ack
14:14 egonw no rush
14:15 Gpox so when resizing the zoom would change?
14:15 egonw no, the scale factor would
14:16 egonw the zoom would change upon middle mouse scroll or so
14:16 egonw not sure what the zoom default mouse action for zooming is on OS/X :)
14:16 egonw resizing a table cell in the mol table, should change the scale factor
14:16 Gpox but if you are zoomed in and resize what then
14:17 egonw depends on application behavior...
14:17 Gpox :)
14:17 egonw I see your point...
14:17 egonw the current, requested behavior in bioclipse is to not adapt the scale factor
14:17 egonw but I think the zoom factor should not change either
14:18 egonw because the molecule does not get larger/smaller...
14:18 egonw but I think I understand where your thinking is coming from...
14:19 egonw because you still have set scale-to-fit... the only option to not change the molecule size itself, is to change the zoom
14:19 egonw is that what is currently implemented?
14:19 egonw so, we basically need a initialScaleFactor parameter?
14:21 Gpox well there is the world to screen transforma and then the zoom transform
14:21 egonw and the first involves the scale factor
14:21 Gpox yes
14:21 egonw the second the zoom factor
14:22 Gpox the first is based on canvas to atomcontainer bounds if scale to fit is set
14:25 egonw right
14:25 Gpox one could have the option to set the view to show it at some length like 10Å but that is still dependatn on the canvas size
14:26 egonw yes
14:27 Gpox just need to make the API clean so it is easy to atchiev both
14:28 egonw what d'you think of Renderer.java line 773
14:28 egonw wrt to what we just discussed?
14:28 Gpox got a patch for RenderModel.getRenderingParameter(...)
14:28 egonw oh, ok
14:28 * egonw is rewriting more rendering params into the NWO
14:29 Gpox you want to be able to do model.getRenderingParameter( AtomRadius.class ).setValue( 8.0 );
14:29 egonw because it does not require much thinking
14:29 egonw yes, that looks brilliant
14:29 egonw that's exactly what I want, but was unable to get right myself
14:29 egonw Gpox++ (for his Java generics wizardly skills)
14:30 Gpox the ? in generics is dangerous because it is not specific
14:31 egonw yeah, but I could not get something with T working, and had no clue what I was doing wrong...
14:31 egonw the ?-stuff at least made it compilable...
14:36 Gpox fix pushed to pele
16:20 edrin why is it not possible to run knime within bioclipse?
16:20 egonw theoretically, or practically?
16:20 edrin (because of spring) but is it really not possible?
16:20 edrin both
16:21 egonw spring is a problem for running bioclipse plugins in knime
16:21 egonw also the other way around?
16:22 edrin that's the question
16:22 edrin it would be cool to run kinme as a plugin in bioclipse
16:24 egonw have you tried it?
16:24 egonw ask Bernd or Thorsten if they think it should be possible...
16:25 edrin did not try it yet
17:10 edrin left #bioclipse
17:10 edrin joined #bioclipse
17:21 edrin left #bioclipse
17:26 egonw joined #bioclipse
17:38 annzi joined #bioclipse
18:53 CIA-71 joined #bioclipse
18:56 sneumann joined #bioclipse
22:19 egonw joined #bioclipse

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