Camelia, the Perl 6 bug

IRC log for #bioclipse, 2009-02-16

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

All times shown according to UTC.

Time Nick Message
07:15 egonw joined #bioclipse
07:38 edrin joined #bioclipse
07:39 edrin hi morning
07:41 egonw hi edrin
07:41 edrin hi egonw
07:42 egonw can you put those (or a few illustrative) scripts on github?
07:43 edrin yes, but we should have the coffee machine sample running at ws1
07:44 egonw ack
07:45 edrin they are showing really much of the xws4j API
07:45 edrin i used them for development from bc1, now they are running in bc2.
07:48 edrin egonw: I moved all the thing from disc to intro now, I hope this reads better now?
07:48 egonw please write the abstract and methods section
07:49 egonw only then can we judge what needs to go into the intro and not
07:49 edrin ok
07:53 sneumann joined #bioclipse
08:02 Gpox joined #bioclipse
08:18 Gpox joined #bioclipse
08:22 egonw hi Gpox
08:22 egonw Gpox: what's your work plan for this week?
08:23 Gpox not sure
08:34 sneumann hi there
08:34 sneumann has this plugin arrived in BC2 yet ?
08:34 sneumann http://wiki.bioclipse.net/ind​ex.php?title=RdfEditor_plugin
08:35 zarah sneumann's link is also http://tinyurl.com/c343uo
08:35 sneumann or is it still 1.x ?
08:36 egonw that's not ready yet
08:47 rojasm joined #bioclipse
08:49 egonw Gpox: meanwhile...
08:50 egonw I am working on the pref stuff for JCP in BC2
08:53 egonw rojasm: did you see Gilleain's blog about rendering of LPs and radicals?
08:54 egonw rojasm: and Zest has good means of laying out graphs...
09:04 rojasm yes
09:04 rojasm impresive
09:05 rojasm what about Duan Lian with his jsmoleditor
09:06 rojasm will it be competence for jchempaint?
09:06 rojasm I don't think competence is the best word mm
09:19 egonw competition
09:19 egonw yes, I think it might very well be
09:19 egonw just like JMEEditor, etc are too
09:26 egonw but only in web applications...
09:26 egonw moreover...
09:26 egonw JCP does way more...
09:32 rojasm yes that is true. But I  think cdk should focus to be more accessible from the browser like now you are doing with the web service.
09:36 egonw stefan is doing new applets...
09:37 egonw and the new design allows quite small applets too
09:39 egonw cool
09:39 egonw jcpprop manager works :)
09:42 masak joined #bioclipse
09:49 sneumann egonw: can JCP do hierarchical reactions ?
09:49 egonw no, but rojasm stuff can
09:49 sneumann Think of a -> b + ( C -> D + E)
09:49 egonw sneumann: oh...
09:50 egonw did you mean that kind of RDF??
09:50 egonw the reaction SDF so to say?
09:50 egonw not resource description framework?
09:50 sneumann yup
09:50 egonw oh...
09:50 sneumann no ...
09:50 egonw ummm...
09:50 egonw that RDF plugin does what you need
09:50 egonw rojasm: ping
09:50 egonw it's currently in outdated, because it needs updating for the new JChemPaint
09:51 egonw while rojasm's rdf plugin still depends on cdk10
09:51 egonw but that should be portable now...
09:51 egonw won't have time for that this week myself...
10:04 egonw finish metware first
10:04 egonw and comment on CMLSpect4MB
10:09 jonalv joined #bioclipse
10:09 jonalv oh hai
10:10 egonw hej jonalv
10:10 maclean joined #bioclipse
10:10 jonalv how are you?
10:10 egonw me?
10:10 egonw really, really busy...
10:11 egonw but close to finishing the first pref window for JCP settings in BC2
10:11 jonalv sounds good
10:11 * jonalv is studying for that test on thursday...
10:42 egonw blah...
10:43 maclean blah?
10:43 egonw right
10:43 egonw I got all the pref code in place (except applying the prefs upon opening a new editor)
10:43 egonw but
10:43 maclean but...
10:43 egonw It changes the settings once...
10:43 egonw but not back
10:43 egonw :(
10:43 maclean :(
10:44 maclean hmmm
10:44 egonw RendererModel model = this.getRendererModel(editor);
10:44 egonw model.setShowAromaticity(getShowAromaticity());
10:44 egonw I mean...
10:44 egonw what's wrong with that?
10:44 egonw why would that code set it only once??
10:44 maclean no idea
10:44 maclean hmm
10:45 egonw I mean... it *is* called twice...
10:45 egonw so , that's not the problem :(
10:45 maclean this doesn't seem to be the problem that Rajarshi came across (non-shared instances)
10:46 maclean what happens in getRendererModel(...)/
10:46 maclean s#/#?#
10:47 egonw OK... the Preferences is broken :(
10:47 egonw it always returns false...
10:47 maclean your class?
10:48 egonw guess so...
10:48 egonw not sure...
10:49 maclean a bug in eclipse would be more difficult to fix :/
10:49 egonw indeed
10:49 egonw can't imagine that would be the problem...
10:49 maclean really? I can :) - even eclipse has bugs!
10:50 egonw hahaha...
10:50 egonw found the problem
10:50 maclean good
10:50 egonw problem due to that all activitators are called Activator !
10:50 maclean ah!
10:50 egonw bloody idiot who suggested that...
10:50 egonw it was using the Activator of net.bioclipse.ui
10:50 maclean ohhhh
10:50 egonw so, it would not find the correct prefs
10:51 maclean naturally
10:51 egonw and never find any JCP prefs..
10:51 egonw so always returned false :)
10:53 egonw another one such bug
10:54 egonw OK, fixed
10:54 maclean yay
10:55 egonw OK, it only does 2 properties right now...
10:55 egonw but the rest is just more of the same
10:56 maclean hah! "the rest are left as an exercise for the user" :)
10:56 egonw hehehehe
10:56 egonw junior jobs
10:56 maclean hmmm. I forgot zoom.
10:57 maclean although, since i can't get !fitToScreen to work...
10:57 egonw oh, btw...
10:57 egonw relating to that...
10:57 maclean hmmm?
10:57 egonw drawing on an empty pane is working in JCP, right?
10:57 maclean yup
10:57 egonw can you talk to Gpox about that?
10:58 maclean sure
10:58 egonw so, to make that work in BC2 too
10:58 egonw tack
10:58 maclean it's drawing at (0,0)
10:58 maclean so it's probably to do with setting up the transform
10:58 egonw yes
10:59 maclean that jsMolEditor is bloody great, by the way
10:59 maclean and it contains a mix of code from all sorts of people :)
11:00 egonw yes, I envision that something like that may be more useful in NMRShiftDB than a JCP applet
11:00 egonw then again...
11:00 egonw no forget it...
11:01 egonw NMRShiftDB needs highlighting etc...
11:01 egonw I still like it to have hover over boxes with shifts...
11:01 egonw those yellow thingies
11:01 maclean ToolTips
11:02 egonw right :)
11:02 maclean the RenderModel has Map<IChemObject, String> for that
11:02 egonw Gpox: ping...
11:02 maclean it's not used at the moment, though
11:03 egonw mmm... me feels a SpecMolEditor patch peeking around the corner
11:04 CIA-55 bioclipse: gilleain * r8843 /bioclipse2/trunk/plugins/net.biocli​pse.seneca/src/net/bioclipse/seneca/ (2 files in 2 dirs): BestStructureView now compiles again, and is cleaned up, but is far from working...
11:04 maclean certainly
11:04 egonw ah crap...
11:04 egonw if days only had 48 hours :(
11:04 maclean :}
11:11 maclean I can't reproduce bug 170
11:11 maclean which depends on bug # 467
11:11 maclean @bug 170
11:11 zarah http://tinyurl.com/aalgxr
11:11 maclean @bug 467
11:11 zarah http://tinyurl.com/c6x6os
11:12 * masak has slightly bad conscience over using tinyurl for the above task
11:13 egonw maclean: how so?
11:13 rojasm sorry sneumann I was out
11:13 maclean egonw: I can open emptyFile.mdl just fine
11:14 egonw masak: yes, having the title would be much more useful
11:14 egonw maclean: can you draw on the empty pane?
11:14 masak egonw: I was mainly talking about not creating a tinyurl.
11:14 egonw without having the atom in the top right corner?
11:14 maclean no, but that's 467, not 170
11:14 egonw maclean: yes, but
11:14 masak egonw: having the title would be nice, but a bit more work.
11:14 egonw 170 is open, because the template need to create files with zero atoms again...
11:15 maclean I don't understand. "emptyFile.mdl" has no atoms
11:15 egonw maclean: renamed 170
11:16 egonw should be clearler now
11:16 maclean ok
11:16 egonw the zero atom files does open
11:16 egonw but then you end up in situation 467
11:16 maclean ohhhhh
11:16 egonw so, the template actually has two atoms right now...
11:16 egonw to force proper scaling :)
11:16 maclean riiiigh
11:16 maclean +t
11:17 maclean hmmm. this is my fault.
11:17 maclean no, wait...
11:19 CIA-55 bioclipse: carl_masak * r8844 /bot/trunk/zarah.pl:
11:19 CIA-55 bioclipse: [zarah] now has better regex for the 'tell me' case
11:19 CIA-55 bioclipse: ...making it possible to send messages to meklund.
11:19 CIA-55 bioclipse: carl_masak * r8845 /bot/trunk/zarah.pl: [zarah] no longer tiny-izes individual pz ticket urls
11:20 masak zarah: reboot time, dear.
11:20 * egonw is rewriting his local git history
11:20 egonw found I added some output to STDOUT in a commit a while back
11:20 zarah joined #bioclipse
11:20 egonw so, I remove that output to STDOUT
11:20 maclean Revisionism!
11:20 egonw then rebase --interactive, and changing the patch order, and squashing ...
11:20 egonw the removal of STDOUT with the commit that introduced it :)
11:21 egonw ending up in a cleaner patch which does *not* introduce the output to STDOUT in the first place
11:21 egonw @pz 170
11:21 zarah http://pele.farmbio.uu.se/cgi-b​in/bugzilla/show_bug.cgi?id=170
11:21 egonw masak: yes, that's short enough anyway...
11:22 masak not as nice, but more environment-friendly :)
11:22 masak tinyurl urls are a scare resource.
11:22 egonw indeed
11:23 maclean egonw: should an empty file start with a gray background, or white?
11:23 egonw white
11:23 maclean ok
11:24 egonw "HowTo: Dual Booting the N810"
11:24 egonw come on... I mean... a dual boot phone??
11:25 * masak doesn't see why not
11:25 egonw oh, wait... can you call me back in a few minutes, then I'll reboot into linux so that we can live editor/share that document
11:25 egonw something like that?
11:26 masak possibly.
11:26 egonw this is funny
11:26 masak a phone is much more than a device for making calls these days.
11:27 maclean not mine
11:27 egonw masak: you cannot even single boot your phone!
11:27 egonw :)
11:27 masak :)
11:27 masak I can single boot all my mobile phones.
11:28 egonw hahaha
11:28 egonw the complete set of them
11:28 * masak likes vacuous truths
11:29 * egonw commiting the JCP Global Props patch
11:29 maclean egonw : shouldn't the Bioclipse JCP be painting the entire ChemModel, not just one atom container?
11:29 CIA-55 bioclipse: egonw * r8846 /bioclipse2/trunk/plugins/ne​t.bioclipse.cdk.jchempaint/ (9 files in 5 dirs): Added a preference page + manager for global JCP properties
11:31 CIA-55 bioclipse: egonw * r8847 /bioclipse2/trunk/plugins/net.bioclipse.cd​k.jchempaint/src/net/bioclipse/cdk/jchempa​int/widgets/JChemPaintEditorWidget.java: Apply global JCP properties when a file is opened
11:33 egonw maclean: yes
11:33 maclean hmmm
11:33 egonw ok, file a bug report...
11:33 egonw that might not be fixed in 2.0, but surely 2.1
11:33 egonw rojasm's RDF editor needs it anyway...
11:33 maclean indeed
11:34 maclean and painting reactions is a whole _other_ can of worms
11:34 maclean I anticipate bugs...
11:35 maclean (especially when editing a reaction, where you can have a chem-model with a separate set of molecules, and a reactants and products molecule set :)
11:35 * egonw is heading for lunch now
11:35 egonw bbl
11:35 egonw maclean: yes, indeed
11:36 egonw but the feature concept can accomodate that...
11:36 egonw we'll have an experimental Reaction Feature
11:36 maclean ah, like a branch
11:36 egonw no, more like a label
11:36 egonw anyway...
11:36 egonw hungry now
11:36 egonw bbl
11:37 maclean smackt gut
12:02 egonw back
12:26 * maclean is getting there, but needs lunch
13:09 maclean Gpox : around?
13:09 Gpox maclean: yes
13:09 maclean how do you feel about the editor not being fit to screen, by default?
13:10 maclean (no the viewer, the editor)
13:10 maclean it makes creating molecules from scratch much easier...
13:10 Gpox thats good
13:10 Gpox it would be nice to fit them when they are first opend
13:11 maclean true, hmmm.
13:11 maclean it's just that single bonds have huuuuge highlight circles
13:12 maclean but fitting existing structures is good...
13:12 Gpox set the scale at 5 or 8 bond lenths for small molecules maby
13:12 maclean well I was thinking that
13:12 maclean some sort of sigmoid scaling, as with font sizes :/
13:13 maclean too complicated for now, though :(
13:13 maclean for example, the ring bonds in this screenshot:
13:13 maclean http://bp0.blogger.com/_h6tAVZXYvfs/SJNEvh-5tUI​/AAAAAAAAABU/eAXL6byWug4/s1600-h/sdfeditor.png
13:13 zarah maclean's link is also http://tinyurl.com/bcntt4
13:16 masak maclean: aye, those ring bonds need to be spaced relative to the ring size...
13:16 Gpox a simple way is if atomBounds.width < RendererModel.getAvarageBondLength()*8
13:16 maclean Gpox : could do, yes. something like that
13:17 maclean masak : they are - maybe that's an old screen shot...
13:17 masak ah.
13:18 maclean basically, it currently interpolates between the ring vertices and ring center, using the ringProportion paramter
13:18 Gpox highlight circles should not be scaled but relative to the highlighted object
13:19 maclean Gpox : they are scaled - but then they are zoomed in on.
13:20 Gpox the highlight could be generated in the atom/bond generator witha flag, it dont use any extra functionality so it could be in the basic ones
13:20 maclean so, if the scale is 10, and the zoom is 2, an OvalElement of width 0.2 becomes 40 px
13:20 Gpox ok highlights shoul not be scaled :)
13:20 maclean yes, yes they should
13:20 Gpox *zoomd
13:20 maclean ah. maybe not zoomed. well,
13:22 Gpox i have some ideas on highlighting/selections being put in the rendering tree with thier corsponding atom/bond
13:23 maclean in an ElementGroup?
13:24 maclean so Group(LineElement, OvalElement) and Group(TextElement, OvalElement) ?
13:24 Gpox yes, AtomGroup
13:24 maclean ok ,why?
13:26 Gpox to be able to customize how they are rendered
13:26 maclean hmmm, okay
13:27 maclean at the moment, most customisation is done in generation
13:27 maclean except font size
13:28 Gpox right now i have problem with highlighted atoms as they hava a white background but i need a background to cover the bonds
13:28 maclean yes, this is an annoyance
13:28 Gpox it would be done in the generator
13:29 maclean oh, right yes - as a group.
13:30 maclean so, have a Group(Text, Rectangle[white|gray]), say
13:30 maclean er, by that ad-hoc syntax I mean a group of the atom symbol and a backing shape
13:30 Gpox yes
13:31 Gpox with a flag not to highlight if you dont want it
13:31 maclean where would the flag live?
13:32 Gpox the Rendering properties realy needs to be broken up in different parts
13:32 maclean yes, indeed it does
13:33 Gpox in the generator maby, Highlihgtgenerator extends basic generator highlight = true, instead of a flag in the constructor
13:34 maclean I didn't understand that, sorry.
13:35 Gpox override shouldHighlight() from BasicAtomGenerator
13:35 maclean ahhh, ok I think i see
13:35 Gpox but that is only becaus we use that pattern now to add it
13:35 maclean yes.
13:36 Gpox BasicAtomGenerator, HighlightGenerator ...
13:36 Gpox ocourse we can't set it to be above or below like it is now in the JCP application
13:37 Gpox I have to think more about it
13:38 maclean sure.
13:38 Gpox Or maby wait untill reactios are implemented to get more general use cases
13:39 maclean oh, btw : I made a highlightShapeFilled parameter in RendererModel.
13:39 maclean yes, reactions will be...fun
13:39 maclean especially since I saw sneumann ask about nested reactions earlier :)
13:39 egonw well... a reaction schema...
13:39 sneumann :-;
13:39 egonw maclean: see the support rojasm added in the CDK
13:40 maclean egonw: merged recently, or still in a branch?
13:40 Gpox @pz 166
13:40 zarah http://pele.farmbio.uu.se/cgi-b​in/bugzilla/show_bug.cgi?id=166
13:41 egonw no, that's part of 1.2.x for a while already
13:41 maclean ah, right
13:44 maclean another thing to paint, oh joy.
13:48 egonw but I think we can better use different approach for reaction schema
13:48 egonw general gef stuff..
13:48 egonw and use the renderer to individual reactions, not full schema
13:48 maclean fair 'nuff.
13:49 maclean but swing-jcp...
14:07 egonw ah, true...
14:07 egonw with the applet: there is little what a bit of JS can do
14:11 maclean yes, but there's the application, too. ah, well - that's for the future :)
14:23 egonw maclean, Gpox: does RendererModel have defaults?
14:23 egonw that is...
14:23 maclean yes
14:23 egonw if I instantiate a new one, and get a value... it'll be ok?
14:24 maclean see the RenderingParameter class
14:24 maclean all variables should be instantiated with a value
14:24 maclean (and a reasonable one, at that :)
14:24 egonw Gpox: you are using that?
14:24 egonw Gpox: e.g. margin?
14:25 Gpox not right now
14:26 egonw OK, as I will introduce a number of these things in the prefs page...
14:26 egonw let's see how it goes...
14:26 Gpox there needs to be a overview of what is there, so i can take them in to account when rendering
14:26 egonw Gpox: working on that :)
14:29 maclean margin is used in the renderer, so yes you are using it :)
14:35 maclean although it doesn't seem to be :/
14:36 maclean or, it does the width, but not the height
14:40 jonalv joined #bioclipse
14:40 maclean heh : double w = box.getWidth(); double h = box.getWidth()... (copy/paste errors :)
14:54 masak that's one of the reasons I like writing that as double w = box.getWidth(), h = box.getHeight();
14:54 masak makes me think, and makes the code a little less verbose.
14:57 maclean bah - it's now just verbose in width, rather than verbose in height :)
14:57 masak :)
14:57 maclean (okay, so one less double, of course)
14:58 masak that's what I mean.
14:58 masak and I do make a line break after the comma.
14:58 masak and line things up, of course.
14:58 maclean ah, well, then like it again.
14:59 masak it's one of the few cases in Java when you can actually be a bit terser than usual.
15:12 egonw OK, margin is in now too...
15:12 egonw that did require using some EPL code from IBM
15:12 egonw DoubleFieldEditor
15:14 maclean Is there a floating point slider, too?
15:15 maclean Swing only has an int valued one :(
15:15 egonw no, it does not seem so :(
15:15 egonw same for SWT
15:16 maclean grmph.
15:16 egonw maclean: interesting patch...
15:16 egonw should I test 8852?
15:16 maclean er... cia didn't mention it, but if its the renderer one, then sure
15:16 egonw :)
15:16 egonw Better handling of empty and single-atom molecules, by using a default scale for such cases.
15:17 maclean guess you get emails?
15:17 egonw yes
15:17 maclean that one, yes
15:17 egonw btw, we need green fluorides
15:17 egonw brown bromides
15:17 egonw and pink fluorides
15:18 egonw is the IAtomColorer interface used?
15:18 maclean err....
15:18 maclean I /thought/ so
15:18 egonw good
15:18 egonw than the default colorer might need a patch or so :)
15:19 maclean model.getAtomColor(atom, Color.BLACK);
15:19 maclean where Color.BLACK is default, I suppose.
15:19 maclean (in BasicAtomGenerator)
15:19 egonw model == RendererModel ?
15:20 egonw and RendererModel has .setAtomColorer(IAtomColorer)?
15:20 * egonw will check...
15:20 * maclean is also checking
15:21 maclean colorAtomsByType is true by default
15:21 egonw that's good
15:21 maclean oh - btw, a 'restore defaults' button could be good...
15:22 egonw BC2: done
15:22 egonw that's why I asked about a new RendererModel() :)
15:22 maclean so, yes iff colorAtomsByType is true, then the IAtomColorer is used
15:23 maclean which, by default is a CDK2DAtomColors
15:23 maclean .o0(ahhh a class with 2D in its name that slipped my purge!)
15:25 maclean ..which has no definition for Cl, Br, or F.
15:25 maclean (you did meen green Cl, and pink F, right?)
15:25 egonw yes
15:25 egonw and brown Br
15:26 maclean hmmm we also have a CDKAtomColors class which says its for 3D (!!)
15:27 maclean and it has Cl, but magenta
15:27 egonw those come from Jmol, I guess...
15:27 maclean ahh and good old CPK.
16:14 Gpox maclean: SelectSquareModule takes a IChemModelEventRealyHandler in the constructor, I set one in the my IChemModelRelay which is also passed in.
16:14 Gpox any reason it cant get it directly?
16:14 maclean shk3 made that change, I didn't really think about it
16:15 maclean I just added another constructor so that your reflection stuff didn't complain
16:15 Gpox yeah, i added square selection now :)
16:16 maclean so, what change do you suggest?
16:17 Gpox the IChemModelRelay has a setEventHanler but no get, it is probably from there you want to get it anyways unless i am missing something
16:18 Gpox or just another constructor so my reflection dosen't break ;)
16:18 maclean oh, right, you mean could the module ask the relay for the relay handler.
16:18 maclean I'll check
16:19 maclean weird I thought I added another constructor...
16:19 Gpox There is for the SeletModule
16:21 maclean ohhh. right, it was the select module, of course.
16:22 maclean no, you're right - a getEventHandler in the ChemModelRelay is the best approach, and remove the constructor from the SelectModule that takes both.
16:22 maclean (so long as egonw is okay with it?)
16:23 egonw I need some explanation into the problem...
16:24 egonw why would you want a getEventHandler...
16:24 maclean Selections need to listen to the handler
16:24 egonw setEventHandler is to register yourself as listener
16:24 egonw you should not have access to listeners...
16:24 egonw instead, whatever you do should trigger an event passed around *by the hub*
16:25 egonw instead of passing around events yourself...
16:26 maclean hmmm. okay, I'll fix the selections
16:26 maclean including removing :
16:26 maclean rectangle.setRect(rectangle.getX(), rectangle.getY(),p.x-rectangle​.getX(),p.y-rectangle.getY());
16:26 Gpox it could be packaged protected
16:28 Gpox if ICHemModelRelay and IchemModelEventRealyHandler, IControllerModel moved from the render module to the control one
16:29 maclean yeah, I wondered about that
16:29 maclean but I don't think package protected is the way to go
16:30 Gpox you are right
16:31 egonw Gpox: those three classes sounds like the should be in control, indeed
16:38 maclean Okay, so the problem is that there is no class that implements IChemModelEventRelayHandler...
16:39 maclean or, no,  the module should implement that?
16:40 Gpox i implement it in jcpeditor
16:41 maclean yeah, I'm just checking swing-jcp
16:42 maclean riight, the JChemPaintPanel implements it
16:47 maclean aha. the hub never calls selectionChanged, as it has no idea about selections.
16:49 egonw maclean: ummm... not upon setSelection() ?
16:49 rojasm joined #bioclipse
16:49 maclean nope
16:49 egonw maclean: btw, tested that patch...
16:50 egonw and the first atom now shows up in the middle of the screen
16:50 egonw but upon adding a second (bonded) atom...
16:50 egonw ... it seems to rescale
16:50 egonw loosing the 8 bond-length dimension of the screen
16:50 egonw OK, enough for today...
16:50 egonw going home (5m away from the laptop) and make dinner
16:51 maclean ok
17:15 maclean zarah: tell egonw I have a solution; it involves adding a method select(ISelection) to the hub, so that the hub could then call the selectionChanged method.
17:15 zarah Consider it noted.
17:15 maclean tack
19:47 egonw @msg
19:47 zarah egonw: 2 h 32 m 37 s ago, maclean said I have a solution; it involves adding a method select(ISelection) to the hub, so that the hub could then call the selectionChanged method.
19:48 egonw @tell maclean sounds good... where does ISelect come from?
19:48 zarah Consider it noted.
19:48 maclean oh, well it's in the cdk.renderer.selection pacakge
19:48 zarah maclean: You have new messages. Write '/msg zarah @messages' to read them.
19:49 maclean it could stand to be renamed to "ICDKSelection", also to avoid naming conflict with eclipse ISelection.
19:50 egonw I suggest IChemObjectSelection
19:51 maclean Okay.
19:51 maclean I'm currently trying to resolve some of sam adams bugs
19:51 egonw ok
19:51 egonw one thing is arom ring depiction
19:52 maclean yes - Arvid didn't use the RingGenerator, so rings won't show as circles. Easily fixed.
19:52 egonw e.g. bug #504
19:52 egonw yes, important fix
19:52 egonw I'll assign to you
19:52 egonw then
19:53 maclean ah, no that doesn't fix that.
19:53 egonw if you can easily fix it... OK?
19:53 maclean I tried.
19:53 egonw oh...
19:53 maclean it's weird - the outline says that they are aromatic
19:53 maclean and if I add a phenyl (with the RingGenerator added) I get a circle on the added phenyl
19:54 maclean anyway, yes - assign it to me, I might as well try fixing it - just warning you that I may come back to you for advice...
19:55 egonw no worries
19:55 egonw that's fine
19:55 egonw mom...
19:55 egonw need to finish the dishes
19:55 maclean okay :)
20:02 egonw back
20:03 maclean okay, I see that the Ola's revolt over bond moving is gathering pace :/
20:03 maclean I've just been trying to resolve it, but it's very hard
20:03 maclean as it is very fundamental to the rendering
20:04 egonw which one is that?
20:05 maclean @bug 494
20:05 egonw @pz 494
20:05 zarah http://pele.farmbio.uu.se/cgi-b​in/bugzilla/show_bug.cgi?id=494
20:05 zarah http://pele.farmbio.uu.se/cgi-b​in/bugzilla/show_bug.cgi?id=494
20:05 maclean "When adding/moving atoms, other atoms change position slightly"
20:05 maclean I maintain that it is due to my implementation being TOO good :p
20:07 egonw ah, that thing...
20:07 egonw yes, fit-to-screen is typically unwanted in editing mode...
20:07 edrin dman
20:07 egonw hej edrin
20:07 edrin today my cells died
20:08 edrin because i made the medium pH 11.5
20:08 edrin damn
20:08 edrin hi egonw
20:08 edrin I have to repeat the experiment
20:09 edrin and put less NaOH to the medium
20:10 edrin when they are dead they are dead
20:12 egonw yes, no frankenstein that will bail you out...
20:12 maclean As a trained biologist, I can confirm that dead == dead.
20:13 edrin i tried a ph shift from ~6 to 8.5
20:14 edrin and added "35 mM" NaOH as indicated in some paper...
20:14 edrin put this was not a good idea
20:14 edrin i should have considered that this concentration is just stupid
20:14 edrin it entirely depends on the medium you add it to
20:15 edrin at least it is just the fungus
20:16 maclean what is your experiment?
20:17 edrin to induce/inhibit reaction of my fungus on ph shift
20:18 maclean ah
20:18 edrin the fungus should show a certain reaction on the shift
20:18 edrin and this i want to test
20:18 maclean but not too large a shift, clearly :/
20:19 edrin yeah
20:19 edrin they all were dead
20:19 edrin i checked with propidium iodide
20:20 edrin i could see the nuclei
20:20 edrin thus PI entered the cells
20:20 edrin and then i check the ph
20:20 edrin and noticed
20:20 edrin 11.5
20:20 edrin ok
20:20 edrin not good
20:20 edrin will try again
20:20 edrin now i will go home
20:20 maclean wast it meant to be µM?
20:20 maclean oh, actually I have to get the bus ack!
20:21 egonw oops
20:21 edrin heh
20:21 edrin :)
20:21 egonw I hope he did not miss it...
20:21 edrin he reacted fast
20:21 edrin less then 1 minute
20:21 edrin than
20:39 edrin left #bioclipse
21:06 maclean joined #bioclipse
22:43 edrin joined #bioclipse
23:08 CIA-55 bioclipse: gilleain * r8852 /bioclipse2/trunk/cdk-externals/trunk/​org.openscience.cdk.render/src/org/ope​nscience/cdk/renderer/Renderer.java: Better handling of empty and single-atom molecules, by using a default scale for such cases.

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