Camelia, the Perl 6 bug

IRC log for #cdk, 2009-06-11

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

All times shown according to UTC.

Time Nick Message
04:07 Gpox joined #cdk
05:33 bag joined #cdk
08:36 jbrefort joined #cdk
10:20 sneumann joined #cdk
12:07 CIA-42 cdk: gilleain * r14607 / (5 files in 4 dirs): More general implementation of atom labelling, allowing more combinations of charge/isotope/atomnumber/implicit hydrogens.
12:12 shk3 joined #cdk
13:29 jbrefort joined #cdk
14:02 egonw joined #cdk
14:39 zarah joined #cdk
14:39 CIA-42 cdk: gilleain * r14608 /cdk/branches/jchempaint-primary/src​/main/org/openscience/cdk/renderer/ (3 files in 2 dirs): Made reaction-specific generators a constructor argument, so that they can be added (or not) as necessary. Also changed two generators to make it possible to do this.
14:43 CIA-42 cdk: gilleain * r14609 /cdk/branches/jchempaint-primary/src/main​/org/openscience/cdk/renderer/generators/ (BasicAtomGenerator.java ExtendedAtomGenerator.java): Display isotope information. Does not yet show for carbons unless carbons are explicitly shown.
14:43 maclean joined #cdk
14:44 maclean hey shk3.
14:44 maclean I committed the isotope stuff, but I realise that it does not show them unless the atom symbol is shown.
14:45 maclean part of the problem is the convoluted logic for determining what is or isn't shown.
14:45 maclean it's not very maintainable or extensible, but I can't think of a better way at the moment.
14:46 shk3 that's the same problem as with numbers, right?
14:48 maclean well it works now for numbers
14:49 CIA-42 cdk: shk3 * r14610 /cdk/branches/jchempaint-primary/src​/main/org/openscience/cdk/renderer/ (6 files in 2 dirs): made the "show reaction boxes" setting to have an actual effect
14:49 shk3 ah ok
14:49 maclean oh, right, so the isotopes could be at the atom positions, too?
14:49 maclean hmmm.
14:49 maclean diagrams with isotope numbers, and more than 10 carbons could get very confusing...
14:50 maclean oh well.
14:50 CIA-42 cdk: shk3 * r14611 /jchempaint/trunk/src/main/org/opens​cience/jchempaint/RenderPanel.java: using the new reaction generation
14:53 shk3 are isotopes number only?
14:53 shk3 or always with symbol?
14:53 shk3 13 or 13C?
14:53 shk3 should be at the center anyway, I think.
14:54 maclean nooo
14:54 maclean they are generally 13C where the 13 is superscript top-left.
14:55 maclean what I have not done is to fix labels at nice looking positions
14:56 maclean like numbers should usually be NE/NW/SE/SW, not N/E/S/W.
14:56 maclean but it's the other way round with implicit hydrogens.
14:58 shk3 I am happy it's there
15:04 CIA-42 cdk: gilleain * r14612 /jchempaint/trunk/src/main/org/opens​cience/jchempaint/RenderPanel.java: Minor tidying.
15:04 maclean Errr. shk3 : did you just open a bug on isotopes not being rendered?
15:06 maclean I'll make it a little more specific, with comments...
15:06 shk3 no
15:06 shk3 I assigned it to you
15:06 shk3 since you are working on that
15:08 maclean hmmm null pointer exception.
15:16 maclean if you make a 5-membered ring, then add a bond to one of the atoms, then right-click on the added atom (the terminal CH3) you get a NPE.
15:19 shk3 trying
15:21 shk3 seems to be specific for atoms in chains
15:22 maclean hmmm
15:23 maclean also : you made the AtomAtomMappingModule, yes? Because the @author tag says Niels Out...
15:24 shk3 yes
15:25 shk3 the problem is update of these tags is not done in jcp projects
15:25 shk3 it needs to be set up
15:25 shk3 can't remember exactly
15:25 shk3 what was the problem
15:25 shk3 it wasn't a just do it thing
15:25 maclean oh, right. I don't care, particularly, I just noticed it :)
15:26 maclean I see what you mean about chains
15:26 maclean if you add multiple atoms to the ring, any of them trigger a NPE.
15:26 maclean it's the 2D coordinates that are null.
15:29 maclean actually, for that matter, Niels Out has apparently written the whole of the ControllerHub... :)
15:30 maclean Aha! There is a call to builder.newAtom(atomType) in the addAtomWithoutUndo method. That might be it.
15:40 CIA-42 cdk: gilleain * r14613 /cdk/branches/jchempaint-primary/src/main/o​rg/openscience/cdk/renderer/Renderer.java: Removed unused imports
15:42 maclean oh fucking hell. updateAtom is a really horrible method to be calling so often D:
16:01 egonw yes, should not be called too often
16:22 CIA-42 cdk: gilleain * r14614 /cdk/branches/jchempaint-primar​y/src/main/org/openscience/cdk/ (8 files in 2 dirs): Use the bounds calculation methods from the renderer.
18:06 sneumann_ joined #cdk
18:24 timvdm joined #cdk
20:27 jbrefort joined #cdk
21:01 bag joined #cdk

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