Camelia, the Perl 6 bug

IRC log for #cdk, 2010-09-06

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

All times shown according to UTC.

Time Nick Message
03:31 azeem left #cdk
03:32 azeem joined #cdk
05:46 egonw joined #cdk
05:51 egonw left #cdk
06:05 bag_ joined #cdk
06:45 bag_ left #cdk
06:52 sneumann joined #cdk
07:08 Gpox joined #cdk
07:22 egonw joined #cdk
08:24 s9asad joined #cdk
08:24 maclean joined #cdk
08:26 egonw hi s9asad, maclean
08:26 s9asad Hi egonw
08:26 maclean hi egonw,
08:26 maclean just mailing you...
08:33 egonw left #cdk
08:35 egonw joined #cdk
08:36 egonw maclean: and why is the latter bad?
08:36 maclean a) it exposes the implementation
08:36 maclean b) it means you can't have IRenderers that don't use IRenderingElement
08:37 egonw ok, makes sense
08:37 egonw I'll apply anyhow
08:37 maclean Ok.
08:38 maclean It does have the advantage that it will now be possible to pass in an instance of (say) IRenderer<IAtomContainer> to a MoleculeSetRenderer constructor.
08:39 maclean So that, if you have a custom Reaction renderer, for example...
08:39 maclean :)
08:40 egonw that sounds cool
08:40 egonw that sort of makes the MoleculeSetRendere the de facto layout engine, not?
08:40 egonw as it only takes care of putting the molecules in the right place...
08:41 egonw mmm... I will have to write a tutorial on that in my blog... can't say I used the moleculesetrenderer myself yet
08:42 maclean Me either! I probably should....
08:43 maclean As for layout, I was thinking about that on the 2&1/4 hr journey to work.
08:43 maclean One way to do it would be:
08:44 maclean IMoleculeSetLayout layout = new GridMoleculeSetLayout(3, 4, ...///some other params
08:44 maclean layout.apply(moleculeSet);
08:44 maclean renderer.render(moleculeSet, drawVisitor, layout);
08:44 egonw 2.15??
08:44 egonw one way, daily?
08:44 egonw damn, I thought my situation in NL was bad
08:44 maclean One way, currently weekly :)
08:45 egonw oh, that's nothing then
08:45 egonw :)
08:45 egonw that layout API looks very nice
08:46 maclean It's okay. the proof will be in the eating.
08:47 Gpox Shouldn't AtomContainerRenderer be able to render IMolecules?
08:47 Gpox AtomContainerRenderer implements IRenderer< ? extends IAtomContainer>
08:48 Gpox *IMolecule
08:48 maclean hmmm
08:48 egonw it should, and it can, not?
08:48 maclean Well it would be nice if it did, but as usual with IMolecule/IAtomContainer, I expect difficulties.
08:49 maclean I have no idea.
08:49 egonw IMolecule extends IAtomContainer?
08:49 Gpox It might be ok becaus the definition is internal to the renderer
08:49 egonw yes, that's what I would have thought too...
08:49 egonw surely IRenderer<IMolecule> is quite different from IRenderer<IAtomContainer>
08:50 egonw mm.... would make an interesting unit test :)
08:50 egonw I need to finish two book chapters today
08:50 egonw and review a data set
08:50 egonw so won't do much coding myself today
08:50 egonw but will apply patches that come in
08:51 egonw (BTW, one big thing is that I need to clean up that 13-unsorted-patches branch...)
08:51 egonw the number of patches it getting ridiculously long
11:06 maclean left #cdk
11:47 s9asad left #cdk
12:08 maclean joined #cdk
12:19 s9asad joined #cdk
12:48 maclean left #cdk
13:12 maclean joined #cdk
13:15 egonw maclean: hej
13:15 maclean hej
13:15 egonw I am close to submitting a new set of patches to pele
13:15 egonw that is...
13:15 maclean 14?
13:15 egonw a cleaned up set of patches
13:15 egonw rather then new patches
13:15 egonw putting them in the right order...
13:15 maclean ah, good
13:15 egonw no, cleaned up 13-unsorted-patches
13:16 maclean 13-sorted-patches?
13:16 egonw yeah, something like that :)
13:17 maclean Ok, I can review them in a while, but will probably have to re-check out pele's branches.
13:17 maclean also : you are meant to be writing book chapters! >:{
13:20 egonw yeah, I know
13:20 egonw I am doing that too, but all in parallel
13:20 egonw pushing patches to Pele
13:20 egonw there are not important patches in 13 anymore
13:20 egonw just administrative stuff
13:22 egonw rerunning Nightly
13:23 egonw branch 13 compiles
13:23 egonw and so does branch 1
13:23 egonw not sure about the others yet :)
13:23 egonw but I think most patches are pretty 'split up' now...
13:24 egonw maclean: so, when I get a patch by you, I have my git magic split it up in one patch for each branch
13:24 egonw which works like a charm
13:24 maclean ah, right. sorry forgot to do that.
13:24 egonw no worries
13:24 maclean at least you had a (rare!) opportunity to use the script...
13:25 egonw yeah, I have a quite a few scripts in use, actually
13:25 egonw these I use most: updatePele, rebaseAll, splitCommit, splitPatch
13:25 egonw well, the only other one is makePaches
13:26 egonw which I use when making releases :)
13:26 egonw maclean: ok, while you are around... one question
13:26 maclean oh?
13:26 egonw I think one design mistake sneaked in... want your opinion
13:26 maclean ok
13:26 egonw Gpox: please comment too
13:27 egonw IGeneratorElement should work with world coordinates, not screen coordinates, but I think it is currently doing the last...
13:27 egonw correct?
13:27 egonw ok, will make some coffee first now
13:27 maclean hmmm. will look at it
13:31 egonw one question is, how this relates to drawing text...
13:31 egonw which requires fonts...
13:33 maclean FONTS!! Argh! the accursed word!
13:33 maclean And, yes, probably two-pass rendering.
13:34 maclean (which, BTW, can be done by a visitor that first calculates/draws text boxes, then draws other elements accordingly)
13:37 egonw maclean: I'm now implementing this feature request: http://sourceforge.net/tracker/?func=detail&amp​;aid=3047009&amp;group_id=20024&amp;atid=370024
13:37 zarah egonw's link is also http://tinyurl.com/2vvkhre
13:37 egonw and will then make release 16 later this week
13:37 egonw after that CDK 1.2.7
13:37 egonw and CDK 1.3.6
13:38 egonw then cdk-jchempaint 17 based on 1.3.6
13:38 maclean ah, ok
13:38 egonw but not today
13:38 maclean :)
13:38 maclean no, not today
13:55 maclean egonw : was there a way to see the pele patches online? I thought that there was, but can't remember where...
13:56 egonw yes
13:56 egonw mom
13:57 egonw http://pele.farmbio.uu.se/cgi-bin/gitweb.cgi
13:57 zarah egonw's link is also http://tinyurl.com/qzsq4u
13:57 egonw make sure to look at the cdk-jchempaint.git/
13:57 egonw repos
13:57 egonw I should remove the old, outdates ones, I guess...
13:58 maclean ah right, takk
14:09 egonw ok last pushes to Pele for today
14:09 maclean ok
14:09 egonw with an implementation of the feature request
14:09 maclean excellent.
14:22 CIA-98 blueobelisk: pansanel * r306 /structures/trunk/ (CMakeLists.txt xml/l10n.xml):
14:22 CIA-98 blueobelisk: Thai translation added (many thanks to
14:22 CIA-98 blueobelisk: John Francis Lee)
14:53 maclean left #cdk
14:54 maclean joined #cdk
15:09 sneumann left #cdk
15:12 maclean left #cdk
15:43 jbrefort joined #cdk
16:09 s9asad left #cdk
16:28 Gpox left #cdk
18:37 egonw left #cdk
18:44 egonw joined #cdk
19:24 CIA-98 cdk: Egon Willighagen master * r6209a48 / pom.xml : Added the maven build file (closes #3042475) - http://bit.ly/bCS55d
20:22 egonw left #cdk
20:29 egonw joined #cdk
20:37 jbrefort left #cdk
22:04 bag_ joined #cdk
22:20 egonw left #cdk
22:23 bag_ left #cdk

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