Camelia, the Perl 6 bug

IRC log for #cdk, 2011-09-14

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

All times shown according to UTC.

Time Nick Message
03:57 egonw joined #cdk
05:07 _ilbot joined #cdk
05:07 Topic for #cdk is now The Chemistry Development Kit (http://cdk.sf.net) | JUnit CDK 1.2.7: 15690, 76 fails + 15 errors (0.58%) | Logs at http://moritz.faui2k3.org/​irclog/out.pl?channel=cdk | CDK News Feed: http://friendfeed.com/rooms​/chemistry-development-kit
05:13 sneumann joined #cdk
05:20 egonw joined #cdk
05:46 egonw_ joined #cdk
05:47 egonw__ joined #cdk
05:57 egonw joined #cdk
06:01 egonw_ joined #cdk
06:21 sneumann joined #cdk
06:40 egonw_ joined #cdk
07:04 sneumann joined #cdk
07:04 jonalv joined #cdk
07:11 Gpox joined #cdk
07:27 bag joined #cdk
07:48 bag joined #cdk
07:57 bag joined #cdk
09:05 jbrefort joined #cdk
11:03 jbrefort joined #cdk
11:18 maclean joined #cdk
11:18 maclean 'noon
12:10 egonw hi maclean
12:10 maclean hi egon
12:11 s_wolf joined #cdk
12:11 * maclean is making SMSD patches
12:12 egonw one sec...
12:12 egonw as in: 'rebasing' ?
12:12 maclean as in, splitting into topics.
12:12 egonw did you see what I suggested?
12:12 egonw and what Asad agreed to?
12:12 maclean Matcher refactoring, VF2 refactoring, etc
12:12 egonw (I think he did, that is...)
12:13 maclean I saw the discussion yes.
12:13 egonw how do I link that to what you just said?
12:13 maclean ER.
12:13 maclean I mean "Er..."
12:13 egonw was just wondering who got a hearth attack :)
12:14 maclean Heart attack?
12:14 egonw what do you need a ER for then?
12:15 maclean Oh, hoho. Sorry I have a cold, so am a little slow :)
12:15 maclean But just to check : splitting into separate patches is preferable?
12:15 maclean As in, preferable to you submitting bug reports and then us (well, me) replying?
12:16 egonw and others I hope...
12:16 egonw but umm...
12:16 egonw my point is how does applying the current patch fit in, as I suggested?
12:17 maclean Not at all, I'm afraid. I don't see any gain in working with a 'failed' patch.
12:17 egonw how so failed?
12:17 maclean I'm considering it like a biochemical reaction : that patch is like a transition state, I'm interested in the overall changes.
12:18 maclean From current master to what is in asad/SMSD (on github).
12:18 egonw mmm... well, before just now I was not aware of this change of plans, and still waiting for an ack on the SMSD patch rebase...
12:18 maclean Failed = 1Mb, with STDOUTS, changes to non-relevant classes.
12:19 egonw why not just apply this dump, and then start from that?
12:19 egonw or does it makes things worse?
12:19 maclean Well it's actually more work. For everyone - isn't it?
12:19 maclean Because there are more changes since then.
12:19 egonw in case, I like your comment in the patch tracker, saying you reject the patch as is
12:19 maclean Ok.
12:20 maclean Oh, btw : I'm assuming that you agreed about rejecting the nitrogen patch. I could have put it up for more general review, but it seemed conclusive.
12:20 maclean (I'm not sure if there is proper handling of azides, though, come to think of it).
12:21 egonw azide should be supported
12:21 egonw I'm pretty sure I added a test case for that a long time ago
12:23 maclean Oh, ok. Seems like there are resonance structures, though N=N=N to N#N-N{-} or something.
12:24 maclean Anyway, I guess atom types get fixed when there are actual cases where it doesn't work : I'm not going to chase up on potential new ones :)
12:26 egonw I guess we're good for the next year or so :)
12:26 maclean Heh, hope so.
12:26 egonw seriously...
12:26 egonw I'm gonna write some unit tests, to test for consistency
12:26 egonw and for missing detail, like hybridization
12:26 egonw there will be bug reports
12:26 egonw (and not just for the new ones)
12:27 egonw but I think in master I'll have to fix something else too:
12:27 egonw split hybridization from geometry
12:27 maclean That sounds reasonable.
12:28 maclean It was kind of confusing me, the difference between s2d4 (or whatever) and octahedral (or whatever).
12:28 egonw yeah, bad design decision by me
12:28 maclean Are there many hybridization states? Or are they only semi-bounded?
12:28 maclean As in, can there be new ones?
12:29 egonw bounded
12:29 egonw not too many
12:29 egonw yeah, there can be new ones
12:29 egonw hybridization is just a model
12:29 egonw in reality, sp3 orbitals cannot be isolated
12:30 egonw the atomic orbital hybridize into molecular orbitals
12:30 egonw that said, they have a tendency to match that model relatively well
12:30 egonw and simplifies talking about it
12:30 maclean Right, makes sense.
12:31 maclean When you say "relatively well", do you mean there can be expert disagreement about it - eg in strange molecules?
12:32 maclean (I saw this example the other day of a copper-oxygen bond in a crytand)
12:32 egonw expert disagreement not so much
12:32 egonw more that the model does not always fit reality well
12:32 maclean (http://www.rsc.org/Publishing/ChemScience/V​olume/2006/06/bonding_in_small_spaces.asp)
12:32 zarah maclean's link is also http://tinyurl.com/6d7h8h7
12:33 maclean But then you could have more than one match within the model to reality.
12:33 egonw yeah, nice example
12:33 egonw here the H3O+ model doesn't fit reality anymore
12:33 maclean Hmm, yes.
12:33 egonw leading to seemingly contradicting things
12:36 maclean Oh dear : 1.8 is out...
12:37 egonw of what?
12:38 maclean minecraft :)
12:45 maclean anyway, should go for a late lunch. later!
12:55 s9asad joined #cdk
14:46 egonw joined #cdk
14:56 egonw_ joined #cdk
15:10 egonw__ joined #cdk
15:18 egonw joined #cdk
15:21 egonw_ joined #cdk
15:21 jbrefort joined #cdk
15:26 maclean joined #cdk
15:51 jonalv joined #cdk
17:17 sneumann joined #cdk
17:27 egonw joined #cdk
20:03 zarah joined #cdk
20:03 CIA-38 joined #cdk
20:05 jonalv joined #cdk
20:18 alchmiste joined #cdk

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