Camelia, the Perl 6 bug

IRC log for #cdk, 2011-02-08

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

All times shown according to UTC.

Time Nick Message
04:57 egonw joined #cdk
05:26 egonw left #cdk
06:12 egonw_ joined #cdk
06:30 egonw__ joined #cdk
06:34 egonw_ left #cdk
06:42 egonw joined #cdk
06:42 egonw__ left #cdk
06:45 egonw_ joined #cdk
06:47 bag__ joined #cdk
06:48 egonw left #cdk
06:58 egonw__ joined #cdk
06:59 egonw_ left #cdk
07:05 egonw joined #cdk
07:07 jbrefort joined #cdk
07:08 egonw__ left #cdk
07:22 egonw left #cdk
07:28 bag__ left #cdk
07:29 egonw joined #cdk
07:53 Gpox joined #cdk
08:36 sneumann_ joined #cdk
09:33 sneumann_ left #cdk
10:11 maclean joined #cdk
10:11 maclean morning
10:14 egonw hi maclean!
10:14 maclean hi egonw
10:14 egonw saw your email
10:14 egonw will try to reply asap, but not before tonite, I think
10:16 maclean np, whenever.
10:27 jbrefort left #cdk
11:14 maclean Doh. The chlorines were due to "CL", rather than "Cl"...
11:14 egonw oh, yeah, I ahve seen that before...
11:14 egonw (in different contexes)
11:28 maclean Heh heh - CB5 : a metallecarborane (Co atom connected to two C/B icosohedral clusters). Y U NO TYPE? :)
11:28 egonw :)
11:28 egonw that reminds me...
11:29 egonw I still need a good web app where people can report atom types...
11:30 maclean Would they need any 'evidence'? Like paper refs, NMR data, etc..
11:30 egonw I would probably like to see such fields...
11:31 egonw but it's not required for the CDK right now...
11:35 maclean Well, there's only 1,601 nulls left and many are metal clusters. Oh, and some tertiary nitrogens.
11:37 egonw charge, or uncharged?
11:37 egonw oh, I now also spotted your blog post :)
11:39 maclean :)
11:39 maclean Charge doesn't seem to matter.
11:42 maclean nulls @ https://github.com/gilleain/h​etdict/blob/master/nulls.txt
11:42 maclean (largeish file - 120kb)
11:52 egonw maclean: can you add the PDB id for those? perhaps a HTML file with links to the DB?
11:52 egonw well, if you like to do it fancy :)
11:52 egonw why not a popup with Jmol with some script to zoom in on the mistyped atom?
11:52 maclean Hehe, let's not get too fancy now...
11:53 maclean Have been using PDBSum, PDBeChem, and Protopedia to look at these.
11:54 egonw maclean: http://chem-bla-ics.blogspot.com/2010/07/s​cripts-logs-as-htmlrdfa-mix-free-text.html
11:54 maclean Protopedia has an impressive thing where the hetgroups are listed in the sidebar, and clicking on them highlights in the Jmol window.
11:54 zarah egonw's link is also http://tinyurl.com/34eh2vr
11:54 egonw if you output that log like that, I can run SPARQL on it to create summaries
11:54 egonw like the most failing element
11:56 maclean hmmm. could do.
12:06 maclean anyway, now : lunch.
12:07 maclean left #cdk
12:28 jbrefort joined #cdk
12:33 Gpox left #cdk
12:35 Gpox joined #cdk
12:52 egonw left #cdk
13:17 egonw joined #cdk
14:29 maclean joined #cdk
14:42 egonw_ joined #cdk
14:44 egonw left #cdk
14:44 egonw_ is now known as egonw
15:00 egonw left #cdk
15:07 Gpox left #cdk
15:15 egonw joined #cdk
15:17 maclean right, egonw, you asked for it : https://gist.github.com/816566
15:17 maclean (test output - best format I can do at the moment...)
15:18 egonw almost perfect
15:19 egonw you need to change the resource URI for either the compound or the atom
15:19 egonw and I need to catch my public transport
15:19 maclean ok
15:19 egonw will be back online in about 30 minutes
15:19 egonw maclean: line 8 and 11
15:20 maclean I saw, it's annoying. Actually all rdf-a is annoying.
15:20 egonw this is not RDFa, btw
15:20 egonw this is RDF/XML
15:21 egonw bbl
15:21 maclean ok
15:21 egonw (but well done!)
15:26 egonw left #cdk
15:41 egonw joined #cdk
15:47 egonw maclean: you can use a XSLT stylesheet to have your browser convert it to HTML
15:47 egonw check the source of this page:
15:48 egonw http://rdf.openmolecules.net/
15:50 maclean I suppose I could, yes, But I don't think that the structure of the rdf/whatever is correct.
15:51 maclean the hierarchy is meant to be Doc : = Hetgroup*. Hetgroup := Atom*. Atom := {AtomId, AtomSymbol, Height2Sig}.
15:52 maclean In pseudo-BNF.
15:52 egonw_ joined #cdk
15:52 egonw left #cdk
15:53 maclean In any case, there seems to be almost no benefit to sprinkling URIs on some data, and wrapping it in mysterious tags.
15:58 egonw__ joined #cdk
15:59 egonw_ left #cdk
16:18 egonw joined #cdk
16:22 egonw__ left #cdk
16:25 maclean left #cdk
16:30 maclean joined #cdk
16:34 egonw left #cdk
16:45 egonw joined #cdk
16:48 maclean egonw : it seems like you use PHP to make your RDF? (I've been using jena)
16:49 egonw correct
16:49 egonw php makes it easier to mess up :)
16:51 egonw_ joined #cdk
16:52 maclean Well, jena seems to make it impossible to do what you want.
16:53 egonw_ hahaha
16:54 egonw left #cdk
16:54 egonw_ bbl
16:56 maclean ok
16:59 egonw_ left #cdk
17:00 jbrefort left #cdk
17:36 Gpox joined #cdk
18:10 _ilbot joined #cdk
18:10 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
18:30 cvala23 joined #cdk
18:31 cvala23 hey can I get some help on a few questions?
18:31 cvala23 hello?
18:32 cvala23 left #cdk
18:43 egonw_ joined #cdk
18:48 maclean Aaarrrgghhgh! Parsing CIF files is melting my head! eg : "HOH'" "'HOH" (that's HOH-prime and prime-HOH...)
18:48 maclean g2g
18:48 maclean left #cdk
19:02 Gpox left #cdk
19:41 bag joined #cdk
20:37 jbrefort joined #cdk
21:19 bag left #cdk
21:48 jbrefort left #cdk
22:02 bag joined #cdk
22:32 egonw_ left #cdk
23:48 bag left #cdk

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