Camelia, the Perl 6 bug

IRC log for #bioclipse, 2009-03-05

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

All times shown according to UTC.

Time Nick Message
06:12 masak joined #bioclipse
07:26 rojasm joined #bioclipse
07:31 Gpox joined #bioclipse
07:32 egonw joined #bioclipse
07:35 rojasm joined #bioclipse
07:50 CIA-55 bioclipse: goglepox * r9204 /bioclipse2/trunk/plugins/net.bioclipse.cd​k.ui.sdfeditor/src/net/bioclipse/cdk/ui/sd​feditor/editor/MoleculeTableViewer.java: Check if table is disposed before trying to refresh
07:52 CIA-55 bioclipse: goglepox * r9205 /bioclipse2/trunk/cdk-externals/trun​k/org.openscience.cdk.render/src/org​/openscience/cdk/renderer/selection/ (IChemObjectSelection.java ShapeSelection.java): Moved clear() form select(IAtomContainer) to select(IChemModel). Removed select(IAtomContainer) from interface and made it private. Fixes bug #633
07:52 zarah bug #633 | http://tinyurl.com/cwol45
08:24 olass joined #bioclipse
08:26 egonw olass: I have not seen jonalv yet, so I guess that is a reasonable assumption
08:27 olass ok, ordered
08:27 olass I have an account so very easy for me
08:27 egonw thanx
08:27 olass 2-5 days to arrive
08:27 egonw no, not true
08:27 olass usually 2
08:27 egonw this book has yet to be printed
08:27 egonw they will not have it on stock
08:27 egonw the get the book from lulu.com
08:28 egonw a print-on-demand service
08:28 egonw Angel has on his website 10-15 days
08:28 olass they claim "Skickas från oss inom 2-5 vardagar"
08:28 egonw (unless your website actually stockpiled Jmol books... which I find hard yo believe, but would be cool)
08:29 egonw ok, so maybe they did :)
08:31 CIA-55 bioclipse: goglepox * r9206 /bioclipse2/trunk/plugins/net.bioclipse.cdk.​jchempaint/src/net/bioclipse/cdk/jchempaint/ (3 files in 3 dirs):
08:31 CIA-55 bioclipse: Updates JCP outline with chemmodel from hub
08:31 CIA-55 bioclipse: resolves bug #641 with some side effects
08:31 zarah bug #641 | http://tinyurl.com/asxn32
08:34 egonw Gpox: what are the side effects?
08:35 Gpox It is related to fragments all separate atomcontainers in the chemmodel shows up in the outline
08:35 zarah Gpox: You have new messages. Write '/msg zarah @messages' to read them.
08:35 egonw ic
08:35 egonw bioclipse is not booting for me right now... no clue why, but cannot look at things... so was wondering...
08:46 jonalv joined #bioclipse
08:47 jonalv o/
08:47 egonw hi jonalv
08:48 egonw jonalv: olass ordered the Jmol book
08:48 egonw ok, just saw your mail
08:48 jonalv oki
09:14 CIA-55 bioclipse: goglepox * r9207 /bioclipse2/trunk/ (2 files in 2 dirs): Rightclick delete clears the selection.
09:14 CIA-55 bioclipse: goglepox * r9208 /bioclipse2/trunk/cdk-externals/trunk/or​g.openscience.cdk.control/src/org/opensc​ience/cdk/controller/ControllerHub.java: Update selection when structure changes
09:17 shk3 joined #bioclipse
09:23 CIA-55 bioclipse: goglepox * r9209 /bioclipse2/trunk/plugins/net.bio​clipse.cdk.jchempaint/plugin.xml: Changed 'Add Bond' toolbar button from push to radio fixes bug #629
09:23 zarah bug #629 | http://tinyurl.com/bn7amp
09:28 masak http://sqlanywhere.blogspot.com/2​008/03/unpublished-mysql-faq.html
09:28 zarah masak's link is also http://tinyurl.com/6nf88k
09:33 egonw Gpox: should not more buttons reflect whether they are pushed or not?
09:34 Gpox they are, just not synced with the hub
09:34 Gpox @pz 514
09:34 zarah http://pele.farmbio.uu.se/cgi-b​in/bugzilla/show_bug.cgi?id=514
09:35 egonw ack
09:39 olass egonw: you at bmc?
09:39 egonw no
09:40 egonw too distracted to do bioclipse hacking there
09:40 olass ok
09:40 olass where is maclean==gillean?
09:40 olass has he stopped working on JCP?
09:40 olass vaccation?
09:40 egonw I guess grant writing too
09:41 jonalv I guess shk3 might know...
09:41 olass right... so he'll be back
09:41 olass shk3?
09:41 shk3 he's focusing on his work
09:42 olass gillean has several bugs assigned, I'm sure he would not keep it that way if he didn't inted to work on them soon
09:42 shk3 I can ask him
09:42 olass shk3: work other than jcp you mean?
09:42 olass I assume JCP is not his hobby
09:42 egonw but not his core business either
09:42 shk3 well, sort of
09:43 egonw which is CASE
09:43 olass and I'm sure JCP aligns well with your and Christoph's work
09:43 egonw it's a side job for all of us, I think
09:43 olass well, CASE requires a structure editor, right?
09:43 egonw olass: no, not really
09:44 egonw there are many structure editors, OS and proprietary
09:44 egonw no one gets payed for JCP development
09:44 egonw no grants involved in this
09:44 egonw just a necessity
09:44 egonw you can't get grants for JCP development
09:44 olass publication > reputation + users > grants
09:44 olass I know
09:44 egonw no, not for JCP
09:45 olass Peter Ertl is quite famous for JME...
09:45 egonw well, 1) that was 30 years ago
09:45 egonw 2) I think he has many other skills for which he is actually being payed instead
09:45 olass :)
09:45 olass in 30 years from now...
09:46 olass I know
09:46 egonw Then the Blue Obelisk will be the first 'organization' to get the Nobel prize
09:46 olass I can live with that
09:46 jonalv okey let's stop walking around the actual topic. JCP is the real blocker for a Bioclipse 2 release. If we can't work as much as we need on it then maybe we just have to accept that and move the release data again?
09:47 olass Gpox is working on it
09:47 olass but if Gilleain cannot work on JCP at all...
09:47 olass I did not get that impression from shk3
09:47 * jonalv dont know enough about it to actually be helpfull :(
09:48 olass jonalv: maybe focus on your litt project then? :-)
09:48 jonalv olass: I am
09:48 * olass wants jonalv back in the coding loop
09:48 shk3 gilleain was dragged away from case too much
09:48 shk3 so he now tries to focus on that
09:48 shk3 I am afraid if he will start working on jcp just "a little bit" Christoph
09:49 shk3 will fear he will stop doing case altogether
09:49 olass shk3: well, I need a clear answer
09:49 olass will he do nothing on JCP?
09:49 olass else I cannot plan
09:50 olass if he can do some, e.g. one day per week, that would be great
09:50 olass but this is his and Christophs decision
09:50 jonalv olass: sounds to me like you should just ask Christoph straight out. ISn't that the german way btw?
09:50 shk3 sorry, can't say that
09:50 shk3 yes, might be a good idea
09:50 olass ok, will do
09:52 shk3 what essential stuff is missing in jcp right now?
09:52 shk3 undo/redo (will do that)
09:53 shk3 merging when moving
09:53 shk3 custom placement of atoms (arvids former solution + a minimum length should do it, if I understood correctly)
09:53 shk3 anything else?
09:57 egonw olass: current JCP is beyond the original 2.0 basic editing requirements...
09:57 egonw olass: this has distracted from bug fixing, and making clean APIs
09:57 egonw olass: please decide what editing features you want to release in 2.0
09:57 olass did you not see my email?
09:57 egonw undo/redo sneaked in, for example...
09:57 olass yes, that is important
09:58 olass egonw: my email was just to clear this up
09:58 egonw ok, good
09:58 egonw so, that defines the limits...
09:58 olass and shk3 had another release blocker
09:58 olass which I totally agreed on
09:58 egonw then I do not think the RC next friday is feasible
09:58 olass reactions is IMO 2.1
09:59 olass egonw: ok
09:59 olass what do you expect for RC1?
09:59 olass another week? 2?
09:59 egonw well, the introduction of reactions in jchempaint-primary causes several of the bugs in the current code
10:00 egonw because the design was not clean enough for that yet, as Stefan indicated
10:00 olass can we remove it?
10:00 olass and put it in a branch?
10:00 olass maybe too hard...
10:00 egonw it was a really bad trade off
10:00 egonw yeah, it is too hard
10:00 olass ok
10:00 egonw so much good stuff entered at the same time
10:01 olass right
10:01 olass interpreting shk3's answer from the mail, undo/redo was only 2-3 days
10:01 masak olass: no need to apologise for pushing deadlines, I think.
10:01 egonw the major problem I see is that we now have to make ugly work arounds for fitting the CDK controller IChemModel model which supports even reactions with the IAtomContainer of CDKMolecule
10:01 masak olass: software development is hard.
10:01 olass ;)
10:02 olass egonw: I understand
10:02 olass masak: true
10:02 egonw egonw: I'm sorry, but could not keep the jcp-prim development under control...
10:02 masak olass: http://en.wikipedia.org/wiki/Hofstadter%27s_law
10:02 zarah masak's link is also http://tinyurl.com/3eauk4
10:02 egonw too many people pulling for features
10:02 egonw so that there was too litte time to think about making a good API
10:03 shk3 egonw talking to himself ;-)
10:03 shk3 I discussed that with gilliean yesterday
10:03 egonw shk3: and in general... I think you pretty well know where this comes from :)
10:03 shk3 in a way, we would need a hub implementation which only a SetOfMolecules
10:03 egonw shk3: even worse...
10:04 egonw we even need one which just has Molecule
10:04 shk3 this would also make class sizes smaller
10:04 egonw that's the most common use case anyway: draw a single structure
10:04 shk3 yes, might be
10:04 egonw and exactly the one we have in BC2
10:04 shk3 so modularity of the hub is an issue
10:04 egonw yes
10:04 shk3 the issue in that context
10:05 egonw OK, back to grant writing now...
10:06 egonw if I manage to get the two grants done this week, I have time for BC2/CDK/JCP stuff next week again
10:06 egonw olass: PS I'll invite you for VRNT this afternoon...
10:07 shk3 good luck
10:08 shk3 one question:
10:08 shk3 jcp in bc has menu items like "update implicit hydrogens"
10:08 egonw shk3: please write a simple patch for the one API change in jcp-prim for IReactionSet, and submit that for review for 1.2
10:08 shk3 is there handling of implicit hydrogens in jcp?
10:08 egonw the manipulator changes are less of a problem
10:08 egonw but the IReactionSet patch needs to be singled out...
10:09 shk3 will do so
10:16 egonw shk3: next week I will work on modularizing the controller
10:17 egonw as it really seems to be a limiting factor in development now...
10:17 egonw that should also help with making the 38kB applet
10:22 CIA-55 bioclipse: goglepox * r9210 /bioclipse2/trunk/cdk-externals/trunk/org.op​enscience.cdk.control/src/org/openscience/cd​k/controller/ChangeFormalChargeModule.java: Use action on IChemModelRelay to change formal charge fixes bug #545
10:22 zarah bug #545 | http://tinyurl.com/c9y3av
10:30 egonw rather interesting read for over lunch: http://boscoh.com/protein/bioinformatics-in-t​he-near-concurrent-future?commented=1#c001037
10:30 zarah egonw's link is also http://tinyurl.com/be3uua
10:35 ilogger2_ joined #bioclipse
10:36 Gpoks joined #bioclipse
10:36 egonw olass: is martin around?
10:36 olass yes
10:37 olass in office
10:42 egonw Gpox: I'm routing the reports stefan is filing to you, as at least some of them are close to things you are working on...
10:42 Gpox ok
10:42 egonw please decide if it's 2.0 or 2.1 material
10:44 egonw olass: I am not going to make the öl tonite... :( not feeling energetic enough for that, and a lot I need to finish
10:44 egonw (this is rare for me, to skip such an important social team building event, because of work to be done)
10:45 olass egonw: ok
10:57 CIA-55 bioclipse: goglepox * r9211 /bioclipse2/trunk/cdk-externals/tr​unk/org.openscience.cdk.control/sr​c/org/openscience/cdk/controller/ (ControllerHub.java CycleSymbolModule.java): Moved configuring of atom to ControlerHub.setSymbol, and made CycleSymolModule call it fixes bug #544
10:57 zarah bug #544 | http://tinyurl.com/bwn54n
11:26 masak jonalv: http://en.wikipedia.org/wiki/Doomsday_argument
11:26 zarah masak's link is also http://tinyurl.com/2mgqm3
11:28 shk3 egonw: it looks like the core classes (like ReactionSet) in 1.2 are untested.
11:29 shk3 or is there something special I missed?
11:29 egonw let me check...
11:29 egonw it's a bit special set up...
11:29 egonw to test things properly...
11:30 egonw booting Eclipse for CDK development...
11:30 egonw hang on
11:31 egonw do you have ReactionSetTest.java open?
11:32 egonw well, if you do... notice the 'extends' into AbstractReactionSetTest
11:32 egonw that class is where most of the testing happens
11:33 egonw shk3: check the email archives on a lenghty discussion on these things
11:33 * egonw is away for lunch
11:37 shk3 ah, thanks
11:49 shk3 olass: another major point before the release is documentation.
11:49 shk3 must plugins are still undocumented.
11:49 shk3 i. e. have not help pages.
11:49 olass I know
12:03 CIA-55 bioclipse: goglepox * r9212 /bioclipse2/trunk/cdk-externals/trunk/org.o​penscience.cdk.control/src/org/openscience/​cdk/controller/AlterBondStereoModule.java: Changed stereo module to call ChemModelRelay for changing stereo order fixes bug #547
12:03 zarah bug #547 | http://tinyurl.com/bxowbk
12:13 CIA-55 bioclipse: ospjuth * r9213 /bioclipse2/trunk/plugins/se.​uu.farmbio.hivpredrt.model/: Initial import of model for HIV RT predictions. Client will come later.
12:15 CIA-55 bioclipse: ospjuth * r9214 /bioclipse2/trunk/plugins/se​.uu.farmbio.hivpredrt.model/ (46 files in 11 dirs): Model which is online now on ws1.bmc.uu.se.
12:36 edrin joined #bioclipse
12:38 edrin hollq
12:46 masak edrin: hollq!
13:12 jonalv egonw, olass: we don't have a save sdf file method do we?
13:12 egonw we do
13:13 egonw saveMolecules()
13:13 jonalv egonw: oh
13:13 jonalv egonw: that one is not published is it?
13:13 olass egonw: about the smartsMatchingView and where to place it...
13:14 olass is it OK to place it in JCP?
13:14 olass it requires JCP...
13:14 egonw yes, guess so
13:14 olass ok, doing that for now
13:14 olass feel free to move it at a later point if you like
13:14 egonw jonalv: no, it does not seem published
13:15 egonw jonalv: I also see it has no progressmonitor
13:15 egonw both should be fixed
13:15 jonalv egonw: I think it's the IChemFormat taht makes things troublesome
13:15 egonw please file bug report
13:15 egonw jonalv: why?
13:15 egonw olass: yes...
13:15 jonalv egonw: how do you get one of those?
13:15 jonalv egonw: from the js-console
13:15 egonw cdk.getFormat()
13:15 olass I think they should be hidden from the user
13:15 egonw cdk.getFormat(String)
13:15 olass too complicated
13:16 olass but I guess power users need them...
13:16 jonalv couldn't the method just take a String?
13:16 olass my point
13:16 egonw no
13:16 jonalv it was too complicated even for me... :)
13:16 egonw but there is:
13:16 olass and for me too
13:16 egonw cdk.guessFormatFromExtension()
13:16 jonalv or cdk.saveSDFFile
13:17 jonalv but maybe we want to support _very_ many formats one day...
13:17 egonw jonalv, olass: yes, we really need a direct brain/bioclipse link...
13:17 egonw so that the console shows you which methods match whatever task you think about
13:18 olass egonw: we're trying to be productive and come up with solutions
13:18 olass do not take offense for this
13:18 jonalv What we really want is popup help in the console just like in the Eclipse java editpor...
13:18 olass that would be great
13:18 olass and very helpful
13:18 jonalv masak: you can fix that right? :)
13:19 * masak backlogs
13:19 olass just backlog 3 lines!
13:19 masak aye. :)
13:19 masak of course it can be done.
13:19 masak but that doesn't mean I know how.
13:20 masak I need to learn more about SWT and text widget.
13:20 masak s/widget/widgets/
13:20 masak in the meantime, please file things as feature tickets.
13:21 olass egonw,jonalv: just for the above discussion: we said we should sit down and brainstorm and try to make the scripting API more stringent after 2.0
13:21 olass right now we should be happy if things work
13:21 olass even if tha API can be approved
13:21 olass but we can still discuss things :-)
13:21 masak right now things don't always work... :/
13:21 olass right, then it must be fixed
13:21 jonalv olass: yea I guess so, right now it doesn't though. Which was what started this discussion...
13:22 edrin olass: you wrote, the sample script for blast did not work...
13:22 edrin what is the result?
13:22 olass what did you get?
13:22 egonw jonalv: what doesn't work?
13:22 jonalv egonw: I can not create an SDF file from the console
13:22 olass edrin: worked with you?
13:22 jonalv egonw: or rather from js
13:22 edrin could not test yet
13:22 olass ok
13:22 egonw jonalv: check this gist:
13:23 edrin olass: just wanted to mention, it takes a few minutes to complete :)
13:23 egonw http://gist.github.com/72430
13:23 egonw jonalv: that script worked this weekend
13:23 edrin olass: what does the script output?
13:23 olass edrin: sorry, no time now for this
13:23 edrin ok
13:24 olass might have worked, just didnät know it takes a few mins
13:24 olass got no result
13:24 olass so might have run
13:24 jonalv egonw: ah, cdk.createSDFile. I missed that method :)
13:24 jonalv I was looking for save...
13:24 jonalv egonw: thanks\
13:24 egonw jonalv: indeed... I did say saveMolecules()
13:24 egonw my mistake
13:24 jonalv np
13:24 egonw so, you mean saveMolecules() is broken?
13:25 egonw if so, please file a bug report
13:25 edrin olass: ok... so in case you have a bc instance open, you could run it again and look in the window again 5 min later ;)
13:25 egonw and please run CDKManagerPluginTest
13:25 jonalv egonw: I have, it is not published
13:25 edrin olass: i did this for demoing delayed services ;)
13:28 egonw shk3: I think your last bug report was fixed already... when did you last update?
13:29 egonw Gpox made a change for the Add button this morning
13:29 shk3 might have missed that
13:29 shk3 I will update
13:31 edrin olass: the HIV servlet thing, does it use xws4j ?
13:31 edrin on the client side?
13:31 olass yes
13:31 edrin cool :)
13:31 edrin really cool
13:32 edrin olass: I like this
14:07 egonw jonalv++
14:32 jonalv mm karma, why?
14:32 egonw for writing a bioclipse script that at least one user liked
14:32 jonalv wohow, one user liked it?
14:33 egonw and even suggested to put it on Gist
14:33 egonw like me
14:33 * jonalv reads emails now :)
14:33 egonw :)
14:38 jonalv egonw: I never seeem to learn the procedure for doing those gist and delicious linkings---
14:38 jonalv egonw: are there some instructions somehwere_
14:38 egonw mmm... no, but will write them up now...
14:39 * jonalv just realised that egonw might be busy and shouldn't be distracted...
14:39 egonw that's fine...
14:40 egonw it's time for coffee anyway...
15:07 jonalv what date was the bheta 3 reelase?
15:08 egonw 28/2
15:11 jonalv egonw: seems you did some stuff the day after which I am using in my script...
15:12 egonw sorry... can't decode that
15:13 egonw what do you mean?
15:13 egonw http://wiki.bioclipse.net/index.php?title=Gist
15:13 jonalv egonw: you added a cdk method the day after the release which I am using in my script. Peter can't run it since he is on beta3...
15:13 zarah egonw's link is also http://tinyurl.com/djz6rf
15:13 egonw ah, ic... yes, correct
15:14 egonw yes, we should make sure to put an update of the CDK feature online tomorrow
15:14 jonalv yup
15:14 egonw which would also have all the fixes by Gpox
15:15 jonalv good
15:37 jonalv egonw: okey now I have tagged the cite as bioclipse and gist
15:37 egonw jonalv++
15:37 jonalv egonw: I am not sure that did the trick though...
15:37 egonw I don't see it show up
15:37 egonw http://delicious.com/tag/bioclipse+gist
15:38 zarah egonw's link is also http://tinyurl.com/5gqjet
15:39 jonalv egonw: so what more must I do_
15:39 jonalv egonw: http://delicious.com/jonalv
15:40 egonw ah, it was just timing it seems...
15:40 egonw it showed on your private page
15:41 egonw and I now see it on the other apge
15:41 jonalv okey
16:02 CIA-55 bioclipse: shk3 * r9215 /bioclipse2/trunk/cdk-externals/tr​unk/org.openscience.cdk.control/sr​c/org/openscience/cdk/controller/ (4 files in 2 dirs): Updated some changes from jcp-primary (not all)
16:05 CIA-55 bioclipse: shk3 * r9216 /bioclipse2/trunk/plugins/net.bioclipse.cdk.​jchempaint/src/net/bioclipse/cdk/jchempaint/ (13 files in 3 dirs): Undo/Redo works for all toolbar/menu actions. The manager stuff is a bit tricky, will email because of that.
16:19 edrin left #bioclipse
17:05 edrin joined #bioclipse
17:05 edrin hi again
17:11 * egonw is having dinner first now, will continue grant/paper writing afterwards
17:25 masak http://xprogramming.com/blog​/2009/01/30/context-my-foot/
17:25 zarah masak's link is also http://tinyurl.com/cf2txa
17:25 masak a really good read.
17:27 masak oh, and there's a reply to that one as well: http://www.satisfice.com/blog/archives/174
17:27 zarah masak's link is also http://tinyurl.com/atxm8t
17:29 CIA-55 bioclipse: shk3 * r9217 /bioclipse2/trunk/plugins/net​.bioclipse.cdk.ui.sdfeditor/ (4 files in 2 dirs): filename set by default (bug #656)
17:30 zarah bug #656 | http://tinyurl.com/d5xbgl
17:37 CIA-55 bioclipse: shk3 * r9218 /bioclipse2/trunk/plugins/net.bioclipse.chemoinfo​rmatics/src/net/bioclipse/chemoinformatics/conten​tlabelproviders/MoleculeFileContentProvider.java: directories with . no longer in MoleculeFileContentProvider
17:56 CIA-55 bioclipse: shk3 * r9219 /bioclipse2/trunk/plugins/net.biocli​pse.cdk.ui.sdfeditor/src/net/bioclip​se/cdk/ui/wizards/NewSdfWizard.java: avoids problems with selection
17:56 CIA-55 bioclipse: shk3 * r9220 /bioclipse2/trunk/plugins/net.bioclipse.chemoinfo​rmatics/src/net/bioclipse/chemoinformatics/conten​tlabelproviders/MoleculeFileContentProvider.java: special request, nicer code
18:16 ilogger2 joined #bioclipse
18:31 CIA-55 bioclipse: shk3 * r9221 /bioclipse2/trunk/plugins/ (5 files in 3 dirs): default names set
18:52 CIA-55 bioclipse: shk3 * r9222 /bioclipse2/trunk/plugins/net.bioclipse.chemoinfo​rmatics/src/net/bioclipse/chemoinformatics/conten​tlabelproviders/MoleculeFileContentProvider.java: handling of 5d content type
19:11 CIA-55 bioclipse: shk3 * r9223 /bioclipse2/trunk/plugins/net.bioclipse.ui.tests/ (9 files in 3 dirs): had a go on tests for filetypes

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