Camelia, the Perl 6 bug

IRC log for #bioclipse, 2007-10-10

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

All times shown according to UTC.

Time Nick Message
06:58 EskilA joined #bioclipse
07:14 olas joined #bioclipse
07:43 masak joined #bioclipse
07:54 jonalv joined #bioclipse
08:16 edrin joined #bioclipse
08:16 edrin hi again
08:17 edrin jonalv, did you get my chat request?
08:31 edrin anyone here?
08:32 edrin olas:
08:32 edrin ?
08:38 edrin left #bioclipse
08:40 edrin joined #bioclipse
08:40 edrin hello again
08:40 viklund joined #bioclipse
08:41 edrin anyone available?
08:41 viklund In what sense?
08:42 viklund edrin?
08:43 edrin viklund: hi, sorry
08:44 viklund hello!
08:44 edrin viklund: are you lokated at a university?
08:44 viklund yes, I'm at EBC
08:44 viklund in Uppsala
08:46 edrin I need a newly released article. but unfortunately my uni has no access to the journal. Uppsala has I think. could you maybe download the article for me and send it to me via email?
08:46 viklund sure
08:46 edrin http://www.ncbi.nlm.nih.gov/sites/entrez?db=pu​bmed&cmd=Retrieve&dopt=AbstractPlus&am​p;list_uids=17921348&itool=pubmed_DocSum
08:47 edrin there is a link on this page
08:47 edrin right upper corner, but maybe you know PubMed anyway
08:47 viklund yes I do ;)
08:47 edrin hehe
08:48 viklund so, what's your email-address?
08:49 edrin i sent it in a private chat to you
08:49 viklund sent
08:50 edrin thank you
08:50 viklund np.
08:51 edrin it's a pitty that my university does not access this journal. I have this problem
08:51 edrin several times
08:51 viklund I wonder for how long there will be non open-access journals
08:52 viklund They seem to be disappearing... or, at least changing to an open access model
08:52 edrin slow migration
08:52 viklund yes, but a migration nonetheless
08:52 edrin i hope so
09:23 olas hi edrin, hi viklund
09:23 viklund hi
09:24 olas edrin: is everything fine with you?
09:43 viklund masak?
09:44 viklund do you know a quick way of generating every possible pair of two lists in perl?
09:45 viklund preferably without two nested for loops... I think I know how to do it with haskell and scanl...
09:50 edrin olas: a lots of stuff todo
10:10 viklund ahh, finally...
11:47 masak viklund: there's no shortcut, no
11:48 viklund masak: I just wanted it a bit more beautiful than nested fors... which I managed
11:48 masak goodie
11:48 * masak is in a meeting
11:48 viklund btw. humpus has random starting positions now
12:04 masak viklund: woot
12:16 viklund joined #bioclipse
12:48 viklund masak: I just thought of a much nicer way of generating a list of random numbers that are unique
12:49 viklund So I couldn't help myself, I went and implemented it...
12:49 viklund It's more or less like this:
12:49 viklund take 5 $ nub $ randomRs (2::Int, gridLength) seed
12:49 viklund yep, randomRs generates an infinite list of random numbers!
12:50 viklund haskell++
12:59 bjarni joined #bioclipse
12:59 bjarni foo
13:00 viklund bar
13:01 bjarni t�nkte jag skulle joina s� jag ser vad ni h�lls med
13:06 olas joined #bioclipse
13:09 jonalv hi olas
13:10 jonalv should the user manager be: net.bioclipse.userManager? or whats the naming convention?
13:10 olas sounds reasonable
13:11 jonalv it should be camelCase then?
13:11 olas net.bioclipse.usermanager preferably
13:12 olas I think the convetion is lowercase
13:12 olas but only 95% sure
13:12 jonalv and not something like user_manager or user-manager?
13:13 olas no
13:13 olas net.bioclipse.usermanager IMO
13:13 olas IMHO I should say :-)
13:13 jonalv right so net.bioclipse.usermanager it is then
13:13 c-coder I suggest "uman"
13:14 olas uman is not bad, but a little non-informative
13:14 bjarni that was my c-doer joke of the day
13:14 bjarni c-coder*
13:14 olas cool
13:14 olas Will tell ChanServ to ban c-coder from now on :-)
13:15 bjarni :D
13:15 jonalv Im not sure I am looking forward to tomorrows... :)
13:15 olas why not?
13:15 * jonalv didn't think it was that funny...
13:15 bjarni :(
13:15 jonalv :)
13:17 jonalv but yet again the next c-coder joke could be something great!
13:17 CIA-31 bioclipse: biocoder * r3940 /trunk/playground/plugins/net.bioclipse.analysis/: Initial import.
13:20 CIA-31 bioclipse: biocoder * r3941 /trunk/playground/plugins/net.bioclipse.analysis/:
13:46 viklund jonalv: nitpick: that wasn't CamelCase it was humpBack...
13:53 jonalv I was waiting for someone to correct me. I don't klnow the difference so I made a guesss... :)
13:54 jonalv safely assured that if I guessed wrong someone would correct me :)
14:02 viklund ;)
14:47 olas how do you display an image in SWT? What control should be used?
14:48 olas PNG or preferably SVG
14:48 olas but without batik...
14:48 olas I assume png is preferred
14:48 olas jonalv: any idea?
14:49 olas EskilA: any idea?
14:50 olas found it: http://www.eclipse.org/articles/Artic​le-SWT-images/graphics-resources.html
14:51 jonalv olas, not really. It would be really cool if we could show SVG easily
14:55 olas have not found any other solution than using the Swing JSVGCanvas in batik
14:55 olas and embed that in SWT
14:56 olas sounds very much doable
14:56 olas but not an SWT widget
14:56 olas why do you want it?
14:56 jonalv olas?
14:57 jonalv btw have you asked on #Eclipse?
15:00 EskilA Label works fine for showing images
15:00 jonalv but svg?
15:02 EskilA aha, dunno about native support
15:04 olas true
15:04 olas read on newsgroup that no svg support ifor Image
15:06 olas eho is test?
15:06 olas eho=who
15:06 olas a test bot
15:06 olas hmm
15:06 olas wonder what it is testing...
15:12 jonalv I think it is one of the logging bots...
15:13 olas still testing?
15:13 jonalv no idea but obviously yea. Always in beta? :)
15:17 CIA-31 bioclipse: ospjuth * r3942 /trunk/playground/plugins/net.bioclipse.taverna/: Removed old taverna plugin
15:18 CIA-31 bioclipse: ospjuth * r3943 /trunk/playground/plugins/net.bioclipse.taverna/: Initial import of Tavverna 1.6.2 converted into an Eclipse plugin. needs Maveniser to build.
15:19 CIA-31 bioclipse: ospjuth * r3944 /trunk/playground/plugins/net.bioclipse.taverna/ (21 files in 12 dirs): Initial import of Tavverna 1.6.2 converted into an Eclipse plugin. needs Maveniser to build.
15:20 CIA-31 bioclipse: ospjuth * r3945 /trunk/playground/plugins/​net.bioclipse.taverna.ui/: Initial import of Taverna GUI. Not working yet.
15:20 CIA-31 bioclipse: ospjuth * r3946 /trunk/playground/plugins/​net.bioclipse.taverna.ui/ (17 files in 9 dirs): Initial import of Taverna GUI. Not working yet.
15:21 olas left #bioclipse
15:29 CIA-31 bioclipse: jonalv * r3947 /trunk/playground/plugins/ (4 files in 4 dirs): copied net.bioclipse.keyring to net.bioclipse.usermanager and net.bioclipse.keyring.tests to net.bioclipse.usermanager.tests
15:38 EskilA joined #bioclipse
15:47 CIA-31 bioclipse: jonalv * r3948 /trunk/playground/plugins/ (28 files in 11 dirs): Fixed naming changes after copying
15:48 CIA-31 bioclipse: jonalv * r3949 /trunk/playground/plugins/net.bioclipse.userm​anager.tests/tests/net/bioclipse/usermanager/ (KeyRingTest.java UserManagerTest.java): more Keyring -> UserManager renaming
15:52 CIA-31 bioclipse: jonalv * r3950 /trunk/playground/plugins/ (2 files in 2 dirs): switched dependency to User Manager instead of Keyring
16:00 masak bjarni: hi!
16:00 bjarni h�j
16:01 masak did you know that they sometimes say "hej" in Poland as well?
16:01 bjarni No.
16:02 EskilA cool, now I know one word in polish :p
16:03 masak but mainly they say "cześć"
16:03 EskilA how do you spell (sorry for botching this up) jak sie masj correctly?
16:04 masak "jak się masz?"
16:04 masak ...I think
16:04 EskilA ok, now I can have a 10 second conversation :)
16:05 masak :)
16:05 masak just say "hi, how are you?" and then lean back and listen
16:05 EskilA and nod a lot
16:05 masak yeah, that could work
16:05 jonalv :)
16:20 bjarni Cholera, kto rzucil jagody!? >:(
16:22 masak bjarni: jak rzucil jagody!
16:22 bjarni >:((
16:22 * bjarni -> movie
16:22 masak have fun.
16:58 CIA-31 bioclipse: jonalv * r3951 /trunk/playground/plugins/ (11 files in 5 dirs): Read through the tests and bettered some stuff and changed things that should work different in the Usermanager compared to how they worked in the Keyring.
17:08 masak shk3, if you arrive before I'm gone, I'll be back here in the channel within an hour
17:09 masak just going to have some supper
17:48 shk3_ joined #bioclipse
17:49 shk3_ hi masak
18:15 masak hi shk3_
18:15 shk3_ hi
18:15 masak so, let's get to work :)
18:16 shk3_ yeah
18:16 masak there are two things that I'd like to prepare for today
18:16 masak maybe not do them completely, but at least straighten out the details so that they can be done
18:17 masak 1. fix the 'properties' columns of the table
18:17 shk3_ what's the problem with that?
18:17 masak well,
18:18 masak I have a feeling it contains a lot of unused code right now
18:18 masak and doesn't work
18:18 masak it's like it's in-between models
18:18 shk3_ I have a look, but I need to get into the code again
18:18 masak and I'm not sure how to design it so that it works well
18:18 masak shk3_: sure
18:18 masak I'll give you the hints I think I have
18:19 shk3_ ok, one problem is we don't know which properties are there without looking at all the files.
18:19 masak if I understand correctly from the code and from our discussions
18:19 masak shk3_: that's a problem
18:19 masak if I understand correctly, there's supposed to be multiple columns
18:19 shk3_ (plus we don't know what the user wants to see, but we can't solve that. the user needs to resize columsn or so for that)
18:19 masak one for each property
18:19 shk3_ yes, that was what I had
18:19 masak did it ever work like that?
18:19 masak or was it a goal?
18:20 shk3_ think so
18:20 shk3_ but not sure
18:20 masak ok
18:20 masak because right now, columns are hard-coded to 2
18:20 masak "molecule" and "property"
18:21 masak and frankly, I feel less uneasy about that layout than about a dynamic multi-column layout
18:21 masak that's my main problem right now, I think
18:21 shk3_ yes, indeed
18:21 masak I ought to be able to fix the multi-column layout, but I'm not sure I want to
18:21 shk3_ the dynamic columns are tricky
18:21 masak and not very intuitive
18:22 shk3_ if I remember correctly, it was not possible to remove columsn
18:22 shk3_ so I only added them
18:22 masak that's a problem too, yes
18:22 shk3_ and if number of properties decreased, there were empty columns.
18:22 masak one could always delete the table and start over upon update
18:22 shk3_ what's the alternative?
18:22 masak but I'm still not sure it's the right way to go
18:22 masak would it be possible to cram all properties into one cell?
18:22 shk3_ yes and no
18:22 masak or would that just be too much?
18:23 shk3_ sure we can do it, but then columns can't be resized
18:23 masak why noe?
18:23 masak s/e/t/
18:23 shk3_ and sorted by (which was I goal I had in the long run)
18:23 masak right
18:23 shk3_ well, I you have all in one
18:23 shk3_ you can only resize that column, but not in item in there
18:24 shk3_ which would be separated by | or someting like this?
18:24 masak hm
18:24 masak or maybe one line, one prop
18:24 shk3_ couldn't we have multiple rows in the cell?
18:24 masak or just newlines
18:24 shk3_ yes, since the picure is higher than one line anyway
18:24 masak there's plenty room because there's a molecule next door
18:24 masak yes
18:24 shk3_ sounds good
18:24 masak great!
18:24 masak sorting is out, though
18:25 masak ...except maybe not
18:25 masak could still do sorting by property somehow
18:25 masak another thing that bothers me is that properties are not very well used in the example files
18:25 shk3_ yes, sorting could be done
18:25 shk3_ you just needed a UI element for it
18:25 masak which makes me wonder if people are really going to appreciate the effort made on supporting a lot of property functions
18:26 masak but who knows
18:26 masak maybe seeing the properties will make people care about them
18:26 shk3_ you could go through all properties and change by one if clicking on header or so
18:26 shk3_ well, the idea was qsar stuff
18:26 masak that's a possibility
18:26 shk3_ calculate xlogp or whatever for a directory and view them by the viewer
18:27 shk3_ I think the calculate is in bioclipse, but it just writes to file.
18:27 masak you could bring up a menu of what to sort on when clicking on the header
18:27 shk3_ (would be a nice example for the paper, btw)
18:27 masak what paper?
18:27 shk3_ the one ola is preparing
18:27 masak I'll tell him that :)
18:27 shk3_ are you involved?
18:28 masak not really
18:28 shk3_ just an idea
18:28 masak yes
18:28 masak but it sounds good
18:29 masak ok, 2. change the images in the table to widgets, and make the table itself a widget for possible inclusion into other modules
18:29 masak maybe I can actually handle that one on my own
18:29 masak but it seemed important to get (1) out of the way first
18:29 shk3_ what widget would you use instead of pictures?
18:29 masak there's now a SingleMolecule widget
18:30 masak egonw++ developed it
18:30 shk3_ which is using cdk for rendering?
18:30 masak think so
18:30 shk3_ well, if it is working shouldn't be hard to put it in
18:30 masak exactly
18:30 masak and that makes for shared code, which is nice
18:30 shk3_ I never built a widget, but putting the table in a widget shouldn't be hard either
18:30 masak it's mostly a way of distributing the load across classes
18:31 masak yes, so then the view would basically contain the widget
18:32 shk3_ ok, so you do that.
18:32 * masak will
18:32 shk3_ shall we have a look at the properties together?
18:32 masak sure, why not?
18:32 masak sounds like the closest thing on hand
18:33 shk3_ you do no longer need that lookup-all-properties before, do you?
18:33 shk3_ just put all properties in the cell with \r\n between
18:33 masak right
18:33 masak ...which possible simplifies other code as well
18:33 shk3_ there should be an order in there, but that could be alphabetically
18:34 masak ...which means we have to collect all props from a file first
18:34 shk3_ probably we should repeat property: before each line
18:34 shk3_ why?
18:34 masak because they might not come alphabetically in the file
18:35 masak and we can't sort them until the last one is read
18:35 shk3_ ah, from one molecule
18:35 shk3_ yes
18:36 shk3_ but that's not a problem, I think
18:36 masak no
18:36 shk3_ just use a sortedmap for it
18:36 shk3_ ok, just had a look at the code
18:36 shk3_ the assignEditors method is doing the properties
18:36 shk3_ that can go away
18:36 masak good
18:37 masak I was wondering about that
18:37 shk3_ so can everything building possiblekeys
18:37 shk3_ i think
18:37 masak yes, probably
18:37 masak sounds good for simplicity
18:37 shk3_ instead of this:
18:37 shk3_ for ( Object key : possibleKeys )
18:37 shk3_ assignEditors( mol, item, (String) key, ++l );
18:37 shk3_ you just need to build the string and set it to the tableeditor
18:37 masak yes
18:38 shk3_ shall I try?
18:38 masak please go ahead
18:38 masak one problem is that there are no good examples of property use in the molecule files
18:38 shk3_ I wil check in if ready
18:39 masak oh, thiamin is pretty non-lame
18:39 masak it has Source and SMILES formula
18:39 shk3_ the table has got two columns already, right?
18:39 masak but right now, only the last is shown
18:39 masak shk3_: yes
18:39 masak Structure and Properties
18:41 masak they're created along with the table
18:42 * shk3_ away to the loo
18:45 masak hm, there's an unused variable `showproperties`
18:46 masak what's the idea behind it?
18:49 shk3_ the idea was to offer switch on/off for the properties, but I remove it
18:49 masak :)
18:56 shk3_ I commit something which compiles but is not tested
18:56 masak sounds good
18:56 masak I can test it
18:56 shk3_ I am going to test now as well
18:57 shk3_ ok, need to update first
18:57 masak :)
18:57 masak hope you don't get a lot of conflicts
18:58 shk3_ could be worse
18:58 CIA-31 bioclipse: shk3 * r3952 /trunk/plugins/net.bioclipse.cdk.ui/src/net/b​ioclipse/plugins/views/Structure2DView.java: refactored 2dviewer
18:58 * masak updates
18:58 shk3_ ok
18:58 shk3_ testing
18:59 masak yes
18:59 masak hm
18:59 masak try thiamin.mol
19:00 masak it should show two properties
19:00 masak the one named Source and the SMILES line
19:00 masak on my box, it shows neither
19:01 shk3_ I do not even get the molecule or table
19:01 shk3_ do you?
19:01 masak yes
19:01 shk3_ hm, so old problems
19:01 masak right click and "show all structures"
19:01 shk3_ I forgot the set the string to the table
19:01 masak ok
19:01 shk3_ doing that.
19:02 masak while I remember it, there's an okMessage method which I remember writing
19:02 masak just a convenience for the two instances of dialog boxes that the view issues
19:03 masak one with the message: "It seems molecules in here have neither 2d nor 3d coordinates. We cannot display them."
19:03 masak and one with "It seems molecules in here have only 3d coordinates. They were flattened into 2d, and may not look good."
19:03 shk3_ new commit
19:03 CIA-31 bioclipse: shk3 * r3953 /trunk/plugins/net.bioclipse.cdk.ui/src/net/b​ioclipse/plugins/views/Structure2DView.java: refactored 2dviewer
19:04 masak I'm thinking that these two cases may be bad enough to alert the user, but maybe not bad enough to alert the user with a dialog box
19:04 masak what do you think?
19:04 shk3_ we have got the console, but right now there is much noise in there
19:04 shk3_ but it would be right for that, I think
19:04 masak would it be a better idea perhaps to display the messages in the status bar?
19:04 masak or the console, that works too
19:04 shk3_ status bar sounds good
19:05 masak :)
19:05 masak so either works
19:05 masak I'll work that into the class then
19:05 shk3_ hm, now the structure is in properties
19:05 masak it's mainly because I think it could become an annoyance to frequent users of the view
19:06 shk3_ definitly
19:06 shk3_ or you could make a "show not again" box
19:06 shk3_ with preferences
19:06 shk3_ as in eclipse
19:06 masak that's just pushing the problem on the user, though
19:06 masak I'm not sure the structure is in properties, actually
19:07 shk3_ I think I did not get the point with the table editors
19:07 masak look what happens if you scroll horizontally
19:07 shk3_ hm, looks like here
19:07 shk3_ nothing
19:07 masak the structure stays in place while the table moves
19:07 shk3_ not for me
19:07 masak ok
19:07 shk3_ structure moves with the table
19:07 masak hm, different behaviour again :)
19:08 * masak diffs
19:08 shk3_ I found this layout() call is necessary if added components at runtime
19:08 shk3_ perhaps it helps here?
19:09 masak never hurts to try
19:09 shk3_ trying...
19:09 masak but I'm curious about what you changed from the last rev to make this effect...
19:10 shk3_ don't think i did something
19:10 shk3_ code is pretty much the same
19:10 masak huh
19:11 shk3_ well, I did something
19:11 masak I think I see your changes
19:11 shk3_ but nothing should cause that
19:12 masak they really shouldn't have caused this
19:12 masak no, exactly
19:12 shk3_ try resizing the first columns in the table
19:12 masak very local in one method
19:12 masak ok, hold on
19:12 shk3_ yes, just setting another text in the second column
19:13 masak upon resizing, nothing happens
19:13 masak I will now try commenting out your changes from last rev and running again
19:13 shk3_ for me the content of the second column stays where it is
19:13 shk3_ goes into first, so to say
19:14 masak without your lines, everything is normal
19:14 masak as expected
19:14 masak no props though
19:14 masak changing back
19:15 masak oh, I can see the properties if I scroll a little
19:15 masak they're on the same line and not alphabetical, but they're there
19:15 masak too bad the moleculy gets thrown out of place, though
19:17 shk3_ the keys should be alphabetical, but thats a minor problem
19:17 masak ok, so logically the displacement of the molecule must have been caused by the editor.setEditor call
19:17 masak but how?
19:17 masak it looks fine to me
19:18 shk3_ ah, properties are on the far end
19:18 shk3_ I just moved that call from the deleted method to addtotable
19:19 shk3_ I am reusing the tableeditor, but that should be ok?
19:19 masak think so
19:19 shk3_ I forgot the line break, commited that
19:19 CIA-31 bioclipse: shk3 * r3954 /trunk/plugins/net.bioclipse.cdk.ui/src/net/b​ioclipse/plugins/views/Structure2DView.java: refactored 2dviewer
19:20 masak got it, thx
19:21 masak I'd say this is perfect except for the new interesting displacement behaviour
19:21 masak whoever fixes it first gets a cookie :)
19:22 masak hm, if I use a new editor, the molecule doesn't jump
19:22 masak on the other hand, I don't get any properties any more :(
19:23 masak why is the world so full of compromises?
19:23 shk3_ ok, so the editor stuff is tricky
19:23 masak clearly
19:23 shk3_ just used it by example right now
19:23 shk3_ perhaps we need some reading about it
19:23 masak maybe
19:25 masak actually, when reading TableEditor source, there's an indication that they might be one-use-only
19:25 shk3_ sounds logical
19:25 shk3_ was wondering about that
19:26 masak still, it doesn't explain why when I use a new one I still don't get any props text
19:28 masak ha! got it working!
19:28 shk3_ no
19:28 masak cookie is mine
19:28 masak :)
19:28 shk3_ (::)
19:28 shk3_ that's a cooke with choclat flakes
19:29 masak no doubt
19:29 masak here it comes
19:29 CIA-31 bioclipse: carl_masak * r3955 /trunk/plugins/net.bioclipse.cdk.ui/src/net/b​ioclipse/plugins/views/Structure2DView.java:
19:29 CIA-31 bioclipse: [Structure2DView.java]
19:29 CIA-31 bioclipse: * fixed minor molecule displacement bug
19:31 shk3_ great
19:31 shk3_ only problem for me:
19:31 shk3_ the line breaks don't show as line breaks
19:31 masak no?
19:32 shk3_ for you?
19:32 masak well,
19:32 shk3_ I get dices
19:32 masak I think \n does
19:32 masak because it looked right
19:32 masak no, actually it was probably \r that looked right
19:32 masak and \n looked like a space
19:32 shk3_ I did \r\n
19:32 masak yes
19:32 masak that actually surprised me a little
19:33 masak because usually Java code only needs \n
19:33 shk3_ I always think it can't do harm
19:33 shk3_ java is only \n
19:33 shk3_ ?
19:33 shk3_ really
19:33 shk3_ so we use that
19:33 shk3_ no problem
19:33 shk3_ trying
19:33 masak think so
19:33 * masak tries too
19:34 masak I'm not sure what meaning \r would have within a text field anyway
19:34 masak no, it's as I thought
19:34 masak \n looks like a space over here
19:34 masak but \r did the trick
19:34 masak how about you?
19:35 shk3_ I get one dice now
19:35 shk3_ showing the :: side
19:35 masak :)
19:35 masak no, actually \r didn't work either
19:35 shk3_ works better in sweden
19:35 masak seems you were right
19:35 masak \r\n it is
19:35 masak at least it works here
19:36 masak strange, though
19:36 masak maybe it's a Mac thing
19:37 shk3_ may be
19:37 shk3_ but: i get the structure showing from start now
19:37 shk3_ that's good
19:37 shk3_ googling for the line break stuff
19:37 masak excellent
19:37 masak I will look into the widget stuff in the meantime
19:39 shk3_ Text text = new Text( table, SWT.MULTI); and \r\n works
19:40 CIA-31 bioclipse: shk3 * r3956 /trunk/plugins/net.bioclipse.cdk.ui/src/net/b​ioclipse/plugins/views/Structure2DView.java: properties are multi line now
19:42 masak works very well
19:42 masak the properties also now keep inside the visible part of the table
19:43 shk3_ looks good
19:43 shk3_ much better than too many columsn
19:43 shk3_ for me, the text is on top.
19:43 masak oki
19:44 shk3_ should I try to center that?
19:44 masak vertically?
19:44 shk3_ yes
19:44 masak no, I don't think so
19:44 shk3_ ok
19:44 masak text usually goes at the top of cells
19:44 masak for some reason
19:45 masak ok, so I will attempt to exchange the Image things (and the methods it depends on) with egonw++'s new JChemPaintView
19:46 shk3_ great
19:48 masak shk3_: what's the relation between an IMolecule and an IAtomContainer?
19:49 shk3_ IMolecule inherits from IAtomContainer
19:49 shk3_ so do rings etc.
19:50 masak suits me fine in this case
19:50 masak I haven't figured out how to solve the special case with SDF buttons yet
19:50 shk3_ afaik Molecule doesn't really add anything to AtomContainer
19:50 masak ok
19:50 masak ...but the rest of the integration is done
19:50 shk3_ great
19:50 masak I'm testing once before comitting
19:51 masak works like a charm
19:51 masak comitting...
19:52 CIA-31 bioclipse: carl_masak * r3957 /trunk/plugins/net.bioclipse.cdk.ui/src/net/b​ioclipse/plugins/views/Structure2DView.java:
19:52 CIA-31 bioclipse: [Structure2DView.java]
19:52 CIA-31 bioclipse: * delegated drawing capabilities to JChemPaintView
19:52 masak there's probably a few methods that can be removed now
19:52 masak I will take a look
19:54 masak ...no, doesn't seem like it
19:55 masak basically, the button feature keeps a lot of code in there
19:55 masak and I don't see how to get rid of it without removing the buttons
19:55 masak will have to think about that
19:56 shk3_ hm, we could have an extra control, like an "open" button or so under the widget
19:56 masak shk3_: I was thinking along the same lines, yes
19:57 masak as an extra bonus, maybe one could get both SDF and non-SDF molecules to behave the same
19:59 shk3_ yes. feel free to do what you think
20:01 masak thank you. but maybe not tonight :)
20:01 masak gotta save something for later
20:02 masak but thank you for the help with coding+thinking, this helped clear things up
20:05 shk3_ thank you
20:06 shk3_ it was a pleasure
20:06 masak :)
20:06 masak likewise
20:06 masak I should do more of these late-night hacking sessions...
20:06 masak I'm seldom this focused in the daytime
20:08 shk3_ it really helps to do things together
20:08 masak yes
20:08 masak cross-national pair programming
20:08 shk3_ it sounds a bit stupid if you hear about xp having teams of two all the time,
20:08 shk3_ but it has its merits.
20:08 shk3_ I had that at work as well, it really can help.
20:08 masak nod
20:12 masak 'night
20:13 shk3_ see you

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