Camelia, the Perl 6 bug

IRC log for #bioclipse, 2008-03-25

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

All times shown according to UTC.

Time Nick Message
00:25 CIA-22 bioclipse: rklancer * r5169 /bioclipse2/trunk/plugins/ (15 files in 11 dirs): little fixups to logging, logic, spelling (recieve->receive, GENEBANK->GENBANK, firt->first). Added util function to print throwable stack trace directly to logger (at debug level)
00:46 CIA-22 bioclipse: rklancer * r5170 /bioclipse2/trunk/plugins/net.bioclipse.core/sr​c/net/bioclipse/recording/RecordingAdvice.java: teeny fix to logging statement
06:46 CIA-22 bioclipse: Annzi * r5171 /bioclipse2/trunk/plugins/net.bioclipse.moss/ (15 files in 5 dirs): Modification in help content
07:13 egonw joined #bioclipse
08:18 jonalv joined #bioclipse
08:33 Annzi joined #bioclipse
08:40 olas joined #bioclipse
08:40 olas and good morning to you all
08:41 olas hope everyone had a nice easter holiday
08:47 olas left #bioclipse
08:47 olas joined #bioclipse
09:45 egonw joined #bioclipse
09:45 egonw upsalla: thanx for the toffifee !
09:46 egonw it was nice
09:49 olas :-)
09:49 olas glad you liked it
09:49 jonalv haha
09:49 olas and the diploma?
09:49 olas did you frame it?
09:51 egonw no, not yet
09:52 egonw have to find a frame small enough
09:52 egonw my other diploms are much larger :)
09:55 egonw seriously, I much appreciate it!
10:14 olas :-D
13:01 masak joined #bioclipse
13:08 egonw joined #bioclipse
13:09 EskilA joined #bioclipse
13:22 rojasm joined #bioclipse
13:33 steinbeck joined #bioclipse
13:37 rklancer joined #bioclipse
13:37 rojas1 joined #bioclipse
13:37 rojas1 left #bioclipse
13:37 rojas1 joined #bioclipse
13:41 rklancer hi all
13:41 rklancer jonalv...
13:41 masak hi rklancer
13:41 rklancer hi masak
13:42 jonalv hi rklancer
13:42 rklancer hi. got a question for you...
13:42 rklancer I made a bunch of commits I want to make sure I know what I'm doing :)
13:42 rklancer did you svn update?
13:43 rklancer masak, you too
13:43 rklancer :)
13:43 masak rklancer: on what? bc2?
13:44 rklancer bc2 yeah
13:44 masak oki
13:44 rklancer in particular I rearranged the logic in RecordingAdvice...
13:44 masak nice!
13:44 masak why?
13:46 rklancer i think the if-then statements became a little confusing as recordingadvice got changed around
13:46 masak hm, I'll review the change
13:46 rklancer i was reading the code to try to understand the big picture and I think i spotted an opportunity to rearrange the if-thens to be clearer.
13:46 rklancer thanks!
13:46 rklancer check out the TODO comment too & let me know what you think
13:47 masak sure.
13:47 masak offhand, it sounds like you did the right thing
13:48 olas hi rklancer
13:48 rklancer hi olas!
13:48 olas nice too see your commits
13:48 * olas must run off to prepare for coffe-break
13:48 rklancer thanks...hope i did the right thing :)
13:48 * jonalv is starting up Eclipse now
13:49 masak rklancer: re if statement. definitely the right thing
13:49 olas I'll look at it asap
13:49 * olas is afk
13:49 masak rklancer: re TODO -- if we reach that point, something is probably wrong, yes
13:49 masak so it should most likely be a warning
13:50 rklancer do you think it should be an assert too? (for development time)
13:50 masak maybe. I don't do asserts myself
13:50 masak never got into the habit
13:51 rklancer I'm trying to :)
13:51 masak jonalv is head honcho for manager code, so it's his call, I guess
13:52 masak but, trust the anarchy. it's svn so nobody will bite you if you just go ahead and add it in the meantime
13:52 rklancer "trust the anarchy" ... I like!
13:52 masak audreyt++
13:55 jonalv We should never get there so I think assertment might be an idea
13:55 jonalv although I think getting there would be hard... :)
13:55 masak jonalv: that's why an assert makes sense
13:56 masak you assert things that you think are true
13:56 jonalv yea, assertment is fine by me
13:58 rklancer ok...
13:59 CIA-22 bioclipse: rklancer * r5172 /bioclipse2/trunk/plugins/net.bioclipse.core/sr​c/net/bioclipse/recording/RecordingAdvice.java: assert recorded method is always on IBioObject or IBioclipseManager
14:00 masak rklancer: looks nice
14:00 rklancer ...btw, clearing that up is also helping me to understand what's going on
14:00 masak that's an acheivement in itself
14:01 rklancer well i think it's neat code
14:01 masak I'm a bit wary of the whole recording structure nowadays -- it's way cool but also a bit brittle
14:01 rklancer so I'm happy the more i understand of it ;)
14:01 masak I don't like how we force everyone and everything to use our data types
14:01 masak BioList, for example
14:02 masak a BioList is just an ArrayList that you have to use in Bioclipse
14:03 rklancer I'll keep that in mind ... one thought I had re managed access to biojava/cdk/etc is that all plugin writers should be able to extend that managed access
14:03 rklancer so if they need function x, write the manager that reaches through to do function x
14:04 masak that's the general idea, yes
14:04 rklancer but i def. see your point .... and all i can say is that i don't know enough yet
14:04 rklancer cool
14:04 masak I wasn't always wary. and I'm always prepared to be impressed by the impressive parts
14:05 masak the part of recording that actually works, for example
14:06 rklancer i do think it's a great form of documentation and usability btw
14:06 masak recording? ideally, yea
14:06 rklancer even if you never use recorded scripts directly, by showing them you gradually expose savvy users to how programmatic access works
14:06 rklancer so I'm sold
14:07 masak yes, that is nice
14:10 jonalv so can assertions be removed with some flag when building or what?
14:10 rklancer yeah
14:10 rklancer actually you need a flag to turn them on
14:10 jonalv okey
14:10 jonalv And Eclipse uses that flag when running as default or?
14:11 rklancer no
14:11 rklancer you have to add -ea (-enableassertion) flag to jvm opts
14:11 rklancer -opts
14:12 rklancer +args
14:14 jonalv oh well, I really don't need it but interesting to know
14:14 rklancer i've added a few asserts, just added -ea flag to my launch config but not my bioclipse.product file
14:15 masak is it possible that Debug mode in Eclipse adds -ea for you?
14:15 masak we were just discussing that here in the office
14:15 rklancer no I just tested it
14:16 masak oki
14:26 CIA-22 bioclipse: rklancer * r5173 /bioclipse2/trunk/plugins/net.bioclipse.core/sr​c/net/bioclipse/recording/RecordingAdvice.java: future-proofed warning/assertion message in RecordingAdvice
14:51 edrin joined #bioclipse
14:51 edrin hi#
14:52 rklancer hi edrin
14:52 edrin hi rklancer
14:53 masak hi edrin
14:54 edrin hi masak
15:11 masak log4j:WARN No appenders could be found for logger (net.bioclipse.ui.Activator)
15:11 masak rklancer: what does that mean?
15:11 rklancer basically it means logging wasn't configured properly
15:11 egonw that the configuration has not worked
15:11 rklancer do you have the net.bioclipse.log4jconfig project in your launch config?
15:12 masak no
15:12 rklancer download it :)
15:12 * masak downloads that
15:12 masak btw, is net.bioclipse.logging still in use?
15:12 rklancer not really
15:13 rklancer there's some code there we can borrow from
15:13 masak so I remove that?
15:13 rklancer but I think there are still dependencies...lemme think
15:13 masak ok, I'll leave it for now
15:14 rklancer 'k ... I'll get back to you about that
15:14 egonw why does Eclipse not report it is missing the .log4jconfig plugin?
15:14 egonw that is, why is there no dep defined?
15:16 rklancer probably because I wasn't thinking :)
15:16 rklancer hmm
15:17 masak rklancer: I just want to let you know that to me it seems that you have done the most thinking around here the last few weeks
15:17 * masak appreciates it
15:19 CIA-22 bioclipse: rklancer * r5174 /bioclipse2/trunk/plugins/net.bioclipse.ui/ (4 files in 3 dirs): more logging fixup ... search & destroy of Throwable.printStackTrace() (replaced with net.bioclipse.core.util.LogUtils.debugTrace())
15:20 rklancer masak, thanks for your kind words ... if that's true it's just because I'm still in the learning phase so I have to think lots
15:21 rklancer egonw, actually I'm not sure how to define a dependency on the log4jconfig project (speaking of being in the learning phase...)
15:21 rklancer b/c it's a fragment project I can't just add a require-bundle type dependency ... and it doesn't export any packages
15:24 egonw ah...
15:24 egonw interesting problem indeed :)
15:24 egonw not sure how to do that either
15:33 CIA-22 bioclipse: rklancer * r5175 /bioclipse2/trunk/ (11 files in 9 dirs): r5174 applied to more projects - more logging fixup ... search & destroy of Throwable.printStackTrace() (replaced with net.bioclipse.core.util.LogUtils.debugTrace())
15:48 CIA-22 bioclipse: rklancer * r5176 /bioclipse2/trunk/ (13 files in 13 dirs): Removal of all references to net.bioclipse.logging. No apparent ill effects.
15:48 rklancer masak, there's your answer :)
15:49 masak neato
15:50 rklancer btw, masak, you use osx right?
15:51 masak I do.
15:52 rklancer have you seen that you can double-click the bioclipse.log file to open in Console.app? Nice b/c it updates in R/T like doing tail -f bioclipse.log
15:52 masak ooh
15:52 masak I did not know that
15:53 rklancer bonus for cool os x people :)
15:53 masak where's bioclipse.log?
15:53 rklancer should be in your home dir
15:53 egonw what does b/c stand for?
15:53 rklancer you can edit that
15:53 rklancer "because"
15:53 egonw ah, ok
15:54 rklancer oh didn't mean to mystify!
15:54 masak nope, I don't have a bioclipse.log
15:55 masak maybe because till now I didn't have that log4jconfig thing
15:55 rklancer hmm... what prints to the console when you start bioclipse?
15:55 masak text? bioclipse?
15:55 masak I'm not sure I understand
15:55 rklancer well yeah after you start bioclipse
15:56 rklancer either in debug mode with -consolelog turned on
15:57 masak I believe I do have -consolelog turned on
15:57 rklancer or by starting bioclipse.app/Contents/MacOS/bioclipse -consolelog
15:57 rklancer does it say anything like "INFO - Probable log file location: /Users/masak/bioclipse.log"
15:57 rklancer ?
15:57 masak nope.
15:58 masak jost that same warning as before
15:58 rklancer oh...
15:58 masak log4j:WARN Please initialize the log4j system properly.
15:58 masak I do have log4jconfig now
15:58 masak do I need to do something with it?
15:58 rklancer you also added it to your run or debug configuration?
15:59 masak rklancer: hm... god point...
15:59 masak s/god/good/
15:59 masak ah...
15:59 masak works much better now :)
16:00 * masak forgets obvious steps sometimes
16:00 * rklancer understands the feeling
16:00 masak I was _very_ pleasantly surprised a few days ago when I realized that ^S does incremental search in Eclipse
16:01 masak that is, using the Emacs keymap
16:01 masak the incremental search is not perfect, but it's there, and that's very nice
16:02 masak hm, bc2 actually puts the log file in my homedir
16:02 rklancer not optimal
16:02 masak that's not very socially accepted...
16:02 masak no
16:02 masak http://www.codinghorror.com​/blog/archives/001032.html
16:03 rklancer thanks :)
16:05 quennie joined #bioclipse
16:07 rklancer masak,
16:07 rklancer hmm.
16:07 rklancer I want to put the log in the root of the installation location (if we were os x only I suppose it could go in Application Support)
16:08 rklancer but osgi.install.area == "/Applications/eclipse" when self-hosting
16:09 masak hold on... busy right now
16:10 rklancer okay
16:24 masak oki, avail again
16:24 masak unfortunately, I don't have a direct answer
16:24 masak but surely there must be a way to find a path to put it?
16:25 rklancer one possible path is the osgi instance area (== your workspace location)
16:25 masak I like that better
16:26 CIA-22 bioclipse: Annzi * r5177 /bioclipse2/trunk/plugins/net.bioclips​e.moss/src/net/bioclipse/moss/wizards/ (MossContextProvider.java MossWizard.java ParametersPage.java): Restrictions on how you write atoms and molecules were added
16:26 rklancer It would be nice to be able to use the install location
16:26 rklancer but that has tradeoffs too
16:26 CIA-22 bioclipse: carl_masak * r5178 /bioclipse/trunk/plugins/net.bioclipse.base2/sr​c/net/bioclipse/base2/dialogs/LoginDialog.java:
16:26 CIA-22 bioclipse: [LoginDialog.java]
16:26 CIA-22 bioclipse: * dimensions and coordinates
16:27 masak rklancer: yes
16:27 Annzi left #bioclipse
16:27 rklancer at least on all platforms you should be able to write to the workspace
16:28 masak ok, I have a difficult SWT question
16:28 masak it concerns TreeViewers and ITreeContentProvider
16:28 masak basically, I want the same behavior as subclipse has
16:29 masak I'm in the subclipse source right now, digging for answers
16:29 masak but maybe someone here knows
16:29 masak the behavior is: 1. I expand a directory 2. it says "Pending..." 3. after a while the real contents show up
16:30 masak how do I do that? I suppose I need threads? is there a working example somewhere?
16:36 * masak asks on #eclipse
16:55 jonalv joined #bioclipse
17:02 * masak gets help on #eclipse
17:02 masak \o/
18:19 rklancer joined #bioclipse
18:35 edrin left #bioclipse
19:54 egonw joined #bioclipse
21:54 edrin joined #bioclipse

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