Camelia, the Perl 6 bug

IRC log for #bioclipse, 2008-03-14

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

All times shown according to UTC.

Time Nick Message
06:23 thomas_ku joined #bioclipse
07:43 rojasm joined #bioclipse
07:51 egonw joined #bioclipse
08:09 olas joined #bioclipse
08:09 olas morning all
08:09 olas A beutiful day here on IRC
08:13 egonw oh god... what did they do to you?
08:14 egonw the basterds
08:14 viklund` olas: agreed
08:24 olas egonw: stepford?
08:24 * olas was in stepford yesterday
08:25 * olas is not a stepford wife
08:25 egonw sounds familiar
08:25 egonw right...
08:25 egonw ic :)
08:56 jonalv joined #bioclipse
09:32 olas joined #bioclipse
12:54 egonw joined #bioclipse
13:53 rklancer joined #bioclipse
13:57 rojas1 joined #bioclipse
14:04 rklancer hi ola
14:04 rklancer olas
14:04 rklancer rather
14:08 jonalv hi rklancer
14:08 rklancer hi jonalv
14:09 jonalv all well?
14:09 rklancer yup
14:09 rklancer amused by the website Dr. Steinbeck's student discovered
14:09 jonalv yea
14:09 rklancer it's rather charmingly clueless
14:10 jonalv well I have checked around a bit more on it. THey have a bunch of courses and stuff. Makes me wonder...
14:10 rklancer a bunch of courses?
14:11 jonalv <http://www.infocomlaboratories.com/dbtrain.htm> for example
14:11 rklancer hmm
14:11 rklancer maybe
14:12 rklancer the website has all kinds of little clues that it's one or two guys trying to sound bigger than they are and desperately hoping someone will hire them for something they sort of know how to do
14:13 rklancer registered to john and claudia gallagher of prescott arizona
14:13 rklancer btw
14:13 jonalv well yea from our oint of view it's rather obcious :)
14:13 olas looks pathetic IMHO
14:13 jonalv *obvious
14:13 olas rklancer: how do I get rid of the verbos e logging output from OSGI?
14:13 rklancer "This position is for Account Openers who are seeking to leverage a lifetime of sales contacts with minimum effort and maximum ROI. Commission rates of up to 10%. If you are seeking a position from your home or home office with realistic hours and full time commission, please forward your resume and credentials to  humanresources@InfoComLaboratories.com."
14:14 rklancer olas
14:14 jonalv I couldnt help but compare to Genetta...
14:14 rklancer quick fix is to set debug level to info
14:14 olas in what docu?
14:15 rklancer (translation: we can't pay you anything or, like, give you an office, but if by some miracle you happen to sell something, we might give you up to 10% of that. PLEASEPLEASEPLEASEagreetodoit..."
14:15 rklancer olas, in n.b.ui.Activator
14:15 rklancer I want to set up a logging config file today
14:16 jonalv rklancer, sounds great
14:16 * jonalv is getting rid of bugs one by one in Brunn
14:17 rklancer after thinking about it, I bet the best use of time is to set up a logging config file in the base directory (rather than in a plugin directory, which will probably end up jar'd) and just let developers edit it
14:17 rklancer rather than do anything fancy
14:18 rklancer although we still want to listen for events on the platform log and write them to the bioclipse/log4j log
14:18 rklancer what do you think?
14:18 jonalv would that mean that you turn of debugging of the complete application at once?
14:18 rklancer yes
14:18 jonalv I thik I can live with that.
14:19 rklancer or on a package by package/plugin by plugin basis
14:19 jonalv oh that's even beter
14:20 rklancer to do that I'd have to remove references to n.b.logging, although I think there's very little code that uses it right now anyway
14:20 jonalv yes I think so too
14:21 jonalv so the log properties file wouldn't be in a jar file?
14:21 rklancer one thing is I'm not sure how to tell PDE Export to put a particular file in the base directory.
14:22 rklancer right
14:22 jonalv I am not sure how that would work...
14:22 jonalv ..either
14:22 rklancer after all you'd want to be able to edit it after you've exported a product
14:22 jonalv very true
14:31 rklancer olas: I'm going to commit a change right now that changes level to INFO. OK?
14:34 CIA-22 bioclipse: rklancer * r5093 /bioclipse2/trunk/plugins/net.bioclipse.ui/ (META-INF/MANIFEST.MF src/net/bioclipse/ui/Activator.java): quick fix for verbose logging from Spring -- sets threshold to INFO in code. Obvious edit in n.b.ui.Activator.start() will change you change back to DEBUG or whatever you like
14:35 rklancer btw, jonalv, I figured out what was going on with that Spring auto-startup issue I mentioned on, what, Monday?
14:36 jonalv okey? cool
14:36 rklancer yeah, it turns out that when Eclipse loads bundles they are labeled with the START_TRANSIENT bit
14:37 rklancer which means they aren't autostarted the next time the framework starts up
14:37 CIA-22 bioclipse: ospjuth * r5094 /bioclipse2/trunk/plugins/net.bioclipse.services/ (4 files in 3 dirs): Updated Services Extension Point and added several extensions.
14:37 rklancer but when you call Bundle.start() with no arguments, it doesn't set the START_TRANSIENT bit
14:37 rklancer so the bundle gets automatically started every time from then on
14:38 jonalv ok that is a strange behavior...
14:38 CIA-22 bioclipse: ospjuth * r5095 /bioclipse2/trunk/plugins/net.bioclips​e.services/src/net/bioclipse/services/ (27 files in 6 dirs): Adadpted model and view for new services extension point.
14:38 rklancer and yes this is the documented behavior. I really slapped my forehead when I figured that one out.
14:38 rklancer well I think it's a mismatch between OSGI and Eclipse assumptions
14:38 jonalv yea so it seems
14:39 rklancer in non-Eclipse OSGI platforms like maybe fancy cellphones, I guess it makes sense for the default behavior to be "start this bundle every time"
14:40 rklancer so I set the line that activates the spring bundle extender to use the Bundle.start(int options) method with options=START_TRANSIENT
14:42 jonalv ah okey
14:42 jonalv good
14:51 olas rklancer: how about not showing even INFO from OSDI by default?
14:54 rklancer olas: are you referring to Spring-OSGI?
14:54 olas yes
14:54 rklancer OK, I'll try this:
14:54 olas for similar to: 5802 [SpringOsgiExtenderThread-8] INFO org.springframework.osgi.context.sup​port.OsgiBundleXmlApplicationContext  -
14:54 olas which I don't want to see by default
14:54 rklancer Logger.getLogger("org.springfra​mework").setLevel(Level.WARN);
14:55 olas ok
14:55 olas thx
14:55 rklancer and then make a config file so you don't have to muck around in code to do this stuff. Since I figure jonalv or anyone working closely to figure out the Spring stuff might want to see spring logging :)
14:56 rklancer should be, "and then I'll make a config file"...
14:56 rklancer :)
14:59 CIA-22 bioclipse: rklancer * r5096 /bioclipse2/trunk/plugins/net.bioclipse.​ui/src/net/bioclipse/ui/Activator.java: quick fix to change Spring-OSGI logging level to WARN while retaining DEBUG logging for everything else. Will add file configuration soon so we can stop mucking around changing logging levels in code!
15:01 olas thx a lot
15:01 olas just cluttered my other debug outs...
15:01 olas working on a blog that will demo services
15:01 jonalv cool
15:02 CIA-22 bioclipse: rklancer * r5097 /bioclipse2/trunk/plugins/net.bioclipse.​ui/src/net/bioclipse/ui/Activator.java: logging again: small edit to use log4j api consistently--allowed me to remove two imports
15:03 CIA-22 bioclipse: ospjuth * r5098 /bioclipse2/trunk/plugins/net.bioclip​se.services/src/net/bioclipse/service​s/views/ServicesLabelProvider.java: Better display of impls
15:03 rklancer nice
15:06 rklancer olas: I'd like to make a small edit to n.b.services manifest, to remove some log4j packages that normal code should never need to use
15:07 rklancer is ok?
15:08 olas sure
15:08 olas go on
15:08 rklancer 'k...
15:11 CIA-22 bioclipse: rklancer * r5099 /bioclipse2/trunk/plugins/net.biocl​ipse.services/META-INF/MANIFEST.MF: Removing extraneous log4j packages from manifest to keep things neat and clear
15:11 olas great
15:13 CIA-22 bioclipse: rklancer * r5100 /bioclipse2/trunk/plugins/net.bi​oclipse.ui/META-INF/MANIFEST.MF: oops -- should have rolled into 5099 -- removing org.apache.log4j.spi from manifest since I already removed references to it from the code (physician, heal thyself)
15:29 CIA-22 bioclipse: ospjuth * r5101 /bioclipse2/trunk/plugins/net.bioclipse.services/ (8 files in 3 dirs): Documenting the EP and cleanup of model.
16:00 olas http://bioclipse.blogspot.com/
16:00 olas left #bioclipse
17:23 rklancer left #bioclipse
18:59 GrantSparks joined #bioclipse
19:02 GrantSparks Hi, I'm a professional java software engineer who wants to help out with bioclipse.  Who is the best person to send an email asking to join the project?
19:03 GrantSparks I've got the source for bioclipse2 checked-out and building/running, so I'm ready to go.
19:27 GrantSparks (I just sent my introduction to the dev mailing list)
21:04 rklancer joined #bioclipse
21:04 rklancer left #bioclipse
21:55 GrantSparks joined #bioclipse
22:08 GrantSparks joined #bioclipse

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