Camelia, the Perl 6 bug

IRC log for #bioclipse, 2008-10-31

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

All times shown according to UTC.

Time Nick Message
07:05 thomasku joined #bioclipse
07:28 |thomasku| joined #bioclipse
07:55 thomasku joined #bioclipse
08:15 Gpox joined #bioclipse
08:23 egonw joined #bioclipse
08:50 masak joined #bioclipse
10:13 edrin joined #bioclipse
10:14 egonw hi edrin!
10:14 edrin hi
10:14 zarah hi edrin
10:14 edrin hi egonw
10:14 egonw edrin: I'm writing a blog on XWS
10:15 egonw got a CDK service workgin :)
10:15 edrin oh :)
10:15 egonw found one problem
10:15 egonw the schema for the input I composed triggered a NullPointerException
10:16 edrin oh?
10:16 edrin can you paste xws4j console log in paste bin
10:16 egonw https://cdk.svn.sourceforge.n​et/svnroot/cdk/cdk-xws/trunk
10:16 edrin ?
10:16 egonw there is a client and the service
10:17 egonw please check the schema of the calculateMass
10:17 egonw though I think I have it right...
10:17 egonw anyway...
10:17 egonw I wrote a patch to work around this problem
10:17 egonw with some details
10:17 egonw will email this
10:17 edrin ok
10:17 edrin will have to work soon#
10:17 egonw ok
10:20 egonw sent
10:20 edrin egonw: i checked
10:20 edrin the schema
10:20 edrin it is a similar problem as with ola's schema
10:21 edrin The 'minOccurs' attribute is not allowed. The 'maxOccurs' attribute is not allowed.
10:21 edrin try this: http://schneegans.de/sv/
10:21 egonw ah, copied that from your example code...
10:22 egonw sorry...
10:22 edrin :)
10:22 egonw should have verified the schema myself
10:22 egonw apologies for that
10:22 edrin example in iodata xep?
10:22 edrin no prob
10:22 edrin i wasn't aware of this, too
10:22 edrin is it a mistake in the xep? should fix this
10:22 egonw xws-coffee-machine-sample/src/coff​ee/machine/GetCoffeeCanStatus.java
10:23 egonw the minOccurs is used there too
10:23 edrin oh
10:23 edrin that's wrong then
10:23 egonw just copied your example, and made changes step by step :)
10:23 edrin well
10:23 edrin i wonder why xmlbeans did nott complain
10:24 egonw it doesn't... not for the output
10:24 egonw but it does for the input
10:24 edrin yes, the input is the only schema that is "really" used by the component
10:25 edrin it is used to validate input before it is passed to the function
10:25 edrin you can also turn component's off if you want; in config
10:26 egonw ok
10:26 egonw but this is a rather good featue
10:26 egonw will fix my schema stuff
10:58 Gpox joined #bioclipse
11:36 egonw edrin: http://chem-bla-ics.blogspot.com/2008​/10/next-generation-asynchronous.html
11:36 egonw ok, going for lunch now
11:36 zarah egonw's link is also http://tinyurl.com/5poh8e
11:38 stain why doesn't the <iq> message have a namespace?
11:39 stain also in that message then <mass> is inside the urn:xmpp:tmp:io-data namespace
11:39 egonw bad copy/paste I think
11:39 egonw yeah, there a NS got lost too
11:40 stain it should come out as <mass xmlns=""> or something
11:40 egonw need to check my XSD ...
11:40 egonw I think that's why it is failing
11:40 stain it get's a bit simpler if you use prefixes instead of nested xmlns-es :-)
11:41 stain so <iq from="cdk.ws1.bmc.uu.se"> .. is cdk.ws1.bmc.uu.se the jabber ID? I didn't know you could have jabber-IDs without @ ;)
12:53 Julio joined #bioclipse
12:53 Julio left #bioclipse
12:54 egonw stain: where can I download the t2 beta?
12:57 masak egonw++ # "sorry that it took 10 minutes, but I'm running Vista..."
12:58 egonw I should have added:
12:58 egonw "actually, this is just to convince you into buying our cloud computing products"
12:58 masak yes, that might be true.
12:59 masak actually, the term "cloud computing" is exceedingly silly
12:59 masak on the same level as AJAX and Web 2.0
12:59 egonw well, except that AJAX is actually useful
13:00 stain egonw: the secret link is http://www.mygrid.org.uk/dev/wiki/dis​play/developer/Taverna+2.0+beta+1+RC1
13:00 zarah stain's link is also http://tinyurl.com/5ryajv
13:00 stain the official answer is to wait one week :)
13:00 stain as it's the release cnadidate it might not be the one going out.. for sure it won't be, as we forgot the Save results button
13:00 egonw heheheh :)
13:00 stain egonw: I have an example for an activity here now
13:01 stain not the bits about adding it to the service palette, though.. but should show you how it works
13:01 egonw oh, cool
13:01 masak egonw: I didn't say AJAX isn't useful. "cloud computing" is useful, too. it's just that they're managerial rebrandings of old concepts, and that's silly.
13:02 egonw stain: what does it look like?
13:03 stain like http://rpc268.cs.man.ac.uk/myacti​vity-for-t2v2.0b1-2008-10-31.zip
13:03 zarah stain's link is also http://tinyurl.com/6fo385
13:07 egonw ok, not that much different from a LocalWorker
13:08 egonw so, how would I compile/install this thingy?
13:08 egonw eclipse, export as jar?
13:08 egonw would that be enough?
13:08 stain mvn clean install
13:09 egonw issued....
13:09 egonw ok, downloading stuff now :)
13:09 stain to then include it  in your Taverna you would need to edit plugins/plugins.xml
13:09 egonw OK, that's not too difficult :)
13:09 stain let's hope it doesn't put you off..
13:09 egonw it's almost done downloading t2
13:10 stain basically make a new <plugin> anuy
13:10 stain yeah, iut only downloads APIs and stuff, so it shouldn't require too much
13:10 egonw ah, just the regular ecplispe project like plugins.xml ?
13:10 egonw or actually something custom?
13:11 stain make a new <plugin> entry there that lists your artifact (just copy the groupid/artifactid/version from the pom.xml and put inside a <dependency>) - and include a <repository> to file://$HOME/.m2/repository/
13:11 stain well, not $HOME, but replace with whatever is appropriate
13:11 egonw :)
13:11 egonw sure
13:12 stain oh, if you downloaded the Mac-version it might be a bit tricky to find the plugins.xml - you can edit $HOME/Library/Application support/t2workbench-something/plugins/plugins.xml instead of navigating inside the .app
13:12 egonw ah, no...t hat's fine...
13:12 egonw i'm ubuntu ibex
13:13 egonw ok, the GUI has not changed much
13:14 egonw nice to see biomart show up
13:14 egonw so, that Localworker folder is all BeanShell stuff?
13:15 egonw this is only T2 enactor, right?
13:15 egonw nothing special I have to do to use the new enactor, correct?
13:16 stain what do you mean.. special for whome?
13:16 egonw I just ran a workflow with the 'Ask' activity :)
13:16 egonw never mind
13:16 stain yeah, it's a bit silly that it's listed first :)
13:16 stain it's quite a lot of new code
13:16 stain the workbench is also new because it deals with the new API for defining workflows
13:17 stain we're not allowed to call it T2 anymore, it's just Taverna 2 now (and soon just Taverna)
13:17 egonw oh, sorry to hear that...
13:18 stain well, it's just to avoid confusion
13:18 stain although t2 is a new "magic piece" our users shouldn't have to know
13:18 stain for them it should still just be Taverna, right
13:18 egonw I understand...
13:18 egonw right
13:19 stain but if you are thinking about your asynchronous activity
13:19 egonw so, if these 'Localworkers' are BeanShells... how do I edit them?
13:19 stain select one, and then click configure down in the left corner
13:20 egonw ah, ok...
13:20 egonw overlooked that button
13:20 stain you would only edit the one in your workflow, so feel free to customize it
13:20 egonw right clicked on the node
13:20 stain yeah, iut should also be in right-click
13:20 stain but not yet ;(
13:20 stain we're slowly moving away from requiring right-clicks
13:21 egonw [INFO] BUILD SUCCESSFUL
13:22 stain but if you see your ExampleActivity-code, see how it does referenceService.register and then callback.receiveResult ?   Well, you don't need to do that from within that run() method - you could instead just do it as a handler that is listening on all your xmpp-messages coming back
13:22 stain so you could have just 1 thread for 100s of concurrent calls
13:22 egonw ok, called away from keybaord...
13:22 stain I'll go for lunch ;)
13:22 egonw I'll play with it over the weekend...
13:22 egonw and with a bit of luck, I have a nice sheet for the EMBRACE meeting on Monday :)
13:23 stain have fun :-)   And please report any emergency-issues with the rc1
13:23 egonw thanx very, *very* much!
13:23 stain it shoudl be possible to add comments on that wiki page
13:23 egonw okidoki
13:23 egonw cu later
13:24 CIA-52 bioclipse: carl_masak * r7525 /bioclipse2/trunk/manager-apidocs/ (. find-published-classes):
13:24 CIA-52 bioclipse: [manager-apidocs]
13:24 CIA-52 bioclipse: * find-published-classes
13:24 CIA-52 bioclipse: ** a minimal script that finds the class files to be processed
13:51 olass joined #bioclipse
13:51 olass hi
13:51 zarah hi olass
15:04 edrin hi olass
15:04 edrin egonw: nice to read it !
15:21 edrin egonw: sorry, I was away
15:55 egonw edrin: ping
15:55 edrin ping
15:55 edrin egonw: I am here!
15:55 egonw glad you liked to blog
15:56 egonw stain, edrin: have you 'met' before?
15:56 egonw edrin: was talking to stain earlier on how we could use xws in Taverna
15:56 egonw you showed me around, and will give it a go asap...
15:57 edrin No, I don't think so despite sitting in the same channel
15:58 edrin ah, yes, was reading chat log right now
15:58 egonw stain: edrin is they guy to ask about details of XMPP-webservice stuff
15:59 edrin stain: are you involved in Taverna development??
16:00 charlgren joined #bioclipse
16:07 egonw got to go
18:22 charlgren joined #bioclipse
19:07 edrin left #bioclipse

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