Perl 6 - the future is here, just unevenly distributed

IRC log for #bioclipse, 2014-09-03

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

All times shown according to UTC.

Time Nick Message
04:47 egonw joined #bioclipse
04:56 zaetnick joined #bioclipse
06:41 egonw joined #bioclipse
07:01 sneumann joined #bioclipse
07:25 jonalv joined #bioclipse
07:58 jonalv joined #bioclipse
08:01 samuell_ joined #bioclipse
08:29 Gpox joined #bioclipse
08:36 commitbot joined #bioclipse
08:36 commitbot [bioclipse.ds] goglepox pushed 1 new commit to develop: http://git.io/QZqnvQ
08:36 commitbot bioclipse.ds/develop 2d08519 Arvid Berg: Add tabbed properties contributor to ds
08:36 commitbot left #bioclipse
09:22 olass joined #bioclipse
12:11 olass joined #bioclipse
13:16 Gpoks joined #bioclipse
13:30 olass joined #bioclipse
14:24 commitbot joined #bioclipse
14:24 commitbot [bioclipse.cheminformatics] egonw opened pull request #47: Added method variants without IReturner that can be used from new InChIM... (develop...8-directInChIManagerCalls) http://git.io/8lvaYg
14:24 commitbot left #bioclipse
14:59 egonw jonalv: your comments on this pull request welcome too!
15:00 jonalv egonw: that looks stange
15:00 jonalv *strange
15:01 egonw please elaborate...
15:01 egonw I was unsure how to do it, and wondering...
15:01 jonalv you should never run the constructor for a manager
15:01 egonw in other managers that is normal...
15:01 egonw but not proper then?
15:02 jonalv you should _always_ use the one built by Spring
15:02 zaetnick joined #bioclipse
15:02 egonw (we should have a code clean up session then :)
15:02 jonalv or rather, egonw you are free to do what you want but I will not be held responsible for what doesn’t work :)
15:02 egonw jonalv: OK, thanks! I will update my code then
15:03 jonalv egonw: don’t you get a logger warning with your code?
15:03 commitbot joined #bioclipse
15:03 commitbot [bioclipse.cheminformatics] egonw closed pull request #47: Added method variants without IReturner that can be used from new InChIM... (develop...8-directInChIManagerCalls) http://git.io/8lvaYg
15:03 commitbot left #bioclipse
15:04 egonw phww.... I stopped looking at them... many, many logger warnings about manager call misuse in many managers :)
15:04 egonw [09/03/14 17:02] <egonw> (we should have a code clean up session then :)
15:04 jonalv aha
15:04 jonalv well, at least my warnings are working then :)
15:05 egonw hehehehe
15:05 jonalv but when you write a return tatement that means I can not run that method as a job
15:05 jonalv only void methods can be run as jobs
15:05 egonw yes, understood
15:06 jonalv hence the warning
15:06 jonalv it works
15:06 jonalv only it will be run in the same thread
15:06 egonw I will keep an eye on those plenty of calls where it says it won't repeat it the next 100 seconds or so...
15:06 egonw those are very frequent
15:06 egonw so, this is the proper code, right?
15:06 egonw ICDKManager cdk = net.bioclipse.cdk.business.Activator.getDefault().getJavaCDKManager();
15:06 egonw (depending on the manager you need)
15:06 jonalv egonw: yea and that will be made som much nicer in Bioclipse 3
15:06 jonalv one day
15:07 egonw :)
15:07 egonw jonalv++
15:07 egonw thanks for your peer review!
15:09 egonw PS, this is the code I just added (for @olass): openphacts.getURI(cdk.fromSMILES("CC(=O)OC1=CC=CC=C1C(O)=O"))
15:09 egonw that code requires me to get an InChI for the IMolecule
15:11 egonw jonalv: question... if you have time...
15:11 jonalv okey
15:11 egonw how does this practically work? in terms of monitors and jobs:
15:11 egonw (in some manager method)
15:11 egonw IRDFManager rdf = net.bioclipse.rdf.Activator.getDefault().getJavaManager();
15:11 egonw rdf.importFromString(store, rdfContent, "Turtle");
15:12 egonw the calling method already has a monitor, but that cannot be reused by the IRDFManager for a subtask of some sort...
15:12 jonalv ah the subtask problem
15:12 egonw instead, it will just be spawning just an additional monitor?
15:12 jonalv oh it’s been so long
15:13 jonalv Gpox correct me if I am wrong. But I think since you are using the Javamanager, your call to importFromString will be run as a separate job
15:14 egonw yeah, I'd guess so too :)
15:14 egonw ok, thanks
15:14 jonalv if you do the Javascriptmanager instead it will be run in the same thread
15:15 jonalv no progressmonitotr then though. For that I recommend setting status by subtask
15:15 jonalv and give a String
15:16 jonalv something like: “Doing blah (1 of 20000 done expecting to be finished in 45 days…)
15:16 jonalv egonw: there are some examples of that
15:16 jonalv can’t remember where though
15:16 jonalv egonw: look for referenses to  TimeCalculator.java
15:20 egonw ack... may try to find for that...
15:20 egonw BTW, this is that warning many, many managers trigger:
15:20 egonw 17:19:47.362 WARN  [n.b.m.b.AbstractManagerMethodDispatcher] rdf.importFromString is not void or returning a BioclipseJob. But implementation takes a progress monitor. Can not run as Job. Running in same thread. This message will not be repeated withing the next 120 seconds
15:21 jonalv yea that’s it
15:21 egonw so, that's OK to get that?
15:22 egonw but then it should not be a WARN, but INFO, not?
15:22 jonalv egonw: it’s not OK
15:22 jonalv it means someone is using the manager system wrong
15:23 egonw but, but...
15:23 jonalv yea it works
15:23 jonalv hence a warning
15:23 egonw this is teh current code behind it:
15:23 jonalv warnigns are fine to ignore but they genrally mean there is a better way of doing it :)
15:23 egonw IRDFManager rdf = net.bioclipse.rdf.Activator.getDefault().getJavaManager();
15:23 egonw rdf.importFromString(countStore, turtle, "Turtle");
15:23 jonalv that’s my view of how they should be used at least :)
15:23 egonw what's wrong about that?
15:23 egonw or
15:23 egonw what's the better way of doing it?
15:24 egonw use the JS manager version?
15:24 jonalv wait a moment
15:24 jonalv that method importFromString, does it return something?
15:25 egonw yes
15:25 egonw well...
15:25 egonw one sec...
15:25 jonalv _that_ is the problem..
15:25 egonw IRDFManager has this:
15:25 egonw public IRDFStore importFromString(
15:25 egonw IRDFStore store, String rdfContent, String format
15:25 egonw ) throws IOException, BioclipseException, CoreException;
15:26 egonw so, yes, also
15:26 jonalv egonw: is it on github?
15:26 egonw yes
15:26 jonalv that looks like a case where you want that IReturner thingie
15:27 egonw this is the implementation version, RDFManager:
15:27 egonw public IRDFStore importFromString(IRDFStore store, String rdfContent,
15:27 egonw String format, IProgressMonitor monitor)
15:27 egonw so, if I get such WARN, the manager should be using IReturner ?
15:27 jonalv @google how to make a manager 2
15:27 zarah jonalv: http://www1.salary.com/Project-Manager-II-Salary.html
15:28 jonalv nope
15:28 jonalv zarah—
15:28 egonw @google how to make a bioclipse manager 2
15:28 zarah egonw: http://wiki.bioclipse.net/index.php%3Ftitle%3DHow_to_make_a_manager_2
15:28 jonalv eya thatäs it
15:28 jonalv zarah++
15:28 egonw http://wiki.bioclipse.net/index.php?title=How_to_make_a_manager_2
15:29 jonalv egonw: http://wiki.bioclipse.net/index.php?title=How_to_make_a_manager_2#Running_manager_methods_as_jobs_which_return_a_result
15:29 egonw ok, so, that "pattern" should be used when I get this kind of WARN?
15:29 egonw I may fix such instances, because that  WARN message has been bugging me for a long time
15:29 jonalv yea eiher that or the one below for doing “partial results"
15:30 egonw but this shows up in *many* managers ...
15:30 egonw yeah, got it...
15:30 egonw IReturner for things with List<>
15:30 jonalv maybe we should update the warning with a link to that page? :)
15:30 egonw yes, would be good!
15:30 jonalv egonw: would you do the honors? :)
15:31 egonw I hope I remember... have to head home now...
15:31 egonw I'll keep the browser tab open...
15:31 jonalv (I haven’t even got a runnable Eclipse workspace with Biclipse…)
15:31 egonw that should remind me...
15:31 jonalv egonw++
15:31 egonw :(
15:31 egonw that explains why the number of commits has been so low in Uppsala recently ...
15:31 jonalv yea hope to get back to it real soon
15:31 egonw ok, thanks for the pointers!
15:32 jonalv yea something about getting phd and producing p-valeus :)
15:32 jonalv *values
15:32 egonw :)
16:42 Gpox joined #bioclipse
17:26 Gpox joined #bioclipse
18:00 olass joined #bioclipse
18:01 Gpox joined #bioclipse
19:30 zaetnick joined #bioclipse
20:04 olass joined #bioclipse
22:05 zaetnick joined #bioclipse

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