Camelia, the Perl 6 bug

IRC log for #bioclipse, 2010-01-28

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

All times shown according to UTC.

Time Nick Message
06:12 egonw joined #bioclipse
06:18 sneumann__ joined #bioclipse
06:37 egonw sneumann__: http://wiki.bioclipse.net/index​.php?title=Metabolomics_Feature
06:37 zarah egonw's link is also http://tinyurl.com/ykvsd5v
06:43 egonw joined #bioclipse
06:44 egonw sneumann__: or better: http://wiki.bioclipse.net/index.​php?title=Bioclipse.Metabolomics
06:44 zarah egonw's link is also http://tinyurl.com/yamu9ts
07:30 egonw joined #bioclipse
08:01 sneumann__ joined #bioclipse
08:07 Gpox joined #bioclipse
08:20 olass joined #bioclipse
08:24 egonw moin
08:24 zarah hi egonw
08:44 annzi joined #bioclipse
08:48 olass egonw: I have big problems exporting 2.2.0
08:48 olass it worked just fine for RC3
08:48 egonw wow
08:49 olass but I have now struggled for many many hours and it refuses to build
08:49 olass it gets into some infinite loop
08:49 olass I have not been able to track down th origin :(
08:49 egonw what are the clues?
08:49 egonw are there any?
08:49 olass not really..
08:49 olass after an hour of hard work I get an out of memory exception
08:50 olass after an hour of hard processor work I get an out of memory exception
08:50 egonw not good
08:50 olass it should take 5 minutes
08:50 olass indeed
08:50 olass I downloaded a fresh eclipse, same problem
08:50 egonw ok, what about exporting the bare minimum first
08:50 egonw the platform
08:50 egonw and then add other features one by one?
08:50 olass that is what I always do
08:50 olass problem is
08:50 olass it works to export just windows(!)
08:51 olass something is strange here
08:51 egonw yeah, can't make sense of that sentence
08:51 olass I will set up a new workspace now, just to be sure
08:51 egonw "it works to export just windows(!)" ?
08:51 olass mac, linux fails
08:51 egonw ah, now I get it
08:51 olass export win32 works :(
08:51 egonw just *for* windows
08:51 egonw ok :)
08:51 olass right
08:51 egonw I'm back with you :)
08:52 egonw and your exporting from mac?
08:52 egonw so, it's not delta being messed up
08:52 olass indeed
08:52 olass I am nowadays quite experienced with tracking down errors in exporting, but this is a new situation
08:53 olass and it worked just a few days ago!
08:53 olass I suspect some circular dependency that never completes
08:53 olass but how can I tell, and how can I find out?
08:55 olass is there some way to list all changes from a point, for example using git?
08:55 olass I will try to narrow down to a feature first
08:55 olass will keep you posted
08:55 olass but I am now quite frustrated
08:56 egonw did you tag?
08:56 olass nope
08:56 olass (damn)
08:56 egonw well, you can still do that...
08:56 egonw pinpoint the last commit
08:56 egonw and write down the hash
08:56 olass I guess I can do that by date..
08:56 egonw right...
08:56 egonw git log --since=
08:56 egonw then later:
08:56 egonw git log ThatHash..
08:57 egonw for all commits since that commit
08:57 egonw or something like that...
09:05 olass egonw: how can I make a checkout from a certain commit?
09:05 olass so that I create a branch startingfrom an earlier commit
09:05 olass so that I create a branch from the commit that worked (RC3)
09:05 olass any ideas?
09:06 egonw umm
09:06 egonw git checkout COMMITHASH, I  think
09:06 olass I have now tried a fresh workspace and only the core feature, still same infinite loop
09:06 olass ok, will try that
09:06 egonw ah, no
09:06 egonw well, quite like that
09:06 egonw but create a branch from it...
09:07 egonw git checkout -b branchedFromFoo COMMITHASH
09:07 egonw that's better
09:07 egonw than you can also add the tag
09:15 egonw olass: it would be a bit involving... but rather educational...
09:15 egonw once you know that you can still export RC3
09:15 egonw (I suggest from a workspace with only bioclipse.core)
09:15 egonw you know a commit that works, and one that does not...
09:15 egonw that begs for 'git bisect'
09:16 egonw quickest way to pinpoint the troublesome commit
09:16 olass yes, sounds good
09:16 olass only two commits differ in core
09:16 egonw oh wow
09:18 egonw annzi: where's your blog? and what's your github account?
09:19 egonw annzi: please add your blog to this wiki page: http://wiki.bioclipse.net/inde​x.php?title=MoSS_in_Bioclipse
09:19 zarah egonw's link is also http://tinyurl.com/ygkfs5y
09:20 egonw and I'll add it to http://planet.bioclipse.net too
09:20 annzi egonw: annziproject.blogspot but there is nothing on there yer
09:22 annzi egonw: my username for github is annzi, but I still need help with setting up my github
09:22 annzi egonw: will you be att bmc on Monday?
09:22 egonw yes
09:24 olass egonw: ping
09:25 olass I have a lot of untracked files in my repo... any way to get rid of them all?
09:25 olass they are zombies from the build process..
09:25 egonw git clean
09:26 egonw -d for dirs too
09:26 egonw -f for force
09:26 egonw -ff to force remove of dirs
09:26 olass so removes everything not under git control?
09:26 egonw right
09:27 olass ok, I will now build only core from the commit in RC3 and see if that works
09:28 olass if it works I will bisect
09:28 * olass is afraid the problem goes deeper...
09:28 egonw annzi: you have now write permission too https://github.com/egonw/bioclipse.moss
09:28 egonw olass: yes, that's what I fear too
09:28 egonw annzi: use this git clone URL:
09:29 annzi egonw: I need an introduction to everything, see examples an such, is there time for that on Monday?
09:29 egonw git@github.com:egonw/bioclipse.moss.git
09:29 egonw what about right now?
09:29 egonw olass, Gpox: what git client can you recommend? gitx for visualization, command line for the rest?
09:30 olass that's what I use
09:30 egonw annzi: do you have a Bioclipse source workspace yet? (on your new laptop)
09:30 olass I suppose egit could be useful too
09:30 egonw yes: git@github.com:egonw/bioclipse.moss.git
09:30 egonw bugger
09:31 egonw egit -> http://www.eclipse.org/egit/
09:31 olass I seldom use egit
09:31 annzi egonw: yes I have, and I ran MoSS yesterday, It acutally worked
09:31 jonalv joined #bioclipse
09:31 egonw I already like it because the eclipse navigator will show the active branch :)
09:31 egonw annzi: :)
09:31 olass yes, that's what I use it for too :)
09:31 egonw perfect
09:31 egonw annzi: are you comfy on the command line?
09:32 egonw with plain text editors?
09:32 egonw if so, I like you to edit this file:
09:32 egonw /path/to/bioclipse.moss/.git/config
09:32 egonw and look up this section:
09:32 egonw [remote "origin"]
09:33 egonw there, change the url to:
09:33 annzi egonw: I would not say that I am at the moment
09:33 egonw url = git@github.com:egonw/bioclipse.moss.git
09:33 egonw well, I guess a GUI editor will work too :)
09:33 jonalv oh hai all
09:33 zarah hello jonalv, you fantastic person you
09:35 * jonalv wonders when the release will be
09:35 egonw jonalv: ola is trying to fix export trouble
09:36 jonalv egonw: ah okey that's strange. What has changed since last time?
09:36 egonw that's what he is trying to figure out right now
09:37 * jonalv is lookign at the size of our old releases. Why 2.0RC3 268 MB, RC5 270 MB and 2.0 469 MB...
09:37 jonalv hm RC3 was 404 MB
09:38 egonw looking forward to a blog item about that
09:38 egonw that's a tid bit large, I'd say :)
09:38 jonalv so it went 404 then down to 282 and then down to 270 and then back up to 469
09:38 egonw if that is really needed... I'd say we look for torrents...
09:39 jonalv egonw: ah never mind me I am being stupid. Or rather the sourceforge interface is
09:39 jonalv apharently I was looking at summed up sizes for folders
09:39 jonalv that was not so easy to understand.... eh zarah?
09:42 olass egonw: I can export RC3
09:42 olass so clearly one of the commits broke it
09:42 olass *SIGH*
09:42 * olass will try git bisect
09:43 samuell joined #bioclipse
09:45 edrin how is the current git integration in eclipse? can i now sync with repo from eclipse GUI ?
09:45 egonw olass: what's the working commit hash?
09:45 egonw edrin: yes, with egit
09:45 edrin i think this was not possible last time i tried...
09:45 egonw see that link earlier
09:45 egonw oh, I thought it was
09:46 olass egonw: mom
09:46 olass egonw: working hash is d33b85ff4868e554d84d1564feaf01a395c7d252 in core
09:46 olass how do I get the differences with git bisect?
09:46 egonw tack
09:47 egonw not
09:47 edrin egonw: actually i really disliked to sync from console a repo...
09:47 egonw but you jump to an intermediate commit
09:47 egonw edrin: try egit
09:47 edrin :/
09:47 olass how do I see what's in a commit?
09:48 egonw git show hash
09:50 edrin Does this mean i cannot update (outdated) local with (actual) repo using egit? http://wiki.eclipse.org/EGit​/User_Guide/Remote#Pulling_.28merge_or_rebase.29_new_changes_from_upstream
09:50 zarah edrin's link is also http://tinyurl.com/ya4ccsu
09:52 edrin egonw: maybe you know the answer :D?
09:53 egonw ah, it seems so
09:53 * olass thinks he has pinpointed the error
09:53 olass WEEHOOW!!
09:53 * olass built 2.2.0
09:53 olass eclipse--
09:53 egonw what's the problem?
09:53 olass eclipse--
09:54 egonw eclipse--
09:54 olass eclipse--
09:54 zarah eclipse--
09:54 olass stupid stupid eclipse
09:54 olass the graphical product editor did not reflect the underlying product XML file
09:54 egonw ah
09:54 olass I have refreshed 1000 times
09:54 jonalv edrin: what is it you are having trouble with?
09:54 olass but still it did not
09:55 olass eclipse cost me 10 hours of work now :(
09:55 jonalv olass: maybe you should file them a bug?
09:55 egonw yeah, I noticed I need to close editors often, to get it properly refreshing
09:55 olass hell now, it is not reproducible
09:55 olass hell no, it is not reproducible
09:55 jonalv oh
09:55 olass GRRRR
09:55 olass *morr*
09:55 * jonalv hates non reproducable bugs...
09:56 olass *grumf*
09:56 olass *chew chew*
09:56 * olass is chewing on his laptop
09:56 olass *swallow*
09:56 * olass feels better now
09:59 CIA-59 bioclipse.core: Ola Spjuth 2.2.x * r8d1dc69 / features/net.bioclipse.platfo​rm_feature/bioclipse.product :
09:59 CIA-59 bioclipse.core: Removed VM arguments from product file.
09:59 CIA-59 bioclipse.core: Eclipse did not show these in GUI which resulted in an infinite loop. - http://bit.ly/cTVdmM
10:00 olass at least this is my theory
10:00 * olass is now trying to export a full 2.2.0
10:00 olass we shall know in a few minutes if it works
10:00 olass but it explains why windows actually did work :)
10:02 egonw yes, makes sense indeed...
10:02 * egonw mumbles hudson ...
10:07 egonw jonalv: http://cisrg.shef.ac.uk/shef2010
10:07 zarah egonw's link is also http://tinyurl.com/ycdldhk
10:07 egonw jonalv: there are bursaries if you have a poster
10:07 egonw deadline in three days, but no worries about retracting your abstract if you do not get a bursary
10:08 egonw does not include travel, though
10:08 jonalv egonw: threee days?!
10:08 egonw not sure about hotel
10:08 egonw ah, an abstract is just 10 lines of text
10:08 egonw with a few references
11:02 olass jonalv: are you a member of Smålands nation?
11:02 olass if not, join!
11:02 jonalv olass: no I am not
11:02 olass sad to hear
11:03 olass they have so many travel grants
11:03 olass I remember saysing this a while back
11:03 jonalv olass: yea it sounds familiar
11:03 olass I advice you to switch to Smålands asap, if you want to apply for travel grants
11:04 olass you need to be member for a year before you can apply
11:04 olass but much money and very high chances for getting trips funded
11:04 jonalv olass: okey, how do I join them then?
11:04 olass go to Smålands and ask
11:05 * jonalv is not even sure he knows where Smalands nation is
11:05 * olass and Martin has been to Las Vegas and on a week long conference on a cruise ship in the Mediterranian on Smålands expense
11:05 olass not bad actually
11:06 jonalv no I guess not
11:08 * egonw wishes that applied to post-docs too :(
11:19 * olass too
11:21 egonw :)
11:29 masak joined #bioclipse
11:43 * jonalv @ lunch
12:39 * olass is uploading 2.2.0 to SF
12:42 egonw olass++
12:43 masak \o/ olass++
12:43 * egonw is rebooting eclipse into his 2.4 workspace
12:43 jonalv \o/ olass++
12:43 olass \o/
12:43 olass let us rejoice when we see that it works..
12:43 egonw :)
12:43 jonalv and there was much rejoicing...
12:43 jonalv :)
12:44 masak wai~!
13:09 egonw annzi: please check: http://www.myexperiment.org/search?​type=workflows&query=kind:%28Bi​oclipse%20Scripting%20Language%29
13:09 zarah egonw's link is also http://tinyurl.com/ydkolms
13:09 egonw I have uploaded a few RDF/SPARQL BSL scripts there
13:10 edrin jonalv: i would prefer synchronizing with git repo via the eclipse GUI... makes it much more easy to commit/update singe files
13:10 annzi egonw: will do
13:11 annzi egonw: found a tutorial about SPARQL, have you seen this http://www.cambridgesemantics.com/​2008/09/sparql-by-example/#%281%29
13:11 zarah annzi's link is also http://tinyurl.com/4thmkl
13:11 egonw edrin: the second 'workaround' might do well
13:11 jonalv edrin: oh but why do you want to update single files? (Is that even possible?)
13:11 edrin with svn it was
13:11 egonw edrin: just 'reset' when you have no local changes
13:12 edrin ok
13:12 jonalv edrin: oh I didn't even realise it was possible with svn
13:12 jonalv edrin: but why? If you don't update everythign you could be running old code together with new code...
13:13 edrin yes, but i preferred to run always current svn for development, this way my code will be compatible with next release
13:16 jonalv edrin: okey you lost me. IF you prefer to always run current version then where is the problem again? IS it only then you don't like the terminal? Ican totally relate to that bu sadly enough not help you... :(
13:17 egonw edrin: updating is no problem with egit, if you do not have local changes
13:18 edrin ok
13:18 edrin well, maybe i just try to get current thing checked out and then see how it works with console...
13:19 egonw yes, it's easy
13:23 egonw git++ (for allowing me to fix a stupid bussines typo, so that it never shows up in history other then the IRC logs :)
13:34 jonalv (irc logs)++ //The only uneditable history these days
13:36 egonw yeah! 'man moss' :)
13:37 CIA-59 bioclipse.moss: Egon Willighagen master * re16354e / .gitignore : Ignore bin/ folders autogenerated by Eclipse - http://bit.ly/ddOJB5
13:37 CIA-59 bioclipse.moss: Egon Willighagen master * r3fa996e / (27 files in 10 dirs): Added templates for the MOSS manager and matching plugin tests - http://bit.ly/aWmLls
13:37 CIA-59 bioclipse.moss: Egon Willighagen master * r3192062 / (5 files): Added the MOSS feature - http://bit.ly/aqJfhZ
13:38 egonw no methods yet, that's next
13:56 egonw_ joined #bioclipse
15:15 egonw joined #bioclipse
15:20 egonw olass: ping
15:20 olass pong
15:21 egonw can you schedule a date/time for deciding what we want in 2.4 ?
15:21 olass will do
15:21 olass me is thinking
15:21 egonw in particular, I like to know if we go for Java6 or not
15:22 olass cannot say right now
15:22 olass need to investigate
15:22 olass but thanks for the reminder
15:22 olass we should bring it up for discussions
15:22 olass pros/cons
15:22 olass another matter
15:22 olass egonw: is there a wiki page for your endevour in semantic web with Bioclipse?
15:23 egonw no for semantic web
15:23 egonw +t
15:23 egonw but there is for RDF, etc...
15:23 olass I miss it on http://wiki.bioclipse.net/index.php​?title=Category:Bioclipse_Projects
15:23 zarah olass's link is also http://tinyurl.com/yjgt76j
15:23 olass no link there
15:23 egonw right... lot's of 'projects' do not use the project box yet
15:23 egonw and I am certainly guilty there...
15:24 olass if you should choose only one wiki page or blog post, which one would you recommend sending to people who are interested in semantic web and Bioclipse?
15:25 egonw looking
15:26 egonw ok, let me write up one :)
15:26 egonw there are many blogs around this topic
15:26 egonw will summarize them
15:28 egonw something else: Social Web: http://chem-bla-ics.blogspot.com/2009​/08/bioclipse-enters-social-web.html
15:28 zarah egonw's link is also http://tinyurl.com/yb4db7z
15:30 edrin hm :) with a hook you could modify bioclipse to report any click, drag, &drop to twitter...
15:30 egonw edrin: yes, we have been discussion that :)
15:30 edrin "egon clicked the "new" icon ..."
15:31 edrin :)
15:31 egonw "vscreen reports to have a succesful HIV hit: InChI=1/..."
15:32 olass egonw: that would probably not be something you'd like to disclose...
15:32 egonw indeed...
15:32 egonw we need a plugin for epo.eu too :)
15:33 edrin ehm egonw: how do i find out if alpha-tocopherol is soluble in ethanol ?
15:33 egonw edrin++
15:33 egonw the Open Notebook Solubility project is exactly about that
15:34 edrin good
15:34 egonw try the onssolubility manager
15:34 edrin where is it ?
15:34 egonw pull in the data from the google spreadsheet
15:34 egonw Bioclipse Social Feature, IIRC
15:34 egonw on the update site
15:34 edrin oh
15:34 edrin the bad thing is i need to know now, and have no bc installed and no admin access...
15:36 egonw http://spreadsheets.google.com/ccc?key=0Ak1R8T6​wt4YQcGx3d3VmcDMwaGZxMHVkbkVtUkQxYVE&hl=en
15:36 zarah egonw's link is also http://tinyurl.com/nf7828
15:36 egonw not sure if it is listed, likely not
15:37 edrin ok
15:37 edrin google says its soluble
15:37 zarah edrin: http://latimesblogs.latimes.com/boost​er_shots/2009/08/its-soluble-fiber-no​t-wheat-bran-or-irritable-bowel.html
15:37 edrin The tocopherols are viscous oils at room temperature, insoluble in water but soluble in certain organic solvents such as ethanol and acetonitrile
15:38 egonw edrin: try this prediction service: http://showme.physics.drexel.ed​u/onsc/models/multisolvent.php
15:38 zarah egonw's link is also http://tinyurl.com/yc87mey
15:38 * olass is waiting for egonw's bioclipse-semantic-web wiki page
15:38 edrin Solute SMILES: ? what is it of ethanol?
15:38 egonw CCO
15:39 olass egonw: should we remove the page http://wiki.bioclipse.net/index.php?title=Projects in favor of http://wiki.bioclipse.net/index.php​?title=Category:Bioclipse_Projects ?
15:39 zarah olass's link is also http://tinyurl.com/68c7ma
15:40 edrin egonw: cool service!
15:40 egonw olass: better... use a query to autogenerate content in title=Projects
15:41 egonw I was about to do that, but decided that I needed to make sure all those currently listed use the Template:Project already
15:41 egonw then we can easily make a table, listing all active projects, etc...
15:41 egonw reducing info duplicaton :)
15:41 olass yes
15:41 olass ok, let us wait some then
15:41 egonw edrin: yeah, I should hook that into the onssolubility manager... feel free to file a bug report
15:42 egonw olass: yes
15:52 * olass is waiting for egonw's bioclipse-semantic-web wiki page
15:52 egonw blog, it will be a blog...
15:52 egonw working on it
15:52 egonw actively, with priority...
15:52 olass blog is just as good
15:54 olass egonw: may I request a nice-looking screenshot from Bioclipse, some features, and also some sort of vision of where the project is heading?
15:54 * olass is also documenting stuff...
15:55 olass egonw: and please link to the bog post on the wiki so it can be accessed from the Category:Projects page
15:55 olass so we have one place to aggregate them all
15:55 egonw http://chem-bla-ics.blogspot.com/2010/01/s​emantic-web-features-in-bioclipse-22.html
15:55 zarah egonw's link is also http://tinyurl.com/ylb64rr
16:06 egonw olass: sorry, no fancy graphics
16:06 egonw not yet...
16:06 olass :(
16:06 egonw it's on my schedule, to get Zest for RDF back online
16:07 olass how about a screenshot from BC with some script that delivers a molecule in 2D?
16:07 olass that is something at least..
16:21 egonw ok, hang on
16:22 egonw now with screenshot
16:22 egonw ok, changing network... brb
16:27 egonw_ joined #bioclipse
16:48 olass egonw_: ping
16:49 egonw_ dinner time
16:49 egonw_ ah... pong
16:53 egonw_ bbl
17:48 egonw_ joined #bioclipse
18:06 egonw joined #bioclipse
18:21 sneumann__ joined #bioclipse
19:41 annzi joined #bioclipse
20:13 edrin left #bioclipse
20:43 sneumann joined #bioclipse

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