Camelia, the Perl 6 bug

IRC log for #bioperl, 2010-05-07

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

All times shown according to UTC.

Time Nick Message
03:48 cawiss joined #bioperl
04:27 luisrodr joined #bioperl
05:43 bag joined #bioperl
06:33 melic joined #bioperl
07:37 bag joined #bioperl
09:35 bag joined #bioperl
11:18 Jun joined #bioperl
14:25 cawiss joined #bioperl
15:53 melic joined #bioperl
16:51 spekki01 joined #bioperl
17:02 deafferret ..o.
17:03 spekki01 So im having some trouble with using esearch, i have an array of 724 accession numbers that im putting into esearch http://codepad.org/WSuJaK1d the error i get is http://codepad.org/p1tkzkWN. I've tried the same code with an array of 529 accession numbers and it worked so im just wondering is there a way to do the esearch in chunks so this error doesn't come up?
17:04 deafferret woo. sounds like a doozy. give me a few minutes
17:04 spekki01 thanks :)
17:04 deafferret and/or try to the bioperl-l mailing list in case someone has seen that before already  ?
17:04 spekki01 i havn't used the mailing list, let me go start looking
17:05 deafferret and i'd try it in bioperl-live if you haven't already
17:05 deafferret in case someone has already fixed it
17:05 spekki01 is that 1.6.1?
17:06 deafferret 1.6.1 is 9 months old. bioperl-live is straight out of version control
17:06 deafferret i'm a programmer at heart so I'm always in bioperl-live, and sometimes its fixed in there already
17:06 spekki01 ok let me upgrade
17:07 deafferret note that bioperl-live may be rough around the edges, so you have to be comfortable with that
17:07 spekki01 oh do you have a working link, the one on http://www.bioperl.org/wiki/Getting_BioPerl under the subversion section that says bioperl-live is broken.
17:07 deafferret the glory and price of life on the bleeding edge
17:08 deafferret ya, it's moving to github. gimme a sec
17:08 spekki01 np
17:08 deafferret http://github.com/bioperl   bioperl-live or bioperl-test
17:09 spekki01 thanks
17:16 spekki01 sry to seem like a noob but on the http://github.com/bioperl/bioperl-live what do i download?
17:22 deafferret git clone http://github.com/bioperl/bioperl-live.git
17:24 spekki01 ?
17:24 deafferret :)
17:24 deafferret git is a version control system. are you on linux/
17:24 deafferret ?
17:24 spekki01 yeah, red hat 4
17:24 deafferret if you type 'git' what happens?
17:25 spekki01 type it where? in a terminal?
17:26 deafferret yup
17:27 deafferret at a shell / command prompt / X-11 window / whatever they call it
17:28 spekki01 well on my current terminal the red hat 4 its command not found, but where i do all my testing ubuntu 8.04 i get The program 'git' is currently not installed.  You can install it by typing:
17:28 spekki01 sudo apt-get install git-core
17:29 spekki01 im gonna install it and see what i get
17:29 deafferret yup. then   git clone http://github.com/bioperl/bioperl-live.git
17:29 deafferret and boom! you've got all of bioperl, all commits for all time, sitting on your computer  :)
17:30 spekki01 cool
17:30 deafferret even *my* commits, you brave soul  :)
17:31 deafferret lucky for you I've barely committed anything
17:33 spekki01 hmm i think i have to configure something im getting git clone http://github.com/bioperl/bioperl-live.git
17:33 spekki01 ops mt
17:33 spekki01 im getting pboutet@ubuntu:~/vkcdbMay4$ git clone http://github.com/bioperl/bioperl-live.git
17:33 spekki01 Initialized empty Git repository in /home/pboutet/vkcdbMay4/bioperl-live/.git/
17:33 spekki01 /usr/bin/git-clone: line 59: curl: command not found
17:34 spekki01 sry for the paste
17:34 deafferret apt-get install curl
17:34 deafferret weird that apt-get didn't do that for you
17:34 deafferret mind you I'm not saying bioperl-live will fix your problem. I don't know if it is relevant or not
17:35 spekki01 no worries its worth a try
17:35 spekki01 and im searching the mailing list still
17:35 spekki01 so hopefully somethign comes up
17:54 spekki01 just double checking i have it installed properly, when i do perl -MBio::Root::Version -e 'print $Bio::Root::Version::VERSION,"\n"'    I get 1.0069 is this the right version number for live?
17:55 deafferret lemme look
17:56 deafferret yup
17:56 deafferret i guess that number is in "pre-1.7" mode
17:57 deafferret as in prepping for a 1.7 release
17:57 deafferret anyhoo, rest assured if you're using bioperl-live now you're on the bleeding edge  :)
18:19 wnprc joined #bioperl
18:46 ende joined #bioperl
18:46 ende Hello
18:47 * deafferret shakes ende
18:47 ende gah!
18:47 deafferret the traditional greeting of my people
18:48 deafferret don't be xenophobic
18:48 spekki01 well i just got back from lunch and my test didn't work its still giving me the error http://codepad.org/EVUm8a6q with that many accession numbers going into esearch. So im wondering is there a way to submit in chunks or something?
18:48 deafferret sudo ende debug that
18:50 ende Anyone going to Galaxy Dev conference?
18:55 rbuels ende: i kind of wish i was
19:05 spekki01 going off of http://www.bioperl.org/wiki/HO​WTO:EUtilities_Cookbook#epost how do i modify this to post in batches of 500?
19:39 ende how many ids do you have?
19:43 ende spekki01 ?
19:43 spekki01 oh hi sry was afk
19:43 spekki01 728
19:46 spekki01 reason i ask about posting in chunks of 500 with epost is that i want to try and apply the same concept to esearch to get around the error of my request URI being to large
19:50 ende yeah
19:50 ende well
19:50 ende assuming you have a giant array to beign with
19:50 ende I would simply break that array into subarrays of 500 each
19:54 spekki01 ok, then im assuming i just do back to back esearch calls with the different arrays. But how is the use history parameter handled. because when i follow the esearch up with a efetch call, will it know about the ui's posted to the history server from the seperate arrays
19:58 spekki01 also another side question how do you rollback from bioperl live back to 1.6.1? do i just reinstall 1.6.1 or do i have to uninstall bioperl live?
19:59 ende on the 2nd question..
19:59 ende generally I keep a stable version installed somewhere in my $PERL5LIB, and I keep a copy of bioperl-live handy else where but only invoke with with a uselib
19:59 ende in case I need to roll back
20:00 ende basically, look at your $PERL5LIb, it will tell you the order that perl will load your libraries
20:00 ende just make sure the version you want to use is the first one loaded
20:00 ende or rather in a location where it'd be the first one loaded
20:00 spekki01 ok
20:38 spekki01 im still a bit confused im atempting to edit the $PERL5LIB and see where bioperl is in it and im not seeing it, what's the default install directory for bioperl?
20:46 deafferret anyone played with this?   http://mbf.codeplex.com/
20:46 deafferret spekki01: Linux distributions all bend perl installation paths to whatever schema makes sense to them
20:47 spekki01 k, ill see where ubuntu puts it
20:47 deafferret I'm not sure I've ever actually installed bioperl, though I've been using it for years
20:48 deafferret PERL5LIB manipulates your @INC, which is what all things Perl use to know what libraries paths to look it to load stuff
20:48 deafferret so if I
20:48 spekki01 yeah that's what im attempting to edit, take out the path where bioperl live went
20:48 deafferret so if I had 30 different versions of the Foo module, I could use PERL5LIB to point to whichever one I wanted at a given time
20:48 spekki01 so it will load bioperl-1.6.1
20:49 deafferret unset PERL5LIB
20:49 spekki01 @INC:
20:49 spekki01 /etc/perl
20:49 spekki01 /usr/local/lib/perl/5.8.8
20:49 spekki01 /usr/local/share/perl/5.8.8 **bio
20:49 spekki01 /usr/lib/perl5
20:49 spekki01 /usr/share/perl5
20:49 spekki01 /usr/lib/perl/5.8
20:49 spekki01 /usr/share/perl/5.8
20:49 spekki01 /usr/local/lib/site_perl
20:49 spekki01 sry for the paste
20:49 deafferret Paste to http://codepad.org/
20:49 deafferret or wherever
20:50 deafferret so ... sorry, what's your question?
20:50 deafferret have you installed 1.6.1 ?  and it's sitting in /usr/local/share/perl/5.8.8 now?
20:50 spekki01 i've installed 1.6.1 and bioperl-live
20:51 spekki01 and my scripts keep using live which i dont want them to do anymore
20:51 spekki01 so im trying to figure out where it is installed so i can remove that path from $PERL5LIB
20:52 spekki01 from browsing around i thin 1.6.1 is in /usr/local/share/perl/5.8.8
20:52 spekki01 but im having trouble locating live
20:53 spekki01 unless im really just not understanding this, which could entierly be the case as well :(
21:02 deafferret oh, you *installed* bioperl-live by running Build.PL ?
21:02 spekki01 yep
21:02 deafferret gotcha. then you'll need to re-install 1.6.1.
21:02 deafferret there's really no going back
21:02 spekki01 ok that works for me, thanks so much
21:03 deafferret if you want to try out bioperl-live temporarily you can just   export PERL5LIB=/home/jhannah/src/bioperl-live
21:03 deafferret and then you'll see when you do    perldoc -l Bio::Seq
21:03 deafferret that the Bio::Seq it's using is    /home/jhannah/src/bioperl-live/Bio/Seq.pm
21:03 deafferret then you unset PERL5LIB
21:04 deafferret and you'rve back to    /usr/local/share/perl/5.8.8/Bio/Seq.pm   (1.6.1)
21:04 spekki01 cool
21:05 deafferret so you tried bioperl-live and that didn't change anything?
21:05 deafferret so you're breaking your requests into 500 or fewer blocks?
21:06 deafferret sorry, only have paying attention
21:06 deafferret damn mortgage
21:07 spekki01 yeah, live didn't fix the error with esearch, but using live is making my script behave a bit different so im gonna switch back to 1.6.1 and rewrite it to use blocks of 500
21:08 deafferret oh... well, when 1.7 comes out you might find that the "a bit different" is now the new "right way" ...  so you might code your stuff to bioperl-live   ... your call
21:08 deafferret the river of time flows ever forward
21:09 deafferret but backwards compat should be maintained, so if it's serious that would surprise me
21:09 spekki01 when is 1.7 expected release? or your best guess.
21:09 deafferret no clue. you might search the mailing list
21:09 deafferret or ask the mailing list  :)
21:09 spekki01 kk
21:09 deafferret I don't really do email either
21:10 deafferret :)
21:10 deafferret email always feels like this infinite backlog of issues I haven't addressed  :)
21:11 spekki01 lol
21:11 spekki01 it sits there and keeps reminding you
21:12 deafferret ya. sigh. i have 2 jobs and way too many hobbies
21:12 deafferret / interests
21:14 spekki01 is one of the jobs working on bioperl?
21:14 spekki01 just curious how you got into it if it is.
21:15 deafferret my backstory:  :)   http://www.bioperl.org/wiki/User:Jhannah
21:16 spekki01 cool
21:16 deafferret ya, I AM pretty awesome. I agree.
21:16 spekki01 lol
21:16 deafferret :)
21:17 * deafferret is an agreeable fellow
21:18 deafferret wow. been doing this stuff for 5 years now
21:18 * deafferret is not getting any younger
21:42 * deafferret sighs at the latest bioperl-l post
21:51 deafferret pyrmiwhateveridine++
21:51 deafferret is command line blast "legacy blast" now?
22:08 spekki01 k im having trouble figuring out how to submit accessoin numbers in batches heres what i have so far http://codepad.org/C7BE9ISn but it's not giving me the correct number of id's when i call get_count; near the end. I have a feeling i should not be creating a new eutil esearch each time but im not sure.
22:08 spekki01 any tips would be helpful
22:13 deafferret hm
22:15 deafferret maybe not all your @genbank_na_acc are valid?
22:15 deafferret try batches of 5
22:16 spekki01 k
22:18 spekki01 k
23:39 dnewkirk Does Bioperl have a function for accessing WebQTL? Specifically with a "let me spam the server with queries" flag?
23:51 * deafferret doesn't know what WebQTL is
23:51 deafferret &

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