Camelia, the Perl 6 bug

IRC log for #bioperl, 2010-08-06

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

All times shown according to UTC.

Time Nick Message
01:34 dnewkirk joined #bioperl
04:17 Lynx_ joined #bioperl
05:15 NiiHiiL joined #bioperl
05:49 bag_ joined #bioperl
05:53 brandi joined #bioperl
05:53 brandi left #bioperl
06:24 brandi1 joined #bioperl
07:08 brandi1 left #bioperl
11:26 dbolser hi
11:27 dbolser does bp_seqfeature_delete work for anyone?
11:27 dbolser I'm just debugging it now
11:27 dbolser the SQL has a weird : that I can't track down for the life of me
11:27 dbolser SELECT f.id,f.object,f.typeid,f.se​qid,f.start,f.end,f.strand
11:27 dbolser FROM feature as f, typelist AS tl
11:27 dbolser WHERE (   tl.id = f.typeid
11:27 dbolser AND     (tl.tag = 'break point:mapping:')
11:27 dbolser )
11:29 dbolser print "foo: '$type'\n";
11:29 dbolser foo: 'break point:mapping'
11:29 dbolser if (ref $type && $type->isa('Bio::DB::GFF::Typename')) { print "wtfoo: '$type'\n";
11:30 dbolser ...
11:30 dbolser wtfoo: 'break point:mapping'
11:30 dbolser OMG?!
11:31 * dbolser resets his prejudice
11:32 dbolser joy
11:32 dbolser $primary_tag = $type->method; # = 'break point:mapping'
11:32 dbolser $source_tag  = $type->source; # = ''
11:33 dbolser is that a bug or what?
11:33 * dbolser prods rbuels with an empty six-pack of redbull
11:33 dbolser ;-)
11:34 rbuels unh wht
11:34 dbolser nm, just the usual debugging bitch rant
11:35 rbuels dbolser: perl -d
11:35 dbolser heh
11:35 dbolser yeah
11:35 rbuels dbolser: $DB::single = 1; is a magic magicness.
11:35 rbuels dbolser: man perldebug
11:35 dbolser bp -kill all GFF refactoring intentions
11:35 rbuels dbolser: also, $SIG{__DIE__} = \&Carp::confess is another chestnut
11:36 dbolser shouldn't you be asleep?
11:36 rbuels almost done.
11:36 dbolser ;-)
11:36 rbuels writing a rant about GMOD and databases all endlessly reprogramming the same stupid crap badly
11:36 rbuels for the 'future work' section
11:36 rbuels well, a 3-sentence rant.
11:36 dbolser cool
11:37 dbolser now I have to work out if this Bio::DB::GFF::Typename is being constructed wrong, or is returning wrong
11:38 dbolser $method ||= '';
11:38 dbolser can't you just write, ... meh.
11:39 * rbuels is bludgeoned by an irritated bloodhound
11:39 rbuels she is like 'wtf are you doing it is nearly 5am'
11:39 rbuels 'and you are on my couch!'
11:39 dbolser lol
11:39 dbolser sounds like our cat at teh weekend
11:40 dbolser wtf are you doing in my house?
11:40 rbuels the dog at this point is groaning at me approximately twice a minute
11:40 dbolser at least he doesn't jump on the kbd
11:41 * dbolser assumes
11:44 dbolser seems the constructor is being called correctly
11:48 dbolser got it
11:48 * dbolser gasps with bug-found relief
11:48 dbolser if ($source eq '' && $method =~ /^([\w-]+):([\w-]*)$/) {
11:48 dbolser er... what if my source or method has a space in it?
11:48 dbolser YOUR BUGGED!
11:52 dbolser Undefined subroutine &Bio::DB::GFF::Typename::new called at ... why?
13:23 luisrodr joined #bioperl
13:51 luisrodr joined #bioperl
14:26 luisrodr joined #bioperl
14:36 dbolser OK, I done debugging on my branch, testing, and committing
14:36 dbolser who should I add to the pull request?
14:38 * dbolser includes kblin because he is here...
14:38 * dbolser includes hyphaltip because he wrote a comment on one of my commits once...
14:39 kblin you prolly want someone from the core team
14:39 dbolser genehack?
14:39 dbolser I think hyphaltip is that core person
14:40 dbolser kblin: is sending a pull request a good idea?
14:40 kblin yeah
14:40 dbolser here goes...
14:40 dbolser should I have rebased?
14:42 dbolser git pull --rebase origin/master
14:42 dbolser fatal: 'origin/master': unable to chdir or not a git archive
14:42 dbolser fatal: The remote end hung up unexpectedly
14:43 dbolser git rebase origin/mastergit rebase origin/master
14:43 dbolser Interactive rebase already started
14:43 dbolser doh
15:09 genehack dbolser: what's up?
15:11 * genehack has now caught up on the scrollback...
15:11 genehack git rebase --abort
15:12 genehack dbolser: ^^
15:12 genehack and then i'm not clear on what you're trying to do.
15:50 spekki0 so I'm getting an error when using Bio::DB::SeqVersion; here is the error it's at the bottom of the page http://codepad.org/uuL9O7vI here is the code im running http://codepad.org/0XeBodqv I'm just wondering if it's something im doing or problems on ncbi's end. I have verbose =>2 set for my bioperl objects to catch warnings that thery produce so the code spams a lot.
16:02 * deafferret freebases
16:02 deafferret err.. rebases
16:03 brandi joined #bioperl
16:08 luisrodr_ joined #bioperl
16:08 CIA-93 bioperl-live: Chris Fields master * r5c18e28 / Bio/Range.pm : tr/// lists must be same length, switch to s/// - http://bit.ly/9WIgtf
16:09 spekki0 hmmm seems to be one record in particular that it keeps breaking on
16:10 deafferret o/`  i sound like a broken record, record, record / a broken record, record, record / all the times i said i was gonna change / im'a sound like a broken record / a broken record, record, record / every time i lie, your ears bleed with pain / form the sounds of a broken record   o/`
16:13 pyrimidine joined #bioperl
16:15 spekki0 lol
16:22 pyrimidine spekki0: not sures what's going on there, but I would chalk it up to NCBI flakiness
16:22 pyrimidine *sure
16:23 spekki0 yeah, im with you on that
16:24 spekki0 i'm just gonna have my script scip over the records that give me trouble i guess
16:24 spekki0 skip*
16:24 pyrimidine catch it in the eval and resubmit (redo)
16:24 spekki0 I do, 5 times and each time it craps out
16:24 spekki0 :(
16:26 pyrimidine odd
16:26 pyrimidine yes, would just mark them then
16:55 Lynx_ joined #bioperl
17:01 Lynx_ Good Morning! I keep getting this error: "MSG: In sequence zoox2_c7354 residue count gives end value 461" when I try to get an alignment from a Blast hit. It's caused somewhere in the LocateableSeq module, but I don't quite get why...
17:03 deafferret Lynx_: show us the error, code    gist.github.com
17:08 Lynx_ http://gist.github.com/511625
17:09 deafferret hmm... can you also provide input so I can try to recreate?
17:10 Lynx_ I just noticed I only get the error on windows, it seems, not on my linux box
17:10 deafferret oh. i no know no windows  :)
17:11 Lynx_ heh
17:12 Lynx_ weird. oh well, it's ok then.
17:13 Lynx_ I got another one though: Depending on the blast file, Bio::Index::Blast either just tells me there is an index file already, as it's supposed to, or gives me "MSG: overwriting a current value stored for seqx" for all the sequences in the blast. Even if I just run the same script twice without changing anything in the blast files.
17:17 pyrimidine sounds like you have two versions of bioperl running
17:17 pyrimidine one on windows box, diff one on linux
17:21 Lynx_ I just updated them both yesterday, so there should not be any difference
17:21 deafferret hmm... I haven't used Bio::Index::Blast. If it works like Bio::Index::Fast and GenBank then I'd need to see your code to try to reproduce your results
17:22 ank joined #bioperl
17:22 pyrimidine afk &
17:50 deafferret there are places without keyboards?   ...  that sucks
17:59 Lynx_ deafferret: now I have something to test: http://gist.github.com/511625 and input: http://scsys.co.uk:8002/47691​?tx=on&submit=Format+it!
18:00 Lynx_ Blast::Index always returns the result for the first sequence in that script, no matter what you give fetch_report
18:02 deafferret what's a good input filename?   input.blastout ?
18:03 Lynx_ yeah, that's ok
18:15 deafferret http://gist.github.com/511719
18:16 deafferret http://github.com/jhannah/​sandbox/tree/master/Lynx_/
18:24 Lynx_ I think I found the bug. If I remember correctly I found it already earlier this year...
18:25 deafferret shoulda fixed it then  :)
18:25 deafferret patches welcome!!  :)
18:26 Lynx_ Indeed, I found that bug, reported it in the mailing list, and fixed it in my lib. but through the update I did yesterday I brought it back...
18:26 Lynx_ Yeah, I thought it had been fixed
18:26 deafferret doh! this time patch github and it's fixed in bioperl-live  :)
18:27 deafferret email is full of lies. you can only trust IRC and git.  :o)
18:28 Lynx_ http://bioperl.org/pipermail/bi​operl-l/2010-March/032473.html
18:28 deafferret lies! all lies!  :)
18:29 Lynx_ hehe
18:29 Lynx_ Chris must have forgotten about it.
18:29 Lynx_ Oh well, inserting one # will work for me now at least.
18:30 deafferret if you can fix the bug, whatever it is, we can get it into the repo today
18:30 deafferret (looks ok from my box, per my gists)
18:31 Lynx_ No, Chris wrote the fix is not so easy, as it breaks parsing of RPS BLAST I guess. It's in his follow up mail.
18:32 deafferret oh. bummer.
18:37 Lynx_ Noone ever seems to use Index::Blast...
18:52 pyrimidine I use Index::Blast
18:52 pyrimidine the indexing is broken with BLAST+ output only
18:53 pyrimidine IIRC
18:55 pyrimidine Lynx_: you work with Heikki?
19:04 Lynx_ pyrimidine: no, not directly
19:04 Lynx_ I'm in marine bio, he's in comp biology
19:04 pyrimidine ah
19:04 rbuels Lynx_: do you use CUGI's marine genomic site/
19:05 Lynx_ pyrimidine: yes, it's blast+ only
19:05 Lynx_ rbuels: no, whats that?
19:05 rbuels Lynx_: http://www.marinegenomics.org/
19:05 rbuels (which is based on GMOD tripal)
19:06 deafferret 200 marine sergeants spit into tubes
19:06 rbuels a series of them?
19:06 pyrimidine rbuels: please tell us how you feel about tripal.  ;-)
19:06 * deafferret ties rbuels' tubes
19:06 pyrimidine OHNOES
19:06 rbuels i have no feelings.
19:07 Lynx_ rbuels: looks nice! even has two corals, that's what we work on
19:07 pyrimidine rbuels is an automaton
19:11 pyrimidine :(
19:11 pyrimidine I think deafferret scared him away
19:15 Lynx_ joined #bioperl
19:46 deafferret :)
19:47 deafferret boogity boogity!
20:10 rbuels eh what
20:10 rbuels i gots meetings!  i gots work!
20:11 rbuels can't sit here and jaw all day.  lan'sakes
20:12 * deafferret gots the vapors
20:13 deafferret lord, child! you be runnin' right along now!
20:27 brandi joined #bioperl
20:27 brandi left #bioperl
21:03 bag_ joined #bioperl
21:06 melic joined #bioperl
22:20 pyrimidine left #bioperl

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