Camelia, the Perl 6 bug

IRC log for #bioperl, 2013-01-28

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

All times shown according to UTC.

Time Nick Message
01:13 byte48 joined #bioperl
09:51 jmf joined #bioperl
13:35 carandraug joined #bioperl
13:37 carandraug hey! I have a small fix for the script "Bio-DB-EUtilities/bp_genbank_ref_extractor.pl". Then I noticed the new repo with the separate eutilities tools. Should I make a pull request to which one?
14:06 byte48 joined #bioperl
14:29 deafferret carandraug: what's the URL to the "new" repo?
14:39 carandraug deafferret, Octave Forge?
14:40 carandraug deafferret, svn, or https protocol?
14:41 carandraug deafferret, go to https://sourceforge.net/p/octave/code/11603/tree/ and click on RW, RO or HTTP buttons for the various options
14:49 carandraug deafferret, apologies. I hadn't slept all night. I thought I was another channel
14:50 carandraug *on another channel where people ask me about Octave (we changed repositories recently)
14:50 carandraug deafferret, here's the "new" repo https://github.com/bioperl/Bio-EUtilities
14:51 carandraug I actually ended up making 2 pull requests, one for each
14:56 deafferret carandraug: ah yes, the gh URL. thanks. multitasking, biab
15:36 deafferret carandraug: ok, I'm back. Looking at github now
15:39 deafferret carandraug: hmm.... https://github.com/bioperl/bioperl-live​/tree/master/scripts/Bio-DB-EUtilities
15:41 carandraug deafferret, it's just some small changes on the POD documentation so it's parsed with pod2latex
15:41 carandraug deafferret, but it seems suboptimal to have the same code in two places...
15:42 deafferret carandraug: ya, so you could just patch it in bioperl-live. I don't see it in TWO places. Looks like it should be moved out of bioperl-live to me.
15:42 hash_table joined #bioperl
15:43 deafferret If it were me, I would patch both repos, moving bioperl-live/scripts/Bio-DB-EUtilities/
15:44 deafferret carandraug: I can perlform the move if you want me too and care :)
15:44 deafferret carandraug: better yet, you could ;)
15:45 deafferret github.com/bioperl/Bio-EUtilities also needs a README, etc.
15:45 carandraug deafferret, what I'm patching it's not in the Bio-EUtilities repo. But seems like it should. But shouldn't the stuff in bioperl-live be removed then? bp_einfo for example
15:45 deafferret carandraug: yes, I think we're both saying the same thing. Things should be moved into the Bio-E repo.
15:46 carandraug deafferret, should we check that that the rest of bioperl-live does not use those modules? And it does. then we have a circular dependency
15:49 deafferret $ ack --perl -l Bio::DB::EUtilities
15:49 deafferret Bio/DB/GenericWebAgent.pm
15:49 deafferret scripts/Bio-DB-EUtilities/bp_einfo.pl
15:49 deafferret scripts/Bio-DB-EUtilities/b​p_genbank_ref_extractor.pl
15:49 deafferret so there's only 3 -- if WebAgent could be separated then you're home free
15:50 deafferret whoever split the repo should have done all that to begin with -shrug-.  presumably just overlooked / ran out of time
15:50 carandraug deafferret, what's the actual status of the project? I've heard many times about ending the monolithic release. How did that end?
15:51 carandraug deafferret, we done it in Octave Forge some years ago and it helped a lot
15:51 deafferret carandraug: well, it never ends. :)  you'd be helping with the patches we've been talking about :)
15:51 deafferret I'm not familiar with Octave Forge
15:52 carandraug deafferret, but is anyone working on it?
15:53 deafferret carandraug: unknown. you'd have to check the git logs / mailing list and ask around
15:53 carandraug deafferret, Octave Forge is a project for Octave packages. Octave is a language like matlab, and octave forge is its toolboxes. We had a monolithic release, all packages had to be ready at the same time and released together. Some were unmaintained so we never did releases. Now they are all separate sub-projects under the same name.
15:54 carandraug bioperl last stable release was almost 2 years ago...
15:54 deafferret ah, cool :)
15:55 deafferret carandraug: ya. I just use github (not CPAN) for BioPerl
15:56 deafferret carandraug: that's another case of someone throwing a ton of time at it to declare a SHA stable and push it
15:57 carandraug deafferret, I use the debian repos
15:57 deafferret carandraug: oh, wow. that would be crazy-super-old then :)
15:58 carandraug deafferret, yeah! The last stable release...
16:00 carandraug deafferret, so I'll make a pull request removing those files. Should I make note somewhere? A NEWS file that I'm not seeing?
16:01 deafferret carandraug: well, do 2 pull requests -- one removing the files from bioperl-live, one adding those files to the other repo
16:01 deafferret carandraug: there's a Changes file, but I wouldn't worry about it for what you're doing
16:02 deafferret /Changes
16:04 carandraug deafferret, is there a way in git to move a file to another repo and keep its history? If there has been no changes on the "new" repo since the fork, maybe would be better to make it anew with all the changes of the all files
16:05 deafferret carandraug: nope. You can't keep history. Don't worry about it.
16:06 deafferret carandraug: in your commit message, say that you're moving FROM REPO blah blah blah so that if someone cares they can go look at history over there
16:06 carandraug deafferret, I think it's possible because there's history in common between the two. But probably not worth the trouble finding how to do it
16:08 deafferret carandraug: ya, I researched that a couple years ago. I wouldn't worry about it. :)
16:08 deafferret carandraug: the history, as it exists now, is public, and should not be changed. so ... just slap the current files in there.
16:09 deafferret carandraug: git lets you do all sorts of evil, but you shouldn't :)
16:12 carandraug deafferret, done. I'll make a pull request now
16:12 carandraug deafferret, can you please accept the ones I made before? They won't apply otherwise. Or you can pull them from my fork
16:13 carandraug deafferret, how do you prefer I do it?
16:16 github [bioperl-live] jhannah pushed 3 new commits to master: https://github.com/bioperl/bioperl-liv​e/compare/9f20929a902e...f242511e5957
16:16 github bioperl-live/master 9bc9b02 Carnë Draug: Fixing POD documentation. Empty line before =cut to avoid problems with pod2X tools
16:16 github bioperl-live/master d6dd39f Carnë Draug: finish moving the Bio-DB-EUtilities to the Bio-EUtilities repository
16:16 github bioperl-live/master f242511 Jay Hannah: Merge pull request #50 from carandraug/master...
16:19 carandraug deafferret, Thank you. There's also this one https://github.com/bioperl/Bio-EUtilities/pull/2 which makes the rest of the move
16:20 deafferret carandraug: merged. apparently that repo doesn't update IRC
16:22 carandraug deafferret, cool. One thing. I made a script some time ago that I named pretty_fasta. It basically reprints a sequence in the fasta format but with whitespace to make it easier to count.
16:23 carandraug deafferret, I have thought of making it an option of Bio::SeqIO::fasta. Should I bother?
16:24 deafferret carandraug: in Perl? Using Bio::SeqIO?
16:24 deafferret carandraug: ah, cool.
16:24 carandraug deafferret, or would just add it to the scripts section be better?
16:24 deafferret carandraug: I'd say YES! If it's helpful I'd make it a feature of SeqIO :)
16:24 deafferret carandraug: if it's helpful to you then it'll probably be helpful to others :)
16:25 deafferret in SeqIO::* is defaintely the place to put it -- there are so many formats already -- adding another one is always appreciated!
16:25 deafferret add a test file too (.t) that exercises your format (in and out)
16:26 carandraug deafferret, what I have at the moment is https://github.com/af-lab/scripts/blob​/master/bioinformatics/pretty_fasta.pl
16:27 carandraug deafferret, the thing, I'm not sure what's the most helpful method for it. A friend in the lab asked, so I wrote her a script. But if it was an option of SeqIO, she doesn't program so would be useless for her
16:28 carandraug deafferret, on the other hand, making it into SeqIO, means the script could then be written in much less lines
16:28 deafferret yes, exactly :)
16:29 deafferret my $out = Bio::SeqIO->new(-format => 'FastaPretty' ...    :)
16:29 deafferret FastaWithLine perhaps
16:32 carandraug deafferret, I was thinking more of still naming the format fasta, but allow for an optional length values
16:35 carandraug -format => 'fasta', -blklength => 10, -linelength => 10
16:36 carandraug there's a hardcoded $WIDTH on SeqIO::fasta.pm
16:41 deafferret carandraug: yes, I like your idea much better :)
16:46 deafferret carandraug: i'll make you collab on github when your next patch looks good to me
18:05 looper joined #bioperl
21:14 hash_table joined #bioperl
22:40 byte48 joined #bioperl
22:43 hash_table joined #bioperl

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