Camelia, the Perl 6 bug

IRC log for #bioperl, 2011-04-13

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

All times shown according to UTC.

Time Nick Message
00:01 carandraug left #bioperl
00:01 carandraug joined #bioperl
00:03 carandraug left #bioperl
00:50 zenman_ left #bioperl
02:18 scottcain joined #bioperl
02:49 scottcain left #bioperl
03:04 scottcain joined #bioperl
03:39 scottcain left #bioperl
04:16 bag_ left #bioperl
09:54 dbolser is someone developing on master?
09:54 dbolser SeqIO is broken and becoup tests fail
09:55 dbolser Author: hyphaltip "small doc change"
09:56 dbolser ahahahahahah
09:56 dbolser sorry, that sounds wrong
09:56 dbolser more like ahhhhhhhhhhhhhhhhhhhhhhhhhh​hhhhhhhhhhhhaaaaaaaaaaaaaa
09:59 * dbolser doesn't know enough git to probe the history of this file
10:00 dbolser does bp have the beer rule?
10:02 dbolser https://gist.github.com/917298
10:02 dbolser whoever commits broken code to the deveopment branch has to buy all devs a beer
10:04 dbolser /topic Be patient!  Sometimes it takes a few minutes for people to notice your question! | http://bioperl.org/wiki/IRC#Getting_help | http://www.bioperl.org/wiki/Using_Git | nopaste to gist.github.com | 16:15 <@rbuels> sometimes, i just want to kiss myself on the lips
10:04 * dbolser lulz
10:05 dbolser SeqIO still fails test in master no less
10:08 Drin joined #bioperl
10:20 kai dbolser: do you know git bisect?
10:35 dbolser no
10:35 dbolser was that a rehtorical question?
10:35 * dbolser goes to man
10:42 dbolser kai: since you're here... I made a big change to a file and commited, now I find I introduced a tiny bug, do I do 'git reset <commit>' / bugfix / git commit ?
10:42 dbolser git reset --soft
10:43 kai did you push that stuff already?
10:43 dbolser into my fork, yes
10:43 kai ok, reset will futz with your history
10:44 kai so you might as well just do your changes, git add them and then do git commit --amend
10:44 * dbolser tries to conjure the explitive to reflect the mix of emotions caused by this statement
10:44 dbolser --amend... ty
10:44 kai which changes the last commit without you having to go through the reset and remembering your commit message
10:45 kai if you want to keep the history intact, it's probably better to just do another commit
10:45 dbolser You should understand the implications of rewriting history if you amend a commit that has already been published. # I certainly don't
10:46 dbolser ok, I'll just commit again with a 'bugfix' message.
10:51 dbolser here is at least one bug introducing commit: https://github.com/dbolser/bioperl-live/comm​it/b6540bac2429e2e9739c11ef64315520f274408d
10:51 dbolser (I didn't use bisect, I just used github)
10:53 kai dbolser: for your github fork I don't think it's going to be an issue
10:53 kai I rewrite the history on my work-in-progress trees occasionally
10:54 kai mainly by using git rebase on them to keep my changes on the top of the list
10:54 kai on a tree like that, git commit --amend is no problem whatsoever
10:58 dbolser kai: I'm submitting pull requests against bp though
10:58 dbolser (just now for example)
10:58 dbolser don't I need my fork history to be clean so that it gets pulled cleanly?
10:58 dbolser g2g, divisional seminar
11:00 kai dbolser: dunno, I don't know the github way to do this
11:01 kai if the patches are applied using cherry-pick or git format-patch; git am; then your local history doesn't matter
11:52 takadonet left #bioperl
12:24 takadonet joined #bioperl
12:28 ManWithNoName joined #bioperl
12:28 ManWithNoName left #bioperl
12:41 takadonet left #bioperl
12:43 takadonet joined #bioperl
12:44 zenman_ joined #bioperl
12:48 * dbolser shuders ... cherry-pick..
12:48 dbolser spells doom in my recolection
12:59 zenman_ left #bioperl
13:23 kai dbolser: why?
13:34 scottcain joined #bioperl
13:45 scottcain left #bioperl
13:51 scottcain joined #bioperl
15:12 pyrimidine dbolser: huh?
15:42 CIA-119 bioperl-live: Dan Bolser master * r1f714ef / Bio/DB/SeqFeature/Store/DBI/mysql.pm : fixing a documentation bug - http://bit.ly/ijcynw
15:42 CIA-119 bioperl-live: Dan Bolser master * rbbaca27 / scripts/Bio-DB-SeqFeature-S​tore/bp_seqfeature_load.PLS :
15:42 CIA-119 bioperl-live: added 'usage' section using pod2usage
15:42 CIA-119 bioperl-live: I wanted to experiment with pod2usage (and pod for that matter), but I ended up
15:42 CIA-119 bioperl-live: doing a lot more than was really necessary. Hopefully this can serve as a good
15:42 CIA-119 bioperl-live: reference for the other bp scripts ;-) One thing I didn't get round to is
15:42 CIA-119 bioperl-live: inserting links to the relevant CPAN pages - http://bit.ly/fhTJ2i
15:42 CIA-119 bioperl-live: Dan Bolser master * rbc97ee4 / (6 files in 2 dirs): changing directory name to be consistent with the others - http://bit.ly/gK5qnT
15:42 CIA-119 bioperl-live: Dan Bolser master * rc2e8033 / scripts/Bio-DB-SeqFeature-S​tore/bp_seqfeature_load.PLS : minor bugfix on my previous commit (why did I push\?) - http://bit.ly/elMZ49
15:42 CIA-119 bioperl-live: Dan Bolser master * rfaff02b / MANIFEST : changing manifest to reflect my file renaming - http://bit.ly/egbfns
15:52 scottcain_ joined #bioperl
15:52 scottcain left #bioperl
15:52 scottcain_ is now known as scottcain
16:15 dbolser ignore me, my git foo is weak
16:15 rbuels dbolser: getting stronger though.
16:18 dbolser if only this page were written as clearly as git docs: http://search.cpan.org/~lds/Bio-B​igFile-1.01/lib/Bio/DB/BigWig.pm
16:18 dbolser that link was weak
16:18 dbolser it's confusing though
16:18 dbolser and dieing when I request a feature out of range
16:18 dbolser Can't call method "next_seq" on an undefined value at /homes/dbolser/perl5/lib/perl5/i386-​linux-thread-multi/Bio/DB/BigWig.pm line 658, <> line 264.
16:19 rbuels sounds like it needs MOAR PATCHES
16:19 dbolser I guess I'll have to stick it in eval and shove it in my debug stack
16:19 dbolser oh I'm good at them... achem
16:20 dbolser it's like, the summary type on a region method is given in the synopsis, then not discussed again
16:20 dbolser (summary type is discussed, but it doesn't respect the region)
16:20 dbolser ((when I try it that is))
16:20 dbolser and for some reason, there are 4129 bins in this region, which I know to be one bin...
16:23 pyrimidine dbolser: re git experience: wax on, wax off
16:24 dbolser pyrimidine: will you slap chris for me?
16:24 * pyrimidine slaps self
16:24 dbolser oh wait... I mean.. whats his name
16:24 dbolser hyphaltip
16:24 * pyrimidine slaps Jason
16:24 dbolser right
16:24 dbolser he checked bugs into master
16:25 pyrimidine didn't notice, but I would have seen it today
16:25 dbolser he made three commits, I commented on one line of one of them
16:26 pyrimidine he won't see the comment unless you @hyphaltip him
16:26 dbolser oic
16:26 pyrimidine at least, I don't think he will
16:26 dbolser well, pull and build test, you'll see everything fail
16:26 pyrimidine dbolser: https://github.com/blog/821-men​tion-somebody-they-re-notified
16:27 dbolser ty
16:28 dbolser the wig stuff is pretty powerfull
16:28 pyrimidine yes, it is
16:29 pyrimidine you mean bigwig/bigbed I assume
16:29 dbolser yes
16:31 pyrimidine dbolser: I'm seeing fails on master branch, actually
16:31 * dbolser attempts a dedpan gaze
16:32 dbolser I feel bad, because one of the commits from Jason is my pull request that you and rbuels were discussing
16:32 dbolser (I've yet to write tests and get back to you)
16:33 dbolser I've subsequently submitted an entirly different pull request
16:33 rbuels dbolser: make sure you close any that are no longer relevant.
16:34 pyrimidine current master branch error: MSG: syntax error at Bio/SeqIO.pm line 373, near "};"
16:34 dbolser I need to get to them, because they are relevant
16:34 dbolser (I guess the one you're talking about)
16:34 dbolser (last comment, rbuels: is this still relevant)
16:34 * dbolser attempts a dedpan gaze in the direction of pyrimidine
16:35 rbuels (i think it's speled deadpan)
16:35 dbolser you no me
16:35 * pyrimidine uses a mirror, kills dbolser, uses dbolser's head to kill Kraken
16:35 * rbuels casts level 12 Deadpan Gaze on dbolser
16:35 * rbuels chuckles
16:36 pyrimidine :)
16:36 pyrimidine we're both full of the snark today
16:36 * dbolser upgrades to steely
16:36 pyrimidine blue steel?
16:37 rbuels ooo ho hoho
16:37 dbolser plz enact the beer rule
16:37 * rbuels has his Magnum almost ready
16:39 * pyrimidine holsters Le Tigre
16:43 * dbolser looks on the mexican standoff iwth a steely deadpan gaze
16:50 * deafferret belches up a little snark
17:03 rbuels wow.
17:53 dbolser rbuels: thanks for the merge
17:53 dbolser and thanks for the pod2usage tip
17:53 dbolser try "bp_seq_feature_load.pl --man" :D
17:53 kblin rbuels: so, happy with the slot count?
17:54 rbuels kblin: it's a lot
17:54 rbuels kblin: not sure if we have enough good apps
17:54 rbuels dbolser: np
17:57 dbolser rbuels: pass all excess slots on to gsod ;-)
17:57 dbolser night
17:57 rbuels gsod?
17:57 rbuels oh, GSoD, i remember now
17:58 kblin left #bioperl
17:58 kblin joined #bioperl
18:09 pyrimidine ok, did no one else catch the test fails on bioperl-live?
18:09 pyrimidine (from those last few commits)
18:11 scottcain pyrimidine: um, there's lots
18:11 scottcain of failures, that is.
18:11 pyrimidine all SeqIO-based
18:12 pyrimidine getting &@!$% tired of commits w/o tests being run first
18:12 rbuels this is not the first time jason has broken msater
18:13 pyrimidine right, but dbolser's commits were supposed to *fix* that
18:13 pyrimidine at least, that's what I assumed from the package label
18:13 rbuels his were all docpatches
18:14 CIA-119 bioperl-live: Chris Fields master * r700949b / Bio/SeqIO.pm : fix typo - http://bit.ly/gDa35G
18:14 pyrimidine still got one fail
18:26 CIA-119 bioperl-live: Chris Fields master * ra063537 / t/SeqIO/SeqIO.t : TODO failing test per bug #3204 - http://bit.ly/ggLrFu
18:29 CIA-119 bioperl-live: Chris Fields master * rc4e20c6 / (12 files in 5 dirs): Merge remote branch 'wkretzsch/master' - http://bit.ly/f9Nmxe
18:48 CIA-119 bioperl-live: Chris Fields master * rb2e55e9 / (9 files in 6 dirs):
18:48 CIA-119 bioperl-live: Merge branch 'topic/bug_3196', fixes #3196 but needs further testing
18:48 CIA-119 bioperl-live: Conflicts:
18:48 CIA-119 bioperl-live:  MANIFEST
18:48 CIA-119 bioperl-live:  t/LocalDB/BioDBGFF.t - http://bit.ly/gpTIQJ
18:49 pyrimidine d'oh, shoulda removed the 'Conflicts' (were minimal)
18:52 pyrimidine rbuels: that last commit should be interesting (removed much of the Bio::Root::Build cruftiness).  Noticed that Build.PL is now giving manifest warnings, though
18:53 pyrimidine c'est la vie
18:53 rbuels eh.
18:53 * rbuels shrugs
19:06 CIA-119 bioperl-live: Chris Fields master * r2e37d52 / t/Tools/Run/RemoteBlast_rpsblast.t : fix skip test count (off by 1) - http://bit.ly/eIT9LQ
19:06 CIA-119 bioperl-live: Chris Fields master * r874a5f8 / Bio/Root/Test.pm : typo - http://bit.ly/ez7NbM
19:13 CIA-119 bioperl-live: Chris Fields master * r85fa2a8 / (MANIFEST MANIFEST.SKIP):
19:13 CIA-119 bioperl-live: Revert "remove MANIFEST/MANIFEST.SKIP (only need for dist release)"
19:13 CIA-119 bioperl-live: This reverts commit e3e94b1525aaa0fae3f5c4ddbbc6d6f9c2516b06. - http://bit.ly/gLnQKx
19:13 CIA-119 bioperl-live: Chris Fields master * rbbf8cb2 / MANIFEST :
19:13 CIA-119 bioperl-live: add back MANIFEST.SKIP; we have a few test files created on the fly that
19:13 CIA-119 bioperl-live: are to be added - http://bit.ly/i9KdzR
19:27 CIA-119 bioperl-run: Jason Stajich master * re9f6a66 / lib/Bio/Tools/Run/Phylo/Phyml.pm : fix bug #3203 - freq not assigned to freq but to kappa in init - http://bit.ly/hgQ2Jm
19:42 CIA-119 bioperl-live: Chris Fields master * r7dfb317 / MANIFEST.SKIP : add a few things to skip that are possibly created during the Build process - http://bit.ly/dFxJeq
19:42 CIA-119 bioperl-live: Chris Fields master * rb9ff027 / (Bio/Root/Build.pm Build.PL): MANIFEST is created dynamically, we leave MANIFEST.SKIP alone (dist-specific) - http://bit.ly/iaWcLV
19:56 perl_splut busy commit day I see
19:58 pyrimidine :)
20:27 CIA-119 bioperl-live: Chris Fields master * r1f5f578 / Build.PL : catch missing EntrezGene and warn re: circ dependency - http://bit.ly/e0o93L
20:32 CIA-119 bioperl-live: Chris Fields master * ra48b6c2 / t/Tree/TreeIO/newick.t : parameter missing actual argument - http://bit.ly/hgd6oX
20:48 CIA-119 bioperl-live: Chris Fields master * r6a71090 / Build.PL : add some spacing for warning - http://bit.ly/ePgzca
20:56 scottcain left #bioperl
20:58 CIA-119 bioperl-live: Chris Fields master * r09a3f36 / t/Tree/TreeIO/newick.t : remove redundant tests per Dave Messina - http://bit.ly/h4FPZ7
20:58 CIA-119 bioperl-live: Chris Fields master * r94e155b / Build.PL : formatting again for warning - http://bit.ly/hqsSbT
20:59 pyrimidine rbuels: think the MANIFEST should be cleaned up with a './Build realclean'?
21:00 pyrimidine not a big deal really
21:01 rbuels erm, i dunno
21:01 pyrimidine easy enough to remove with a 'git clean -f'
21:01 pyrimidine I suppose
21:01 rbuels wait, do you have './Build' generate the manifest?
21:01 rbuels the only time that's needed is for the release tarball
21:01 pyrimidine not at the moment
21:02 rbuels that's why there's a separate Build manifest
21:02 rbuels so most of the time, it doesn't even need to be cleaned up
21:02 pyrimidine right, it's only for releases (./Build dist)
21:02 rbuels i never use build for cleaning, i always use git
21:02 rbuels (and i tend to not put many things in .gitignore)
21:02 pyrimidine same here
21:05 rbuels so i guess my answer to your question is "eh."
21:06 rbuels (meh?)
21:06 pyrimidine feh
21:07 pyrimidine heh, latest Module::Build adds itself to configure_requires
21:07 pyrimidine Module::Build was not found in configure_requires! Adding it now automatically as: configure_requires => { 'Module::Build' => 0.38 }
21:08 pyrimidine running a few test dists, will release 1.6.9 tomorrow
21:08 rbuels pyrimidine++
21:09 scottcain joined #bioperl
21:09 pyrimidine actually, it's technically 1.6.900, but I don't want to scare the bioperl n00bs
21:10 pyrimidine and I would like to go to simple 2.x versioning after GSoC
21:26 scottcain left #bioperl
21:35 CIA-119 bioperl-live: Chris Fields master * rd569ad5 / t/Assembly/core.t : partly squash Assembly warnings - http://bit.ly/fw1hgO

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