Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2015-03-13

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

All times shown according to UTC.

Time Nick Message
00:39 kcranstn joined #opentreeoflife
00:45 jar286 joined #opentreeoflife
01:05 jar286 joined #opentreeoflife
01:31 jar286 joined #opentreeoflife
12:34 kcranstn joined #opentreeoflife
13:10 kcranstn joined #opentreeoflife
13:18 jar286 joined #opentreeoflife
13:46 kcranstn joined #opentreeoflife
14:11 jar286 joined #opentreeoflife
15:45 josephwb kcranstn have you been able to install otcetera?
15:59 mtholder joined #opentreeoflife
15:59 mtholder hi, josephwb
16:02 josephwb hey
16:02 josephwb i cannot get otcetera installed
16:02 josephwb Makefile:464: recipe for target 'ftree.lo' failed
16:02 josephwb not sure why
16:03 josephwb [pulling, trying again]
16:04 josephwb on debian, btw. install instructions seem mac-centric
16:04 josephwb nope, failed again on same spot
16:04 josephwb mtholder ^ (in case you are away)
16:05 mtholder travis is building it
16:06 mtholder any messages that I can work with?
16:06 josephwb hmm. what can i send you?
16:06 mtholder logs
16:06 mtholder I've got to run. but I'll check when I get home.
16:07 josephwb configure log?
16:07 mtholder the text of the terminal would be a good place to start.
16:07 josephwb or what make prints to screen?
16:07 josephwb ok, got that
16:07 josephwb sending...
16:08 josephwb i'll keep busy until you can look at this
16:09 jar286 joined #opentreeoflife
16:10 josephwb this is probably it:
16:10 josephwb cc1plus: all warnings being treated as errors
16:10 josephwb right?
16:11 josephwb e.g. unused variables might be messing me up?
16:17 josephwb commenting out the unused variable allowed me to proceed forward; now stuck on another error.
16:28 josephwb has anyone got otcetera to install?
16:28 jar286 sorry, haven’t looked at it yet
16:30 josephwb ok, thanks. about to give up here.
16:31 jar286 since Mark H is coding in C++, I will start using Lisp.
16:31 josephwb whatever works
16:34 jar286 I can’t tell what it is really
16:34 jar286 there was no announcement, and the readme doesn’t say
16:35 josephwb i just want to vet trees
16:35 josephwb re: unsupported edges
16:35 kcranstn joined #opentreeoflife
16:35 josephwb https://github.com/OpenTreeOfLife/otcetera#checking-for-unnamed-nodes-in-a-full-tree-that-have-no-tree-supporting-them
16:36 josephwb kcranstn: have you got otcetera to install?
16:36 kcranstn no, I haven’t
16:36 kcranstn I haven’t tried
16:36 josephwb ok, i give up then
16:37 jar286 I bet mh will be on irc later today
16:41 jar286 I still don’t get what ‘support for an edge’ means
16:41 jar286 one supports a claim.  what does an edge claim?
16:42 josephwb error from before: edges in synth that are not in any input trees ot taxonomy
16:42 josephwb ot -> or
16:43 jar286 how do you identify an edge in synth with an edge in an input tree? to me those are different edges because they’re in different data structures
16:43 josephwb they are both trees
16:44 josephwb edge between same (or compatible) parent and same (or compatible) child
16:45 jar286 hmm… then ‘sameness’ isn’t unique since there can be multiple compatible nodes
16:46 jar286 and why aren’t we talking about paths instead of edges?
16:46 josephwb i guess we are
16:46 josephwb i mean, yes: we are
16:47 josephwb [not guessing]
16:48 jar286 that’s not what this says https://github.com/OpenTreeOfLife/treemachine/blob/nonsense-1/nonsense/iteb_support_theorem.md#input-tree-edge-bijection-support
16:51 josephwb that is a bit different: that compares trees (edges), but we are working with the graph (paths)
16:51 josephwb a graph path can create a synth edge
16:52 josephwb regardless, i just want to run the damn test!
16:52 jar286 I find it difficult to help out because the language is confusing to me
16:52 josephwb i think it is a bit moot, as (at least the synth method I am working on) cannot possibly deliver an "unsupported" edge
16:53 josephwb i agree
16:53 josephwb re: confusing
16:54 mtholder joined #opentreeoflife
16:54 mtholder hi
16:55 josephwb hey mtholder
16:55 josephwb i don't know where it is getting the -Werror flag from
16:55 josephwb oh, wait. i see
16:55 mtholder how did you configure?
16:55 josephwb reconf-gcc.sh
16:56 josephwb [had to change location of g++, gcc)
16:56 mtholder yeah just remove it from there.
16:56 mtholder -Werror that is
16:56 mtholder i've been compiling w/ clang
16:56 josephwb trying...
16:56 josephwb for shame...
16:57 josephwb clang seems evil
16:57 mtholder why?
16:57 josephwb no openmp support
16:57 josephwb fails on so many things that gcc works on
16:58 josephwb (only on mac)
16:58 josephwb never tried it on linux
16:58 josephwb In file included from ../../otc/otcli.cpp:1:0:
16:58 josephwb ../../otc/otcli.h:48:25: error: 'getTitle' function uses 'auto' type specifier without trailing return type
16:58 josephwb auto getTitle() const {
16:59 josephwb ^
16:59 josephwb ../../otc/otcli.h:48:25: note: deduced return type only available with -std=c++1y or -std=gnu++1y
16:59 josephwb Makefile:464: recipe for target 'otcli.lo' failed
16:59 mtholder you'll definitely need C++14 support
16:59 josephwb ok, conf file has:
16:59 josephwb -std=c++11
17:00 mtholder yeah the g++ version is stale, as I've been using clang++
17:00 mtholder i did compile on both for a while.
17:00 josephwb trying again
17:00 mtholder so it may not be hopeless on g=+
17:00 mtholder g++ that is
17:01 josephwb clang doesn't handle -Werror in the same way?
17:01 josephwb oh, clang conf only has:
17:01 josephwb -Wno-c++98-compat -Weverything -Wpadded -pedantic
17:01 josephwb why are you strict with g++?
17:02 mtholder I started strict with both.
17:02 mtholder the g++ conf file is stale.
17:02 mtholder edit at will
17:02 josephwb -Waddress -Warray-bounds -Wc++11-compat -Wchar-subscripts -Wcomment -Wformat -Wmain -Wmaybe-uninitialized -Wmissing-braces -Wnonnull -Wopenmp-simd -Wparentheses -Wreorder -Wreturn-type -Wsequence-point -Wsign-compare -Wstrict-aliasing -Wstrict-overflow -Wswitch -Wtrigraphs -Wuninitialized -Wno-unknown-pragmas -Wunused-function -Wunused-label -Wunused-value -Wunused-variable -Wvolatile-register-var -Werror -Wno-pragmas
17:02 josephwb ok, will do
17:02 josephwb should i submit an issue?
17:03 josephwb wee! it worked
17:03 mtholder just push it to master if that is the only change you made.
17:03 mtholder thanks
17:03 josephwb will do
17:05 josephwb wee! a push to master without a PR!
17:05 josephwb commit message is just for you, mtholder
17:05 josephwb oh wait, no permission
17:06 josephwb premature glee
17:06 mtholder should be fixed
17:06 josephwb works now
17:06 josephwb thankx
17:29 josephwb hey mtholder
17:29 mtholder yes
17:29 josephwb otcetera takes taxonomy.tre as an arg for a bunch of things
17:29 mtholder yes
17:29 josephwb does it make taxonomy.tre
17:29 mtholder no
17:29 josephwb where does it come from?
17:29 josephwb treemachine?
17:30 mtholder i've been using the output of treemachine that I got from you or Ruchi
17:30 josephwb ok
17:31 mtholder some docs at https://github.com/OpenTreeOfLife/otcetera/tree/master/supertree, but things have been in flux
17:31 josephwb maybe distribute with code?
17:32 mtholder I don't like putting data with code (except test data)
17:32 jar286 +1
17:32 mtholder agree that we should have it posted somewhere
17:32 josephwb right
17:33 josephwb maybe on the ott page?
17:33 mtholder otcetera will eventually generate the pruned version of OTT, but doesn't yet.  Yes the pages of OTT seem like a good spot
17:40 josephwb newickNames has changed recently; is that a problem, or does it look at the ottID?
17:42 jar286 n.b. ott/ott2.8/index.html is generated from a .md and then copied manually
17:43 josephwb ok
17:50 josephwb mtholder: when doing: https://github.com/OpenTreeOfLife/otcetera#checking-for-unnamed-nodes-in-a-full-tree-that-have-no-tree-supporting-them
17:51 josephwb only 1 tree per file?
17:51 josephwb i have them all in 1 file
17:52 mtholder not sure.
17:53 mtholder might work.
18:44 mtholder joined #opentreeoflife
18:45 mtholder josephwb, did you want responses on the "synthesis vetting" thread, or is it moot now that you can compile otc?
18:49 josephwb responses welcome
18:49 josephwb i am going to put some scripts in otcetera to generate taxonomy.tre files, ok?
18:50 josephwb mtholder ^
18:50 mtholder w/ any dependencies?
18:50 josephwb nope
18:50 mtholder sure. go for it.
18:51 josephwb yes, please everyone respond to the email; trying to get everyone on the same page/involved.
19:22 josephwb hey mtholder
19:22 mtholder yes
19:22 josephwb getting error:
19:22 josephwb Unexpected continuation of a label after a quoted string in newick.
19:22 mtholder error or warning?
19:22 josephwb warn
19:22 josephwb is it newick compliant?
19:22 mtholder yes
19:23 josephwb hmm.
19:23 mtholder i think
19:23 josephwb so are my trees
19:23 josephwb i think
19:23 josephwb ;)
19:24 mtholder it just needs the suffix of a label to be numeric. and it treats that as the ott ID for matching across files. In the newicks, that I got, there was munging and odd quoting that made the labels not match exactly.
19:24 mtholder so the numeric suffix is what it goes on.
19:25 josephwb ok
19:25 josephwb my input trees are *only* ottIDs (with no preceeding "_ott"
19:25 josephwb otcetera does not like that
19:26 mtholder that should not be too hard to fix
19:30 josephwb does it not like internal node labels?
19:31 josephwb mtholder ^
19:31 josephwb hold on
19:31 mtholder no it is fine with internal label. If you look through the data dir you'll see some examples that are handled.
19:32 josephwb i am looking at the old synth tree, which did not have proper newick labels
19:35 josephwb arg. i had these trees in R and exported them from there. seems they messed things up.
19:37 josephwb but i am (pretty) sure that my taxonomy tree is kosher...
19:37 josephwb for a quoted string, does it think "_" is wrong?
19:42 mtholder no. it shouldn't care.
19:42 mtholder if it is _ in quotes then it is a _ and not a space, but since it is just looking at the numeric suffix that shouldn't matter.
19:48 josephwb thought so
19:49 josephwb does it fully support the internal ' -> '' ?
19:50 josephwb yeah, got it to finish
19:50 josephwb yeah -> yay
21:00 jar286 joined #opentreeoflife
21:13 jar286 joined #opentreeoflife
21:56 kcranstn joined #opentreeoflife
22:15 jar286 joined #opentreeoflife
22:28 jar286 joined #opentreeoflife
23:33 jar286 joined #opentreeoflife

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