Perl 6 - the future is here, just unevenly distributed

IRC log for #pdl, 2015-04-03

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

All times shown according to UTC.

Time Nick Message
14:59 mohawk sivoais++
17:31 mohawk sivoais, looks to me like projfixp2 is ready to be merged - see sisyphus's email to pdl-devel
17:31 mohawk what's your view?
18:59 opkick [pdl] wchristian merged projfixp2 into master: http://git.io/vemNT
18:59 opkick [pdl] wchristian closed pull request #91: Alien::Proj4 - fix include ordering error (master...projfixp2) http://git.io/jg1O
19:00 opkick [pdl] mohawk2 deleted projfixp2 at d9a474e: http://git.io/vemNK
19:30 sivoais mohawk++: \o/
19:34 mohawk ha ha, thanks
19:34 mohawk hi sivoais
19:34 mohawk sivoais, what's your view of the merge?
19:36 sivoais mohawk: what changed since I last looked at it
19:36 sivoais I remember testing the branch on my box
19:37 mohawk sivoais, this is the totality of the changes: ohawk: what changed since I last looked at it
19:37 mohawk [20:36] <@sivoais> I
19:37 mohawk dohhh
19:37 mohawk https://github.com/PDLPorters/pdl/compare/931dcc0926e1...d9a474e209e0
19:37 mohawk don't know why it did that
19:38 sivoais well, it looks like the same code I saw on the PR before, so looks good to me
19:38 mohawk great
19:50 travis-ci PDLPorters/pdl#165 (master - d9a474e : Ed J): The build passed.
19:50 travis-ci Change view : https://github.com/PDLPorters/pdl/compare/931dcc0926e1...d9a474e209e0
19:50 travis-ci Build details : http://travis-ci.org/PDLPorters/pdl/builds/57078433
20:25 mohawk w00t
20:25 mohawk not entirely a surprise however
20:26 mohawk sivoais, what do you think of, in due course, also zapping PDL's inc/Carp*?
20:27 sivoais sure, that makes sense
20:28 sivoais one thing I want to see later is Docker images and installers for PDL (SciPDL on more than just OSX)
20:28 sivoais adding issues for those things
20:29 opkick [pdl] zmughal opened issue #93: Docker images http://git.io/veYr7
20:30 opkick [pdl] zmughal opened issue #94: remove inc/Carp.pm http://git.io/veYrb
20:30 mohawk however, not before 2.008 ;-)
20:30 sivoais nope, definitely not!
20:30 opkick [pdl] zmughal opened issue #95: SciPDL installer http://git.io/veYol
20:31 mohawk i have the feeling that this next RC will be the actual release
20:34 sivoais we're so close!
20:39 mohawk sivoais, you're a knowledgeable chap - i tried remake, step-debugging wasn't so useful. know of any makefile-dep-tree visualiser tools?
20:40 sivoais makepp
20:40 sivoais pure-perl make :-)
20:40 sivoais outputs to .dot which can be viewed in GraphViz, Gephi, etc
20:40 sivoais <http://makepp.sourceforge.net/>
20:42 mohawk i'm reading https://metacpan.org/pod/makepp - what's the dump-tree arg?
20:46 mohawk wow, the build system looks a bit... special
20:48 mohawk http://makepp.sourceforge.net/gallery/ indicates separate command, "makeppgraph"
21:15 sivoais mohawk: ... well... it is a make(1) replacement... it has to install itself somehow ;-)
21:16 mohawk sivoais, it should use EUMM not its own weird-ish thing ;-)
21:45 mohawk makeppgraph -gr 'm:(^/(bin|(usr/(bin|lib|include)))|((/\.exists|\.xsc|makemakerdflt)$)): && undef $_ || cwd(2)' ; twopi -Tsvg -Goverlap=scale log.dot >/x/what-if.svg
21:45 mohawk (run in Basic/Core after doing top-level "perl Makefile.PL")
21:51 mohawk http://imgur.com/LRrhEyD is the result
21:57 sivoais mohawk: useful?
22:00 mohawk a little
22:00 mohawk it's reminding me of the dep i added from Core.o to things like $(INST_ARCHLIB)$(DFSEP).exists pm_to_blib pdldataswitch.c pdlbadvalinit.c pdl.h pdlcore.h
22:01 mohawk and i think more accurate is that Core.c deps on those
22:01 mohawk but that may be too fine a hair to split
22:01 mohawk basically i'm in that dir doing make clean, then make, and it's not completing
22:01 mohawk if i run make again, then it sometimes does, or sometimes takes a 3rd go
22:02 mohawk or rather, make -j4 has that
22:02 mohawk -j1 consistently works
22:03 mohawk that implies an unexpressed dep
22:07 mohawk sivoais, and if there's a way of making remake express its debugging less uselessly noisily than simple GNU make, i haven't found it
22:16 mohawk if i expressly say "make dynamic" (which my MAKEFLAGS makes have -j4) it works everytime
22:16 mohawk but if i go "make linkext" it doesn't
22:17 mohawk so that seems to be the difference, which looks like a GNU make bug

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