Perl 6 - the future is here, just unevenly distributed

IRC log for #pdl, 2015-06-17

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

All times shown according to UTC.

Time Nick Message
00:10 mohawk sivoais, we probably should mirror SF390 on github issues
20:40 opkick [pdl] zmughal opened issue #124: SF#390 scalar PDL with badvalue always compares BAD with perl scalars http://git.io/vL4Kt
20:42 opkick [pdl] zmughal opened issue #125: SF#391 Operators called in the form "explicit call with trailing 0" give syntax error http://git.io/vL4Kp
20:43 opkick [pdl] zmughal opened issue #126: SF#392 Inline Pdlpp doesn't work in clean build tree http://git.io/vL46Z
20:44 mohawk sivoais++
20:46 sivoais mohawk: the list to add to <https://github.com/PDLPorters/devops/blob/master/NOTES.md>
20:46 mohawk great
20:47 opkick [devops] zmughal pushed 1 new commit to master: http://git.io/vL4P5
20:47 opkick devops/master 3100b1f Zakariyya Mughal: Add a note about BAD value configurations for Travis-CI
20:49 sivoais so far there are 10 tasks per split dist
20:50 sivoais I'd like to look into the travis issues. Perhaps I can figure a way to test PDL-Core directly from git. That way it is always up to date
20:51 sivoais too bad Travis-CI is not flexible enough to handle changes across repositories
20:51 sivoais i.e., any changes to PDL-Core should kick off jobs for PDL-IO-GD, etc.
20:53 sivoais but I think I know of a way to work around that. Just for the convenience of not having to set up another CI server.
20:53 sivoais YAGNI (for now)
20:55 mohawk sivoais, you're probably envisioning some kind of 2-way dep
20:56 mohawk i instead view pdl-core as the c compilter
20:56 mohawk compiler
20:56 mohawk we get it right, with lots of tests
20:56 mohawk then it just gets used in other stuff
20:57 sivoais mm, right. Assume it is correct. Good tests for PDL-Core will help ensure that.
20:57 sivoais but at the same time, I was thinking like how Perl blead is tested against CPAN
20:58 mohawk don't run before you've split the iceberg ;-)
20:58 mohawk we'll burn that bridge when we come to it
21:58 mohawk if pdl 2.012 is sufficiently stable, we should be making sure we have everything in place
21:59 mohawk for the Great Move to pdl-core / pdl
22:10 mohawk in fact, i even want to get pdl::core 2.013 out before we even start merging stuff again
22:10 mohawk even t-c-3
22:11 mohawk because let's not mess with success
22:11 mohawk we'll get pdl::core stable on its own first, which may not happen correctly first time
22:11 mohawk THEN we'll work at making it even better
22:11 mohawk sivoais, thoughts?
22:20 opkick [pdl-fftw3] mohawk2 pushed 1 new commit to master: http://git.io/vLBmH
22:20 opkick pdl-fftw3/master b706b91 Ed J: Dep on PDL::Core not PDL, v0.04
22:20 opkick [pdl-fftw3] mohawk2 tagged v0.04 at master: http://git.io/vLBm7
22:43 sivoais mohawk: sure, makes sense. The only reason I wanted to get t-c-3 in was so that the tests can be split out with PDL-Core
22:43 sivoais but we can move them later. It might be a little tricky, but it is doable
22:44 sivoais mohawk: could you write out what you are proposing in response to <http://thread.gmane.org/gmane.comp.lang.perl.pdl.devel/6534>?
22:45 sivoais in the "PDL's Big Split" thread
22:56 mohawk sivoais, it's not tricky - i did it already on my core-only branch
22:56 mohawk nothing fundamental has changed since my previous email on the subject
22:56 mohawk really all we're waiting for is agreement from chm that 2.012 is all systems go
22:57 mohawk then we pull pdl-core over and dev-release it
23:07 sivoais mohawk: how about you pull it over now to a branch of the pdl-core repo?
23:08 sivoais in preparation for that
23:10 sivoais will test-cleanup-3's patches apply cleanly to the pdl-core repo after the split?
23:14 travis-ci PDLPorters/pdl-fftw3#5 (master - b706b91 : Ed J): The build passed.
23:14 travis-ci Change view : https://github.com/PDLPorters/pdl-fftw3/compare/ae4b97b2fa43...b706b9174c0f
23:14 travis-ci Build details : http://travis-ci.org/PDLPorters/pdl-fftw3/builds/67282103
23:18 mohawk well it wouldn't be a branch of the p-c repo
23:18 mohawk it would be master
23:19 mohawk t-c-3 ought to apply cleanly since it is simply applied to the t/* and apart from some t/*.t being segregated into kitchen-sink, nothing else has changed
23:19 mohawk admittedly, i haven't rebase my core-only branch lately
23:19 mohawk i'm kind of actively waiting
23:19 mohawk if i don't hear by tomorrow however, i will in fact copy it over
23:19 mohawk however not now
23:21 mohawk i'm keen to avoid too much maintaining of two repos in parallel
23:38 sivoais true

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