Perl 6 - the future is here, just unevenly distributed

IRC log for #inline, 2014-12-05

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

All times shown according to UTC.

Time Nick Message
01:05 ingy davido__: back
01:36 ingy sigh
01:38 ingy I think autostubbing needs to put the ::Inline in lib
01:38 ingy or more specifically, next to the caller
01:39 ingy which should always be lib
01:40 ingy or user declares somehow
01:42 ingy -MInline::Module=autostub,blib,::Inline,--quiet
01:45 ingy -MInline::Module=autostub,-lib=blib/lib,-stub=::Inline,-quiet
01:46 davido__ joined #inline
03:55 ingy davido__: you about?
05:35 davido_ joined #inline
05:39 davido_ ingy. is the pairup going?
05:43 ingy yeah
05:46 ingy davido_: can you not get in?
05:46 davido_ yeah, wasn't working, i'll try again.
06:10 daoswald^ingy joined #inline
06:10 davido_ how did you resolve it?
06:11 ingy resolve what?
06:11 davido_ the pairup
06:11 ingy restarted it
06:12 davido_ oh, ok.
06:12 ingy I've done a few things
06:12 davido_ yeah, what's transpired?
06:12 ingy there is now -MInline::Module=autostub and -MInline::Module::makestub
06:13 ingy and they default to using lib
06:13 davido_ ok.
06:13 ingy so that prove -lv will work
06:14 davido_ does import handle it automatically?
06:14 ingy I've been going back and forth on what to do
06:14 davido_ -MInline::Module=autostub just does the right thing.
06:14 davido_ got it.
06:16 ingy I think we should allow people to generate in blib
06:16 ingy and then use prove -blv
06:16 davido_ ok, what about _Inline?
06:16 davido_ Yeah, that's the thing; _Inline doesn't take advantage of the existing toolchain as well.
06:17 davido_ it may be an abuse of blib, but it works.
06:17 ingy nod
06:17 ingy and make clean deletes blib
06:17 davido_ yeah
06:18 davido_ is the only downside that it's an unconventional use of blib?
06:19 ingy I was able to have a few wins along the way
06:20 davido_ brb,
06:20 davido_ :( it'll be a few...
06:21 ingy ok
06:24 davido_ :( kid woke up, aileen out.  be back in a bit.
06:52 GitHub4 [inline-module-pm] ingydotnet pushed 1 new commit to autostub2: http://git.io/oCltpA
06:52 GitHub4 inline-module-pm/autostub2 e39b5f7 Ingy döt Net: Got makestub and autostub working with tests
07:09 GitHub60 [inline-module-pm] ingydotnet pushed 1 new commit to autostub2: http://git.io/K4ke3Q
07:09 GitHub60 inline-module-pm/autostub2 886be03 Ingy döt Net: Update the tutorial
07:09 ingy davido_: I'm quitting
07:09 ingy I'll be on a train at 7:45 PST tomorrow
07:10 ingy I can hack from the train
12:24 ilbot2 joined #inline
12:24 Topic for #inline is now LOG: http://irclog.perlgeek.de/inline | TPF Grant Work in Progress | http://inline.ouistreet.com | Discussion of Inline.pm and ILSMs | https://github.com/ingydotnet/inline-pm
15:21 ingy davido_: moin moin
15:21 daoswald^ingy .
15:28 davido_ HEY!
15:29 davido_ I'm here ingy
15:41 davido_ hi
15:42 davido_ :)
15:43 ingy hey
15:43 ingy was just showing a friend the pairup
15:44 davido_ ok
15:55 davido_ do you need my terminal larger?
15:59 ingy let's try to put out a eumm release!
15:59 ingy with autostub!
15:59 davido_ Lets!
16:00 ingy :)
16:00 davido_ Once we've done that we can look at the POD for Inline::Module to see what needs to be tweaked.
16:00 ingy I did some last night
16:00 ingy but yeah
16:07 davido_ weird. froze. was about to reboot, then came alive again.
16:08 ingy that seemed to fast
16:08 ingy too
16:08 davido_ yeah, couldn't have compiled.
16:09 davido_ lol
16:10 davido_ still
16:10 davido_ that's better but failed.
16:13 ingy we might want to build under blib/_inline/
16:13 davido_ That would probably be fine.
16:15 ingy never seem this
16:15 davido_ what is the "remove the makefile dep" for?
16:16 ingy I get it
16:20 ingy the PERL5OPT/autostub thing is getting in the way
16:20 davido_ that sucks.
16:20 ingy of other perl usages
16:20 davido_ does that take the "export" method off the table?
16:21 ingy probably not
16:21 ingy just need to be smarter
16:21 davido_ i was about to type that.
16:25 ingy terrible internet
16:45 ingy davido_: here's one potential gotcha
16:46 davido_ i'm out of the pairup :(
16:46 ingy if you git clone ingydotnet/acme-math-xs-pm -b eumm
16:46 ingy so am i
16:47 ingy and you cpanm Alt::Acme::Math::XS::EUMM
16:47 ingy and then prove -lv t
16:47 ingy it passes without compilation
16:47 davido_ oh.
16:48 ingy because it picks up the installed ::Inline
16:48 davido_ yeah.  -l should prefer local first, but then since local isn't there, it moves outward
16:48 ingy which is maybe correct, but maybe confusing
16:50 davido_ if we could come up with a line of code authors can put in their tests to ensure they're working with the correct version, that might be worth documenting.
16:51 ingy I'll file an issue
16:51 davido_ it's bound to come up; most authors have installed older versions of whatever work they're developing
16:53 davido_ Inline::Module could detect if $INC{lib/My/Module/Inline.pm} is the right one, maybe.
17:22 GitHub171 [inline-module-pm] ingydotnet opened issue #11: Dev testing can find installed version of stub http://git.io/-3xjOQ
17:26 ingy davido_: I fixed the PERL5OPT thing
17:26 ingy at least for now
17:30 GitHub9 [inline-module-pm] ingydotnet pushed 2 new commits to autostub2: http://git.io/T5JZ9w
17:30 GitHub9 inline-module-pm/autostub2 d06c505 Ingy döt Net: Better error message
17:30 GitHub9 inline-module-pm/autostub2 abe1d18 Ingy döt Net: We don't want PERL5OPT in the environment...
17:32 ingy davido_: also yesterday I got rid of a line from the proxy stub. the 'use base "Inline";'. also I got rid of the eval for the Inline Config.
17:34 ingy I'm going to try to build under blib/_Inline now
17:34 ingy I think this will help the fixblib too
17:35 ingy and issue #9
17:35 ingy also if we build under blib/_Inline I think everything is clear
17:35 ingy not an abuse of blib
18:01 GitHub32 [inline-module-pm] ingydotnet pushed 1 new commit to autostub2: http://git.io/j2t2-w
18:01 GitHub32 inline-module-pm/autostub2 6fda745 Ingy döt Net: Build in blib/Inline/
18:01 ingy davido_: blib/Inline/ works
18:01 ingy now to fix the fixblib
19:00 ingy davido_: it's working pretty great now
19:01 ingy The fix blib doesn't do much now. It only links the .so file to the place where a normal blib wants it
19:02 ingy but it leaves the blib/Inline dir fully intact
19:03 ingy make test and make install ignore the blib/Inline
19:11 GitHub6 [inline-module-pm] ingydotnet pushed 1 new commit to autostub2: http://git.io/DpusiQ
19:11 GitHub6 inline-module-pm/autostub2 4a66c20 Ingy döt Net: Refactor fixblib...
19:27 ingy leont: do you know if `make` and `make test` (under EUMM) kills PERL5LIB?
19:28 ingy I've tried `make -e` as well
19:39 ether PERL5LIB is respected
19:39 ether it has to be, or perlbrew would be unusable.
19:42 GitHub20 [inline-module-pm] ingydotnet pushed 1 new commit to autostub2: http://git.io/WowINw
19:42 GitHub20 inline-module-pm/autostub2 0eb056c Ingy döt Net: Add a DEBUG feature to get more info...

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