Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2014-01-21

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

All times shown according to UTC.

Time Nick Message
04:53 jeffreykegler joined #marpa
04:54 jeffreykegler I have uploaded Marpa-R2 2.079_011 to CPAN, and CPANtesters looks good.
04:54 jeffreykegler As usual, your own testing is appreciated.
04:55 jeffreykegler By the way, I am going to try moving some of the "developer's chit-chat" from the Google Group to this channel.
04:57 jeffreykegler I think that may be better because that way the Google group becomes more focused on stuff of interest outside of a core group.
04:57 jeffreykegler AFK
04:58 jeffreykegler left #marpa
05:23 ronsavage joined #marpa
09:46 LLamaRider joined #marpa
17:49 jeffreykegler joined #marpa
18:40 rns_ joined #marpa
20:06 jdurand joined #marpa
20:47 jeffreykegler left #marpa
21:47 LLamaRider joined #marpa
21:48 LLamaRider hi #marpa folks! Anyone online who can help me with the Marpa installation from source (GitHub)
21:48 LLamaRider and I missed my question mark - ?
21:55 rns_ joined #marpa
21:55 rns_ hi
21:56 rns_ anybody here
21:56 rns_ ?
21:58 LLamaRider ah there you are :)
21:58 LLamaRider so, made some progress, now trying to get "make install" to work for libmarpa
21:58 LLamaRider which seems to be a prerequisite
22:00 rns_ Yep, perhaps you can get around it by make releng in the root dir.
22:00 LLamaRider made some progress, now getting "config.status: error: cannot find input file: `Makefile.in'
22:00 LLamaRider hm, ok, let me try your make target
22:01 LLamaRider that failed, i need a working build for libmarpa first
22:01 LLamaRider it seems to be almost there
22:01 LLamaRider already got the dependencies installed
22:03 rns_ if all deps are here, make releng recurses the tree and builds the docs, libmarpa, and the dist tarball.
22:04 rns_ on cygwin I also had to install autoconf-archive and run autoreconf --install in some dirs where make failed.
22:07 LLamaRider rns_: You're a life saver! The cygwin advice really did the trick.
22:07 LLamaRider I am installing on an Ubuntu VM running inside a Windows 8 , so who knows what's going wrong
22:07 LLamaRider plus my C skills are very rusty
22:08 LLamaRider but now I need 'texi2pdf'. The installation procedure can definitely benefit from some simplification.
22:11 LLamaRider ok, got almost to the end. META.json and META.yml are reported missing and i get an invalid license error
22:12 rns_ This is where it's best to do a fresh clone and make releng in the root dir.
22:12 LLamaRider :) smart
22:14 rns_ http://goo.gl/nXfRfn — marpa-parser thread about building from source on cygwin.
22:16 LLamaRider thanks. Sadly the errors didn't magically solve themselves when I did a fresh clone
22:16 LLamaRider but it's really the end of the make and everything else succeeds so i am hopeful
22:19 rns_ Yes, I had to do 3 to 5 frech clones IIRC.
22:20 LLamaRider yikes, that many
22:22 rns_ I didn't have all deps and when had to start afresh after make failureed .
22:22 rns_ after make failures, that is. :)
22:24 LLamaRider right
22:24 LLamaRider my current problem is to figure out how to satisfy Module::Build's need for a metafile(), since there is none
22:24 LLamaRider there are no META.* files in the repo
22:26 rns_ never seen anything like that; you may also need to have relatively (latest even better :) ) Module::Build and Config::AutoConf
22:27 LLamaRider you're right again, I just went and did cpanm on Module::Build and progress happened!
22:28 LLamaRider and testing... ^_^
22:28 rns_ Congratulations are in order?
22:29 LLamaRider indeed! test suite passed.
22:29 LLamaRider but I got an error saying that your t/sl_panda1.t isn't in the Manifest
22:30 LLamaRider that however I know how to fix :D
22:31 rns_ Sure. And you seem to have the right branch, too.
22:32 LLamaRider ok. all set
22:33 LLamaRider I can sleep in peace now :D
22:33 LLamaRider will have something fun to play with tomorrow
22:33 LLamaRider thanks again for all the help! you really sped me along
22:33 rns_ BTW, using closures if not the fastest way of doing semantics as they are eval'ed.
22:33 LLamaRider eval doesn't have such a horrible overhead in the newer perl versions i think
22:33 LLamaRider also, why would they be evaled?
22:34 rns_ I'm glad you succeeded.
22:34 LLamaRider you just &$closure(arguments), or am I missing something?
22:35 rns_ Hmm, really, let me see.
22:39 rns_ They are eval'ed to catch possible errors, like this:
22:39 rns_ $eval_ok = eval { $result = $closure->( $semantics_arg0, $values ); 1; }; in Value.pm
22:41 rns_ Ok, quality time that was. And it's just the time to hibernate here, too. So, good night and good luck. :) AFK
22:41 LLamaRider I guess that's just for error-handling. I do some of that in a bunch of my Perl projects as well
22:41 rns_ Sure.
22:41 LLamaRider good night!
22:41 LLamaRider and thanks again :) it was very productive.
22:41 rns_ Great to heat that.

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